Discovery of polymethoxyflavones as potential cyclooxygenase-2 (COX-2), 5-lipoxygenase (5-LOX) and phosphodiesterase 4B (PDE4B) inhibitors

Article abstract of DOI:10.1080/10799893.2021.1951756

Non-steroidal anti-inflammatory drugs (NSAIDs) are widely prescribed to treat inflammatory-related diseases, pain and fever. However, the prolong use of traditional NSAIDs leads to undesirable side effects such as gastric, ulceration, and renal toxicity due to lack of selectivity toward respective targets for COX-2, 5-LOX, and PDE4B. Thus, targeting multiple sites can reduce these adverse effects of the drugs and increase its potency. A series of methoxyflavones (F1–F5) were synthesized and investigated for their anti-inflammatory properties through molecular docking and inhibition assays. Among these flavones, only F2 exhibited selectivity toward COX-2 (Selectivity Index, SI: 3.90, COX-2 inhibition: 98.96 ± 1.47%) in comparison with celecoxib (SI: 7.54, COX-2 inhibition: 98.20 ± 2.55%). For PDEs, F3 possessed better selectivity to PDE4B (SI: 4.67) than rolipram (SI: 0.78). F5 had the best 5-LOX inhibitory activity among the flavones (33.65 ± 4.74%) but less than zileuton (90.81 ± 0.19%). Docking analysis indicated that the position of methoxy group and the substitution of halogen play role in determining the bioactivities of flavones. Interestingly, F1–F5 displayed favorable pharmacokinetic profiles and acceptable range of toxicity (IC₅₀>70 μM) in cell lines with the exception for F1 (IC₅₀: 16.02 ± 1.165 μM). This study generated valuable insight in designing new anti-inflammatory drug based on flavone scaffold. The newly synthesized flavones can be further developed as future therapeutic agents against inflammation.

Full text of DOI:10.1080/10799893.2021.1951756

Journal of Receptors and Signal Transduction
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/irst20
Discovery of polymethoxyflavones as potential
cyclooxygenase-2 (COX-2), 5-lipoxygenase (5-LOX)
and phosphodiesterase 4B (PDE4B) inhibitors
Muhd Hanis Md Idris, Siti Norhidayah Mohd Amin, Siti Norhidayu Mohd
Amin, Agustono Wibowo, Zainul Amiruddin Zakaria, Zurina Shaameri,
Ahmad Sazali Hamzah, Manikandan Selvaraj, Lay Kek Teh & Mohd Zaki
Salleh
To cite this article: Muhd Hanis Md Idris, Siti Norhidayah Mohd Amin, Siti Norhidayu
Mohd Amin, Agustono Wibowo, Zainul Amiruddin Zakaria, Zurina Shaameri, Ahmad Sazali
Hamzah, Manikandan Selvaraj, Lay Kek Teh & Mohd Zaki Salleh (2021): Discovery of
polymethoxyflavones as potential cyclooxygenase-2 (COX-2), 5-lipoxygenase (5-LOX) and
phosphodiesterase 4B (PDE4B) inhibitors, Journal of Receptors and Signal Transduction, DOI:
10.1080/10799893.2021.1951756
To link to this article: https://doi.org/10.1080/10799893.2021.1951756
Published online: 29 Jul 2021.
Article views: 22
View supplementary material
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=irst20

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
https://doi.org/10.1080/10799893.2021.1951756
RESEARCH ARTICLE
Discovery of polymethoxyflavones as potential cyclooxygenase-2 (COX-2),
5-lipoxygenase (5-LOX) and phosphodiesterase 4B (PDE4B) inhibitors
Muhd Hanis Md Idrisa , Siti Norhidayah Mohd Amina , Siti Norhidayu Mohd Amina,
Agustono Wibowob, Zainul Amiruddin Zakariac , Zurina Shaamerid , Ahmad Sazali Hamzahd
Manikandan Selvaraje , Lay Kek Teha and Mohd Zaki Salleha
,
aIntegrative Pharmacogenomics Institute (iPROMISE), Universiti Teknologi MARA (UiTM) Selangor Branch, Puncak Alam Campus, Selangor,
c
Malaysia;bFaculty of Applied Science, Universiti Teknologi MARA (UiTM) Pahang Branch, Jengka Campus, Pahang, Malaysia; Department of
d
Biomedical Science, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia (UPM), Serdang, Malaysia; Organic Synthesis
Laboratory, Institute of Science (IOS), Universiti Teknologi MARA (UiTM) Selangor Branch, Puncak Alam Campus, Selangor, Malaysia; School
e
of Engineering, Monash University (Malaysia Campus), Bandar Sunway, Malaysia
ABSTRACT
ARTICLE HISTORY
Received 3 January 2021
Revised 27 June 2021
Accepted 30 June 2021
Non-steroidal anti-inflammatory drugs (NSAIDs) are widely prescribed to treat inflammatory-related dis-
eases, pain and fever. However, the prolong use of traditional NSAIDs leads to undesirable side effects
such as gastric, ulceration, and renal toxicity due to lack of selectivity toward respective targets for
COX-2, 5-LOX, and PDE4B. Thus, targeting multiple sites can reduce these adverse effects of the drugs
and increase its potency. A series of methoxyflavones (F1–F5) were synthesized and investigated for
their anti-inflammatory properties through molecular docking and inhibition assays. Among these fla-
vones, only F2 exhibited selectivity toward COX-2 (Selectivity Index, SI: 3.90, COX-2 inhibition:
98.96 1.47%) in comparison with celecoxib (SI: 7.54, COX-2 inhibition: 98.20 2.55%). For PDEs, F3
possessed better selectivity to PDE4B (SI: 4.67) than rolipram (SI: 0.78). F5 had the best 5-LOX inhibi-
tory activity among the flavones (33.65 4.74%) but less than zileuton (90.81 0.19%). Docking analysis
indicated that the position of methoxy group and the substitution of halogen play role in determining
the bioactivities of flavones. Interestingly, F1–F5 displayed favorable pharmacokinetic profiles and
acceptable range of toxicity (IC50>70 mM) in cell lines with the exception for F1 (IC50:
16.02 1.165 mM). This study generated valuable insight in designing new anti-inflammatory drug
based on flavone scaffold. The newly synthesized flavones can be further developed as future thera-
peutic agents against inflammation.
KEYWORDS
Flavonoids synthesis;
molecular docking; multi-
target inhibitor;
inflammatory; pharmacokin-
etics; cytotoxicity
GRAPHICAL ABSTRACT
Introduction
arachidonic acid from transforming into prostaglandins,
thromboxane, and prostacyclin. The traditional nonselective
non-steroidal anti-inflammatory drugs such as ibuprofen and
naproxen are capable of inhibiting both COX-1 and COX-2
isoforms [5]. Nevertheless, adverse gastrointestinal events
such as gastric mucosal damage and gastroduodenal ulcers
have been reported due to the use of the non-steroidal anti-
inflammatory drugs. More selective COX-2 inhibitors were
developed to minimize the side effects but chronic use of
some of these inhibitors still cause cardiovascular adverse
effects and increase thrombotic risk due to blockage of pros-
taglandin I2 [6]. In addition, PDE4 inhibitors that are used for
treating inflammation have been discontinued due to side
effects such as emesis, mild to moderate nausea, headache,
and diarrhea [7]. The only 5-LOX inhibiting drug, known as
zileuton, was also withdrawn from the market due to its hep-
atotoxicity adverse effect [8].
Inflammation is a natural protective body response to harm-
ful stimuli, such as pathogens, damaged cells, or irritants, to
eliminate initial cause of injury, clear out necrotic cells and
damaged tissue, and initiate tissue repair. However, chronic
uncontrolled inflammation is harmful as it damages host tis-
sues and causes inflammatory diseases such as cancers, dia-
betes, and rheumatoid arthritis [1]. Inflammation is closely
associated with pro-inflammatory enzymes including cycloox-
ygenases (COX), lipoxygenases (LOX), and phosphodiester-
ases (PDE), in addition to other pro-inflammatory mediators
and cytokines including prostaglandin E2, interleukins IL-1,
IL-6, and IL-8, and tumor necrosis factor (TNF-a) [2–4].
Therefore, many studies are focusing on these inflammatory
markers as drug target to treat inflammatory diseases.
Non-steroidal anti-inflammatory drugs (NSAIDs) and COX-
2 inhibitors are the most widely used drug for reducing
Flavone is among the major plant secondary metabolites
inflammation and pain. The drugs inhibit COX and stop that are broadly distributed across the plant kingdom. It
CONTACT Mohd Zaki Salleh
zakisalleh@uitm.edu.my
Integrative Pharmacogenomics Institute (iPROMISE), Universiti Teknologi MARA (UiTM) Selangor,
Puncak Alam Campus, Bandar Puncak Alam 42300, Malaysia
Supplemental data for this article can be accessed here.
ß 2021 Informa UK Limited, trading as Taylor & Francis Group

2
M. H. MD IDRIS ET AL.
consists of a phenyl ring that attaches to chromone at 2-pos- petroleum ether-ethyl acetate 1:2 (100 ml), filtered over a
ition. With a high degree of chemical diversity by modifica- short column of silica gel and washed with petroleum ether-
tions of the chemical backbones, flavone and its derivatives ethyl acetate 1:2 (100 ml). The solvent was evaporated, and
have multiple roles in biological functions and attract great the residue recrystallized from toluene–petroleum ether to
interest from the researchers as privilege structures in drug afford product as a white crystalline solid (44%). 1H NMR
discovery [9, 10]. In recent years, in vitro and in vivo studies (CDCl3, 400 MHz) d: 2.63 (3H, s, CH3), 3.96 and 3.93 (6H, s,
as well as clinical studies have reported flavones as anti- OCH3), 5.28 (1H, s, CH).
inflammatory agent [11–13]. For instance, flavones displayed
the reduction of COX-2 mediated prostaglandin E2 (PGE2)
production and inhibited the activity of 5-lipoxygenase (5-
Synthesis of flavones
Step 1. To a solution of appropriate acetophenone (4 mmol)
LOX) [14]. Another study also reported that natural flavone
and aldehyde (4 mmol) in C2H5OH (25 ml), 20% of aqueous
was able to inhibit formation of nitric oxide (NO) in lipopoly-
KOH (4 mmol) was added. The reaction mixture was stirred at
saccharide (LPS)-activated RAW 264.7 macrophages and indi-
room temperature till completion of reaction (monitored by
cated as potent anti-lipoxygenase inhibitor without
TLC). Then the reaction mass was poured into ice water and
neutralized with aqueous 10% HCl solution. The precipitate
significant cytotoxic effect [15].
The structure–activity analysis showed that the substitu-
was filtered, washed with excess of water, dried, and recrys-
tion at 5 and 7 positions of ring A and at 40 position of ring
tallized from methanol to obtain pure chalcones. Step 2. To
the solution of appropriate chalcones in dimethyl sulfoxide
B and the presence of C2–C3 double bond are crucial in
determining their anti-inflammatory effects. Additionally, the
(10 ml), I2 (catalytical amounts) was added. The reaction mix-
oxidation of flavones into reactive electrophilic quinones can
ture was heated to reflux for 1–2 h, cooled, and poured into
interact with nucleophilic natured thiols and amino groups
of proteins. This leads to the formation of different additional
products that are responsible for their valuable biological
effects [16, 17]. In our previous study, isolated flavones from
water and extracted into EtOAc (3 ꢀ 25 ml). The organic layer
was washed with brine and dried over MgSO4. The solvent
was evaporated to get the product (F1–F5).
Muntingia calabura possessed anti-nociceptive and anti-
inflammatory activity by regulating inflammatory response
through cAMP pathway [18]. The ability of flavones to mimic
cAMP stacking interaction in PDE makes flavones good com-
petitive PDE inhibitors which elevates level of cAMP. Besides,
the functional group at C-40 position also plays important
role in inhibiting PDE [19]. Therefore, the present study is a
continuation of our previous work to further investigate the
inhibitory potential of flavones toward multi-targets of COX-
2, 5-LOX, and PDE4B as well as pharmacokinetic and toxicity
profiles of the synthesized flavones.
5,7-Dimethoxyflavone (5,7-dimethoxy-2-phenyl-4H-chro-
men-4-one) (F1)
1H NMR (CDCl3, 400 MHz): dH) , 7.87 (2H, d, J ¼ 7.6 Hz, Ar H-
20/60), 7.36-7.44 (3H, m, Ar H-30,40,50), 6.69 (1H, s, H-3), 6.54
(1H, d, J ¼ 2.0 Hz, Ar H-8), 6.27 (1H, d, J ¼ 2.0 Hz, Ar H-6), 3.92
(1H, s, OCH3), 3.91 (3H, s, OCH3). 13C-NMR (100 MHz, acet-
one-d6) dC 181.0 (C¼O), 170.3 (C-7), 169.1 (C-2), 159.8 (C-5),
147.9 (C-8a), 132.4 (C-10), 130.9 (C-30/50), 129.4 (C-40), 128.6
(C-20/60), 127.9 (C-5a), 110.7 (C-3), 93.8 (C-6), 89.3 (C-8), 55.7
(CH3), 55.3 (CH3). ESI-TOF HRMS m/z: [M þ H]þ 283.0974 (Calc.
for C17H15O4, 283.0965) (yellow solid, yield: 59%).
Methods
8-Iodo-5,7-dimethoxyflavone (8-iodo-5,7-dimethoxy-2-phe-
nyl-4H-chromen-4-one) (F2)
Chemistry
1H NMR (acetone-d6, 400 MHz): dH 8.18 (2H, m, Ar H-20/60),
7.57 (4H, m, Ar-H-30/40/50/60), 6.75 (1H, s, H-3), 6.68 (1H, s, Ar-
H-6), 4.06 (3H, s, OCH3), 3.96 (3H, s, OCH3). 13C-NMR
(100 MHz, acetone-d6) dC 177.8 (C¼O), 168.4 (C-7), 159.6 (C-
5), 156.2 (C-8a), 154.3 (C-2), 131.4 (C-40), 129.2 (C-30/50), 126.0
(C-20/60), 125.6 (C-10), 115.8 (C-5a), 108.0 (C-3), 92.9 (C-6), 64.0
(C-8), 56.6 (OCH3), 55.9 (OCH3). ESI-TOF HRMS m/z: [M þ H]þ
408.9930 (Calc. for C17H14O4I, 408.9931) (colorless needle
crystal, yield: 85.42%).
Solvents and chemicals were purchase from Sigma-Aldrich
1
(St. Louis, MO) unless otherwise stated. The H and 13C NMR
spectra were registered in CDCl3 with Joel Resonance
ECZ400S [400 MHz (1H) and 100 MHz (13C)] using TMS as the
internal standard. Analytical TLC was performed on silica gel
60F254, Merck (layer thickness 0.25 mm, Merck, Kenilworth,
NJ) and visualized with UV light and KMnO4 as the detect-
ing agent.
Synthesis of iodoacetophenone
About 1.0 M solution of iodine chloride (ICl) in dry dichloro-
methane (DCM) (15.0 ml) was added dropwise to a stirred
6-Methoxyflavone (6-methoxy-2-phenyl-4H-chromen-4-
one) (F3)
solution of the 4,6-dimethoxy-2-hydroxy-acetophenone 1H NMR (acetone-d6, 400 MHz): dH 8.07-8.05 (2H, m, Ar H-20/
(2.55 mmol) in dry dichloromethane (50 ml). The mixture was 60), 7.68 (1H, d, J ¼ 9.1 Hz, Ar H-8), 7.57 (2H, m, Ar H-30/50),
stirred for 5 h, diluted with dichloromethane (100 ml) and 7.48 (1H, d, J ¼ 3.2 Hz, Ar H-40), 7.37 (1H, d, J ¼ 9.1 Hz, Ar H-5),
washed with aqueous Na2S2O3 (2 ꢀ 30 ml), and water 6.82 (1H, s, H-3), 3.90 (3H, s, OCH3). 13C-NMR (100 MHz, acet-
(1 ꢀ 30 ml). The organic layer was dried (MgSO4) and the one-d6) dC 177.0 (C¼O), 163.3 (C-1), 162.9 (C-6), 157.2 (C-8a),
solvent was evaporated. The residue was dissolved in 132.0 (C1’), 131.6 (C-40), 129.2 (C-30/50), 126.3 (C-20/60), 124.6

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
3
(C5a), 123.2 (C-7), 120.0 (C-8), 106.3 (C-5), 104.9 (C-3), 55.4 of binding pose. Full ligand conformational and partial pro-
(OCH3). ESI-TOF HRMS m/z: [M þ H]þ 253.0860 (Calc. for tein flexibility were considered during the docking [28]. One
C16H13O3, 253.0859) (orange solid, yield: 58%).
hundred genetic algorithm (GA) runs were performed for
each ligand. All water molecules and hetero atoms were
omitted from the protein. The binding region was defined in
10 Å radius around the catalytic iron (Fe2_1_A). The binding
interaction patterns of selected compounds in the binding
sites of enzymes were visualized using PyMOL (http://www.
pymol.org/).
4’-Bromo-6-methoxyflavone
(2-(4-bromophenyl)-6-
methoxy-4H-chromen-4-one) (F4)
1H NMR (acetone-d6, 400 MHz): dH 8.02 (2H, d, J ¼ 8.7 Hz, Ar
H-30/50), 7.76 (2H, d, J ¼ 8.7 Hz, Ar H-20/60), 7.68 (1H, d,
J ¼ 8.7 Hz, Ar H-8), 7.47 (1H, d, J ¼ 3.2 Hz, Ar H-5), 7.37 (1H,
dd, J ¼ 9.1, 3.2 Hz, Ar H-7), 6.85 (1H, s, H-3), 3.90 (3H, s,
OCH3). 13C-NMR (100 MHz, acetone-d6,) dC 176.9 (C¼O), 161.5 COX peroxidase assay
(C-1), 157.2 (C-6), 150.8 (C-8a), 132.4 (C-30/40), 131.2 (C-10),
The inhibitory activity of COX was determined by measuring
128.1 (C-20/60), 125.8 (C-5a), 124.6 (C-40), 123.3 (C-7), 119.9 (C-
5), 106.8 (C-5), 104.8 (C-3), 55.4 (OCH3). ESI-TOF HRMS m/z:
[M þ H]þ 330.9960 and 332.9937 (Calc. for C16H12O379Br,
330.9970 and C16H12O381Br, 332.9949) (brown solid,
yield: 64%).
the formation of oxidized N,N,N’,N’-tetramethyl-p-phenylene-
diamine (TMPD) measured spectrophotometrically at 550 nm.
The experiment was carried out as previously described with
modifications [29]. The COX inhibition reaction was per-
formed by a 10 min incubation at room temperature in the
presence of reaction buffer (Tris buffer, 0.1 M, pH 8.0, 150 ml),
heme (2.2 mM in DMSO, 10 ml), human recombinant COX-1 or
COX-2 enzymes (10 ml), and tested compounds (100 mM in
DMSO, 10 ml). The reaction was initiated by adding 20 lL of
freshly prepared TMPD and followed by the addition of 20 lL
arachidonic acid (diluted with ethanol). The COX activity of
compound was compared with celecoxib which is the refer-
ence drug. The absorbance was measured using POLARstar
Omega multi-mode reader (BMG LABTECH GmbH, Offenburg,
Germany) after 5 min of adding arachidonic acid.
6,4’-Dimethoxyflavone (6-methoxy-2-(4-methoxyphenyl)-
4H-chromen-4-one) (F5)
1H NMR (CDCl3, 400 MHz): dH 8.26 (2H, d, J ¼ 9.1 Hz, Ar H-20/
60), 7.66 (1H, d, J ¼ 9.1 Hz, Ar H-8), 7.48 (1H, d, J ¼ 3.2 Hz, Ar
H-5), 7.36 (1H, dd, J ¼ 9.1, 3.2 Hz, Ar H-6), 7.10 (2H, d,
J ¼ 6.9 Hz, Ar H-30/50), 3.91 (3H, s, OCH3), 3.88 (3H, s, OCH3).
13C-NMR (100 MHz, acetone-d6,) dC172.4 (C¼O), 161.1 (C-2),
156.6 (C-40), 150.2 (C-6), 145.0 (C-8a), 137.7 (C-10), 129.7 (C-
5a), 129.4 (C-20/60), 123.5 (C-7), 121.7 (C-8), 120.0 (C-5), 114.1
(C-30/50), 105.1 (C), 103.8 (C-3), 55.5 (OCH3), 54.8 (OCH3). ESI-
TOF HRMS m/z: [M þ H]þ 283.0969 (Calc. for C17H15O4,
283.0965) (white solid, yield: 59%).
5-LOX inhibition assay
The activity of 5-LOX was determined by oxidation of the
nonfluorescent substrates, 20,70-dichlorodihydrofluorescein
diacetate (H2DCFDA) to the highly fluorescent 20,70-dichloro-
flurescein (DCF) product during 5-LOX’s catalytic reaction
[30]. The enzyme assay contained 50 mM Tris (pH 7.5), 2 mM
ethylenediaminetetraacetic acid (EDTA), 2 mM CaCl2, 100 lM
arachidonic acid and 10 lM H2DCFDA (pH 7.5). Prior to the
addition of studied compounds (dissolved in DMSO),
H2DCFDA and enzyme were preincubated for 5 min. Then
the mixture was further preincubated with compounds for
another 10 min. The reactions were initiated by the addition
of arachidonic acid as substrate. The enzymatic reaction was
run for 60 min. Assay plates were read using POLARstar
Omega multi-mode plate reader (BMG LABTECH, Offenburg,
Germany) under 485 nm excitation and 520 nm emission fil-
ters. Fluorescence signals were recorded at 0 and 60 min.
Molecular docking
Prior to molecular docking, the structures of the compounds
were built using Maestro build panels and were optimized
and minimized using LigPrep (v3.5.9) in Schrodinger Small
Molecule Drug Discovery Suite 2017-1. Molecular docking
was performed using Grid-Based Ligand Docking with
Energetics (Glide) (v6.8, Schrodinger 2017-1) for human COX
and PDE as reported before [20, 21] whereas Genetic
Optimization for Ligand Docking (GOLD) 5.2.2 software was
used in molecular docking for human 5-LOX [22, 23]. Briefly,
the X-ray crystallographic structures of human COX-2 (PDB
ID: 3O8Y), 5-LOX (PDB ID: 5F19), PDE4B (PDB ID: 3G45), and
PDE4D (PDE ID: 3G4G) were retrieved from Protein Database
Bank (PDB, https://www.rcsb.org) [24]. Meanwhile, human
COX-1 crystal structure was prepared using homology mod-
eling as described previously [25]. These structures were pre-
pared and optimized using Protein Preparation Wizards. At
the end, the crystal structures were minimized using OPLS3
force field [26]. To conduct Glide, the grid generation was
generated at the center of the ligand with a grid spacing of
12 Å for COXs and 20 Å for PDEs. The molecular docking for
€
€
PDE inhibition assay
The PDE activity of compounds was evaluated using PDE-
GloTM Phosphodiesterase Assay kit and the protocol by
manufacturer was followed (Promega Corporation, Wisconsin,
MI) [31]. Briefly, 1 ml of compounds (final concentration:
COXs and PDEs was performed using extra-precision (XP) 100 mM) were preincubated with 1.5 ml of PDE enzyme in
docking mode of Glide software without applying any con- 384-well white, flat-bottom plate for 5 min at room tempera-
straints [27]. For 5-LOX, docking was performed in GOLD by ture. Then 2.5 ml of 2 mM of cAMP was added to initiate the
using GoldScore as a scoring function to evaluate the quality reaction and incubated for 1 h at 37 ꢁC. After incubation

4
M. H. MD IDRIS ET AL.
period, termination buffer was added and followed by detec-
tion buffer. The mixture was incubated for 20 min at room
Docking scores
The activity of synthesized flavones toward COXs, 5-LOX, and
PDEs were analyzed via molecular docking study (Table 1).
Based on docking study of COX, the structure of all the syn-
thesized flavones able to dock with COX-1 and COX-2 bind-
ing cavities resulting docking scores less than ꢂ7 kcal/mol
but only F2 exhibited selectivity toward COX-2. The reference
drug, celecoxib, however, is the only able to dock inside the
COX-2 binding cavity. In the case of docking to 5-LOX, the
flavones had GOLD fitness score in a range of 13.45 and
54.62 which were slightly less than the reference drug, zileu-
ton (GoldScore ¼ 57.00). F2 in 5-LOX showed the lowest fit-
ness score while F5 possessed the highest fitness score. The
bulky structure of F2 may affect the accessibility of F2 to the
substrate channel in the closed conformation of 5-LOX.
Hence, F2 was unable to bind perfectly at the active sites of
LOX. Small compounds otherwise had good fitness score.
The present study also found that F1–F3 exhibited compar-
able docking scores in PDEs with the reference drug, roli-
pram but only F1 and F3 were more selective to PDE4B as
compared to rolipram. The other flavones showed low bind-
ing affinity to both PDEs with docking scores more than
ꢂ5 kcal/mol.
R
V
temperature before the addition of Kinase-Glo . The mixture
was incubated for another 10 min. Luminescence was meas-
ured using POLARstar Omega multi-mode plate reader (BMG
LABTECH GmbH, Offenburg, Germany).
ADMET prediction
ADMET properties of flavones were predicted using
ADMETlab [32] and QikProp [33] in Schrodinger Small Drug
Discovery Suite 2017-1. On the other hand, ProTox-II (http://
tox.charite.de/protox_II) [34], ACD/i-Lab 2.0 (https://ilab.
acdlabs.com/iLab2) [35], and Pred-hERG 4.1 (http://labmol.
com.br/predherg) [36], a free web-based service were used
to analyze the toxicity profiles of flavones.
Cytotoxicity assay
Cytotoxicity of compounds against RAW 264.7 cells was
determined using MTT assay [37]. 100 ml of compounds (ser-
ial dilution in media, 0–100 mM) was treated with 10,000
cells/well for 24 h at 37 ꢁC in 5% CO2 under 70% humidity.
Then, 10 ml of MTT dye (5 mg/ml in PBS) was added to each
well. The plate was incubated again for 4 h. The supernatant
was then aspirated out and was replaced with 100 ml of lysis
solution (DMSO). After 30 min of incubation, the absorbance
was read at 595 nm and the percentage of cell viability was
calculated. Data were analyzed using sigmoidal dose-
response (variable slope) equation or four-parameter logistic
Inhibitory activity of flavone derivatives
To validate the result of docking, the activities of compounds
were evaluated through inhibition assay. For COXs and PDEs,
only compounds with the SI more than 1 were tested for the
inhibitory activity while all docked compounds to 5-LOX
were proceeded with inhibition assay (Table 2). From the
results, F2 was more selective to COX-2 as compared to
COX-1 which was similar to the result of docking. However,
the selectivity of F2 toward COX-2 was lower than celecoxib
by two-fold but the activity of inhibition against COX-2 was
comparable with celecoxib.
R
R
curve (4PL) in GraphPad PrismV, version 7.03, for WindowsV
(GraphPad Software, La Jolla, CA).
Results and discussion
In PDEs, F1 and F3 were chosen for determining PDE
activity. Although there was a slight difference in SI of dock-
ing scores between F1 and F3, F3 showed four times more
selective to PDE4B than PDE4D in inhibition assay while F1
showed is selective to PDE4D with SI less than 1. Moreover,
SI of PDE4B of F3 was higher than that of rolipram. This indi-
cated that the position of methoxy group in the benzene
ring of chromone played an important role in the selectivity
of flavones where para- position was more favorable in bio-
logical activity than meta-position.
Chemistry
All the compounds were synthesized according to the steps
outlined in Figure 1. The chalcone derivatives (C1, C3, C4,
and C5) were prepared via Claisen–Schmidt condensation
reaction of readily accessible acetophenones and benzalde-
hydes in the presence of 20% sodium hydroxide in ethanol
at room temperature for 24 h. While synthesis of chalcone
derivatives C2 begun through iodonization of 4,6-dimethoxy-
2-hydroxy-acetophenone with iodine monochloride in dry
acetone at room temperature for 5 h. Targeting flavones
For 5-LOX, F5 has the highest inhibition against 5-LOX
(F1–F5) were synthesized by refluxing corresponding chal- but less than zileuton. The inhibition scores were lower for
cones in DMSO in the presence of iodine as catalyst. All the F2, followed by F4 and F3 while no activity of 5-LOX was
synthesized flavones were confirmed on the basis of their observed in F1. The result suggested that para- position of
1
spectroscopic evidence. In the H NMR spectra of F1–F5, ole- methoxy group (F3) was more favorable than meta-position
finic proton attached to the C-3 appeared as a singlet in the (F1) for inhibition of 5-LOX. The substitution of halogen
range of dH 5.69–6.90, while singlet signals from methoxy group, iodine at 8-position in F2 substantially decreased the
protons appeared at dH 3.87–4.03. Besides that, aromatic pro- activity of 5-LOX as compared to F1. The paramount role of
tons attached to the ring A and C appeared as a singlet, halogen in improving drug efficacy was reported earlier
doublet and multiplate in the range of dH 5.28–8.13.
[38,39]. The addition of functional group at para- position in

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
5
Figure 1. Schematic diagram of flavone synthesis. Reaction condition: (i) ICl, dry DCM, room temperature, 5 h; (ii) KOH 20%, ethanol, room temperature, 24 h; (iii)
I2, DMSO, reflux 2 h.
Table 1. Binding score of compounds toward respective targets.
interaction with R120 at the entrance of binding site.
Meanwhile, phenyl group of celecoxib was stabilized by
hydrophobic interactions via strong van der Waals inter-
action in hydrophobic region of binding site which com-
posed of M522, F381, L384, Y385, W387, and A527. In
contrast, the selectivity of F2 was contributed by the add-
ition of iodine group involved in the van der Waals forces
and increased the binding affinity to COX-2 (Table 3) [40]. F2
only occupied a part of COX-2 active site which was mostly
at the hydrophobic region and entrance of binding site.
These explained the difference of selectivity for COX-2 in
both compounds. The orientation of F2 was firmly fixed by
the formation of p–p interaction between phenyl group and
W387 and cation-p interaction between ring A of F2 and
R120. Further, F2 was stabilized by hydrophobic interactions
of active site amino acid V349, Y355, L359, F381, L384, Y385,
W387, F518, M522, V523, A527, and L531 (Figure 2(c)). In
addition, binding pose of F2 in COX-1 remained unchanged
but the absence of side pocket in COX-1 made F2 to move
deeper to hydrophobic region of COX-1. This resulted in the
formation of additional p-p interaction with Y385 along with
W387. The phenyl of F2 was then stabilized by hydrophobic
interaction of F381, L384, Y385, W387, F518, and M522
whereas the chromone part of F2 was stabilized by hydro-
phobic interaction of V116, V349, L352, Y355, L359, I523, and
Docking score (kcal/mol)
GOLD fitness score
Compounds
COX-1
COX-2
PDE4B
PDE4D
5-LOX
Celecoxib
Rolipram
Zileuton
F1
F2
F3
ND
–
–
ꢂ8.019
ꢂ8.351
ꢂ9.316
ꢂ9.460
ꢂ8.883
ꢂ11.869
–
–
–
–
–
–
ꢂ9.368 ꢂ9.576
–
–
57.00
49.99
13.45
53.12
52.97
54.62
ꢂ7.987 ꢂ10.530 ꢂ9.191
ꢂ9.281
ꢂ8.641
ꢂ8.792
ꢂ8.512
ꢂ8.814 ꢂ9.229
ꢂ9.932 ꢂ8.782
ꢂ1.367 ꢂ4.675
ꢂ3.992 ꢂ4.568
F4
F5
ND: not dock; – : not determined.
phenyl ring further enhanced the inhibitory activity of 5-LOX
where methoxy was more favorable (F5) than bromine (F4).
Analysis of binding interactions
The active compounds were further evaluated for their bind-
ing modes in the target sites. The investigation of the bind-
ing interaction of celecoxib in COX-2 active site revealed that
sulfonamide phenyl group of celecoxib occupied the side
pocket of COX-2 binding site and formed hydrogen bond
with H90 (2.4 Å), L352 (2.2 Å), and F518 (2.7 Å) (Figure 2(a)).
The interaction in the side pocket of COX-2 confers selectiv-
ity of celecoxib toward COX-2 compared to compound F2.
Besides, imidazole group of celecoxib formed cation-p L531 (Figure 2(b)).

6
M. H. MD IDRIS ET AL.
Table 2. Percentage of inhibition of enzymes at single concentration (100 mM) of compounds.
Percentage of Inhibition (%)
PDE4B
Compounds
COX-1
COX-2
SI of COX-2
PDE4D
SI of PDE4B
5-LOXa
Celecoxib
Rolipram
Zileuton
F1
F2
F3
13.02 2.84
98.20 2.55
7.54
–
–
–
3.90
–
–
–
–
0.78
–
0.67
–
4.67
–
–
–
–
–
–
–
–
–
–
51.87 0.38
66.51 2.49
–
–
90.81 0.19
NA
20.96 11.20
7.30 6.11
16.99 9.72
33.65 4.74
50.05 4.46
74.31 3.22
25.41 5.71
98.96 1.47
–
–
–
–
–
–
–
–
39.70 5.19
8.50 3.96
F4
F5
–
–
–
–
–
–
Values were presented in mean SD. a: concentration of compounds at 10 mM; SI: selectivity index; NA: not active; – : not determined.
Figure 2. Binding mode of celecoxib in COX-2 (a) active site and F2 in COX-1 (b) and COX-2 (c) active site.
Analysis of binding interaction of rolipram in PDE showed binding and established numerous interactions with residues
that rolipram bind to active sites of PDE4B (Figure 3(a)) and of active sites. The phenylmethoxy ring of rolipram sat in a
PDE4D (Figure 3(b)) in similar orientation and interacted with hydrophobic pocket of PDE4, interacting with Y274, F279,
the same residues. Thus, rolipram had lack of selectivity I582, F585, and F618 in PDE4B and F196, I502, F506, and
toward PDE4B. Rolipram formed two hydrogen bonds with F538 in PDE4D. The pyrrolidone group of rolipram anchored
key interacting residues, glutamine (Q615 in PDE4B and in the direction of highly conserved metal binding pocket
Q535 in PDE4D) (1.8–2.4 Å) which involved in substrate and formed hydrophobic interactions with F279, Y405, L565,

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
7
Table 3. Binding energy of the celecoxib and F2 in COX binding sites.
COX-1/(kcalꢃmol-1)
COX-2/(kcalꢃmol-1)
Comp.
LE
evdw
En
emodel
LE
evdw
En
emodel
Celecoxib
F2
–
–
–
–
ꢂ0.456
ꢂ48.821
ꢂ47.910
ꢂ38.698
ꢂ75.511
ꢂ39.453
ꢂ0.380
ꢂ10.528
ꢂ10.528
23.848
ꢂ0.411
ꢂ38.743
LE: Glide ligand efficiency; evdw: Glide Van der Waals energy; En: Glide energy; emodel: Glide model energy.
Figure 3. Binding mode of rolipram in PDE4B (a) and PDE4D (b) active site, F1 in PDE4B (c) and PDE4D (d) active site and F3 in PDE4B (e) and PDE4D (f) active
site. Zinc and magnesium were depicted in sphere, respectively.

8
M. H. MD IDRIS ET AL.
and I582 in PDE4B and F201 and L485 in PDE4D. On the involving F177, L368, F359, A410, L414, F421, V604, A603,
and L607 (Figure 5(e)).
other hand, the cyclopentyloxy group of rolipram interacted
with Y405, P568, Y575, W578, I582, and F618 in PDE4B and
Y325, P488, Y495, W498, and I502 in PDE4D via van der
Waals interactions. The study also found that hydrogen
bonding plays important role in binding interaction
of rolipram.
On the other hand, the steric effect of halogen group in
F2 caused the orientation of F2 to flip oppositely from F5 in
the active site of 5-LOX. Iodine at ring A then made a con-
tact with N425 (3.3 Å) to form halogen bond and chromone
at ring
B formed hydrogen bond with Q363 (2.5 Å).
Hydrophobic interactions stabilized the position of F2 that
involved F177, A410, L368, L414, and I415 at the phenyl ring
of F2 and F359, F421, A603, V604 and L607 at chromone
group of F2 (Figure 5(b)). The position of F4 in binding
pocket of 5-LOX is also the same with F2 which was sup-
ported by hydrophobic interaction of F177, Y181, L414, L420,
F421, A424, W599, A603, V604, and L607. The interaction was
strengthened by p–p interaction (edge-to-face) between ring
A and phenyl ring of F421 (Figure 5(d)). F3 possessed similar
orientation as F5 but shifted toward polar region of 5-LOX.
This resulted in the formation of hydrogen bond between
chromone part of F3 and Q363 (2.6 Å). It was stabilized by
L368, L414, F421, and L607 at chromone part of F3 and
Y181, L420, A424, W599, A603, and V604 at phenyl part of
F3 (Figure 5(c)).
The binding patterns of F1 and F3 in PDE4B (Figure
3(c,e), respectively) were the same, while there were differen-
ces of binding modes in PDE4D. This is due to the orienta-
tion of I275 in PDE4B and I197 in PDE4D. The orientation of
propane side chain of I275 which buried into the pocket cre-
ates additional space in side pocket of PDE4B. As a result, F1
bound deeper into the side pocket and allow the bulky F1
to fit in whereas smaller side pocket in PDE4D changed the
binding position of F1. The comprehensive exploration of
binding mode of flavones revealed that the heterocyclic
structure of chromone in F1 and F3 had occupied Q pocket
of PDE4B while the ketone group formed strong hydrogen
bond (2.0–2.2 Å) with side chain of Q615. This possibly pre-
vented the interaction between substrate and enzyme, thus
leading to inhibition of PDE4B. Moreover, both flavones were
flanked between I582 and F618 and made p–p interaction
with Y274 and F618. Their position was further stabilized by
hydrophobic interaction of conserved residues including
Y274, Y405, L565, I582, F586, and F618.
In PDE4D, the phenyl ring of F1 and F3 were projected to
be at the entrance of active site. However, the orientation of
chromone ring of F1 and F3 were distinctive. The chromone
part of F1 laid on Q pocket of PDE4D (Figure 3(d)) while the
chromone part of F3 occupied the metal binding pocket of
PDE4D (Figure 3(f)). The position of F1 resulted in the estab-
lishment of p–p stacking with Y325 and F538 in opposite dir-
ection which gave more stable binding position in F1 than
ADMET properties
Failing the criteria of ADMET is the major reason for drug
rejection in development of drug. Therefore, this study used
two software to predict ADMET properties of compounds.
QikProp is a commercial software that calculate prominent
ADMET descriptors of compounds. Another software,
ADMETlab is a free web interface software which compre-
hensively performs systematic evaluation of ADMET of com-
pounds based on 288,967 ADMET data entries collected. In
addition to two software, ProTox-II was used to predict the
average lethal dose (LD50) in rodents and toxicity of com-
pounds according to Tox21 criteria. This study also used
Pred-hERG to specifically assess hERG inhibition of com-
pounds while ACD/i-Lab predicts LD50 of compounds in dif-
ferent route of administration and health effects
of compounds.
Data revealed that all compounds successfully conceded
the Lipinski’s rule of five which is a druglikeness assessment
(Table 5) [41]. It was also observed that all compounds had
satisfactory bioavailability as indicated in Tables S1 and S2.
With respect to metabolism of compounds, all compounds
were predicted to be metabolized by CYP1A2, CYP2C9,
CYP2C19, and CYP2D6. Moreover, all compounds had high
probability to act as inhibitor of CYP3A4. This affected the
bioavailability, the toxicity, and the effectiveness of com-
pounds [42]. The half-lives and clearance rates of the com-
pounds were low but within the acceptable range. Though,
the toxicities vary across the compounds.
F3.
Moreover,
hydrophobic
interactions
(Lipophilic
EvdW þ PhobEn) with surrounding residues which include
V193, F196, I197, F201, Y325, L485, P488, Y495, W498, I502,
F506, and F538 increased stabilization of F1 in the binding
pocket of PDE4D (Figure 4). F3 only formed single p–p stack-
ing between ring B and F538. F3 also stabilized itself
through hydrophobic interactions of V193, F196, I197, F201,
Y325, L485, P488, F506, and F538. Therefore, F3 became less
selective to PDE4D as compared to F1.
The prediction of binding interaction in 5-LOX suggested
that hydrogen bond had significant contribution in binding
interaction of zileuton (Figure 5(a)). Hydroxyl group of zileu-
ton formed strong hydrogen bond with Q363 (1.9 Å) while
two hydrogen bonds were formed between ketone group of
zileuton and N425 (2.0 Å) and H600 (2.6 Å). The position of
zileuton in the binding pocket was further stabilized by
hydrophobic interaction of F177, Y181, L368, L414, F421, and
L607 at benzothiophene group and F359, A603, and V604 at
the hydroxyurea group. In contrast, the interaction of F5 in
5-LOX was contributed by Van der Waals forces as indicated
in Table 4 and possessed different binding position as com-
pared to zileuton. Strong hydrophobic interactions were
Human ether-a-go-go related gene (hERG Kþ) potassium
channel involves in modulating electrical activity of the heart
that coordinates heart’s beating. Hence, the blockage of
hERG Kþ can cause fatal arrhythmia which is the main con-
cern in developing new therapeutic drugs. The compounds
formed between F5 and hydrophobic residues of 5-LOX were non-blocker of hERG Kþ channel according to the

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
9
Figure 4. Binding energy of rolipram and selected flavones in PDE4B (a) and PDE4D (b). XP Hbond: ChemScore H-bond pair term; XP PhobEn: hydrophobic enclos-
ure reward; XP LipophilicEvdW: lipophilic term derived from hydrophobic grid potential at the hydrophobic ligand atoms; XP Electro: electrostatic rewards which
includes Coulomb and metal terms.
results shown by two softwares (ADMETlab and Pred-hERG) showed toxicity in liver but may cause severe drug-induced
(Tables S1 and S2). The LD50 values for the acute toxicity
analyses using ADMETlab showed that F2 and F4 had mod-
erate toxicity while F1, F3, and F5 had low toxicity (Table
S1). Contrarily, ProTox-II had classified F2–F5 to class V
(2000 mg/kg < LD50 ꢄ 5000 mg/kg) and F1 to class IV
(300 mg/kg < LD50 ꢄ 2000 mg/kg) based on oral toxicity in
rats with high similarity and prediction accuracy (Table S3).
LD50 plots for each compound calculated using ACD/i-Lab
were also compared (Figure 6(a)). The degree of safety route
of administration from the safest in the compounds was
oral > subcutaneous > intraperitoneal > intravenous. F1 had
the highest LD50 whereas others had similar LD50. From toxi-
cological end point evaluation of compounds, the prediction
showed that all compounds did not cause immunotoxicity.
Ames test which determines the mutagenicity of compounds
was negative in all compounds despite had moderate prob-
liver injury (DILI), the frequent cause of drug withdrawal [43].
This was in line with the prediction of health effects of com-
pounds where F2 had the highest probability to affect liver
besides cardiovascular (CV) and gastrointestinal tract (GIT)
(Figure 6(b)). The compounds may also affect blood and kid-
ney while lung is the least affected organ.
Cytotoxicity
The cytotoxicity data revealed that F3–F5 were not toxic to
the cells at 100 mM (Table 6). Otherwise, F1 was moderate
toxic and F2 had the lowest IC50 among all compounds. The
result proved that the position of methoxy group influenced
the inhibitory properties of the compounds and affected
their cytotoxicity. The flavones with para- position of
ability to cause carcinogenicity. The compounds also did not methoxy exhibited less toxicity in comparison with meta-

10
M. H. MD IDRIS ET AL.
Figure 5. Binding mode of zileuton (a), F2 (b), F3 (c), F4 (d), and F5 (e) in 5-LOX active site.
Conclusions
position of methoxy. The incorporation of halogen group fur-
ther increases the toxicity of flavones (F2). The risk of toxic
potential of halogenated lead compounds had been dis-
cussed before and became a major concern in designing
drug [40,44].
The studied compounds (F1–F5) demonstrated different
degree of selectivity toward different inflammatory targets.
Amid these compounds, F2, F3, and F5 had potential as ini-
tial scaffold for further development of selective inhibition

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
11
Table 4. Binding energy of the zileuton and flavones in 5-LOX binding site.
for COX-2, 5-LOX, and PDE4B, respectively. The potency of
the compounds linked with the position of methoxy group
which need to be retained. However, the presence of iodine
may cause toxic effect as showed in F2. Thus, the replace-
ment of iodine with less reactive halogen group needs to be
considered to decrease the toxicity of the compounds. These
key features need to be emphasized in future optimization
of these promising scaffold.
Compound
Fitness score
Shb_ext
Svdw_ext
Shb_int
Sint
Zileuton
F2
F3
F4
F5
57.00
53.12
13.45
52.97
54.62
2
40.53
36.74
12.44
40.53
42.19
0
0
0
0
0
ꢂ0.74
ꢂ0.23
ꢂ5.09
ꢂ3.03
ꢂ3.39
0.08
1.16
0.27
0
Shb_ext: contribution of protein–ligand hydrogen bond to GoldScore value;
vdw_ext: contribution of van der Waals to GoldScore value; Shb_int: contribu-
S
tion of intracellular hydrogen bonds to GoldScore value; Sint: contribution of
intramolecular strain in the ligand to GoldScore value.
Table 6. Cytotoxicity of flavone derivatives.
Table 5. Drug-likeness properties of flavones derivatives using QikProp.
Compounds
IC50 (mM)
Compounds
Molecular weight
HB Acceptor
HB Donor
Log P
F1
F2
F3
F4
F5
78.61 1.329
16.02 1.165
>100
F1
F2
F3
F4
F5
282.295
408.191
252.269
331.165
282.295
4
4
3
3
4
0
0
0
0
0
3.477
4.082
3.469
4.231
3.477
>100
>100
Figure 6. LD50 of mouse (intraperitoneal, oral, intravenous, subcutaneous) graph plot for F1–F5 (a) and probability of health effects of F1–F5 on organs (b).

12
M. H. MD IDRIS ET AL.
[12]
[13]
Azam S, Ansari P, Rashid M, et al. In vitro anti-oxidant and in vivo
anti-inflammatory activity determination of the methanolic leaves
extract of Millettia pachycarpa. Biomed Res Ther. 2015;2:366–373.
Farzaei MH, Singh AK, Kumar R, et al. Targeting inflammation by
flavonoids: novel therapeutic strategy for metabolic disorders.
IJMS. 2019;20(19):4957.
Acknowledgments
The authors would also like to thank Integrative Pharmacogenomics
Institute (iPROMISE), UiTM (Grant no. 241910/2015/MZS/7) for providing
financial assistance and facilities to carry out related research work. The
in vitro synthesis steps, computational modelled data and efficacy of the
compounds as anti-inflammatory agents had been submitted for patent [14]
application (Application no. PI 2018704123).
Barik R, Sarkar R, Biswas P, et al. 5,7-Dihydroxy-2-(3-hydroxy-4, 5-
dimethoxy-phenyl)-chromen-4-one-a flavone from Bruguiera gym-
norrhiza displaying anti-inflammatory properties. Indian
Pharmacol. 2016;48(3):304–311.
J
[15]
Dzoyem JP, Nkuete AHL, Ngameni B, et al. Anti-inflammatory and
anticholinesterase activity of six flavonoids isolated from
Polygonum and Dorstenia species. Arch Pharm Res. 2017;40(10):
1129–1134.
Disclosure statement
No potential conflict of interest was declared by the authors.
[16]
Chen L, Teng H, Xie Z, et al. Modifications of dietary flavonoids
towards improved bioactivity: an update on structure-activity
relationship. Crit Rev Food Sci Nutr. 2018;58(4):513–527.
Sharma A, Sharma P, Tuli H, et al. Phytochemical and pharmaco-
logical properties of flavonols. In: eLS. Chichester: John Wiley &
Sons, Ltd.; 2018. p. 1–12.
Funding
This work was supported by Ministry of Higher Education (MOHE),
[17]
Malaysia under Transdisciplinary Research Grant Scheme [Grant no. 600-
RMI/TRGS 5/3 (1/2014)].
[18]
Yusof MIM, Salleh MZ, Kek TL, et al. Activity-guided isolation of
bioactive constituents with antinociceptive activity from
Muntingia calabura L. leaves using the formalin test. Evidence-
based complement. Altern. Med. 2013;2013:1–27.
ORCID
[19]
Serafini M, Peluso I, Raguzzini A. Flavonoids as anti-inflammatory
agents. Proc Nutr Soc. 2010;69(3):273–278.
Gaddipati RS, Raikundalia GK, Mathai ML. Dual and selective lipid
Muhd Hanis Md Idris
Siti Norhidayah Mohd Amin
Zainul Amiruddin Zakaria
http://orcid.org/0000-0001-8246-6493
http://orcid.org/0000-0002-3382-0019
http://orcid.org/0000-0001-5525-7821
http://orcid.org/0000-0002-0314-8532
http://orcid.org/0000-0002-6826-1659
http://orcid.org/0000-0001-8604-8598
http://orcid.org/0000-0001-9882-0661
[20]
[21]
[22]
inhibitors of cyclooxygenases and lipoxygenase:
a molecular
Zurina Shaameri
docking study. Med Chem Res. 2014;23:1–14.
Ahmad Sazali Hamzah
Manikandan Selvaraj
Mohd Zaki Salleh
Tripuraneni NS, Azam MA. Pharmacophore modeling, 3D-QSAR
and docking study of 2-phenylpyrimidine analogues as selective
PDE4B inhibitors. J Theor Biol. 2016;394:117–126.
Cole JC, Nissink JWM, Taylor R. Protein-ligand docking and virtual
screening with GOLD. In: Alvarez J, Shoichet B, editors. Virtual
screening in drug discovery. 1st ed. Florida: CRC Press; 2005. p.
388–424.
Jones G, Willett P, Glen RC, et al. Development and validation of
a genetic algorithm for flexible docking. J Mol Biol. 1997;267(3):
727–748.
Berman HM, Westbrook J, Feng Z, et al. The protein data bank.
Nucleic Acids Res. 2000;28(1):235–242.
Md Idris MH, Mohd Amin SN, Selvaraj M, et al. High-throughput
structure-based drug design of chalcones scaffolds as dual inhibi-
tor of cyclooxygenase-2 and microsomal prostaglandin E syn-
thase-1. J Pharm Sci Emerg Drugs. 2018;6:1–14.
Shivakumar D, Williams J, Wu Y, et al. Prediction of absolute solv-
ation free energies using molecular dynamics free energy per-
turbation and the OPLS force field. J Chem Theory Comput. 2010;
6(5):1509–1519.
Friesner RA, Banks JL, Murphy RB, et al. Glide: a new approach
for rapid, accurate docking and scoring. 1. Method and assess-
ment of docking accuracy. J Med Chem. 2004;47(7):1739–1749.
Kutil Z, Kvasnicova M, Temml V, et al. Effect of dietary stilbenes
on 5-lipoxygenase and cyclooxygenases activities in vitro. Int J
Food Prop. 2015;18(7):1471–1477.
Wisastra R, Kok PA, Eleftheriadis N, et al. Discovery of a novel
activator of 5-lipoxygenase from an anacardic acid derived com-
pound collection. Bioorg Med Chem. 2013;21(24):7763–7778.
Pufahl RA, Kasten TP, Hills R, et al. Development of a fluores-
cence-based enzyme assay of human 5-lipoxygenase. Anal
Biochem. 2007;364(2):204–212.
References
[1]
Chen L, Deng H, Cui H, et al. Inflammatory responses and inflam-
mation-associated diseases in organs. Oncotarget. 2018;9(6):
7204–7218.
[23]
[2]
Martel-Pelletier J, Lajeunesse D, Reboul P, et al. Therapeutic role
of dual inhibitors of 5-LOX and COX, selective and non-selective
non-steroidal anti-inflammatory drugs. Ann Rheum Dis. 2003;
62(6):501–509.
[24]
[25]
[3]
[4]
[5]
[6]
Kaminska B. MAPK signalling pathways as molecular targets for
anti-inflammatory therapy-from molecular mechanisms to thera-
peutic benefits. Biochim Biophys Acta. 2005;1754(1–2):253–262.
Zheng XY, Chen JC, Xie QM, et al. Anti-inflammatory effect of
ciclamilast in an allergic model involving the expression of
PDE4B. Mol Med Rep. 2019;19:1728–1738.
Brune K, Patrignani P. New insights into the use of currently
available non-steroidal anti-inflammatory drugs. J Pain Res. 2015;
8:105–118.
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
ꢀ
ꢀ
Khan YS, Gutierrez-de-Teran H, Åqvist J. Molecular mechanisms in
the selectivity of nonsteroidal anti-inflammatory drugs.
Biochemistry. 2018;57(7):1236–1248.
[7]
[8]
Li H, Zuo J, Tang W. Phosphodiesterase-4 inhibitors for the treat-
ment of inflammatory diseases. Front Pharmacol. 2018;9:1048.
Bruno F, Spaziano G, Liparulo A, et al. Recent advances in the
search for novel 5-lipoxygenase inhibitors for the treatment of
asthma. Eur J Med Chem. 2018;153:65–72.
[9]
Tungmunnithum D, Thongboonyou A, Pholboon A, et al.
Flavonoids and other phenolic compounds from medicinal plants
for pharmaceutical and medical aspects: an overview. Medicines.
2018;5(3):16–93.
Promega. 2016. PDE-GloTM Phosphodiesterase Assay. https://
www.promega.com/-/media/files/resources/protocols/technical-
bulletins/101/pde-glo-phosphodiesterase-assay-protocol.pdf.
Dong J, Wang NN, Yao ZJ, et al. ADMETlab: a platform for sys-
tematic ADMET evaluation based on a comprehensively collected
ADMET database. J Cheminform. 2018;10(1):29.
[10]
[11]
Verma AK, Pratap R. The biological potential of flavones. Nat
Prod Rep. 2010;27(11):1571–1593.
ꢀ
Perez-Cano FJ, Castell M. Flavonoids, inflammation and immune
€
QikProp. Schrodinger Release 2017-1: QikProp. New York (NY):
€
Schrodinger, LLC; 2017.
system. Nutrients. 2016;8(10):659.

JOURNAL OF RECEPTORS AND SIGNAL TRANSDUCTION
13
[34]
Banerjee P, Eckert AO, Schrey AK, et al. ProTox-II: a webserver for [39]
the prediction of toxicity of chemicals. Nucleic Acids Res. 2018;
46(W1):W257–W263.
Masunov A. ACD/I-Lab 4.5: an internet service review. J Chem Inf
Comput Sci. 2001;41(4):1093–1095.
Xu Z, Yang Z, Liu Y, et al. Halogen bond: its role beyond drug-
target binding affinity for drug discovery and development. J
Chem Inf Model. 2014;54(1):69–78.
Hernandes MZ, Cavalcanti SM, Moreira DR, et al. Halogen atoms
in the modern medicinal chemistry: hints for the drug design.
Curr Drug Targets. 2010;11:1–12.
Lipinski CA. Lead- and drug-like compounds: the rule-of-five revo-
lution. Drug Discov Today Technol. 2004;1(4):337–341.
Zhang Z, Tang W. Drug metabolism in drug discovery and devel-
opment. Acta Pharm Sin B. 2018;8(5):721–732.
Xu Y, Dai Z, Chen F, et al. Deep learning for drug-induced liver
injury. J Chem Inf Model. 2015;55(10):2085–2093.
[40]
[35]
[36]
Braga RC, Alves VM, Silva MFB, et al. Pred-hERG: a novel web-
accessible computational tool for predicting cardiac toxicity. Mol
Inform. 2015;34(10):698–701.
Meerloo J, Van Kaspers GJL, Cloos J. Cell sensitivity assays: the
MTT assay. In: Cree IA, editor. Cancer Cell Cult. 2nd ed. London:
Humana Press; 2011. p. 237–245.
Makkar F, Chakraborty K. First report of dual cyclooxygenase-2
and 5-lipoxygenase inhibitory halogen derivatives from the thal-
lus of intertidal seaweed Kappaphycus alvarezii. Med Chem Res.
2018;27(10):2331–2340.
[41]
[42]
[43]
[44]
[37]
[38]
Kortagere S, Ekins S, Welsh WJ. Halogenated ligands and their
interactions with amino acids: implications for structure-activity
and structure-toxicity relationships. J Mol Graph Model. 2008;27:
70–177.