Design and synthesis of eugenol derivatives, as potent 15-lipoxygenase inhibitors

Article abstract of DOI:10.1016/j.bmc.2007.10.016

A group of 4-allyl-2-methoxyphenol (eugenol) esters were designed, synthesized, and evaluated as potential inhibitors of soybean 15-lipoxygenase (SLO). Compounds 4c, 4d 4f, 4p, and 4q showed the best IC₅₀ in SLO inhibition (IC₅₀ = 1.7, 2.3, 2.1, 2.2, and 0.017 μM, respectively). All compounds were docked into SLO active site and showed that allyl group of compounds is oriented toward the iron atom in the active site of SLO. It is assumed that lipophilic interaction of ligand-enzyme would be in charge of inhibiting the enzyme activity. The selectivity of eugenol derivatives in inhibiting 15-HLOb was also compared with 15-HLOa by molecular modeling and multiple alignment techniques.

Full text of DOI:10.1016/j.bmc.2007.10.016

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 890–901
Design and synthesis of eugenol derivatives, as potent
15-lipoxygenase inhibitors
Hamid Sadeghian,a Seyed Mohammad Seyedi,a,* Mohammad Reza Saberi,b,
Zahra Arghiania and Mehdi Riazia
aDepartment of Chemistry, Faculty of Sciences, Ferdowsi University of Mashhad, Mashhad 91775-1436, Iran
bSchool of Pharmacy, Pharmaceutical Research Center, Mashhad University of Medical Sciences, BuAli Square,
Mashhad 9196773117, Iran
Received 15 August 2007; revised 22 September 2007; accepted 9 October 2007
Available online 12 October 2007
Abstract—A group of 4-allyl-2-methoxyphenol (eugenol) esters were designed, synthesized, and evaluated as potential inhibitors of
soybean 15-lipoxygenase (SLO). Compounds 4c, 4d 4f, 4p, and 4q showed the best IC50 in SLO inhibition (IC50 = 1.7, 2.3, 2.1, 2.2,
and 0.017 lM, respectively). All compounds were docked into SLO active site and showed that allyl group of compounds is oriented
toward the iron atom in the active site of SLO. It is assumed that lipophilic interaction of ligand-enzyme would be in charge of
inhibiting the enzyme activity. The selectivity of eugenol derivatives in inhibiting 15-HLOb was also compared with 15-HLOa by
molecular modeling and multiple alignment techniques.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
disease (COPD).8 Evidence for the inhibition of 15-LO
in the treatment of vascular disease is, however, most
compelling.9 Both transgenic and knockout studies
implicate a role for 15-LO in atherogenesis.10,11 The en-
zyme is abundantly expressed in macrophages residing
within the atherosclerotic lesion.5 In addition, the imme-
diate products of 15-LO oxidation of AA and linoleic
acid (LA) have been shown to be pro-inflammatory12
and pro-thrombotic.13
Our interest in eugenol derivatives as lipoxygenase
inhibitors emerges from the early work by Naidu and
co-workers, in which the 5-lipoxygenase inhibition of
this natural compound was reported.1 It is well docu-
mented that mammalian lipoxygenases (LOs) are
non-heme iron-containing enzymes responsible for the
oxidation of polyunsaturated fatty acids and esters to
hydroperoxy derivatives.2 There are heterogeneous fam-
ily of enzymes distributed widely throughout the plant
and animal kingdoms,3 and named according to the po-
sition at which a key substrate, arachidonic acid (AA), is
oxidized. Among the mammalian lipoxygenases in-
volved in the etiology of human disease, 5-lipoxygenase
(5-LO) is now well established as a target for reducing
the production of leukotrienes (important particularly
in asthma).4 More recently, 15-lipoxygenase (15-LO)
has emerged as an attractive target for therapeutic inter-
vention.5 15-LO has been implicated in the progression
of certain cancers6,7 and chronic obstructive pulmonary
It is also found that 15-LO is linked to cardiovascular
complications due to participation in oxidative modifi-
cation of low-density lipoproteins (LDL), leading to
the development of atherosclerosis.9
Conrad and colleagues15 showed that in vivo 15-LOa has
antitumor effects in human airway carcinomas and pro-
motes apoptotic pathway. They showed that neoplastic
tissues from human airway carcinomas demonstrated
nonspecific staining for human 15-LOa as compared with
normal tissues. In contrast, in human prostate tumors 15-
LOa was overexpressed as compared with normal adja-
cent tissue,7 and 15-LOb was poorly expressed in prostate
tumors.16 In PC3 cells, 13(S)-HODE, one of the 15-LOa
metabolites, upregulated MAP kinase, whereas in con-
trast 15(S)-HETE, the 15-LOb metabolite, downregu-
lated MAP kinase.17 Taken together, these findings
Keywords: 4-Allyl-2-methoxyphenol; 15-HLOa; 15-HLOb; Protein
modeling; Soybean lipoxygenase; Docking.
*
Corresponding author. Tel.: +98 05118795162; fax: +98
05118795560; e-mail: smseyedi@yahoo.com
Tel.: +98 511 7112611; fax: +98 511 7112596.
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.10.016

H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
891
including the upregulation of 15-LOa within the airway
tissue of smoking patients with chronic bronchitis pro-
vided new evidence of possible acquired abnormalities
linked to airway inflammation. The bronchial epithelium
is clearly a key player in inflammation and structural
changes in airway diseases. Its rich content in 15-LOa–
and 15-LOb–derived products highlights their potential
as new target for therapeutic interventions.
in dentistry and is applied to the oral environment.19 In
addition, eugenol is a flavoring agent in cosmetic and
food products.20 Eugenol has been shown to possess
many medicinal properties such as antispasmodic,21
antipyretic,22 anti-inflammatory,23 and antibacterial
activity.24 Recently it is reported that eugenol inhibits
5-LO enzyme by non-competitive mechanism.1
In this work, seventeen ester derivatives of eugenol 4a–q
were designed, synthesized and their activities were iden-
tified as the mean of IC50 on soybean 15-LO (SLO).
There is reasonable homology between the soybean
LO and the human one (Fig. 1). This homology becomes
Three different strategies have been developed to inhibit
the LO’s pathway.18 They involve (i) redox inhibitors or
antioxidants, which interfere with the redox cycle of 15-
LO, (ii) iron-chelator agents, and (iii) non-redox com-
petitive inhibitors, which compete with AA to bind the
enzyme active site.
˚
more identical (ꢀ50%) within 8A in the active site pock-
et. Obviously soybean enzyme is much more accessible
than the human one. Therefore, one can expect that
the results can be extendable to human LO.
Eugenol (4-allyl-2-methoxyphenol) is naturally occur-
ring phenolic compound in basil, cinnamon, and nut-
meg, and the major component of clove oil. It is
widely used as component of zinc oxide eugenol cement
In this study, (i) common bonding model of 4-allyl-2-
methoxyphenyl carboxylates in SLO active site,
Figure 1. Clustal X (1.81) multiple alignment of SLO (green), 15-HLOa (blue), and 15-HLOb (red). The amino acids in the active site pocket within
˚
8 A are highlighted by yellow background.

892
H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
(ii) QSAR study of inhibitors to propose key features of
this class of inhibitors, and (iii) theoretical potency of
some of these compounds for inhibiting 15-HLOa and
15-HLOb activity are reported.
optimization (Convergence limit = 0.01; Iteration lim-
it = 50; RMS gradient = 0.05 kcal/mol; Fletcher-Reeves
optimizer algorithm) in HyperChem7.5.29,30
Crystal structure of soybean lipoxygenase-3 (arachi-
donic acid 15-lipoxygenase) complex with 13(S)-hydro-
peroxy-9(Z)-2,11(E)-octadecadienoic acid was retrieved
from RCSB Protein Data Bank (PDB entry: 1IK3).
2. Chemistry
4-Allyl-2-methoxyphenyl esters 4a–q (Scheme 1 and Ta-
ble 1), according to the literature,21 started from eugenol
2 and corresponding acid chlorides 3a–q which were
either purchased or prepared (3b–d and 3q) by reaction
of thionyl chloride and corresponding carboxylic
acids.25 All desired esters were synthesized by the action
of acid chlorides in either aqueous solution of sodium 4-
allyl-2-methoxyphenolate (4a, e–q) or hydrochloride
salts of pyridine carboxyl chloride 3b–d (prepared from
pyridine carboxylic acids and thionyl chloride) in dry
pyridine.
3.3. Molecular docking
Automated docking simulation was implemented to
dock 4a–q into the active site of SLO with AutoDock
version 3.0331 using Lamarckian genetic algorithm.32
This method has been previously shown to produce
bonding modes similar to the experimentally observed
modes.30,32–34 The torsion angles of the ligands were
identified, hydrogens were added to the macromolecule,
bond distances were edited, and solvent parameters were
added to the enzyme 3D structure. Partial atomic
charges were then assigned to the macromolecule as well
as ligands (Gasteiger for the ligands and Kollman for
the protein).
Structural assignments of compounds 4a–q were based
upon the spectral and microanalytical data.
3. Molecular modeling, docking, and QSAR study
3.1. Multiple alignment
The regions of interest of the enzyme were defined by
considering Cartesian chart 18.5, ꢁ2.0, and 20.4 as the
central of a grid size of 40, 50, and 40 points in X, Y,
and Z axises. The docking parameter files were gener-
ated using Genetic Algorithm and Local Search Param-
eters (GALS) while number of generations was set to 50.
Compounds 4a–q were each docked into the active site
of LO enzyme and the simulations were composed of
50 docking runs, each of 50 cycles containing a maxi-
mum of 10,000 accepted and rejected steps. The simu-
lated annealing procedure was started at high
temperature (RT = 616 kcal/mol, where R is the gas
constant and T is the steady-state temperature) and
was decreased by a fraction of 0.95 on each cycle.33
The 50 docked complexes were clustered with a root-
Highly conserved amino acids were identified through
multiple alignment on clustalX 1.8126 software. Se-
quences of lipoxygenase (LO) family were selected from
blasted sequences via ExPASY proteomics server27 with
E-value < 0.02. Multiple alignment process was then
carried out on the selected sequences (protein weight
matrix: BLOSUM series, gap penalty = 10%).
3.2. Structure optimization
Structures 4a–q were simulated in chem3D professional;
Cambridge software; using MM2 method (RMS gradi-
ent = 0.05 kcal/mol).28 Output files were minimized un-
der semi-empirical AM1 method in the second
˚
mean-square deviation tolerance of 0.1 A. The program
generated 50 docked conformers of 4a–q corresponding
to the lowest-energy structures. After docking procedure
O
R
OH
O
O
1) NaOH (aq)
2) RCOCl (3a, 3e-q)
O
4a, 4e-q
2
N
O
O
O
O
O
Cl
SOCl2
Eugenol / Py
HCl
N
N
OH
1b-d
3b-d
4b-d
Scheme 1.

H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
893
Table 1. For 4a, 4p, and 4q the IC50 values were calculated from dose–
response curves and values are given as means SEM of three
individual samples
Table 1 (continued)
Compound
Structure
OCH3
IC50 (lM)
O
R
4n
4o
23.0 1.3
O
O
OCH3
Compound
Eugenol
Structure
IC50 (lM)
19.0 1.3
34.6 1.1
4a
4b
4c
4d
4e
4.4 0.3
33.3 1.5
1.7 0.05
2.3 0.2
6.7 0.8
4p
4q
2.2 0.3
0.017 0.001
N
N
in AD3, docking results were submitted to Weblab
Viewerlite 4.035 and Swiss-PdbViewer 3.7 (spdbv)36 for
further evaluations. The results of docking processing
(DGb: estimated free energy of bonding, DGd: final
docked energy, and Ki: estimated inhibition constant)
are outlined in Table 2.
N
F
3.4. QSAR studies
F
QSAR studies were performed for optimized com-
pounds 4a–q in DRAGON 2.1.37 In this study van der
Waals volume (Sv)38 and Moriguchi octanol-water par-
tition coefficient (logP)38 were determined (Table 2).
4f
4g
4h
4i
2.1 0.4
F
7.2 0.5
3.5. Protein modeling
Three-dimensional models of the 15-HLOa and 15-
HLOb sequences were constructed by homology model-
ing. BLAST sequence homology searches were per-
formed in order to identify the template proteins. The
soybean lipoxygenase3 complex with linoleic acid
(PDB entry: 1IK3) was chosen as the template for mod-
eling the proteins. Model building was performed in the
program MODELLER9v139 using model-ligand algo-
rithm. Several models at various refinement levels were
generated and finally the refined structures involving lin-
oleic acid in the active site pocket were minimized under
molecular mechanic AMBER method (RMS gradi-
ent = 1) in HyperChem7.5.29 All models were validated
using the program ERRAT at UCLA.40 The best model
had an Errat score of 78–82%.
Cl
168.0 13.5
Cl
5.2 0.2
Cl
4j
134.3 7.5
77.2 5.1
11.4 1.1
H3C
4k
4l
4. 15-LO inhibitory assessment
CH3
Lipoxygenase activity was measured in borate buffer
solutions (0.1 M, pH 9) using the method described in
the literature,41,42 by measuring the absorbance at
234 nm for 60 s after addition of the enzyme (soybean
15-lipoxygenase), and linoleic acid (final concentration:
134 lM) as substrate at 20 1 ꢁC. The final enzyme
concentration was 167 U/mL. Synthesized substances
were added in DMSO solutions (final DMSO concentra-
CH3
4m
14.6 1.0
(continued on next page)

894
H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
Table 2. Data obtained from docking and QSAR analyses
Compound
Ki
DGb
ꢁ7.75
DGd
ꢁ9.49
Sv
7.79
logP
4a
4b
4c
4d
4e
4f
2.09e-6
2.57e-6
3.06e-6
4.70e-6
2.20e-6
1.39e-6
1.68e-6
2.37e-4
7.22e-7
1.58e-6
4.88e-6
6.14e-7
8.61e-7
1.09e-6
8.41e-7
5.78e-7
9.13e-9
2.255
0.860
0.860
0.860
2.702
2.702
2.702
2.786
2.786
2.786
2.871
2.871
2.871
1.859
1.859
3.516
4.823
ꢁ7.63
ꢁ7.52
ꢁ7.27
ꢁ7.72
ꢁ7.99
ꢁ7.88
ꢁ4.94
ꢁ8.38
ꢁ7.91
ꢁ7.25
ꢁ8.47
ꢁ8.27
ꢁ8.13
ꢁ8.29
ꢁ8.51
ꢁ10.97
ꢁ8.66
ꢁ9.34
ꢁ8.62
ꢁ9.46
ꢁ9.78
ꢁ9.33
ꢁ3.65
ꢁ9.17
ꢁ9.79
ꢁ9.03
ꢁ10.29
ꢁ10.19
ꢁ10.07
ꢁ10.25
ꢁ9.38
ꢁ12.31
7.19
7.19
7.19
7.90
7.90
7.90
8.53
8.53
8.53
9.39
9.39
9.39
9.90
9.90
9.59
14.78
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
4q
DGb, estimated free energy of bonding; DGd, final docking energy, and Ki, estimated inhibition constant; Sv, van der Waals volume of benzene
derivatives; logP, Moriguchi octanol–water partition coefficient.
tion 1%); whereas DMSO was added in control experi-
ments with no inhibitor. The mixture of each inhibitor
and linoleic acid was set as blank sample in testing step.
At least six control test tubes and three tubes for each
inhibitor solution were measured. To ensure constant
enzyme activity throughout the experiment, the enzyme
solution was kept in ice, and controls were measured at
regular intervals. Calculation of enzyme activity was
carried out as previously described42 and IC50 values
were determined by linear interpolation between the
points around 50% activity (Table 1).
Regarding the site-directed mutagenesis reported by
Klinman, Minor, and Holman43,44 and docking pro-
cedure in this study, one might conclude that eugenol
is able to inhibit LO through allyl interaction with
amino acids close to iron atom of the enzyme, mim-
icking enzymic natural substrate (i.e., unsaturated
fatty acid).
Benzoate and cycloalkylate analogs 4b–q were designed,
synthesized, and docked into the active site to support
this mechanism. The esters of 4-allyl-2-methoxyphenol
4a–q showed a broad range of inhibition activity on
the enzyme (IC50 = 0.017–168.0 lM; Table 1). Com-
pound 4q having an adamantanecarboxylate substituent
was the most potent inhibitor at 17 nM, while the nico-
tinate, 3-fluorobenzoate, cyclohexanecarboxylate, and
isonicotinate analogs (4c, 4f, 4p, and 4d, respectively)
presented less activity (IC50 = 1.7, 2.1, 2.2, and 2.3 lM,
respectively). It was interesting to view 2- and 4-chloro-
benzoate analogs (4h and 4j) as weak inhibitors of SLO
(IC50 > 100 lM).
5. Results and discussion
By considering Naidu’s work1 we tested the inhibitory
property of eugenol on the SLO (substrate: linoleic
acid). The results showed IC50 = 34.6 lM for the men-
tioned enzyme. Since a non-competitive mechanism
has been proved for inhibitory activity of eugenol in
inhibition of 5-LO,1 we thought the mechanism might
go through oxidation of hydroxyl group which mimics
the fact of non-competitive theory reported by Naidu.1
To prove this idea the hydroxyl group was protected
by benzoate (a bulky and moderately lipophilic group).
Unexpectedly the benzoate analog 4a showed a better
activity (IC50 = 4.4 lM) which means that the activity
still exists and no other products such as hydroperoxy
are isolated from action of the LO enzyme on 4a as sub-
strate (assuming that hydroperoxy is supposed to be ob-
tained if the redox pathway is blocked and the inhibitor
acts through its allylic group in reaction with the enzyme
active site similar to the oxidation of natural unsatu-
rated fatty acids).à
The experimental results matched with theoretical Ki of
docking study for those models (Table 2) in which allylic
double bond oriented toward iron atom similar to
orientation of linoleic acid (LA) in the active site
(Figs. 2 and 3).43 We generated 50 docked conformers
of 4a–q corresponding to the lowest energy structures
in ADT software. A detailed inspection of each indepen-
dent inhibitor conformer revealed that more than 40%
of docking results had nearly identical orientations in
which allyl group of each inhibitor oriented toward Fe
core (except compounds 4h, 4i, 4k, and 4j: <20%). One
conformer from each ester cluster which had more sim-
ilarity with optimum conformer (lowest Ki) of benzoate
analog (4a) was adopted as the ‘consensus’ structure and
used for further analysis.
à
Twenty milligrams of substrate was reacted with soybean LO enzyme
(1000 U/mL) in 30 mL borate buffer solution (0.1 M, pH 9) at 20 ꢁC
for 1 h. The mixture was then extracted with dichloromethane
(2 · 15 mL) and analyzed by TLC.
It seems that the allyl group and phenyl core have
,
hydrophobic interaction with Ile557, Leu565, Leu773

H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
895
large residues such as Ile or Leu to an Ala opens up
space within the bonding pocket of SLO, leading to al-
tered HÅ transfer kinetics. The Ile557 ! Ala and
Ile572 ! Phe mutants decreased kcat by twofold from
WT (wild type), While Leu565 ! Ala and Leu773 ! Ala
decreased kcat by 60- and 1000-fold, respectively, indi-
cating that these hydrophobic residues (specially
Leu565, and Leu773) contribute significantly to cataly-
sis.44 According to the result of multiple alignment,
three amino acids Ile557, Leu565, and Leu773 are found
to be conserved over all species.
We can also view in Figures. 3 and 4 that the proposed
orientation of docked molecules has hydrophilic and
hydrophobic interaction with conserved His513, Gln514
,
and Gln716. The amino acids Gln514 and Gln716 play a
key role in oxidation potential of Fe3+ via hydrogen
bonding with Asn713 and His518
.
This hydrogen bond
45
network is present in both SLO and 15-RLO (rabbit
15-LO) structures and also plays a steric role in orient-
ing the substrate and inhibitor bound to LO.45 The
C-3–C-8 hydrocarbon tail of LA is flanked by the
hydrophobic portion of the Gln514 and Gln716 (Figs. 2
and 4). Disrupting this bonding pocket by changing
the position of Gln514 and Glu716 may affect the proper
positioning of the substrate for C-H bond cleavage so
that abstraction becomes more rate-limiting (as was ob-
served in the Gln514 ! Ala, Gln716 ! Asn and
Gln716 ! Glu mutants by 4-, 3-, and 6-fold decrease in
kcat from WT SLO, respectively,45). Proposed inhibitory
model of docked molecules has hydrogen bond with
Gln716 via carbonyl group (Fig. 3). This can change
the oxidation potential of ferric ion by disrupting the
hydrogen bond of Gln716 and Asn713. The aromatic
and aliphatic part of carboxylate moiety in eugenol
derivatives is flanked by the hydrophobic portion of
the Gln716 side chain like LA (Fig. 4).
Figure 2. X-ray presentation of SLO active site pocket complex with
13(S)-hydroperoxy-9(Z)-2,11(E)-octadecadienoic acid (green ball and
stick) (PDB entry: 1IK3). The conserved amino acids are presented in
blue and light brown color. Hydrogen bonds are shown by dashed
black lines.
and Ile572, respectively, in such an orientation. The most
critical residues, that is, Ile557, Leu565, Leu773, and Ile572
surprisingly appeared close to the active site (Fig. 2). X-
,
ray presentation of LA into SLO43 indicates that Ile557
˚
,
Leu565, and Leu773 lay within 4–6 A of Fe3+-OH and
both Leus are near the reactive C-11–C-13 of LA (C-
11: hydrogen abstraction site, C-13: oxygenation site).
Although Ile572 is far from Fe3+-OH (at 9 A), still forms
˚
part of the substrate-bonding cavity. Each of these resi-
dues provides a large surface to interact with natural
substrate, particularly Leu565 and Leu773. Mutating
˚
Figure 3. Superimposition of the bonding conformations of 4a–q in colored stick in the active site of SLO within 8 A. The hydrogen bonds between
Gln716 and inhibitors are shown by dashed green lines.

896
H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
ent or heteroatom at para and meta position, respec-
tively (Fig. 8a and b—compound 4j was excluded
because of high deviation). Decreasing of Sv increases
tendency of these compounds in SLO inhibition.
Due to lipophilic interaction area of carboxylate moiety
(Sv and logP) in inhibition of SLO activity, we studied
the tendency of compounds 4a, 4p, and 4q for inhibiting
modeled 15-HLOa and 15-HLOb (these compounds
showed good relation between Ki and IC50 variations
for SLO). The structures of modeled 15-HLOa and 15-
HLOb demonstrate a high level of conservation of the
overall topology. Thus the structures of 15-HLOa could
be superimposed on the 15-HLOb with RMS for the C-a
˚
atoms of around 0.95A. The largest differences between
the 15-HLOa and 15-HLOb were found in the regions of
helix a2, a4-a5, a6-a7, and a15-a16 (residues 141–152,
190–197, 233–246, 257–259, 321–327, and 565–571 for
15-HLOa in contrast with 155–165, 204–207, 243–247,
258–273, 336–339, and 578–588 for 15-HLOb, respec-
tively). These residues do not build up any part of the
substrate binding cleft. In the active site pocket of the
two modeled proteins, the backbone of conserved amino
Figure 4. Solvent surface view of conserved amino acids which have
interaction with 13(S)-hydroperoxy-9(Z)-2,11(E)-octadecadienoic acid
(green stick) and 4q (light brown stick).
˚
acids laying within 8 A of the Fe atom is well fitted but
this is not observed for other conserved amino acids in
this region. Free space of catalytic pocket seems to be
smaller for 15-HLOa in comparison with 15-HLOb. This
lacking comes from steric occupation of aromatic side
chains of Phe352, Phe414, and Tyr551 in the cavity
(Fig. 9b). After docking process on the modeled enzymes,
4q showed better Ki for 15-HLOb than 15-HLOa by 100-
fold in the same orientation which had been proved for
SLO (Table 3 and Fig. 9). It may have something to do
with the smaller space of active site pocket of 15-HLOa
in comparison with 15-HLOb. Considering the lipophilic-
ity of amino acids which surrounded the carboxylate moi-
ety of inhibitors, these amino acids are more lipophilic in
15-HLOb in comparison with 15-HLOa (Table 4). The
lipophilicity was taken from ExPASy27 (ProtScale) by
applying Hphob (Kyte and Doolittle hydropathicity).46
Therefore, we assume 4q a selective inhibitor for
15-HLOb in comparison with 15-HLOa.
The Ki of proposed model of compounds, 4a, 4p, and 4q,
have good relation with IC50 results (Fig. 5). This comes
from tendency of the carboxylate moiety for filling all of
the empty lipophilic space of Val372, Phe576, Ile770, and
Gln716 side chains (Fig. 6). This result can be clarified
by considering lipophilic factor (logP) and van der
Waals volume (Sv) of cyclohexyl, phenyl, and adaman-
tyl groups (Table 2).
It is notable that in each group of isomeric inhibitors,
the compounds with substituent or heteroatom in posi-
tion 2 have lower activity in comparison with other iso-
mers (4h, 4k, and 4b: IC50 = 168.0, 77.2, and 33.3 lM,
respectively). Compound 4e (IC50 = 6.7 lM) does not
follow the above road map probably because of the
hydrogen bonding of fluorine with conserved His513
(Fig. 7). Comparison of the calculated 1D-QSAR data
with IC50 values, showed linear and non-linear relation
between Sv and IC50 values of inhibitors with substitu-
In summary, the present study introduces that large and
lipophilic eugenol esters such as 4q can behave as SLO
inhibitors (IC50 = 17 nM) and also as a selective inhibi-
tor of 15-HLOb when compared with 15-HLOa. The
importance of these compounds could be more high-
lighted when we rank their easy synthesis pathway and
their high yield.
9
R2 = 0.987
4q
8
7
6
5
6. Experimental
6.1. General procedures
4p
4a
Melting points were recorded on an Electrothermal type
9100 melting point apparatus. The 1H NMR (100 MHz)
spectra were recorded on a Bruker AC 100 spectrome-
ter. Elemental analysis was obtained on a Thermo Finn-
igan Flash EA microanalyzer. All measurements of
lipoxygenase activities were carried out using an Agilent
8453 spectrophotometer. The soybean 15-lipoxygenase
5
6
7
8
9
-log Ki
Figure 5. Diagram of ꢁlog IC50 versus ꢁlog Ki for compounds 4a, 4p,
and 4q.

H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
897
Figure 6. (a) Amino acids having lipophilic interaction with 4a, 4p, and 4q (The lipophilic parts of the side chains are distinguished by purple color).
Solvent surface of lipophilic amino acids interacting with carboxylate moiety of 4q, 4p, and 4a is shown in b, c, d, e, f, and g, respectively.
and other chemicals were purchased from Sigma, Fluka,
and Merck Co., respectively.
stirred and refluxed for 1 h. The thionyl chloride was
then evaporated under reduced pressure giving a white
crystalline residue of hydrochloride of 3b–d. The purity
of 3b–d was quite sufficient to use the product directly
for the following synthesis.
6.2. General procedure for preparation of 4-allyl-2-
methoxyphenyl carboxylate 4b-d
Thionyl chloride (10 mL) was added to pyridine carbox-
ylic acids 1b–d (20 mmol, 2.46 g). The mixture was
To a stirred solution of 3b–d (1.78 g, 10 mmol) in dry
pyridine (10 mL) was added eugenol (1.64 g,

898
H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
10 mmol) dropwise at room temperature. The mix-
ture was refluxed in oil bath while stirring for 4 h.
After reaction completion, the pyridine was
evaporated under reduced pressure. The residue was
treated with 5% sodium carbonate (15 mL) and
extracted with dichloromethane (2 · 15 mL). The
organic extract was dried with anhydrous sodium
sulfate, concentrated under reduced pressure, and
crystallized from ethanol to provide the pure desired
compounds 4b–d.
6.3. General procedure for preparation of 4-allyl-2-metho-
xyphenyl carboxylate 4a and 4e-q
The acid chlorides 1a, e–q were synthesized via the
method described for compounds 2b–d.
To a stirred solution of sodium hydroxide (0.48 g,
12 mmol) and eugenol (1.64 g, 10 mmol) in water
(10 mL) were added acid chlorides 3a and 3e–q
(10 mmol) dropwise at room temperature. After 30-
min stirring at room temperature the mixture was ex-
tracted with dichloromethane (2 · 15 mL), washed with
5% sodium carbonate (2 · 15 mL), dried with anhydrous
sodium sulfate, and concentrated under reduced pres-
sure to provide the desired compounds. All products
were crystallized from ethanol except 4h and 4k which
were purified by column chromatography (silica gel 60;
230–400, eluent: chloroform).
Figure 7. Stick view of compound 4e interacting with His513 via
hydrogen bonding of fluorine atom (green color). Hydrogen bonds are
revealed by dashed black lines.
6.3.1. 4-Allyl-2-methoxyphenyl benzoate (4a). White
crystal. Yield: 87%; mp 55–56 ꢁC. IR: 1727 cmꢁ1
1
(C@O). H NMR (CDCl3): d 3.45 (d, 2, J = 6.6 Hz, –
CH2–), 3.80 (s, 3, –OCH3), 5.01–5.25 (m, 2, H2C@),
5.80–6.25 (m, 1, @CH), 6.82 (d, 1, J = 8.4, H-5), 6.85
(s, 1, H-3), 7.10 (d, J = 8.4, H-6), 7.40–7.72 (m, 3, H-
30, H-40, H-50), 8.22 (d, 2, J = 7.9, H-20, H-60). Found
C, 76.21; H, 6.03. C17H16O3 requires: C, 76.10; H,
6.01%.
30
a
Mean
7.956
8.145
Std. Deviation
4n
R2 = 0.9743
20
10
0
4l
6.3.2. 4-Allyl-2-methoxyphenyl 2-pyridinecarboxylate
(4b). Light brown crystal. Yield: 43%; mp 90–91 ꢁC.
4i
4a
1
IR: 1754 cmꢁ1 (C@O). H NMR (CDCl3): d 3.42 (d, 2,
4c
4f
J = 6.6 Hz, –CH2–), 3.80 (s, 3, –OCH3), 5.02–5.25 (m,
2, H2C@), 5.72–6.22 (m, 1, @CH), 6.81 (d, 1, J = 8.4,
H-5), 6.84 (s, 1, H-3), 7.12 (d, 1, J = 8.4, H-6), 7.54
(m, 1, H-50), 7.89 (m, 1, H-40), 8.27 (d, 1, J = 7, H-30),
8.84 (d, J = 4, H-60). Found C, 71.04; H, 5.65; N, 5.25.
C16H15NO3 requires: C, 71.36; H, 5.61; N, 5.20%.
6
7
8
9
10
11
Sv
30
20
10
0
b
Mean
9.527
7.070
Std. Deviation
4o
6.3.3. 4-Allyl-2-methoxyphenyl nicotinate (4c). White
crystal. Yield: 71%; mp 70–71 ꢁC. IR: 1745 cmꢁ1
(C@O). 1H NMR (CDCl3): d 3.42 (d, 2, J = 6.6 Hz,
–CH2–), 3.81 (s, 3, –OCH3), 5.02–5.25 (m, 2, H2C@),
5.72–6.22 (m, 1, @CH), 6.81 (d, 1, J = 8.4, H-5), 6.85
(s, 1, H-3), 7.06 (d, J = 8.4, H-6), 7.45 (dd, 1, J = 7.8,
4.9, H-50), 8.45 (d, 1, J = 7.8, H-40), 8.82 (d, J = 4.9,
H-60), 9.40 (s, 1, H-20). Found C, 71.08; H, 5.58; N,
5.19. C16H15NO3 requires: C, 71.36; H, 5.61; N, 5.20%.
R2 = 0.9733
4m
4g
4a
8
4d
6
7
9
10
11
Sv
6.3.4. 4-Allyl-2-methoxyphenyl isonicotinate (4d). White
crystal. Yield: 73%; mp 56–57 ꢁC. IR: 1762 cmꢁ1
(C@O). 1H NMR (CDCl3): d 3.41 (d, 2, J = 6.5 Hz,
Figure 8. Diagrams of measured IC50 versus van der Waals volume
(Sv) of benzoate moiety of eugenol esters with meta (a) and para
(b) substituent.

H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
899
Figure 9. Docking results of 4q in the active site of SLO (a), 15-HLOa (b), and 15-HLOb (c). The conserved and mutated amino acids are presented
by green and blue color, respectively.
Table 3. The estimated inhibition constant (Ki) of 4a, 4p, and 4q from
docking study on 15-HLOa and 15-HLOb
30, H-50), 8.86 (d, 2, J = 5, H-20, H-60). Found C,
71.19; H, 5.63; N, 5.22. C16H15NO3 requires: C, 71.36;
H, 5.61; N, 5.20%.
Compound
Ki (15-HLOa)
Ki (15-HLOb)
4a
4p
4q
1.90e-6
6.05e-7
1.96e-6
5.48e-6
1.86e-6
3.58e-8
6.3.5. 4-Allyl-2-methoxyphenyl 2-fluorobenzoate (4e).
Light brown crystal. Yield: 63%; mp 55–56 ꢁC. IR:
1729 cmꢁ1 (C@O). 1H NMR (CDCl3): d 3.40 (d, 2,
J = 6.6 Hz, –CH2–), 3.81 (s, 3, –OCH3), 5.01–5.23 (m,
2, H2C@), 5.77–6.26 (m, 1, @CH), 6.60 (m, 2, H-30,
H-50), 6.81 (d, 1, J = 8.5, H-5), 6.84 (s, 1, H-3), 7.06
–CH2–), 3.81 (s, 3, –OCH3), 5.00–5.27 (m, 2, H2C@),
5.80–6.20 (m, 1, @CH), 6.85 (d, 1, J = 8.4, H-5), 6.86
(s, 1, H-3), 7.09 (d, J = 8.4, H-6), 8.00 (d, 2, J = 5, H-

900
H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
Table 4. The Kyte & Doolittle hydropathicity (Hphob) of amino acids
which have direct interaction with carboxylate moiety of inhibitors
J = 6.6 Hz, –CH2–), 3.80 (s, 3, –OCH3), 5.00–5.26 (m,
2, H2C@), 5.80–6.20 (m, 1, @CH), 6.80 (d, 1, J = 8.4,
H-5), 6.84 (s, 1, H-3), 7.10 (d, J = 8.4, H-6), 7.50 (d, 2,
J = 8.7, H-30, H-50), 8.15 (d, 2, J = 8.7, H-20, H-60).
Found C, 67.49; H, 5.02. C17H15ClO3 requires: C,
67.44; H, 4.99%.
15-HLOa
Hphob
15-HLOb
Hphob
Gln211
Phe352
Gln547
Tyr551
Ser552
Val554
Ala557
Pro558
Cys559
Gln589
Met590
Thr593
ꢁ3.50
2.80
Arg221
Phe365
Gln560
Cys564
Ala565
Met567
Leu570
Pro571
Pro572
Val603
Ile604
ꢁ4.50
2.80
ꢁ3.50
ꢁ1.30
ꢁ0.80
4.20
ꢁ3.50
2.50
1.80
6.3.11. 4-Allyl-2-methoxyphenyl 2-methylbenzoate (4k).
1
Light yellow oil. Yield: 18%. IR: 1727 cmꢁ1 (C@O). H
1.90
3.80
1.80
ꢁ1.60
2.50
NMR (CDCl3): d 2.45 (s, 3H, –CH3), 3.39 (d, 2,
J = 6.5 Hz, –CH2–), 3.85 (s, 3, –OCH3), 5.00–5.20 (m,
2, H2C@), 5.79–6.23 (m, 1, @CH), 6.79 (d, 1, J = 8.5,
H-5), 6.83 (s, 1, H-3), 7.04 (d, J = 8.5, H-6), 7.22–7.45
(m, 3, H-30, H-40, H-50), 8.08 (d, 1, J = 7.6, H-60). Found
C, 76.83; H, 6.50. C18H18O3 requires: C, 76.57; H,
6.43%.
ꢁ1.60
ꢁ1.60
4.20
ꢁ3.50
1.90
4.50
3.80
ꢁ0.70
Leu607
(d, J = 8.5, H-6), 7.39 (m, 1, H-40), 7.67 (m, 1, H-60).
Found C, 71.07; H, 5.33. C17H15FO3 requires: C,
71.32; H, 5.28%.
6.3.12. 4-Allyl-2-methoxyphenyl 3-methylbenzoate (4l).
White crystal. Yield: 85%; mp 51–52 ꢁC. IR: 1723 cmꢁ1
1
(C@O). H NMR (CDCl3): d 2.43 (s, 3H, –CH3), 3.41
(d, 2, J = 6.5 Hz, –CH2–), 3.80 (s, 3, –OCH3), 5.02–
5.24 (m, 2, H2C@), 5.81-6.25 (m, 1, @CH), 6.81 (d, 1,
J = 8.6, H-5), 6.84 (s, 1, H-3), 7.06 (d, J = 8.6, H-6),
7.38 (m, 2, H-40, H-50), 8.03 (m, 2, H-20, H-60). Found
C, 76.70; H, 6.46. C18H18O3 requires: C, 76.57; H,
6.43%.
6.3.6. 4-Allyl-2-methoxyphenyl 3-fluorobenzoate (4f).
White crystal. Yield: 75%; mp 61–62 ꢁC. IR:
1744 cmꢁ1 (C@O). 1H NMR (CDCl3):
d
3.41
(d, 2, J = 6.5 Hz, –CH2–), 3.80 (s, 3, –OCH3), 5.00–
5.28 (m, 2, H2C@), 5.78–6.26 (m, 1, @CH), 6.81 (d, 1,
J = 8.4, H-5), 6.85 (s, 1, H-3), 7.06 (d, J = 8.4, H-6),
7.25–7.52 (m, 2, H-40, H-50), 7.96 (m, 2, H-20, H-60).
Found C, 71.15; H, 5.33. C17H15FO3 requires: C,
71.32; H, 5.28%.
6.3.13. 4-Allyl-2-methoxyphenyl 4-methylbenzoate (4m).
White crystal. Yield: 87%; mp 92–93 ꢁC. IR: 1730 cmꢁ1
1
(C@O). H NMR (CDCl3): d 2.45 (s, 3, –CH3), 3.41 (d,
2, J = 6.5 Hz, –CH2–), 3.81 (s, 3, –OCH3), 5.00–5.25 (m,
2, H2C@), 5.80–6.20 (m, 1, @CH), 6.82 (d, 1, J = 8.4, H-
5), 6.84 (s, 1, H-3), 7.07 (d, J = 8.4, H-6), 7.31 (d, 2, J =
8, H-30, H-50), 8.11 (d, 2, J = 8, H-20, H-60). Found C,
76.32; H, 6.41. C18H18O3 requires: C, 76.57; H, 6.43%.
6.3.7. 4-Allyl-2-methoxyphenyl 4-fluorobenzoate (4g).
White crystal. Yield: 79%; mp 56–57 ꢁC. IR:
1732 cmꢁ1 (C@O). 1H NMR (CDCl3): d 3.41 (d, 2,
J = 6.5 Hz, –CH2–), 3.80 (s, 3, –OCH3), 5.02–5.24 (m,
2, H2C@), 5.78–6.23 (m, 1, @CH), 6.81 (d, 1, J = 8.4,
H-5), 6.84 (s, 1, H-3), 7.06 (d, J = 8.4, H-6), 7.18 (m,
2, H-30, H-50), 8.23 (m, 2, H-20, H-60). Found C,
71.17; H, 5.29. C17H15FO3 requires: C, 71.32; H,
5.28%.
6.3.14. 4-Allyl-2-methoxyphenyl 3-methoxybenzoate (4n).
White crystal. Yield: 83%; mp 59–60 ꢁC. IR: 1727 cmꢁ1
(C@O). 1H NMR (CDCl3): d 3.41 (d, 2, J = 6.6 Hz,
–CH2–), 3.80 (s, 3, –OCH3), 3.88 (s, 3, –OCH3), 5.01–
5.23 (m, 2, H2C@), 5.81–6.22 (m, 1, @CH), 6.80 (d, 1,
J = 8.6, H-5), 6.84 (s, 1, H-3), 7.07 (d, J = 8.6, H-6),
7.16 (d, 1, J = 8.8, H-40), 7.40 (t, 1, J = 8.8, H-50), 7.72
(s, 1, H-20), 7.83 (d, J = 8.8, H-60). Found C, 72.41; H,
6.06. C18H18O4 requires: C, 72.47; H, 6.08%.
6.3.8. 4-Allyl-2-methoxyphenyl 2-chlorobenzoate (4h).
Light yellow oil. Yield: 39%. IR: 1741 cmꢁ1 (C@O).
1H NMR (CDCl3): d 3.41 (d, 2, J = 6.5 Hz, –CH2–),
3.84 (s, 3, –OCH3), 5.00–5.26 (m, 2, H2C@), 5.80–6.25
(m, 1, @CH), 6.82 (d, 1, J = 8.4, H-5), 6.85 (s, 1, H-3),
7.10 (d, J = 8.4, H-6), 7.20–7.60 (m, 3, H-30, H-40, H-
50), 8.11 (d, 1, J = 5.9, H-60). Found C, 67.73; H, 5.04.
C17H15ClO3 requires: C, 67.44; H, 4.99%.
6.3.15. 4-Allyl-2-methoxyphenyl 4-methoxybenzoate (4o).
White crystal. Yield: 85%; mp 93–94 ꢁC. IR: 1731 cmꢁ1
(C@O). 1H NMR (CDCl3): d 3.40 (d, 2, J = 6.6 Hz,
–CH2–), 3.80 (s, 3, –OCH3), 3.89 (s, 3, –OCH3), 5.00–
5.25 (m, 2, H2C@), 5.81–6.24 (m, 1, @CH), 6.80 (d, 1,
J = 8.4, H-5), 6.84 (s, 1, H-3), 6.98 (d, 2, J = 8.9, H-30,
H-50), 7.07 (d, J = 8.4, H-6), 8.17 (d, 2, J = 8.9, H-20,
H-60). Found C, 72.32; H, 6.09. C18H18O4 requires: C,
72.47; H, 6.08%.
6.3.9. 4-Allyl-2-methoxyphenyl 3-chlorobenzoate (4i).
White crystal. Yield: 81%; mp 48–49 ꢁC. IR:
1745 cmꢁ1 (C@O). 1H NMR (CDCl3): d 3.40 (d, 2,
J = 6.5 Hz, –CH2–), 3.80 (s, 3, –OCH3), 5.00–5.25 (m,
2, H2C@), 5.80–6.20 (m, 1, @CH), 6.80 (d, 1, J = 8.4,
H-5), 6.84 (s, 1, H-3), 7.08 (d, J = 8.4, H-6), 7.44 (t, 1,
J = 7.6, H-50), 7.60 (d, 1, J = 7.6, H-40), 8.09 (d,
J = 7.6, H-60), 8.19 (s, 1, H-20). Found C, 67.56; H,
4.97. C17H15ClO3 requires: C, 67.44; H, 4.99%.
6.3.16. 4-Allyl-2-methoxyphenyl 1-cyclohexanecarboxyl-
ate (4p). White crystal. Yield: 92%; mp 40–41 ꢁC. IR:
1
1752 cmꢁ1 (C@O). H NMR (CDCl3): d 1.10–2.32 (m,
10, CH2 cyclohexyl), 2.59 (m, 1, CH cyclohexyl), 3.39
(d, 2, J = 6.5 Hz, –CH2–), 3.81 (s, 3, OCH3), 5.02–5.25
(m, 2, H2C@), 5.79–6.27 (m, 1, @CH), 6.75 (d, 1,
J = 8.4, H-5), 6.79 (s, 1, H-3), 6.95 (d, J = 8.4, H-6).
6.3.10. 4-Allyl-2-methoxyphenyl 4-chlorobenzoate (4j).
White crystal. Yield: 88%; mp 82–83 ꢁC. IR:
1743 cmꢁ1 (C@O). 1H NMR (CDCl3): d 3.40 (d, 2,

H. Sadeghian et al. / Bioorg. Med. Chem. 16 (2008) 890–901
901
20. IARC, Monograph, Evaluation of Carcinogenic Risk of
Chemicals to Humans. Lyon, France, 1985; vol. 36, pp 75–
97.
Found C, 74.48; H, 8.10. C17H22O3 requires: C, 74.42;
H, 8.08%.
21. Wagner, H.; Jurcic, K.; Deininger, R. Planta Med. 1979,
37, 9–14.
22. Feng, J.; Lipton, J. M. Neuropharmacology 1987, 26,
1775–1778.
23. Hume, W. R. J. Dent. Res. 1983, 62, 1013–1015.
24. Moleyar, V.; Narasimham, P. Int. J. Food Microbiol. 1992,
16, 337–342.
25. Villani, F. J.; King, M. S. Org. Syn. Coll. 1963, 4, 88–
89.
6.3.17. 4-Allyl-2-methoxyphenyl 1-admantanecarboxylate
(4q). White crystal. Yield: 89%; mp 89–90 ꢁC. IR:
1738 cmꢁ1 (C@O). 1H NMR (CDCl3): d 1.76 (m, 9,
CH, CH2 adamantyl), 2.07 (m, 6, CH2 adamantyl),
3.36 (d, 2, J = 6.6 Hz, –CH2–), 3.78 (s, 3, OCH3),
4.97–5.22 (m, 2, H2C@), 5.82–6.20 (m, 1, @CH), 6.75
(d, 1, J = 8.5, H-5), 6.78 (s, 1, H-3), 6.90 (d, J = 8.5,
H-6). Found C, 77.58; H, 8.03. C21H26O3 requires: C,
77.27; H, 8.03%.
26. Thompson, J. D.; Gibson, T. J.; Plewniak, F.; Jeanmou-
gin, F.; Higgins, D. G. Nucleic Acids Res. 1997, 24,
4876–4882.
Acknowledgments
27. http://us.expasy.org/.
28. ChemDrawꢂ Ultra, Chemical Structure Drawing Standard,
CambridgeSoft Corporation, 100 Cambridge Park Drive,
Cambridge, MA 02140 USA, http://www.cambrigesoft.com.
29. HyperChemꢂ Release 7, Hypercube Inc., http://www.hyper.
com/.
We express our sincere gratitude to Dr. F. Hadizadeh
for software supporting and Dr. A. Sadeghian for statis-
tical analysis studies.
30. Bakavoli, M.; Nikpour, M.; Rahimizadeh, M.; Saberi, M.
R.; Sadeghian, H. Bioorg. Med. Chem. 2007, 15, 2120–
2126.
References and notes
31. Auto Dock Tools (ADT), the Scripps Research Institute,
10550 North Torrey Pines Road, La Jolla, CA 92037-
1000, USA; (http://www.scripps.edu/pub/olson-web/doc/
autodock/); Python, M.F.S.A programming language for
software integration and development. J. Mol. Graphics
Mod. 1999, 17, 57-61.
32. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.;
Hart, W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem.
1998, 19, 1639–1662.
33. Sippl, W. J. Comput. Aided Mol. Des. 2000, 14,
559–572.
34. Dym, O.; Xenarios, I.; Ke, H.; Colicelli, J. Mol. Pharma-
col. 2002, 61, 20–25.
1. Raghavenra, H.; Diwakr, B. T.; Lokesh, B. R.; Naidu, K.
A. Prostaglandins, Leukotrienes Essential Fatty Acids
2006, 74, 23–27.
2. Brash, A. R. J. Biol. Chem. 1999, 274, 23679–23682.
3. Kuhn, H.; Thiele, B. J. FEBS Lett. 1999, 449, 7–11.
4. Larsen, J. S.; Acosta, E. P. Ann. Pharmacother. 1993, 27,
898–903.
5. Schewe, T. Biol. Chem. 2002, 383, 365–374.
6. Kelavkar, U.; Glasgow, W.; Eling, T. E. Curr. Urol. Rep.
2002, 3, 207–214.
7. Kelavkar, U. P.; Cohen, C.; Kamitani, H.; Eling, T. E.;
Badr, K. F. Carcinogenesis 2000, 21, 1777–1787.
8. Zhu, J.; Kilty, I.; Granger, H.; Gamble, E.; Qiu, Y. S.;
Hattotuwa, K.; Elston, W.; Liu, W. L.; Liva, A.; Pauwels,
R. A.; Kis, J. C.; De Rose, V.; Barnes, N.; Yeadon, M.;
Jenkinson, S.; Jeffery, P. K. Am. J. Respir. Cell Mol. Biol.
2002, 27, 666–677.
9. Zhao, L.; Funk, C. D. Trends Cardiovasc. Med. 2004, 14,
191–195.
10. Cyrus, T.; Witztum, J. L.; Rader, D. J.; Tangirala, R.;
Fazio, S.; Linton, M. F.; Funk, C. D. J. Clin. Invest. 1999,
1597–1604.
11. Harats, D.; Shaish, A.; George, J.; Mulkins, M.; Kurihara,
H.; Levkovitz, H.; Sigal, E. Arteriosler. Thromb. Vasc.
Biol. 2000, 20, 2100–2105.
12. Sultana, C.; Shen, Y.; Rattan, V.; Kalra, V. J. J. Cell.
Phys. 1996, 167, 467–487.
35. http://sunfire.vbi.vt.edu/gcg/seqweb-guides/WebLab_Viewer.
html.
36. Swiss-pdbViewer 3.6, Glaxo Wellcome Experimental
Research, http://www.expasy.org/spdbv/.
37. DRAGON 2.1, Milano Chemometrics and QSAR Research
Group, Department pf Environmental Sciences, P.za Della
Scienza, 1-20126 Milano, Italy, http://www.disat.unimib.it/
chem/.
38. Todeschini, R.; Consonni, V. Handbook of Molecular
Descriptors; Wiley-VCH, Weinheim, Germany.
39. Kumar, R.; Pavithra, S. R.; Tatu, U. J. Biosci. 2007, 32,
531–536.
40. http://nihserver.mbi.ucla.edu/savs/.
41. Malterud, K. E.; Rydland, K. M. J. Agric. Food Chem.
2000, 48, 5576–5580.
42. Malterud, K. E.; Farbrot, T. L.; Huse, A. E.; Sund, R. B.
Pharmacology 1993, 47, 77–85.
43. Skrzypczak-Jankun, E.; Bross, R.; Carroll, R. T.; Dun-
ham, W. R.; Funk, M. O., Jr. J. Am. Chem. Soc. 2001,
123, 10814–10820.
44. Knapp, M. J.; Seebeck, F. P.; Klinman, J. P. J. Am. Chem.
Soc. 2001, 123, 2931–2932.
45. Tomchick, D. R.; Phan, P.; Cymborowski, M.; Minor, W.;
Holman, T. R. Biochemistry 2001, 40, 7509–7517.
46. Kyte, J.; Doolittle, R. F. J. Mol. Biol. 1982, 157,
105–132.
13. Setty, B. N.; Werner, M. H.; annun, Y. A.; Stuart, M.
J. Blood 1992, 80, 2765–2773.
15. Chanez, P.; Bonnans, C.; Chavis, C.; Vachier, I. Am.
J. Respir. Crit. Care Med. 2002, 27, 655–658.
16. Shappell, S. B.; Boeglin, W. E.; Olson, S. J.; Kasper, S.;
Brash, A. R. Am. J. Pathol. 1999, 155, 235–245.
17. Hsi, L. C.; Wilson, L. C.; Eling, T. E. J. Biol. Chem. 2002,
277, 40549–40556.
18. Charlier, C.; Michaux, C. Eur. J. Med. Chem. 2003, 38,
645–659.
19. Markowitz, K.; Moynihan, M.; Liv, M.; Kim, S. Oral
Surg. Oral Pathol. 1992, 73, 729–737.