Potential interaction of natural dietary bioactive compounds with COX-2

Article (PDF Available)inJournal of molecular graphics & modelling 30:157-66 · July 2011with66 Reads
DOI: 10.1016/j.jmgm.2011.07.002 · Source: PubMed
Abstract
Bioactive natural products present in the diet play an important role in several biological processes, and many have been involved in the alleviation and control of inflammation-related diseases. These actions have been linked to both gene expression modulation of pro-inflammatory enzymes, such as cyclooxygenase 2 (COX-2), and to an action involving a direct inhibitory binding on this protein. In this study, several food-related compounds with known gene regulatory action on inflammation have been examined in silico as COX-2 ligands, utilizing AutoDock Vina, GOLD and Surflex-Dock (SYBYL) as docking protocols. Curcumin and all-trans retinoic acid presented the maximum absolute AutoDock Vina-derived binding affinities (9.3 kcal/mol), but genistein, apigenin, cyanidin, kaempferol, and docosahexaenoic acid, were close to this value. AutoDock Vina affinities and GOLD scores for several known COX-2 inhibitors significatively correlated with reported median inhibitory concentrations (R² = 0.462, P < 0.001 and R² = 0.238, P = 0.029, respectively), supporting the computational reliability of the predictions made by our docking simulations. Moreover, docking analysis insinuate the synergistic action of curcumin on celecoxib-induced inhibition of COX-2 may occur allosterically, as this natural compound docks to a place different from the inhibitor binding site. These results suggest that the anti-inflammatory properties of some food-derived molecules could be the result of their direct binding capabilities to COX-2, and this process can be modeled using protein-ligand docking methodologies.
Journal of Molecular Graphics and Modelling 30 (2011) 157–166
Contents lists available at ScienceDirect
Journal of Molecular Graphics and Modelling
journal homepage: www.elsevier.com/locate/JMGM
Potential interaction of natural dietary bioactive compounds with COX-2
Wilson Maldonado-Rojas, Jesus Olivero-Verbel
Environmental and Computational Chemistry Group, University of Cartagena, Cartagena, Colombia
article info
Article history:
Received 24 February 2011
Received in revised form 3 July 2011
Accepted 5 July 2011
Available online 12 July 2011
Keywords:
Inflammation
Enzyme inhibition
Docking
Biological activity
abstract
Bioactive natural products present in the diet play an important role in several biological processes,
and many have been involved in the alleviation and control of inflammation-related diseases. These
actions have been linked to both gene expression modulation of pro-inflammatory enzymes, such as
cyclooxygenase 2 (COX-2), and to an action involving a direct inhibitory binding on this protein. In
this study, several food-related compounds with known gene regulatory action on inflammation have
been examined in silico as COX-2 ligands, utilizing AutoDock Vina, GOLD and Surflex-Dock (SYBYL) as
docking protocols. Curcumin and all-trans retinoic acid presented the maximum absolute AutoDock
Vina-derived binding affinities (9.3 kcal/mol), but genistein, apigenin, cyanidin, kaempferol, and docosa-
hexaenoic acid, were close to this value. AutoDock Vina affinities and GOLD scores for several known
COX-2 inhibitors significatively correlated with reported median inhibitory concentrations (R
2
= 0.462,
P < 0.001 and R
2
= 0.238, P =0.029, respectively), supporting the computational reliability of the predic-
tions made by our docking simulations. Moreover, docking analysis insinuate the synergistic action of
curcumin on celecoxib-induced inhibition of COX-2 may occur allosterically, as this natural compound
docks to a place different from the inhibitor binding site. These results suggest that the anti-inflammatory
properties of some food-derived molecules could betheresultof their direct binding capabilities to COX-2,
and this process can be modeled using protein–ligand docking methodologies.
© 2011 Elsevier Inc. All rights reserved.
1. Introduction
Foods have small amounts of bioactive compounds that act as
extra nutritional constituents[1]. Thediversity of thesechemicals is
large and some of themost representative include flavonoids, isoth-
iocyanates, proanthocyanidins, terpenoids, carotenoids, antho-
cyanins, and omega-3 polyunsaturated fatty acids, among many
others [2]. The presence of these natural bioactive molecules in
fruits and foods has been considered relevant, not only due to their
unique organoleptic properties, but also because of their benefi-
cial effects on human health, as demonstrated in numerous studies
[3,4]. A recent review paper by Pan et al. [2], detailed how natu-
ral bioactive compounds exert their anti-inflammatory activities
by modulating gene expression of diverse inflammation-related
genes. However, it is also well known that some anti-inflammatory
molecules carry out their action by directly inhibiting inflammatory
proteins such cyclooxygenase 2 (COX-2) [5]. This enzyme catalyzes
the first step in the synthesis of prostaglandins, thromboxanes and
other eicosanoids in several inflammatory processes [6].
Corresponding author at: Environmental and Computational Chemistry Group,
Faculty of Pharmaceutical Sciences, University of Cartagena, Campus of Zaragocilla,
Cartagena, Colombia. Tel.: +57 5 6698179/6698180; fax: +57 5 6698323.
E-mail addresses: jesusolivero@yahoo.com, joliverov@unicartagena.edu.co
(J. Olivero-Verbel).
Although several natural products have been shown to mod-
ulate COX-2 expression [7–9], it is not clear if those are able to
directly interact with the gene product or its modulating tran-
scription factors. Computational chemistry offers the possibility to
explore these interactions through protein–ligand docking proce-
dures. Docking methods are valuable tools for drug development,
and most current approaches assume a rigid receptor structure
to allow virtual screening of large numbers of possible ligands
and putative binding sites on a receptor molecule [10]. Among
those tools used for this purpose are AutoDock Vina, GOLD and
Surflex-Dock (SYBYL) [11–13]. Docking strategies generate bind-
ing or affinity scores for different sites and poses on targets, and the
protein ‘hits’ identified by using this method can serve as potential
candidates for experimental validation [14,15].
In this study, docking methodologies were used to test the
ability of 29 natural bioactive compounds, isolated from different
food sources, to bind COX-2. In addition, ligands known to bind
COX-2 were submitted to docking protocols to establish relation-
ships between their biological activity and the predicted binding
affinities.
2. Materials and methods
2.1. Protein and ligand structure preparation
Experimental coordinates of three COX-2 structures
(PDB
codes: 1CX2, 1PXX and 1CVU) were obtained from
1093-3263/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jmgm.2011.07.002
158 W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166
Table 1
Examined natural products and their sources.
Compound Dietary source Reference
Apigenin Celery [20]
Tangeretin Citrus peel [21]
Silybinin Milk thistle [22]
Cyanidin Cherries [23]
Delphinidin Dark fruits [24]
Genistein Soybean [25]
Epicatechin and
epigallocatechin-3-gallate
Green tea [26]
Naringenin Citrus peel [27]
Quercetin and kaempferol Broccoli [28]
5-Hydroxy-3,6,7,8,3
,4
-
hexamethoxyflavone
Citrus peel [29]
Curcumin Turmeric powder and
curcuma
[30,31]
Resveratrol Grape skins and red
wine
[32]
[6]-Gingerol and [6]-shogaol Ginger [33]
Carnosol Rosemary [34]
Pterostilbene Blueberries [35]
Benzyl isothiocyanate and phenethyl
isothiocyanate
Cabbage [36]
Sulforaphane Cabbage [37]
Proanthocyanidins Berries [38]
All-trans retinoic acid Carrot, peppers and
broccoli
[39,40]
Menthone Mentha [41]
Lycopene and -carotene Tomato and carrot [42,43]
Lutein Spinach and eggs [44]
Eicosapentaenoic acid and
docosahexaenoic acid
Fish and fish oil [45,46]
Protein Data Bank (PDB) [16] and prepared with SYBYL 8.1.1
package [17]. Anti-inflammatory natural products chosen to
perform this study were those reported to modulate expression of
genes related to inflammation [2]. All these chemicals are present
in foods and vegetables (Table 1), and they have been proven to
have good anti-inflammatory properties. Structures were drawn
with SYBYL 8.1.1 package, exactly as presented by Pan et al. [2], and
optimized using DFT at the B3LYP/6-31G level, and calculations
were carried out with Gaussian 03 package program [18]. The
resultant geometry was translated to Mol2 format with Open Babel
[19]. To determine structural similarities between 1CX2, 1PXX and
1CVU, a molecular superposition was conducted using SYBYL 8.1.1
program.
2.2. Protein–ligand docking calculations
The feasibility of natural compounds to be ligands for COX-
2 structures was evaluated using molecular docking. This was
performed utilizing three different programs that rely on sev-
eral distinct scoring functions to evaluate the performance of the
protein–ligand docking: AutoDock Vina, Surflex-Dock (SYBYL) and
GOLD program.
AutoDock Vina combines some advantages of knowledge-based
potentials and empirical scoring functions: it extracts empiri-
cal information from both the conformational preferences of the
receptor–ligand complex and from experimental affinity measure-
ments. Ligands are ranked based on an energy scoring function and,
to speedup thescore calculation, a grid-based protein–ligand inter-
action is used [11]. The docking site for the ligands on 1CX2, 1PXX
and 1CVU was defined by establishing a cube at the geometrical
center of the native ligand present in each one of the evaluated PDB
structures, with the dimensions 24 × 24 × 24
˚
A, covering the ligand
binding site with a grid point spacing of 0.375
˚
A. The coordinates
X, Y and Z for 1CX2 from center grid boxes were 25.374, 21.657
and 17.292; for 1PXX 27.058, 24.431 and 15.437, and finally for
1CVU 25.277, 22.358 and 49.308, respectively. Ten runs were per-
formed per each ligand, and for each run the best pose was saved.
Finally, the average binding affinity for best poses was accepted as
the binding affinity value for a particular complex.
GOLD utilizes a score function called fitness to rank differ-
ent binding modes. It comprises four terms: the protein–ligand
hydrogen-bond score, the protein–ligand van der Waals score, the
contribution to the fitness due to intramolecular hydrogen bonds
in the ligand and the contribution due to intramolecular strain in
the ligand. It also has a mechanism for placing the ligand in the
binding site using fitting points; and finally, it uses a search algo-
rithm to explore possible binding modes [12]. The docking site
was defined for each structure (1CX2, 1PXX and 1CVU) using the
same coordinates X, Y and Z employed to localize the binding site
with AutoDock Vina. A radius sphere of 10
˚
A was defined around
the geometrical center of the native ligand for each evaluated
protein. For each independent algorithm run, a maximum num-
ber of 125,000 operations were performed. Operator weights for
crossover, mutation, and migration were set in mode auto, the max-
imum distance between hydrogen donors and fitting points was set
to 3.0
˚
A, and non-bonded Van der Waals energies were cut-off at
6.0
˚
A.
The Surflex-Dock module of SYBYL is a molecular docking unit
that performs flexible alignments. Its results are presented as both
docking accuracy and screening utility [13]. The docking procedure
was started with the protomol generation. The protomol was cre-
ated using a ligand-based approach (native ligand for each COX-2
structure). Proto
threshold wasset to 0.5 and proto bloat was left at
0 as a default parameter. For each protein–ligand pair, twenty top
ranked docked solutions were saved and the Surflex-Dock score
presented as the mean for these values.
These docking platforms were also used to calculate docking
scores for COX-2 inhibitors, SC558 and diclofenac, as well as for
the natural substrate arachidonic acid. These molecules were also
obtained from PDB. All protein–ligand docking calculations con-
ducted on COX-2 proteins were performed using the inhibitor
binding site on the crystal structure (PDB: 1CX2 and 1PXX) or the
substrate binding site(PDB: 1CVU). Thesebinding sites arethe same
in these COX-2 structures. In all cases, affinities were reported as
the mean value obtained for 10 docking runs performed per ligand.
2.3. Identification of residues interacting with the natural
bioactive compounds on COX-2 binding site
The identification of protein residues that interact with the
natural bioactive compounds having the greatest affinities was
carried out using LigandScout 3.0 [47]. This program creates
simplified pharmacophores to detect the number and type of pri-
mary existing ligand–residue interactions on the protein active
site.
2.4. Docking validation with biological data for COX-2 inhibitors
The 2D structures and the biological data of 21 COX-2 inhibitors
were obtained from the PubChem chemical library [48] and lit-
erature [49,50]. Docking procedures were performed with three
docking tools: AutoDock Vina, GOLD and Surflex-Dock [11–13],
following the same protocols previously described for studied nat-
ural products. The biological data consisted of median inhibitory
concentrations (IC
50
), and the details of the testing protocols and
materials are available on PubChem BioAssay [48]. The relation-
ship between AutoDock Vina-calculated affinities of inhibitors on
the three tested COX-2 (average values) and experimental activ-
ity data (Log IC
50
) was performed by linear correlation [51], using
Graph Instat Software (Version 3.06, 2003).
W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166 159
2.5. Theoretical approach to study the synergistic effect between
curcumin and celecoxib on COX-2
It has been reported that curcumin acts synergistically with
celecoxib in the inhibition of prostaglandin E2 synthesis by COX-
2 [52,53]. In order to gain insight in this process, we performed
docking simulations on the whole COX-2 (3LN1) structure with
both compounds. Aiming to evaluate if the curcumin shares the
same binding site as celecoxib, a series of 500 AutoDock Vina dock-
ing runs were performed using the following docking parameters.
The docking procedure on the 3LN1 structure was performed by
establishing a cube with the dimensions 60 × 84 × 72
˚
A covering the
whole protein (Chain A), with a grid point spacing of 1.0
˚
A, using as
center of the grid box the protein itself.
3. Results and discussion
3.1. Structural similarities of COX-2 structures
The superpositioning of the 3D COX-2 structures (PDB: 1CX2,
1CVU and 1PXX) as well as the RMSD values for each pair of them
are presented in Fig. 1. As can be seemed, these three-dimensional
structures of COX-2 have only minor differences (sequence iden-
tity >99.5 and RMSD < 0.507
˚
A).
3.2. Docking calculations using AutoDock Vina, GOLD and SYBYL
programs
The docking affinities of natural products for different COX-2,
as calculated by three distinct docking programs are presented in
Table 2. Results indicate that compared to the examined natural
products, AutoDock Vina-calculated binding affinities for SC558,
diclofenac (inhibitors) and arachidonic acid (substrate) were more
consistent in terms of the magnitude of the expected predicted
value, than the values generated for the scores calculated by GOLD
and SYBYL. In the case of GOLD, the presence of the nitrogen seems
to generate conflicting scores (negative values) for diclofenac, and
high variability for binding scores obtained for the different COX-
2 structures. SYBYL, on the other hand, also showed considerable
variability for the scores obtained for the COX-2 structures. There-
Fig. 1. 3D-Superposition of COX-2 structures (1CVU, 1PXX and 1CX2), showing
sequence identity and RMSD values. *RMSD for the binding site.
fore, successive calculations and discussions are referred solely to
results provided by AutoDock Vina.
According to the AutoDock Vina-obtained affinity values
(kcal/mol), several natural compounds are potential ligands for
COX-2, with best scores obtained for PDB: 1CVU, including
curcumin, all-trans retinoic acid (greatest docking scores, with
identical mean absolute affinity value of 9.3 kcal/mol), as well
as genistein, apigenin, cyanidin, kaempferol and docosahexaenoic
acid.
3.3. Interaction between residues in COX-2 and natural products
The complex COX-2 (PDB: 1CVU) with curcumin and all-trans
retinoic acid, as well as the interactions between residues in
the protein binding site and these ligands are shown in Fig. 2.
Both ligands fit into the same binding site (Fig. 2A). The most
important residues on the 1CVU–curcumin complex (Fig. 2B) are
Met113, Val116, Ile345, Val349, Leu359, Leu384, Trp387, Phe518,
Ala527, Val523, and Ser530. Most interactions are hydrophobic and
Fig. 2. 3D structure of COX-2(1CVU)-ligand complexes. (A) COX-2 bound to curcumin or all-trans retinoic acid (box). (B) Residues in the interaction COX-2-curcumin. (C)
Residues in the interaction COX-2-all-trans retinoic acid.
160 W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166
Table 2
Docking results for natural bioactive compound on three COX-2 structures.
Compound Protein name: cyclooxygenase-2 (COX-2)
1CX2 1PXX 1CVU
AV
a
affinity
(kcal/mol)
G fitness S total
score
AV affinity
(kcal/mol)
G fitness S total
score
AV affinity
(kcal/mol)
G fitness S total
score
Curcumin 8.4 51.41 7.60 8.7 52.18 5.44 9.3 52.62 7.03
Silibinin 7.8 25.33 4.67 3.6 44.59 0.79 7.8 37.70 3.97
Apigenin 8.4 47.99 5.24 8.6 49.48 4.96 8.9 48.68 6.08
Genistein 8.4 43.42 4.49 9.1 49.05 5.13 8.8 48.03 5.15
Naringenin 8.3 50.78 6.04 8.4 49.66 5.89 8.6 47.11 5.43
[6]-Shogahol 8.0 54.43 9.10 7.6 49.73 8.34 7.8 53.17 7.02
[6]-Gingerol 8.0 55.73 7.54 7.6 46.76 8.24 7.7 55.00 7.69
Docosahexaenoic acid 7.7 63.84 9.10 7.5 60.62 9.57 8.8 64.05 10.73
Cyanidin 7.6 46.71 5.25 8.1 49.41 3.24 8.9 51.37 5.70
Quercetin 7.6 46.78 5.84 8.1 49.55 4.57 8.8 47.97 6.23
Resveratrol 7.6 45.19 5.11 8.0 47.15 6.49 8.0 46.02 5.31
Eicosapentaenoic acid 7.4 59.86 9.37 7.5 58.10 8.98 8.5 61.63 8.78
Tangeretin 7.5 48.91 3.90 7.7 65.16 6.82 8.1 60.72 5.21
Epicatechin 7.4 43.09 6.32 8.5 47.23 5.19 8.7 46.58 5.68
Kaempferol 7.6 46.93 4.13
7.9 48.48 3.85 8.8 47.29 4.45
Delphinidin 7.1 48.64 5.79 8.1 49.77 3.84 8.4 50.75 6.04
Pterostilbene 6.9 49.10 6.94 7.9 49.03 7.15 8.2 48.09 6.79
All-trans retinoic acid 7.2 36.84 5.08 7.3 39.23 5.36 9.3 46.28 6.90
Carnosol 6.8 17.40 2.81 5.6 44.06 4.33 8.1 48.73 5.56
Menthone 6.3 29.56 3.30 6.6 30.11 4.15 6.6 29.63 4.03
Benzylisothiocyanate 5.9 39.73 2.83 6.0 40.34 3.19 6.1 39.02 2.95
Phenethylisothiocyanate 6.1 44.28 4.42 6.1 41.84 3.48 6.5 38.86 3.29
Epigallocatechin-3-gallate 7.2 52.43 4.88 6.7 54.01 3.26 8.2 59.86 6.43
-carotene 5.7 20.8 6.58 3.3 103.52 5.60 6.1 21.11 6.66
Lycopene 5.3 20.59 5.33 5.6 21.64 3.20 7.6 2.11 4.04
5-Hydroxy-3,6,7,8,3
,4-hexamethhoxyflavone 6.1 48.27 8.33 6.7 55.94 9.59 7.9 49.00 8.06
Sulforaphane 4.4 45.61 3.38 4.7 42.94 4.04 4.8 43.43 3.72
Lutein 3.9 31.69 4.99 3.6 89.77 4.56 5.6 65.32 2.77
Proanthocyanidin B2 1.8 44.04 2.70 3.8 48.54 1.26 4.5 39.79 1.15
SC558 (inhibitor) 10.7 51.75 6.03 10.0 45.21 5.59 10.1 43.44 4.55
Diclofenac (inhibitor) 8.0 122.97 4.42 8.6 119.46 5.46 8.8 117.28 2.24
Arachidonic acid (substrate) 8.0 59.70 8.64 7.5 58.18 9.58 7.8 66.83 10.81
a
Docking scoring function values calculated for each protein: AV, AutoDock Vina; G, GOLD; S, Surflex-Dock (SYBYL).
aromatic in nature, except for Ser 530, which interacts with cur-
cumin through a hydrogen bond. For the 1CVU-all-trans retinoic
acid complex (Fig. 2C), relevant aminoacids are Phe205, Phe209,
Val228, Val344, Tyr348, Val349, Leu352, Tyr385, Trp387, and
Leu534, showing only hydrophobic interactions with the ligand.
Most of these residues have also beenreported forchemicals having
strong interactions with COX-2 [54,55].
The most favorable conformation resulted from the docking of
curcumin into the active site of COX-2 is similar to that experi-
mentally found for the COX-2 substrate arachidonic acid (Fig. 3).
Accordingly, it is plausible to suggest that curcumin may be exert-
ing its action by acting as a competitive inhibitor of arachidonic
acid during prostaglandin E
2
synthesis by COX-2.
Among many natural products with known anti-inflammatory
properties, curcumin is one of the most commonly referenced
[56–59]. It is a phenolic yellow pigment present in curry pow-
der, which has been associated with beneficial effects on human
health as a result of its consumption in food [2]. It has been shown
that curcumin exhibits antioxidant, anti-inflammatory and pro-
apoptotic activities. Other food-related phenolic compounds with
anti-inflammatory properties have also been reported in grapes,
peanuts, blueberries, cranberries and red wine [60].
All-trans retinoic acid is a terpenoid derived from the
mevalonate and isopentenyl pyrophosphate pathway [61]. This
compound has been used for the treatment or alleviation of inflam-
matory diseases [62].
Other molecules that docked into COX-2 were genistein, api-
genin, cyanidin and kaempferol. These are flavonoids commonly
present in foods that have been used for the treatment of many dis-
Fig. 3. Docking conformation of curcumin and arachidonic acid (experimental) on
the active site of COX-2 (1CVU).
eases, mainly due to their anti-allergic, antiviral, anti-inflammatory
and vasodilatory properties [63–67]. Similarly, docosahexaenoic
acid has been reported to possess systemic anti-inflammatory
effects and cardiovascular protection [68].
Although values obtained by docking analysis should be con-
sidered just as a theoretical approximation, this information could
be useful to explore possible mechanisms by which these chemi-
cals behave as anti-inflammatory compounds, in particular if those
could directly bind proteins such as COX-2.
W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166 161
Table 3
Calculated affinities (AutoDock Vina), binding scores values (GOLD and SYBYL) and median inhibitory concentrations [IC
50
] for selected COX-2 inhibitors.
COX-2 inhibitor COX-2 structure
PDB code: 1CX2 PDB code: 1PXX PDB code: 1CVU Mean AV
values
Mean G
values
Mean S
values
IC
50
(M)
Log IC
50
(M)
AV affinity
(kcal/mol)
G fitness S total score AV affinity
(kcal/mol)
G fitness S total score AV affinity
(kcal/mol)
G fitness S total score
Valdecoxib (AID:
162347)
9.5 ± 0.0 65.86 ± 0.02 4.80 ± 0.00 8.5 ± 0.0 61.78 ± 0.02 4.62 ± 0.00 10.0 ± 0.0 62.32 ± 0.02 6.60 ± 0.00 9.4 ± 0.1 63.32 ± 0.34 5.34 ± 0.00 0.005 2.3
Celecoxib (AID:
270014)
10.8 ± 0.0 68.42 ± 0.02 6.63 ± 0.00 9.7 ± 0.0 65.53 ± 0.06 8.14 ± 0.00 9.8 ± 0.1 65.42 ± 0.08 6.52 ± 0.00 10.1 ± 0.1 66.46 ± 0.26 7.10 ± 0.00 0.0022 2.7
Meloxicam (AID:
162326)
7.4 ± 0.1 35.65 ± 0.62 5.96 ± 0.00 7.0 ± 0.0 38.51 ± 0.15 3.90 ± 0.00 7.6 ± 0.0 54.82 ± 0.03 4.14 ± 0.00 7.3 ±
0.1 42.99 ± 1.58 4.67 ± 0.00 0.16 0.8
Piroxicam (AID:
162326)
8.3 ± 0.0 39.06 ± 0.04 3.54 ± 0.00 8.0 ± 0.0 35.07 ± 0.24 3.80 ± 0.00 8.5 ± 0.0 50.56 ± 0.16 4.55 ± 0.00 8.3 ± 0.0 41.56 ± 1.22 3.96 ± 0.00 0.1 1.0
Diclofenac (AID:
313125)
8.0 ± 0.0 122.97 ± 0.07 4.42 ± 0.00 8.6 ± 0.0 119.24 ± 0.24 5.46 ± 0.00 8.8 ± 0.0 117.28 ± 0.11 2.24 ± 0.00 8.4 ± 0.1 119.90 ± 0.11 4.04 ± 0.00 0.02 1.7
Flosulide (AID:
162338)
8.5 ± 0.0 67.40 ± 0.08 7.35 ± 0.00 8.8 ± 0.0 63.56
± 0.14 6.47 ± 0.00 9.4 ± 0.0 62.38 ± 0.10 7.43 ± 0.00 8.9 ± 0.1 64.45 ± 0.40 7.08 ± 0.00 0.021 1.7
Tenidap (AID: 160880) 8.4 ± 0.1 56.04 ± 0.15 6.47 ± 0.00 8.3 ± 0.0 61.57 ± 0.08 4.20 ± 0.00 9.1 ± 0.0 59.80 ± 0.10 5.56 ± 0.00 8.6 ± 0.1 59.14 ± 0.43 5.41 ± 0.00 0.01 2.0
Nimesulide (AID:
162655)
7.6 ± 0.0 57.79 ± 0.10 6.77 ± 0.00 7.6 ± 0.0 56.15 ± 0.13 5.28 ± 0.00 8.2 ± 0.0 55.11 ± 0.24 4.92 ± 0.00 7.8 ± 0.1 56.35 ± 0.22 5.66 ± 0.00 0.015 1.8
Etodolac (AID: 52141) 7.2 ± 0.1 49.14 ± 0.23 7.09 ±
0.00 7.9 ± 0.0 45.72 ± 0.13 6.05 ± 0.00 8.3 ± 0.0 51.35 ± 0.08 6.24 ± 0.00 7.8 ± 0.1 48.74 ± 0.44 6.53 ± 0.00 0.025 1.6
Rofecoxib (AID:
241308)
9.8 ± 0.0 63.49 ± 0.04 6.60 ± 0.00 8.8 ± 0.0 60.60 ± 0.03 7.47 ± 0.00 9.8 ± 0.0 59.57 ± 0.19 6.37 ± 0.00 9.3 ± 0.1 61.20 ± 0.31 6.81 ± 0.00 0.032 1.5
Dup 697 (AID: 162346) 9.9 ± 0.2 73.47 ± 0.04 6.18 ± 0.00 9.5 ± 0.0 69.63 ± 0.09 5.88 ± 0.00 9.4 ± 0.0 66.28 ± 0.15 5.27 ± 0.00 9.6 ± 0.1 69.80 ± 0.55 5.78 ± 0.00 0.01 2.0
L-745337 (AID:
162346)
9.3 ± 0.0 62.69 ± 0.18 6.92 ± 0.00 10.9 ± 0.0 63.73 ± 0.03 6.19 ± 0.00 9.6 ± 0.0 64.33 ± 0.08 5.90 ± 0.00 9.9 ± 0.1 63.58 ± 0.14 6.34 ± 0.00 0.02 1.7
SC558 Filizola [49] 10.7 ± 0.2 51.75 ± 0.25 6.03 ± 0.00 10.0 ± 0.1 45.21 ± 0.36 5.59 ± 0.00 10.1 ± 0.2 43.44 ± 0.26 4.55 ± 0.00 10.3 ± 0.1 46.80 ± 0.68 5.39 ± 0.00 0.0093 2.0
NS 398 (AID: 46852) 7.6 ± 0.0 60.55 ± 0.07 6.23 ± 0.00 7.7 ± 0.0 57.99 ± 0.05 5.18 ± 0.00 8.5 ± 0.0 59.26 ± 0.06 7.39 ± 0.00 7.9 ± 0.1 59.93 ± 0.10 6.27 ± 0.00 0.19 0.7
SC-58125 (AID:
162346)
10.2 ± 0.2 69.83 ± 0.10 6.72 ± 0.00 9.6 ± 0.2 67.76 ± 0.08 6.40 ± 0.00 9.7 ± 0.0 66.09 ± 0.08 6.58 ± 0.00 9.8 ± 0.1 67.89 ± 0.29 6.57 ± 0.00 0.05 1.3
CID: 10459826 (AID:
254745)
7.0 ± 0.1 60.72 ± 0.64 6.28 ± 0.00 7.3 ± 0.0 27.84 ± 1.31 7.27 ± 0.00 10.6 ± 0.0 51.30 ± 0.41 9.24 ± 0.00 8.3 ± 0.3 46.62 ± 2.61 7.60 ± 0.00 0.05 1.3
CID: 10895294 (AID:
162484)
9.4 ± 0.0 73.59 ± 0.03 7.25 ± 0.00 7.6 ± 0.0 69.78 ± 0.08 7.55 ± 0.00 9.3 ± 0.0 68.08 ± 0.19 6.49 ± 0.00
8.8 ± 0.2 70.48 ± 0.43 7.10 ± 0.00 0.034 1.5
CID: 9885354 (AID:
162507)
10.6 ± 0.0 68.32 ± 0.03 7.49 ± 0.00 9.4 ± 0.0 65.24 ± 0.06 8.29 ± 0.00 9.3 ± 0.0 65.07 ± 0.12 6.23 ± 0.00 9.8 ± 0.1 66.21 ± 0.28 7.32 ± 0.00 0.013 1.9
2,3-
Diarylcyclobutenone
methylsulfone
Dewitt [50]
9.4 ± 0.0 66.20 ± 0.03 6.64 ± 0.00 8.7 ± 0.0 64.91 ± 0.09 5.34 ± 0.00 9.2 ± 0.0 64.09 ± 0.06 6.45 ± 0.00 9.1 ± 0.1 64.09 ± 0.17 6.14 ± 0.00 0.003 2.5
2,3-Diarylphenyl
sulfonamide Dewitt
[50]
11.0 ± 0.0 66.32 ± 0.05 6.80 ± 0.00 10.7 ± 0.0 65.70 ± 0.03 7.27 ± 0.00
11.4 ± 0.0 63.38 ± 0.09 6.93 ± 0.00 11.0 ± 0.1 65.10 ± 0.24 7.00 ± 0.00 0.002 2.7
2,3-
Diarylthiazolotriazole
methylsulfone
Dewitt [50]
8.9 ± 0.0 66.62 ± 0.03 7.23 ± 0.00 9.3 ± 0.0 72.88 ± 0.05 6.42 ± 0.00 9.5 ± 0.0 67.86 ± 0.07 4.86 ± 0.00 9.3 ± 0.0 69.12 ± 0.50 6.17 ± 0.00 0.01 2.0
162 W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166
Fig. 4. Correlation between docking theoretical data for inhibitors on COX-2 struc-
tures (1CX2, 1PXX and 1CVU) and their half maximal inhibitory concentration
(Log IC
50
). (A) AutoDock Vina, (B) GOLD and (C) Surflex-Dock. The regression
line is shown for illustrative purposes. The GOLD score value for diclofenac
(119.90 ± 0.11) was not included in the analysis.
3.4. Relationship between biological activity of COX-2 inhibitors
and protein–ligand docking data
In order to determine if affinity values calculated by AutoDock
Vina, as well as the scores calculated by GOLD and Surflex-Dock,
could be utilized as an indication of the likeliness of a compound
to behave as a COX-2 inhibitor, a group of 21 active compounds
with confirmed inhibition activity, reported in PubChem BioAssay
database [48], were docked to COX-2 (PDB: 1CX2, 1PXX and 1CVU).
The PubChem chemical structure identifier (CID), biological activ-
ity (IC
50
), AutoDock Vina affinity values, GOLD and Surflex-Dock
scores for these compounds, and the biological activity (Log IC
50
)
are shown in Table 3. The relationships between biological activity
and docking data are presented in Fig. 4. Results suggest that for all
examined docking tools,COX-2activity follows alinear relationship
only with binding affinity (AutoDock Vina) and the docking scores
from GOLD, being highly significant with the first one. Although the
magnitude of the correlation was moderate (R
2
= 0.462, P < 0.001),
this value is similar to that obtained for other docking studies [69].
Moreover, data showed that ligands with absolute affinities
greater than 10 kcal/mol have a better chance of interaction with
COX-2. For instance, celecoxib, SC558, and 2,3-diarylphenyl sul-
fonamide have absolute affinity values greater than 10 kcal/mol
and low IC
50
s. However, molecules with absolute affinities val-
ues around 9 kcal/mol have also a good probability of acting as
COX-2 inhibitors. This is reassured when biological data is revised
for our food-derived COX-2 inhibitors that presented best affinity
values. Median inhibitory concentrations (IC
50
) tested in different
cell lines for curcumin (range 2–15 M) [70–75], all-trans retinoic
acid (20.5 M) [76], genistein (range: <15–200 M) [77–79], api-
genin (range: 8.04–50 M) [77,80], cyanidin (range: 40–90 M)
[81,82], kaempferol (range: <15–50 M) [77,80], docosahexaenoic
acid (range: 9.8–30 M) [83,84], naringenin (7.9 ± 1.9 M) [85],
[6]-shogahol (2.1 M) [86], resveratrol (range: 3.06 M) [87],
eicosapentaenoic acid (7.1 M) [83] are supporting evidence that
these compounds can modulate COX-2 activity not only at mRNA
but also at the protein level.
3.5. Docking curcumin and celecoxib on COX-2
It is known that some of the chemicals studied here can modu-
late COX-2 activity not only by competitive inhibition, but also by
allosteric binding [52,53]. It has been shown that curcumin pro-
duces a synergistic effect with celecoxib, a highly selective COX-2
inhibitor, almost abolishing all enzyme activity [52,53]. In order to
determine if this additive process occurs due to curcumin (both the
keto and the enol forms) binding on a site different from that used
by celecoxib, a series of 500 AutoDock Vina docking runs were per-
formed on the protein isolated from the complex COX-2-celecoxib
(PDB: 3LN1), and the results are presented in Fig. 5. Celecoxib
docks onto COX-2 (PDB: 3LN1) on two different sites (Fig. 5A). As
expected, the most favorable was the active site of COX-2 (binding
frequency, bf, 96.8%) (Fig. 5B), as found in the crystal structure of
the celecoxib:COX-2 complex (PDB: 3LN1). An additional site (bf,
3.2%) was detected by the docking simulations, but it is less ener-
getically favorable (8.8kcal/mol vs. 11.2 kcal/mol). On the other
hand, in addition to the active site (celecoxib site), curcumin in the
keto form prefers two additional (allosteric) sites on COX-2 (Fig. 5C)
with binding frequencies of 38.6% and 38.2% (Fig. 5D). A different
trend is observed for the enol form of curcumin. This form does
not dock on the active site at all when the whole protein (3LN1) is
used as docking surface (Fig. 5E), and it docks mainly to site 2 (bf,
94.12%), and in a minor grade to site 3 (bf, 5.88%) (Fig. 5F); how-
ever these interactions are less favorable than those detected for
the keto form.
It is important to keep in mind that results from docking the
keto and enol forms of curcumin on the whole protein surface
(3LN1) are different from those acquired when the enol form is
docked directly into the activesite of 1CX2, 1PXX and1CVU. Inthese
last cases, the absolute binding affinities were greater by approx-
imately 1–2 kcal/mol. The docking of the keto form of curcumin
onto the active site of COX-2 generates not only different affinity
values depending on the site, but also distinct spatial orientations.
These last changes could require additional docking energy and this
could be a reason explaining why this curcumin form prefers the
other binding sites, where the docking implies less inner molecular
consumption.
Docking runs (n = 100) for celecoxib and the two curcumin
forms, performed using the three docking tools examined in this
work, on the three binding sites predicted for curcumin (keto form,
PubChem) using AutoDock Vina are shown in Table 4. Results
showed that AutoDock Vina, GOLD and Surflex-Dock predicted that
celecoxib prefers only the known inhibitor binding site (Site 1). In
the case of curcumin, all three docking tools suggested that both
forms of this natural product can at some point interact with any
of the three binding sites. However, there are minor changes in
the preferences based on the curcumin form and the docking tool
used. Taken together, these results suggest that independent from
the tautomeric state of curcumin, it has the ability to interact with
COX-2 on a binding site different from celecoxib.
This
in silico evaluation of curcumin binding on COX-2 offers a
plausible explanation for the synergism observed for celecoxib and
W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166 163
Table 4
Binding affinity (AutoDock Vina) and binding score values (GOLD and Surflex-Dock) for curcumin (keto and enol forms) and celecoxib (inhibitor) on different predicted
binding sites (1, 2, and 3) on COX-2.
Compound Site 1 Site 2 Site 3
AV (kcal/mol) G fitness S total score AV (kcal/mol) G fitness S total score AV (kcal/mol) G fitness S total score
Celecoxib 11.9 ± 0.1 68.54 ± 0.02 9.50 ± 0.00 9.0 ± 0.0 61.04 ± 0.13 6.05 ± 0.00 7.2 ± 0.0 60.87 ± 0.14 4.53 ± 0.00
Curcumin (keto) 8.4 ± 0.0 56.51 ± 0.26 6.71 ± 0.00 8.8 ± 0.0 52.58 ± 0.12 7.24 ± 0.00 8.0 ± 0.0 52.64 ± 0.14 8.75 ± 0.00
Curcumin (enol) 8.6 ± 0.2 48.39 ± 0.29 7.40 ± 0.00 8.7 ± 0.0 51.97 ± 0.12 9.58 ± 0.00 8.3 ± 0.0 51.52 ± 0.12 7.40 ± 0.00
AV, AutoDock Vina; G, GOLD; S, Surflex-Dock (SYBYL).
Fig. 5. Celecoxib (A) and curcumin (keto, C; enol E) binding sites on COX-2, and ligand binding site preferences for each one of them (B, D, and F, respectively). *The affinity
values (mean ± standard deviation, n = 500) in kcal/mol obtained for each protein–ligand complex are shown in below site.
164 W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166
curcumin to inhibit the action of the enzyme. It also showed that
the size of the used docking grid can have profound differences in
the results. However, it was clear that for both keto and enol forms,
a binding site different from the activesite ispreferred by curcumin,
although this process is less energetically favorable.
Although the mechanisms involved in the anti-inflammatory
action of chemicals present in edible plants may comprise dis-
tinct pathways, some of the compounds examined here are
known for their actions on the regulation of transcription factors
such as nuclear factor-kappa B (NF␬␤) [88,89], signal transduc-
ers and activation of transcription-1 (STAT-1) [90], peroxisome
proliferator-activated receptor gamma(PPAR)[91], NF-E2-related
factor-2 (Nrf2) [92], and also in the inhibition of mitogen-activated
protein kinase (MAPK) (ERK, JNK, and p38) phosphorylation [93],
among many other targets. These mechanisms may indeed alter the
expression of COX-2. However, as shown here, it may be equally
important to consider their direct action at the protein level, in
order to have a better knowledge of their pharmacological benefits.
In addition, it is clear that computational chemistry is a powerful
tool that speeds up and lowers the cost of those approaches leading
to find therapeutic agents to promote human health.
4. Conclusion
In silico docking calculations performed with AutoDock Vina
showed that binding affinities obtained for some natural com-
pounds on COX-2, such as curcumin and all-trans retinoic acid, are
of similar magnitude than those generated for known inhibitors
of this protein. Affinities from AutoDock Vina and scores given by
the docking software GOLD showed significant correlations with
experimental data for COX-2 inhibition. Docking studies performed
with curcumin and celecoxib, this last a synthetic inhibitor of
COX-2, suggest that curcumin may be able to bind this protein
both competitively and allosterically. Therefore, natural products
present in the diet are important not only as transcriptional regu-
lators of COX-2, but also they may modulate its enzyme activity to
control inflammatory processes.
Acknowledgements
The authors wish to thank Colciencias, Bogotá (Colombia), and
the University of Cartagena, Cartagena (Colombia) for their finan-
cial support (Grant 110745921616, 2009); as well as the program
to support research groups, sponsored by the Vice-Rectory for
research of the University of Cartagena (2009–2011).
References
[1] D.D. Kitts, Bioactive substances in food: identification and potential uses , Can.
J. Physiol. Pharmacol. 72 (1994) 423–424.
[2] M.H. Pan, C.S. Lai, S. Dushenkov, C.T. Ho, Modulation of inflammatory genes
by natural dietary bioactive compounds , J. Agric. Food Chem. 57 (2009)
4467–5447.
[3] L. Bohlin, U. Göransson, C. Alsmark, C. Wedén, A. Backlund, Natural products in
modern life science , Phytochem. Rev. 9 (2010) 279–301.
[4] C.L. Shen, J.K. Yeh, J. Cao, J.S. Wang, Green tea and bone metabolism , Nutr. Res.
29 (2009) 437–456.
[5] A.K. Chakraborti, S.K. Garg, R. Kumar, H.F. Motiwala, P.S. Jadhavar, Progress in
COX-2 inhibitors: a journey so far , Curr. Med. Chem. 17 (2010) 1563–1593.
[6] J.R. Kiefer, J.L. Pawlitz, K.T. Moreland, R.A. Stegeman, W.F. Hood, J.K. Gierse,
A.M. Stevens, D.C. Goodwin, S.W. Rowlinson, L.J. Marnett, W.C. Stallings, R.G.
Kurumbail, Structural insights into the stereochemistry of the cyclooxygenase
reaction , Nature 405 (2000) 97–101.
[7] R. Wilken, M.S. Veena, M.B. Wang, E.S. Srivatsan, Curcumin: a review of anti-
cancer properties and therapeutic activity in head & neck squamous cell
carcinoma , Mol. Cancer 10 (2011) 12.
[8] G. Frasca, A.M. Panico, F. Bonina, R. Messina, L. Rizza, G. Musumeci, P.
Rapisarda, V. Cardile, Involvement of inducible nitric oxide synthase and
cyclooxygenase-2 in the anti-inflammatory effects of a red orange extract in
human chondrocytes , Nat. Prod. Res. 24 (2010) 1469–1480.
[9] M. Jin, S.J. Suh, J.H. Yang, Y. Lu, S.J. Kim, S. Kwon, T.H. Jo, J.W. Kim, Y.I. Park,
G.W. Ahn, C.K. Lee, C.H. Kim, J.K. Son, K.H. Son, H.W. Chang, Anti-inflammatory
activity of bark of Dioscorea batatas DECNE through the inhibition of iNOS and
COX-2 expressions in RAW264.7 cells via NF-B and ERK1/2 inactivation , Food
Chem. Toxicol. 48 (2010) 3073–3079.
[10] A. May, M. Zacharias, Accounting for global protein deformability during
protein–protein and protein–ligand docking , BBA Proteins Proteomics 1754
(2005) 225–231.
[11] O. Trott, A.J. Olson, AutoDock Vina: improving the speed and accuracy of dock-
ing with a new scoring function, efficient optimization and multithreading , J.
Comput. Chem. 31 (2010) 455–461.
[12] M.L. Verdonk, J.C. Cole, M.J. Hartshorn, C.W. Murray, R.D. Taylor, Improved
protein–ligand docking using GOLD , Proteins: Struct. Funct. Gen. 52 (2003)
609–623.
[13] A.N. Jain, Surflex-Dock 2.1: robust performance from ligand energetic model-
ing, ring flexibility, and knowledge-based search , J. Comput. Aided Mol. Des.
21 (2007) 281–306.
[14] Y. Tang, W. Zhu, K. Chen, H. Jiang, New technologies in computer-aided drug
design: toward target identification and new chemical entity discovery , Drug
Discov. Today Tech. 3 (2006) 307–313.
[15] M. Utsintong, T.T. Talley, P.W. Taylor, A.J. Olson,O. Vajragupta, Virtual screening
against alpha-cobratoxin , J. Biomol. Screen. 14 (2009) 1109–1118.
[16] RCSB PDB Protein Data Bank | Home, http://www.pdb.org/pdb/home/home.do.
[17] SYBYL Molecular Modeling Software, Version 8.1 , Tripos, St. Louis, MO, USA,
2007.
[18] M.J. Frisch, G.W. Trucks, H.R. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheese-
man Jr., J.A. Montgomery, T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P.
Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Strat-
mann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K.
Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski,S.Dapprich,
A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari,
J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Ste-
fanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith,
M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. John-
son, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian03 , Gaussian, Inc.,
Pittsburgh, PA, USA, 2003.
[19] R. Guha, M.T. Howard, G.R. Hutchison, P. Murray-Rust, H. Rzepa, C. Steinbeck,
J.K. Wegner, E. Willighagen, The blue obelisk—interoperability in chemical
informatics , J. Chem. Inf. Model. 46 (2006) 991–998.
[20] Y. Yao, W. Sang, M. Zhou, G. Ren, Phenolic composition and antioxidant activi-
ties of 11 celery cultivars , J. Food Sci. 75 (2010) 9–13.
[21] W. Stuetz, T. Prapamontol, S. Hongsibsong, H.K. Biesalski, Polymethoxylated
flavones, flavanone glycosides, carotenoids, and antioxidants in different cul-
tivation types of tangerines (Citrus reticulata Blanco cv. Sainampueng) from
Northern Thailand , J. Agric. Food Chem. 58 (2010) 6069–6074.
[22] L. Abenavoli, R. Capasso, N. Milic, F. Capasso, Milk thistle in liver diseases: past,
present, future , Phytother. Res. 24 (2010) 1423–1432.
[23] A. Sari
´
c, S. Sobocanec, T. Balog,B.Kusi
´
c, V. Sverko, V. Dragovi
´
c-Uzelac, B. Levaj, Z.
Cosi
´
c, Z.M. Safranko, T. Marotti, Improved antioxidant and anti-inflammatory
potential in mice consuming sour cherry juice (Prunus cerasus cv. Maraska) ,
Plant Foods Hum. Nutr. 64 (2009) 231–237.
[24] S.D. Castellarin, G. Di Gaspero, Transcriptional control of anthocyanin biosyn-
thetic genes in extreme phenotypes for berry pigmentation of naturally
occurring grapevines , BMC Plant Biol. 30 (2007) 7–46.
[25] G. Basini, S. Bussolati, S.E. Santini, F. Grasselli, The impact of the phyto-
oestrogen genistein on swine granulosa cell function , J. Anim. Physiol. Anim.
Nutr. (Berl.) 94 (2010) 374–382.
[26] P. Wang, W.J. Aronson, M. Huang, Y. Zhang, R.P. Lee, D. Heber, S.M. Henning,
Green tea polyphenols and metabolites in prostatectomy tissue: implications
for cancer prevention , Cancer Prev. Res. 3 (2010) 985–993.
[27] L.T. Lien, H.S. Yeh, W.T. Su, Effect of adding extracted hesperetin, naringenin
and pectin on egg cholesterol, serum traits and antioxidant activity in laying
hens , Arch. Anim. Nutr. 62 (2008) 33–43.
[28] H.R. Vasanthi, S. Mukherjee, D.K. Das, Potential health benefits of broccoli—a
chemico-biological overview , Mini Rev. Med. Chem. 9 (2009) 749–759.
[29] C.S. Lai, S. Li, C.Y. Chai, C.Y. Lo, C.T. Ho, Y.J. Wang, M.H. Pan, Inhibitory
effect of citrus 5-hydroxy-3,6,7,8,3
,4
-hexamethoxyflavone on 12-O-
tetradecanoylphorbol 13-acetate-induced skin inflammation and tumor
promotion in mice , Carcinogenesis 28 (2007) 2581–2588.
[30] S. Kaura, N.H. Modib, D. Pandac, N. Roya, Probing the binding site of curcumin in
Escherichia coli and Bacillus subtilis FtsZ—a structural insight to unveil antibac-
terial activity of curcumin , Eur. J. Med. Chem. 45 (2010) 4209–4214.
[31] A. Rasyid, A.R. Rahman, K. Jaalam, A. Lelo, Effect of different curcumin dosages
on human gall bladder , Asia Pac. J. Clin. Nutr. 11 (2002) 314–318.
[32] H. Kalantari, D.K. Das, Physiological effects of resveratrol , Biofactors 36 (2010)
401–406.
[33] C.J. Weng, C.F. Wu, H.W. Huang, C.T. Ho, G.C. Yen, Anti-invasion effects of
6-shogaol and 6-gingerol, two active components in ginger, on human hep-
atocarcinoma cells , Mol. Nutr. Food Res. 54 (2010) 1618–1627.
[34] W.A. Bernardes, R. Lucarini, M.G. Tozatti, M.G. Souza, M.L. Andrade Silva, A.A.
da Silva Filho, C.H. Martins, A.E. Miller Crotti, P.M. Pauletti, M. Groppo, W.R.
Cunha, Antimicrobial activity of Rosmarinus officinalis against oral pathogens:
relevance of carnosica acid and carnosol , Chem. Biodivers. 7 (2010) 1835–1840.
W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166 165
[35] J.A. Alosi, D.E. McDonald, J.S. Schneider, A.R. Privette, D.W. McFadden, Pteros-
tilbene inhibits breast cancer in vitro through mitochondrial depolarization
and induction of caspase-dependent apoptosis , J. Surg. Res. 161 (2010) 195–
201.
[36] J.M. Visanji, S.J. Duthie, L. Pirie, D.G. Thompson, P.J. Padfield, Dietary isothio-
cyanates inhibit Caco-2 cell proliferation and induce G2/M phase cell cycle
arrest, DNA damage, and G2/M checkpoint activation , J. Nutr. 134 (2004)
3121–3126.
[37] K. Skupinska, I. Misiewicz-Krzeminska, K. Lubelska, T. Kasprzycka-Guttman,
The effect of isothiocyanates on CYP1A1 and CYP1A2 activities induced by
polycyclic aromatic hydrocarbons in Mcf7 cells , Toxicol. In Vitro 23 (2009)
763–771.
[38] B.A. Déziel, K. Patel, C. Neto, K. Gottschall-Pass, R.A. Hurta, Proantho-
cyanidins from the American cranberry (Vaccinium macrocarpon) inhibit
matrix metalloproteinase-2 and matrix metalloproteinase-9 activity in human
prostate cancer cells via alterations in multiple cellular signaling pathways , J.
Cell. Biochem. 111 (2010) 742–754.
[39] M. Theodosiou, V. Laudet, M. Schubert, From carrot to clinic: an overview
of the retinoic acid signaling pathway , Cell. Mol. Life Sci. 67 (2010) 1423–
1445.
[40] G.L. Ambrosini, N.H. de Klerk, L. Fritschi, D. Mackerras, B. Musk, Fruit, vegetable,
vitamin A intakes, and prostate cancer risk , Prostate Cancer Prostatic Dis. 11
(2008) 61–66.
[41] S.A. Muhammad, S. Muhammad, A. Waqar, P. Masood, Y. Raghav, Chlorinated
monoterpene ketone, acylated -sitosterol glycosides and a flavanone gly-
coside from Mentha longifolia (Lamiaceae) , Phytochemistry 59 (2002) 889–
895.
[42] J.T. Vogel, D.M. Tieman, C.A. Sims, A.Z. Odabasi, D.G. Clark, H.J. Klee, Carotenoid
content impacts flavor acceptability in tomato (Solanum lycopersicum) , J. Sci.
Food Agric. 90 (2010) 2233–2340.
[43] P.A. Thürmann, J. Steffen, C. Zwernemann, C.P. Aebischer, W. Cohn, G. Wendt,
W. Schalch, Plasma concentration response to drinks containing beta-carotene
as carrot juice or formulated as a water dispersible powder , Eur. J. Nutr. 41
(2002) 228–235.
[44] B.L. Burns-Whitmore, E.H. Haddad, J. Sabaté, K. Jaceldo-Siegl, J. Tanzman, S.
Rajaram, Effect of n-3 fatty acid enriched eggs and organic eggs on serum
lutein in free-living lacto-ovo vegetarians , Eur. J. Clin Nutr. 64 (2010) 1332–
1337.
[45] J.D. Mills, J.E. Bailes, C.L. Sedney, H. Hutchins, B. Sears, Omega-3 fatty acid sup-
plementation and reduction of traumatic axonal injury in a rodent head injury
model , J. Neurosurg. 114 (2010) 77–84.
[46] N.J. Mann, S.L. O’Connell, K.M. Baldwin, I. Singh, B.J. Meyer, Effects of seal oil and
tuna-fish oil on platelet parameters and plasma lipid levels in healthy subjects
, Lipids 45 (2010) 669–681.
[47] G. Wolber, T. Langer, LigandScout: 3-D pharmacophores derived from protein-
bound ligands and their use as virtual screening filters , J. Chem. Inf. Model. 45
(2005) 160–169.
[48] PubChem, http://pubchem.ncbi.nlm.nih.gov.
[49] M. Filizola, J.J. Perez, A. Palomer, D. Mauleón, Comparative molecular modeling
study of the three-dimensional structures of prostaglandin endoperoxide H2
synthase 1 and 2 (COX-1 and COX-2) , J. Mol. Graph. Model. 15 (1997) 290–
300.
[50] D.L. Dewitt, Cox-2-selective inhibitors: the new super aspirins , Mol. Pharm. 55
(1999) 625–631.
[51] Z. Zhou, Y. Wang, S.H. Bryant, Computational analysis of the cathepsin B
inhibitors activities through LR-MMPBSA binding affinity calculation based on
docked complex , J. Comput. Chem. 30 (2009) 2165–2175.
[52] S. Lev-Ari, L. Strier, D. Kazanov, L. Madar-Shapiro, H. Dvory-Sobol, I. Pinchuk, B.
Marian, D. Lichtenberg, N. Arber, Celecoxib and curcumin synergistically inhibit
the growth of colorectal cancer cells , Clin. Cancer Res. 11 (2005) 6738–6744.
[53] S. Lev-Ari, L. Strier, D. Kazanov, O. Elkayam, D. Lichtenberg, D. Caspi, N. Arber,
Curcumin synergistically potentiates the growth-inhibitory and pro-apoptotic
effects of celecoxib in osteoarthritis synovial adherent cells , Rheumatology 45
(2006) 171–177.
[54] C. Selvam, S.M. Jachak, R. Thilagavathi, A.K. Chakraborti, Design, synthesis, bio-
logical evaluation and moleculardockingofcurcumin analogues as antioxidant,
cyclooxygenase inhibitory and anti-inflammatory agents , Bioorg. Med. Chem.
Lett. 15 (2005) 1793–1797.
[55] K.E. Furse, D.A. Pratt, N.A. Porter, T.P. Lybrand, Molecular dynamics simulations
of arachidonic acid complexes with COX-1 and COX-2: insights into equilibrium
behavior , Biochemistry 45 (2006) 3189–3205.
[56] S.J. Wu, K.W. Tam, Y.H. Tsai, C.C. Chang, J.C. Chao, Curcumin and saikosaponin
a inhibit chemical-induced liver inflammation and fibrosis in rats , Am. J. Chin.
Med. 38 (2010) 99–111.
[57] E. Sikora, G. Scapagnini, M. Barbagallo, Curcumin, inflammation, ageing and
age-related diseases , Immun. Ageing 7 (2010) 1.
[58] S.J. Moghaddam, P. Barta, S.G. Mirabolfathinejad, Z. Ammar-Aouchiche, N.T.
Garza, T.T. Vo, R.A. Newman, B.B. Aggarwal, C.M. Evans, M.J. Tuvim, R. Lotan,
B.F. Dickey, Curcumin inhibits COPD-like airway inflammation and lung cancer
progression in mice , Carcinogenesis 30 (2009) 1949–1956.
[59] J.S. Jurenka, Anti-inflammatory properties of curcumin, a major constituent of
Curcuma longa: a review of preclinical and clinical research , Altern. Med. Rev.
14 (2009) 141–153.
[60] G. Galati, O. Sabzevari, J.X. Wilson, P.J. O’Brien, Prooxidant activity and cellular
effects of the phenoxyl radicals of dietary flavonoids and other polyphenolics ,
Toxicology 177 (2002) 91–104.
[61] T. Akihisa, K. Yasukawa, M. Yamaura, M. Ukiya, Y. Kimura, N. Shimizu, A.
Koichi, Triterpene alcohol and sterol ferulates from rice bran and their anti-
inflammatory effects , J. Agric. Food Chem. 48 (2000) 2313–2319.
[62] K. Leelawat, K. Ohuchida,K. Mizumoto, C. Mahidol, M. Tanaka, All-transretinoic
acid inhibits the cell proliferation but enhances the cell invasion through up-
regulation of c-met in pancreatic cancer cells , Cancer Lett. 224 (2005) 303–
310.
[63] R.R. Li, L.L. Pang, Q. Du, Y. Shi, W.J. Dai, K.S. Yin, Apigenin inhibits allergen-
induced airway inflammation and switches immune response in a murine
model of asthma , Immunopharmacol. Immunotoxicol. 32 (2010) 364–370.
[64] E.M. Vela, K.A. Knostman, J.M. Mott, R.L. Warren, J.N. Garver, L.J. Vela, R.L.
Stammen, Genistein, a general kinase inhibitor, as a potential antiviral for
arenaviral hemorrhagic fever as described in the Pirital virus-Syrian golden
hamster model , Antiviral Res. 87 (2010) 318–328.
[65] B.A. Watkins, K. Hannon, M. Ferruzzi, Y. Li, Dietary PUFA and flavonoids as
deterrents for environmental pollutants , J. Nutr. Biochem. 18 (2007) 196–
205.
[66] V. Calderone, S. Chericoni, C. Martinelli, L. Testai, A. Nardi, I. Morelli, M.C.
Breschi, E. Martinotti, Vasorelaxing effects of flavonoids: investigation on the
possible involvement of potassium channels , Naunyn Schmiedebergs Arch.
Pharmacol. 370 (2004) 290–298.
[67] K.M. Lee, K.W. Lee, S.K. Jung, E.J. Lee, Y.S. Heo, A.M. Bode, R.A. Lubet, H.J. Lee, Z.
Dong, Kaempferol inhibits UVB-induced COX-2 expression by suppressing Src
kinase activity , Biochem. Pharmacol. 80 (2010) 2042–2049.
[68] M. Massaro, A. Habib, L. Lubrano, S. Del Turco, G. Lazzerini, T. Bourcier,
B.B. Weksler, R. De Caterina, The omega-3 fatty acid docosahexaenoate
attenuates endothelial cyclooxygenase-2 induction through both NADP[H] oxi-
dase and PKC inhibition , Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 15184–
15189.
[69] A.A. Hare, L. Leng, S. Gandavadi, X. Du, Z. Cournia, R. Bucala, W.L. Jor-
gensen, Optimization of N-benzyl-benzoxazol-2-ones as receptor antagonists
of macrophage migration inhibitory factor (MIF) , Bioorg. Med. Chem. Lett. 20
(2010) 5811–5814.
[70] A.M. Gonzales, R.A. Orlando, Curcumin and resveratrol inhibit nuclear factor-
kappaB-mediated cytokine expression in adipocytes , Nutr. Metab. 5 (2008)
17.
[71] S. Lev-Ari, Y. Maimon, L. Strier, D. Kazanov, N. Arber, Down-regulation of
prostaglandin E2 by curcumin is correlated with inhibition of cell growth and
induction of apoptosis in human colon carcinomacell lines , J. Soc.Integr. Oncol.
4 (2006) 21–26.
[72] B. Du, L. Jiang, Q. Xia, L. Zhong, Synergistic Inhibitory effects of curcumin and
5-fluorouracil on the growth of the human colon cancer cell Line HT-29 ,
Chemotherapy 52 (2006) 23–28.
[73] S. Lev-Ari, A. Starr, A. Vexler, V. Karaush, V. Loew, J. Greif, E. Fenig, D. Aderka,
R. Ben-Yosef, Inhibition of pancreatic and lung adenocarcinoma cell survival
by curcumin is associated with increased apoptosis, down-regulation of COX-
2 and EGFR and inhibition of Erk1/2 activity , Anticancer Res. 26 (2006)
4423–4430.
[74] S. Gafner, S.K. Lee, M. Cuendet, S. Barthélémy, L. Vergnes, S. Labidalle, R.G.
Mehta, C.W. Boone, J.M. Pezzuto, Biologic evaluation of curcumin and struc-
tural derivatives in cancer chemoprevention model systems , Phytochemistry
65 (2004) 2849–2859.
[75] J. Hong, M. Bose, J. Ju, J.H. Ryu, X. Chen, S. Sang, M.J. Lee, C.S. Yang, Modulation of
arachidonic acid metabolism by curcumin and related -diketone derivatives:
effects on cytosolic phospholipase A2, cyclooxygenases and 5-lipoxygenase ,
Carcinogenesis 25 (2004) 1671–1679.
[76] B.H. Kim, K.S. Kang, Y.S. Lee, Effect of retinoids on LPS-induced COX-2 expres-
sion and COX-2 associated PGE2 release from mouse peritoneal macrophages
and TNF- release from rat peripheral blood mononuclear cells , Toxicol. Lett.
150 (2004) 191–201.
[77] Y.C. Liang, Y.T. Huang, S.H. Tsai, S.Y. Lin-Shiau, C.F. Chen, J.K. Lin, Suppression
of inducible cyclooxygenase and inducible nitric oxide synthase by apigenin
and related flavonoids in mouse macrophages , Carcinogenesis 20 (1999)
1945–1952.
[78] V.P. Dia, M.A. Berhow, E. Gonzalez de mejia, Bowman-birk inhibitor and
genistein among soy compounds that synergistically inhibit nitric oxide and
prostaglandin E2 pathways in lipopolysaccharide-induced macrophages , J.
Agric. Food Chem. 56 (2008) 11707–11717.
[79] F. Ye, J. Wu, T. Dunn, J. Yi, X. Tong, D. Zhang, Inhibition of cyclooxygenase-2
activity in head and neck cancer cells by genistein , Cancer Lett. 211 (2004)
39–46.
[80] Y.C. Liang, S.H. Tsai, D.C. Tsai, S.Y. Lin-Shiau, J.K. Lin, Suppression of inducible
cyclooxygenase and nitric oxide synthase through activation of peroxisome
proliferator-activated receptor-Q by flavonoids in mouse macrophages , FEBS
Lett. 496 (2001) 12–18.
[81] H. Wang, M.G. Nair, G.M. Strasburg, Y.C. Chang, A.M. Booren, J.I. Gray, D.L.
DeWitt, Antioxidant and antiinflammatory activities of anthocyanins and their
aglycon, cyanidin, from tart cherries , J. Nat. Prod. 62 (1999) 294–296.
[82] N.P. Seeram, Y. Zhang, M.G. Nair, Inhibition of proliferation of human can-
cer cells and cyclooxygenase enzymes by anthocyanidins and catechins , Nutr.
Cancer 46 (2003) 101–106.
[83] T. Ringbom, U. Huss, Å. Stenholm, S. Flock, L. Skattebøl, P. Perera, L. Bohlin,
COX-2 inhibitory effects of naturally occurring and modified fatty acids , J. Nat.
Prod. 64 (2001) 745–749.
[84] G. Calviello, F. Di Nicuolo, S. Gragnoli, E. Piccioni, S. Serini, N. Maggiano,
G. Tringali, P. Navarra, F.O. Ranelletti, P. Palozza, n-3 PUFAs reduce VEGF
166 W. Maldonado-Rojas, J. Olivero-Verbel / Journal of Molecular Graphics and Modelling 30 (2011) 157–166
expression in human colon cancer cells modulating the COX-2/PGE2 induced
ERK-1 and -2 and HIF-1a induction pathway , Carcinogenesis 25 (2004)
2303–2310.
[85] Y. Takano-Ishikawaa, M. Gotoa, K. Yamakia, Structure–activity relations
of inhibitory effects of various flavonoids on lipopolysaccharide-induced
prostaglandin E2 production in rat peritoneal macrophages: comparison
between subclasses of flavonoids , Phytomedicine 13 (2006) 310–317.
[86] E. Tjendraputra, V.H. Tran, D. Biu-Brennan, B.D. Roufogalis, C.C. Duke, Effect of
ginger constituents and synthetic analogues on cyclooxygenase-2 enzyme in
intact cells , Bioorg. Chem. 29 (2001) 156–163.
[87] H. Cao, R. Yu, Y. Tao, D. Nikolic, R.B. van Breemen, Measurement of cyclooxy-
genase inhibition using liquid chromatography–tandem mass spectrometry , J.
Pharm. Biomed. Anal. 54 (2011) 230–235.
[88] Y.J. Surh, K.S. Chun, H.H. Cha, S.S. Han, Y.S. Keum, K.K. Park, S.S. Lee,
Molecular mechanisms underlying chemopreventive activities of anti-
inflammatory phytochemicals: down-regulation of COX-2 and iNOS through
suppression of NF-␬␤ activation , Mutat. Res. 480–481 (2001) 243–
268.
[89] L. Kole, B. Giri, S.K. Manna, B. Pal, S. Ghosh, Biochanin-A, an isoflavon, showed
anti-proliferative and anti-inflammatory activities through the inhibition of
iNOS expression, p38-MAPK and ATF-2 phosphorylation and blocking NF-␬␤
nuclear translocation , Eur. J. Pharmacol. 653 (2011) 8–15.
[90] E.Y. Chung, B.H. Kim, J.T. Hong, C.K. Lee, B. Ahn, S.Y. Nam, S.B. Han, Y.
Kim, Resveratrol down-regulates interferon--inducible inflammatory genes
in macrophages: molecular mechanism via decreased STAT-1 activation , J.
Nutr. Biochem. (2010), doi:10.1016/j.jnutbio.2010.07.012.
[91] H.M. Wang, Y.X. Zhao, S. Zhang, G.D. Liu, W.Y. Kang, H.D. Tang, J.Q. Ding,
S.D. Chen, PPARgamma agonist curcumin reduces theamyloid-beta-stimulated
inflammatory responses in primary astrocytes , J. Alzheimers Dis. 20 (2010)
1189–1199.
[92] R. Garg, S. Gupta, G.B. Maru, Dietary curcumin modulates transcriptional reg-
ulators of phase I and phase II enzymes in benzo[a]pyrene-treated mice:
mechanism of its anti-initiating action , Carcinogenesis 29 (2008) 1022–1032.
[93] J. Ahn, H. Lee, S. Kim, T. Ha, Curcumin-induced suppression of adipogenic dif-
ferentiation is accompanied by activation of Wnt/beta-catenin signaling , Am.
J. Physiol. Cell Physiol. 298 (2010) 1510–1516.
    • "The relationship between the biological activity (logIC50) and the binding scores for these inhibitors on DNMT1 are shown inFig. 4. The correlation analysis indicated the inhibition of DNMT1 activity follows a highly dependence with calculated binding scores for these compounds (r = 0.83, p value < 0.0001; for binding affinity vs logIC50, r = 0.65, p value < 0.0001; for Total score vs logIC50), with correlation coefficients comparable to those reported in other studies for these validations [56,59,60,73] as shown inFig. 4a and b. "
    [Show abstract] [Hide abstract] ABSTRACT: DNA methyltransferase inhibitors (DNMTis) have become an alternative for cancer therapies. However, only two DNMTis have been approved as anticancer drugs, although with some restrictions. Natural products (NPs) are a promising source of drugs. In order to find NPs with novel chemotypes as DNMTis, 47 compounds with known activity against these enzymes were used to build a LDA-based QSAR model for active/inactive molecules (93% accuracy) based on molecular descriptors. This classifier was employed to identify potential DNMTis on 800 NPs from NatProd Collection. 447 selected compounds were docked on two human DNA methyltransferase (DNMT) structures (PDB codes: 3SWR and 2QRV) using AutoDock Vina and Surflex-Dock, prioritizing according to their score values, contact patterns at 4Å and molecular diversity. Six consensus NPs were identified as virtual hits against DNMTs, including 9,10-dihydro-12-hydroxygambogic, phloridzin, 2',4'-dihydroxychalcone 4'-glucoside, daunorubicin, pyrromycin and centaurein. This method is an innovative computational strategy for identifying DNMTis, useful in the identification of potent and selective anticancer drugs. Copyright © 2015 Elsevier Inc. All rights reserved.
    Full-text · Article · Jun 2015
    • "Also, IC 50 values of some of the already used inhibitors of tyrosinase were compared to their obtained docking binding energies. There is an acceptable relation (with the correlation value of R 2 = 0.53) between the estimated docking binding energy and the biologically obtained IC 50 values which is quite approbate and even better value than those reported in other similar docking studies (Hare et al., 2010; Maldonado-Rojas & Olivero-Verbel, 2011). According toTable 1, beside a non-significant deviation seen when comparing the values, the obtained binding energies reasonably get along with IC 50 scores. "
    [Show abstract] [Hide abstract] ABSTRACT: Tyrosinase, a widely spread enzyme in micro-organisms, animals, and plants, participates in two rate-limiting steps in melanin formation pathway which is responsible for skin protection against UV lights' harm whose functional deficiency result in serious dermatological diseases. This enzyme seems to be responsible for neuromelanin formation in human brain as well. In plants, the enzyme leads the browning pathway which is commonly observed in injured tissues that is economically very unfavorable. Among different types of tyrosinase, mushroom tyrosinase has the highest homology with the mammalian tyrosinase and the only commercial tyrosinase available. In this study, ligand-based pharmacophore drug discovery method was applied to rapidly identify mushroom tyrosinase enzyme inhibitors using virtual screening. The model pharmacophore of essential interactions was developed and refined studying already experimentally discovered potent inhibitors employing Docking analysis methodology. After pharmacophore virtual screening and binding modes prediction, 14 compounds from ZINC database were identified as potent inhibitors of mushroom tyrosinase which were classified into five groups according to their chemical structures. The inhibition behavior of the discovered compounds was further studied through Classical Molecular Dynamic Simulations and the conformational changes induced by the presence of the studied ligands were discussed and compared to those of the substrate, tyrosine. According to the obtained results, five novel leads are introduced to be further optimized or directly used as potent inhibitors of mushroom tyrosinase.
    Full-text · Article · Mar 2014
    • "In vitro assays and animal studies have shown anti-COX activity of curcumin [20, 23, 33], which has been attributed to direct binding of curcumin mainly with COX-1 [17, 20] and to a lesser extent with COX-2 [20, 23]. Molecular docking studies support this opinion by proposing that curcumin can bind directly to the active site [17, 20] or an allosteric site on COX-2 [34]. However, the anti-COX-2 activity of curcumin may also be driven by indirect modulation at transcriptional levels via suppression of the NF-κB pathway [23, 27]. "
    [Show abstract] [Hide abstract] ABSTRACT: Curcumin is the major component of the yellow dye of turmeric, an Indian spice that is extracted from the rhizome of the tropical plant Curcuma longa, which belongs to the Zingiberaceae family. In this chapter, we discuss the pharmacological activities of curcumin and explore the molecular bases for these activities. We then review the proven and potential clinical uses of curcumin. Curcumin is well known historically as a curative agent, but its mechanism of action is intricate. The pharmacological effects of curcumin appear to be the result of a synergism of networks of weak biochemical interactions with multiple biological targets in interrelated signaling pathways. These targets include enzymes such as cyclooxygenase, lipoxygenase and protein kinases, and transcription factors such as NF-κB, STAT and Nrf2. Modulation of these molecules influences downstream affectors that produce the antiinflammatory, antioxidant, chemopreventive, anticancer, and antimicrobial activities of curcumin. These effects have been examined in clinical trials of curcumin for pain and inflammatory diseases, cancer, Alzheimer disease, cardiovascular diseases, and diabetes. The trials have used variable doses of curcuminwith different frequencies and duration, with the general conclusion that high doses of curcumin at the level of grams are required to obtain therapeutic effects.
    Chapter · Jan 2014 · Journal of biomolecular Structure & Dynamics
Show more

  • undefined · undefined
  • undefined · undefined
  • undefined · undefined