Simple heteroaryl modifications in the 4,5-diarylisoxazol-3-carboxylic acid scaffold favorably modulates the activity as dual mPGES-1/5-LO inhibitors with in vivo efficacy

Article abstract of DOI:10.1016/j.bioorg.2021.104861

Microsomal prostaglandin E₂ synthase-1 (mPGES-1), 5-lipoxygenase (5-LO) and 5- lipoxygenase-activating protein (FLAP) are key for biosynthesis of proinflammatory lipid mediators and pharmacologically relevant drug targets. In the present study, we made an attempt to explore the role of small heteroaromatic fragments on the 4,5-diarylisoxazol-3-carboxylic acid scaffold, which are selected to interact with focused regions in the active sites of mPGES-1, 5-LO and FLAP. We report that the simple structural variations on the benzyloxyaryl side-arm of the scaffold significantly influence the selectivity against mPGES-1, 5-LO and FLAP, enabling to produce multi-target inhibitors of these protein targets, exemplified by compound 18 (IC₅₀ mPGES-1 = 0.16 μM; IC₅₀ 5-LO = 0.39 μM) with in vivo efficacy in animal model of inflammation. The computationally modeled binding structures of these new inhibitors for three targets provide clues for rational design of modified structures as multi-target inhibitors. In conclusion, the simple synthetic procedure, and the possibility of enhancing the potency of this class of inhibitors through structural modifications pave the way for further development of new multi-target inhibitors against mPGES-1, 5-LO and FLAP, with potential application as anti-inflammatory agents.

Full text of DOI:10.1016/j.bioorg.2021.104861

Bioorganic Chemistry 112 (2021) 104861
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Simple heteroaryl modifications in the 4,5-diarylisoxazol-3-carboxylic acid
scaffold favorably modulates the activity as dual mPGES-1/5-LO inhibitors
with in vivo efficacy
a
a
b
a
c
Tu g˘ ba Gürses , Abdurrahman Ol g˘ aç , Ulrike Garscha , Tu g˘ çe Gür Maz , Nur Banu Bal ,
c
a
d
,
e
b
,
d
a,*
Orhan Uluda g˘ , Burcu Çalı s¸ kan , Ulrich S. Schubert , Oliver Werz , Erden Banoglu
a
Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Gazi University, Yenimahalle, 06560 Ankara, Turkey
b
Department of Pharmaceutical/Medicinal Chemistry, Institute of Pharmacy, Friedrich-Schiller-University Jena, Philosophenweg 14, D-7743 Jena, Germany
Department of Pharmacology, Faculty of Pharmacy, Gazi University, Yenimahalle, 06560 Ankara, Turkey
c
d
e
Jena Center for Soft Matter (JCSM), Friedrich Schiller University Jena, Philosophenweg 7, 07743 Jena, Germany
Laboratory of Organic and Macromolecular Chemistry (IOMC), Friedrich Schiller University Jena, Humboldtstrasse 10, D-07743 Jena, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords:
Microsomal prostaglandin E2 synthase-1 (mPGES-1), 5-lipoxygenase (5-LO) and 5- lipoxygenase-activating
protein (FLAP) are key for biosynthesis of proinflammatory lipid mediators and pharmacologically relevant
drug targets. In the present study, we made an attempt to explore the role of small heteroaromatic fragments on
the 4,5-diarylisoxazol-3-carboxylic acid scaffold, which are selected to interact with focused regions in the active
sites of mPGES-1, 5-LO and FLAP. We report that the simple structural variations on the benzyloxyaryl side-arm
of the scaffold significantly influence the selectivity against mPGES-1, 5-LO and FLAP, enabling to produce multi-
target inhibitors of these protein targets, exemplified by compound 18 (IC50 mPGES-1 = 0.16 µM; IC50 5-LO =
2
Microsomal prostaglandin E synthase-1
5
5
-lipoxygenase
-lipoxygenase-activating protein
Isoxazole
Inflammation
0
.39 µM) with in vivo efficacy in animal model of inflammation. The computationally modeled binding structures
of these new inhibitors for three targets provide clues for rational design of modified structures as multi-target
inhibitors. In conclusion, the simple synthetic procedure, and the possibility of enhancing the potency of this
class of inhibitors through structural modifications pave the way for further development of new multi-target
inhibitors against mPGES-1, 5-LO and FLAP, with potential application as anti-inflammatory agents.
1
. Introduction
Drug discovery efforts on effective relief of chronic inflammatory
the center of extensive research efforts for next-generation NSAIDs with
the aim of circumventing COX-related side effects [3,4]. mPGES-1 is the
terminal enzyme in the arachidonic acid (AA) cascade, which is func-
tionally coupled to COX-2, and catalyzes the transformation of prosta-
conditions including pain have been an ongoing challenge for medicinal
chemists for many years. Traditionally, cyclooxygenase (COX)-inhibit-
ing non-steroidal anti-inflammatory drugs (NSAIDs) represent an
important therapeutic class for the alleviation of pain and inflammation
associated with a number of pathologies such as arthritis, Alzheimer’s
disease, atherosclerosis, and cancer. However, the prolonged utilization
of these drugs may limit their therapeutic benefits due to severe
gastrointestinal or cardiovascular side effects [1,2]. Therefore, there is
still growing interest for alternative pharmacological approaches for the
discovery of improved anti-inflammatory drugs devoid of the side effects
inherent to traditional NSAIDs.
2 2
glandin (PG)H to pro-inflammatory PGE (Fig. 1) [5]. Since mPGES-1
as well as COX-2 is strongly up-regulated under inflammatory condi-
tions, its inhibition would selectively interfere with the production of
2
mPGES-1-derived pro-inflammatory PGE while keeping the biosyn-
thesis of house-keeping PGs unaffected (Fig. 1). To date, although a
broad spectrum of in vitro SAR data on numerous classes of mPGES-1
inhibitors has been reported, only a limited number of mPGES-1 in-
hibitors are available with in vivo analgesic and anti-inflammatory ef-
ficacy in animal models [5,6]. In addition, two mPGES-1 inhibitors,
LY3023703 (1) by Eli Lilly and GRC-27864 (2) from Glenmark Phar-
maceuticals, completed Ph-I clinical studies for the treatment of
The microsomal prostaglandin E
2
synthase-1 (mPGES-1) has been in
*
Corresponding author.
E-mail address: banoglu@gazi.edu.tr (E. Banoglu).
https://doi.org/10.1016/j.bioorg.2021.104861
Received 4 October 2020; Received in revised form 2 January 2021; Accepted 21 March 2021
Available online 24 March 2021
0
045-2068/© 2021 Elsevier Inc. All rights reserved.

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
inflammation and pain [7–9]. Therefore, selective inhibition of mPGES-
also demonstrated that this quinolinyl derivative (8b) was able to
1
preserves its status to be a promising approach for the design of
2
intervene with PGE production by targeting mPGES-1, therefore
effective anti-inflammatory drugs lacking NSAID related side effects.
Besides COXs and mPGES-1, 5-lipoxygenase (5-LO) and 5-LO-acti-
vating protein (FLAP) are intensively studied therapeutic targets,
which are key for production of leukotrienes (LTs) from AA, and
involved in inflammatory and allergic processes (Fig. 1) [10,11]. 5-LO in
conjunction with FLAP catalyzes the initial step of LT biosynthesis by
resulting in a multitarget compound within pro-inflammatory lipid
mediator biosynthesis. Encouraged by the promising in vitro pharma-
cological profile of this multitarget inhibitor 8b (BRP-187) [25], we
herein randomly explored the effectiveness of various commercially
available heteroaromatic residues in place of quinoline ring in BRP-187
to better understand the role of this part with regard to the activity
tangling among the target proteins mPGES-1, 5-LO and FLAP (see
Table 1).
converting AA to instable LTA
chemotactic LTB and vasoactive cysteinyl-LTs by the action of LTA
hydrolase and LTC synthase, respectively. Preclinical and a few clinical
4
, which in turn is used to produce
4
4
4
studies on the inhibition of LT pathway have demonstrated several
beneficial pharmacological effects, such as suppression of inflammation
and allergy-induced bronchoconstriction [11]. Although a vast number
of 5-LO inhibitors have been developed for anti-inflammatory purposes,
they have not reached the pharmaceutical market for reasons such as
lack of selectivity and mechanism-based side effects. So far, zileuton (3)
is still the only compound, which is approved as drug for treatment of
asthma. In this respect, FLAP inhibitors as broad-spectrum anti-LT
agents are more advanced, and several of them, i.e., fiboflapon (4,
GSK2190915), AZD5718 (5) and BI665915 (6) were reported to be in
various stages of preclinical and clinical studies for inflammation-
related diseases such as asthma, chronic obstructive pulmonary dis-
ease (COPD) and atherosclerosis (Fig. 2) [12–14].
2
. Results and discussion
2
.1. Chemistry
Synthesis of compounds 17–26 was performed by a multi-step re-
action (Scheme 1) according to the previously published procedures
[
24]. In brief, commercially available p-hydroxyacetophenone was
protected by benzylation (1) prior to the reaction with diethyl oxalate to
generate keto-enol ester 2. The obtained keto-enol ester 2 was refluxed
with hydroxylamine in ethanol to produce isoxazole intermediate 3,
which was brominated at 4-position (4) with N-bromosuccinimide in the
presence of a catalytic amount of ceric ammonium nitrate. The resulting
brominated derivative underwent a palladium-catalyzed Suzuki cross-
coupling reaction with p-chlorophenylboronic acid to afford 5. Deben-
zylation of 5 by catalytic hydrogenation furnished the starting inter-
mediate 6, which was subsequently used to produce desired final
compounds 17–26 through first the alkylation of the phenolic hydroxyl
and then the hydrolysis of the ester group.
Meanwhile, single therapy with COX-inhibiting NSAIDs was shown
4
to elevate levels of chemotactic LTB , which also contributes to the in-
flammatory process and also to the gastric toxicity observed with
NSAIDs use [15]. Several lines of studies indicated that dual inhibitors,
which are able to block equally well both the COX and 5-LO pathways
might have superior anti-inflammatory activity and more gastrointes-
tinal safety as compared to the therapy with single inhibitors of each
pathway [15]. Indeed, licofelone (7), a balanced inhibitor of mPGES-1,
FLAP and COX-1, have demonstrated higher anti-inflammatory activity
with greater safety profile and has been evaluated in phase-III clinical
studies [16], indicating that multitarget strategy in the AA pathway
might be the most promising therapeutic solution for inflammatory
diseases.
2.2. Analysis of bioactivity and SAR
To assess the inhibitory potential of the title compounds on mPGES-1
2 2
activity (transformation of PGH to PGE ), a cell-free assay using the
microsomal fractions of interleukin (IL)-1β-stimulated A549 cells as the
enzyme source was applied [26]. However, previous studies on various
mPGES-1 inhibitors demonstrated that they may potentially interact
with 5-LO or FLAP, therefore producing compounds with multitarget
properties within the AA cascade [26–30]. In addition, it is suggested
Within the frame of our research efforts towards the development of
better anti-inflammatory drug candidates targeting 5-LO, FLAP and
mPGES-1 [17–23], we previously identified novel dual inhibitors of PG
and LT pathway based on the core structure of 4,5-diarylisoxazole-3-car-
boxylic acid (8) targeting FLAP and 5-LO [24,25]. Our initial studies
disclosed that the installation of various substituted phenyl or quinoline
rings as R groups on 8 tangled the selectivity of this core structure be-
tween 5-LO (8a) and FLAP (8b) (Fig. 2) [24]. Follow up studies with 8b
that simultaneous interference with the formation of both LTs and PGE
2
might offer added value over single interference with any of these tar-
gets in terms of higher anti-inflammatory efficacy and with reduced side
effects, i.e., the almost complete lack of gastric toxicity, the most trou-
blesome side effect of NSAIDs [15,31]. Biosynthesis of LTs is initiated by
2 2
cytosolic phospholipase A (cPLA )-mediated release of AA, which is
Fig. 1. Schematic illustration of the arachidonic acid cascade. Targeted enzymes (mPGES-1 and 5-LO/FLAP) are indicated.
2

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
Fig. 2. Representative chemical structures of FLAP, mPGES-1 and multi-target inhibitors.
human neutrophils. Analyzed 5-LO products include 5-H(P)ETE and all
trans-isomers of LTB , as well as LTB in intact cells. MK-886 with an
IC50 = 2 M (cell-
Table 1
4
4
Inhibition of mPGES-1 and 5-LO by 4,5-diarylisoxazole-3-carboxylic acid
μ
M for mPGES-1 [35] and zileuton with an IC50 = 0.58 μ
based) and 0.80 (cell-free) for 5-LO [24] used as reference drugs.
In a first screening round, all compounds were tested against mPGES-
derivatives.
1
at a concentration of 10
μM. As shown in Table 1, compounds 17–19
and 26 profoundly inhibited mPGES-1 activity, however all other de-
rivatives (20–25) were not significantly active at a concentration of 10
mPGES-1
r.a. @ 10
mPGES-1
IC50 M]
IC50
[ M]
μ
μM
[
μ
μ
M. More detailed analysis of 17–19 and 26 in concentration–response
R
PMNL
Purified 5-
LO
(
%)
studies revealed IC50 values in the range of 0.16–0.25
μM (Table 1),
which clearly outperforms MK886, indicating that the heteroaryl frag-
ments might have a strong influence for directing the mPGES-1 inhibi-
tory activity in this scaffold. Intriguingly, among these heteroaryl-
1
7
8
44.7 ± 4.0
34.4 ± 0.8
0.25 ±
.04
0.16 ±
.02
0.3 ±
2.8 ± 0.8
0
0.07
1
2.7 ±
0.39 ± 0.1
0
0.5
substituted compounds (17–26), the 5-CF
most active mPGES-1 inhibitor (IC50 = 0.16
potent 5-LO inhibitor with IC50 = 0.39 M, demonstrating the most
favorable dual inhibitor pharmacological profile with sub-micromolar
activity. In regard to 5-LO inhibition, installation of 6-CH -pyridine
17), benzotriazole (19), benzodioxol (25) and 5-Cl-benzothiophene
26) as heteroaromatic counterparts led to decreased activities against
-LO enzyme (IC50 = 1.5 - >10 M), while keeping the potency against
mPGES-1 (IC50 = 0.16–0.25 M). Among them, the benzotriazole 19
behaved as a selective mPGES-1 inhibitor with an IC50 of 0.16
3
-furan-substituted 18, the
μM), was also the most
1
2
9
0
39.6 ± 4.9
68.4 ± 8.6
0.16 ±
.07
>10
>10
>10
μ
0
>10
3.1 ± 0.1
3
(
(
2
1
90.8 ± 2.8
85.0 ± 6.2
>10
0.8 ±
>10
5
μ
0
.1
μ
2
2
2
3
>10
>10
2.2 ±
.8
>10
>10
μM
0
(Table 1). Surprisingly, replacement of benzotriazole in 19 with 1-pro-
pyl-imidazopyridine resulted in a derivative (20) with selective inhibi-
tory activity against isolated 5-LO. Another interesting observation
102.5 ± 9.0
101.8 ± 8.3
>10
made with the replacement of 6-CH
3
-pyridin-2-yl in 17 (dual mPGES-1/
2
4
>10
2.7 ±
5.1 ± 3.3
5
-LO inhibitor) with 2-CH -pyridin-3-yl resulting in an analogue 24 with
3
0
.3
selectivity against 5-LO pathway. A significant and concentration-
dependent suppression of cellular 5-LO product biosynthesis was
2
2
5
6
43.7 ± 6.0
24.8 ± 1.6
3.1 ± 0.3
0.21 ±
1.7 ±
.3
2.5 ± 1.2
1.5 ± 0.8
0
found for 1-ethyl-pyrazole 21 (IC50 = 0.4
μ
M) and also for 1,3-dimethyl-
1.1 ±
pyrazole 22 (IC50 = 2.2 M), although these compounds hardly inhibi-
μ
0
.07
0.6
ted 5-LO and mPGES-1 in the cell-free assay. This suggests that for
suppression of 5-LO product formation in intact cells, 21 and 22 may
primarily act at other targets than the 5-LO enzyme, presumably on
FLAP.
MK886
–
–
2.2
0.08
–
0.6
Zileuton
–
–
subsequently converted to LTA
4
by activated 5-LO with the aid of FLAP
All in all, based on the observed biological activity data, 18 is the
most efficient dual inhibitor of mPGES-1 and 5-LO activities, 21 and 22
may act as direct FLAP inhibitors, while 19 and 20 behave as potent and
selective mPGES-1 and 5-LO inhibitors, respectively. Besides 18, com-
pounds 17, 25 and 26 also act as dual inhibitors of mPGES-1 and 5-LO
activity, although with a moderate inhibitory activity on 5-LO. Hence,
we aimed to rationalize the biological results through molecular
that may facilitate access of 5-LO to AA (Fig. 1) [32]. It is known that
FLAP is essential for cellular 5-LO product formation from endogenous
AA, but it is dispensable for 5-LO activity in cell-free systems [33,34].
Therefore, we also analyzed the inhibition of 5-LO product formation in
a “FLAP-independent” cell-free assay using purified human recombinant
5
-LO enzyme as well as in a “FLAP-dependent” cell-based assay using
3

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
Scheme 1. Reagents and conditions: (a) Benzylchloride, K
2
CO3, MeCN, reflux; (b) (CO
2
Et)
2
, Na, EtOH, rt; (c) NH
2
OH.HCl, EtOH, reflux; (d) NBS, CAN, AcCN; (e) 4-
Chlorophenylboronic acid, Pd(PPh
LiOH⋅H O,THF:H O (1:1).
3
)
2
Cl
2
, Na
2
CO
3
, 1,4-Dioxane, H
2
2 2 3
O; (f) Pd/C, H , HCl, EtOH:THF (3:2); (g) appropriate halides, K CO , MeCN, reflux; (h)
2
2
modelling studies.
interactions but divergent and inconclusive for less potent derivatives.
The binding specificity of the active compounds for each of their
corresponding protein was conferred by taking into consideration of the
binding interactions as described in the following for each target. For
instance, while carboxylate, p-chlorophenyl and isoxazole-5-phenyl
moieties of 17–19, 25 and 26 shared similar binding interactions at
the mPGES-1 active site, different heteroaryl fragments generated slight
changes in their binding modes. While isoxazole-5-phenyl core of the
compounds are found facing through the membrane location, carbox-
ylate fragments make ionic interactions with ARG52 and HIS53 amino
acids, p-chlorophenyl moieties form Van der Waals interactions with
2
2
.3. Molecular modeling studies
.3.1. Docking studies
Molecular docking studies were conducted on the reported crystal
structures of mPGES-1 (PDB code 5TL9 [36]), 5-LO (PDB code 3O8Y
37]) and FLAP (PDB code 2Q7M [38]) for all the molecules in this study
[
to reveal the potential molecular interactions between ligands and
proteins. The interaction regions within the active site can be divided
into four sub-pockets for these series of molecules, i.e., one hydrophilic
and three hydrophobic sub-pockets, and docking-based common inter-
action analysis against mPGES-1, 5-LO, and FLAP is summarized in
Table 2. It should be noted that the isoxazole-5-phenyl group is facing to
the lipophilic tail of membrane for mPGES-1, which is not included in
the docking studies. The predicted binding modes of the compounds
with mPGES-1 (Fig. 3), 5-LO (Fig. 4) and FLAP (Fig. 5 and S1) reveal
that the head carboxylate groups form electrostatic interactions (i.e., salt
bridges) with basic amino acids, and the chlorophenyl moieties are
LEU39 and
π
–
π
stacking with HIS53 (Fig. 3). In addition, compounds
1
7–19 and 26 generated only hydrophobic contacts in their related sub-
pocket, while compound 25 formed a new hydrogen bond with GLN134
that slightly weakened the hydrophobic interaction network in that re-
gion resulting in an undesired decrease to the compound’s inhibitory
effect for mPGES-1.
For 5-LO, compounds 17, 18, 24–26 make ionic interactions to
HIS600 with their carboxylate moiety and share the same interaction
network with p-chlorophenyl group with residues TYR181, LEU414 and
ALA603. Meanwhile, the isoxazole-5-phenyl core makes Van der Waals
interactions with alkyl chains of the surrounding amino acids and
hydrogen bond with ASN425, heteroaryl fragments show minor differ-
ences in their binding orientations. Although these fragments mainly
directed to a hydrophobic groove making hydrophobic contacts,
π π
-
stacking and van der Waals (vdW) interactions. However, small struc-
tural variations at the heteroaryl part apparently have complex conse-
quences for compounds with distinct selectivity. Positioning the
heteroaromatic moieties in a correct orientation is beneficial to establish
the required hydrophobic contacts as well as filling the binding pocket of
three proteins. Accordingly, the predicted binding modes of the active
compounds were similar with regard to the observed sub-pocket
form hydrophobic contacts with the surrounding residues, 5-CF
part in 18 showed an orthogonal multipolar interaction with GLN363
Fig. 4).
Most of the interaction network with carboxylate and p-chlorophenyl
moieties were also shared between the compounds 21–22 and FLAP
Fig. 5 and S1). For example, while carboxylate groups form ionic
3
-furan
(
Table 2
(
Molecular docking-based common multi-target interaction analysis of 4,5-diary-
lisoxazole-3-carboxylic acid derivatives.
interaction with LYS116 and isoxazole-5-phenyl core forms hydropho-
bic interactions with ILE113, p-chlorophenyl moiety makes Van der
Waals interactions with LYS116 and LEU120 amino acids. Heteroaryl
fragments form hydrophobic interactions with surrounding amino acids
(Fig. 5). The docking runs for compounds 20 and 23, which contain o-
substituents on the heteroaryl moiety, were failed and did not generate
any docking result for all targets, due to their increased volume that
sterically blocked fitting the compounds while keeping the proteins’
active site rigid. Additionally, an interaction matrix together with the
IC50 values is given for the reported compounds and targets in the
supporting information section (Tables S1, S2 and S3). Hence, docking
studies demonstrates the binding specificity of each compound for its
interacting protein supporting the hypothesis that heteroaryl fragments
might help to identify target specific or dual inhibitors.
Fragment
Target
mPGES-1
5-LO
FLAP
Interaction Type
Carboxylate
ARG52
HIS53
LEU39
HIS600
HIS28
Ionic Interaction
Hydrophilic P.
p-chlorophenyl
Hydrophobic-1P.
LYS116
LYS116
LEU120
Membrane*
LEU414
ALA603
vdW Interaction
HIS53
TYR181
GLN363
PHE421
ASN425
LEU607
LEU368
ALA410
HIS372
π – π Stacking
Isoxazole-5-phenyl
Membrane
ILE113
vdW Interaction
vdW Interaction
Hydrophobic-2P.
Heteroaryl
ILE32
VAL61
ALA63
TYR112
Hydrophobic-3P.
TYR130
GLN134
*
not present in docking study.
4

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
Fig. 3. Three-dimensional model of interactions of 17 (A), 18 (B), 19 (C), 25 (D), and 26 (E) with the mPGES-1 (PDB code 5TL9) binding site. Sticks in orange
represents ligand, white represents chain 1, green represents chain 2, blue represents GSH. Legend of protein ꢀ ligand interactions as identified by PLIP 1.4.4 [39]
and visualized with PyMOL 2.3 [40] (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
2
.3.2. Molecular dynamic studies
considering ligand-induced conformational changes of mPGES-1 and 5-
LO. Each copy of the molecular simulation showed similar interactions
during the simulations (Fig. 6). The rationalization of the 5-LO binding
mode was obtained considering the fundamental amino acids in the
active site of both enzymes as described previously [24,42–44].
The results indicate that the interaction network obtained by 100 ns
of MD simulations with four copies revealed new important interactions,
which can unveil the potency of 18. Some of the interactions, which
were observed in docking simulations, were not maintained (<10%),
and some additional interactions were observed during the simulations.
As a result of MD studies, we observe that 18 perfectly fits into the active
sites of both proteins. However, compared to the docking results, the
ionic contacts with HIS53 of mPGES-1 and HIS600 of 5-LO are not
maintained, and the carboxyl group is stabilized by ionic interactions
with ARG52 of mPGES-1 and ASN425 of 5-LO as well as water-mediated
hydrogen bonding with ARG52 and SER127 of mPGES-1 and TYR181,
ALA424 and TRP599 of 5-LO (Fig. 7, Table 3). In addition, the formation
Since compound 18 with 5-CF
3
-furan side arm demonstrated note-
M) and 5-
M), the binding interactions of 18 with both proteins
worthy dual inhibitor potency for both mPGES-1 (IC50 = 0.16
LO (IC50 = 0.39
μ
μ
were further evaluated by molecular dynamic (MD) simulations (100 ns)
to take into account potential ligand-induced conformational changes.
Since the available mPGES-1 x-ray structure (PDB code 5TL9 [36]) does
not include membrane coordinates, we analyzed the binding modes of
mPGES-1/18 complex by means of MD simulations by including mem-
brane (taken from OPM Database [41]) in the simulation systems. By
inclusion of the membrane bilayer, it was possible to observe the
membrane’s effect on the binding in which the adopted techniques have
been widely used for the analysis of other membrane-embedded pro-
teins. We used the human 5-LO crystal structure (PDB code 3O8Y [37]),
for rationalizing the inhibitory activity of 18 by means of molecular
docking and MD simulations to 5-LO. We combined docking studies with
four copies of MD simulations to investigate the binding modes of 18
5

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
Fig. 4. Three-dimensional model of interactions of 17 (A), 18 (B), 24 (C), 25 (D), and 26 (E) with the 5-LO (PDB code 3O8Y) binding site. Sticks in orange represents
ligand, white represents residues in chain 1. Legend of protein ꢀ ligand interactions as identified by PLIP 1.4.4 [39] and visualized with PyMOL 2.3 [40] (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
of
π
-
π
interactions of 5-CF
3
-furan with TYR130 of mPGES-1, and HIS367
model of inflammation that is widely used for screening novel anti-
inflammatory compounds. The inflammatory response is usually quan-
tified by an increase in paw size (edema) which is maximal around 5 h
post-carrageenan injection and is known to be modulated by specific
inhibitors of the AA inflammatory cascade. The results demonstrate that
a significant reduction of paw edema from 90 to 360 min as compared to
control group was observed in animals dosed with both the dual mPGES-
1/5-LO inhibitor 18 (10 and 30 mg/kg) and the non-selective COX in-
hibitor indomethacin (10 mg/kg) used as reference control (Fig. 8).
These results indicate that the anti-inflammatory effectiveness of 18 was
comparable or even superior over the reference indomethacin. More-
over, 18 was further investigated for gastric toxicity upon acute
administration to determine the ulcerogenic potential at the time in-
terval in which the compound showed the highest anti-inflammatory
activity (270 min). At doses of 10 and 30 mg/kg, 18 caused weak
damage to the gastric mucosa (ratio of ulceration; control (DMSO and
saline) 0/6; Indo 4/6 at 10 mg/kg; 18 1/6 at 10 mg/kg, and 2/6 at 30
and HIS372 of 5-LO along with hydrogen bonds with THR364, ARG411
and orthogonal multipolar interactions with ASN407 and ALA410 resi-
dues clearly help to stabilize 18 at the active site of both proteins.
Moreover, p-chlorophenyl fits to the hydrophobic groove by making
additional hydrophobic contacts with TYR130 of mPGES-1 and TYR181
of 5-LO, which is a gatekeeper amino acid along with PHE177 at the
active site entrance of 5-LO. All the observed interactions with the oc-
cupancy values equal or higher than 10% were shown in the MD
interaction figures (Fig. 7) and summarized in Table 3.
2
.4. In vivo efficacy of compound 18 in animal model of inflammation
In view of the high potency of 18 for suppression of both PGE and LT
2
biosynthesis, we attempted to evaluate the anti-inflammatory effec-
tiveness in vivo in the carrageenan-induced hind paw edema model in
mice [45]. The carrageenan-induced paw edema is a well-known acute
6

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
Fig. 5. Three-dimensional model of interactions of 21 (A), and 22 (B) with the FLAP (PDB code 2Q7M) binding site. Sticks in orange represents ligand, white
represents residues in chain 1, green represents residues in chain 2. Legend of protein ꢀ ligand interactions as identified by PLIP 1.4.4 [39] and visualized with
PyMOL 2.3 [40] (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 6. Root-mean-square deviation (RMSD) values of 100 ns MD simulations of compound 18 fit to mPGES-1 (A) and 5-LO (B) backbone (each color presents one
copy of the simulation).
Fig. 7. Binding mode analysis of compound 18 during interaction with mPGES-1 (A, PDB code 5TL9) and 5-LO (B, PDB code 3O8Y). Main interactions are rep-
resented schematically with their occupancies calculated in the time window 0–100 ns. Sticks in orange represents ligand, white represents residues in chain 1, green
represents residues in chain 2, blue represents GSH. Red spheres represent bridging water molecules. Interaction types are given in Table 3 to obtain a clearer view.
(
For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
mg/kg).
biological targets within the AA cascade, namely mPGES-1, FLAP and 5-
LO. In view of the good congruity between the results from the biological
assays and the prediction of the molecular docking calculations, a
satisfactory explanation of the putative binding modes in three protein
targets for the 4,5-diarylisoxazol-3-carboxylic acid-based compounds
was provided. Biological assays disclosed that the decorations on the
heteroaryl side arm direct the activity against mPGES-1, FLAP and 5-LO,
3
. Conclusions
We have applied a convenient strategy for the synthesis of a small set
of 4,5-diarylisoxazol-3-carboxylic acids for exploring the role of the
heteroaryl fragments on the activity switch among closely related
7

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
Table 3
MD-based common interaction analysis of 18 with mPGES-1 and 5-LO. The occupancy values of vdW interactions were not calculated (n.c.), because of the additional
contribution of several atoms that exist in the nature of the interaction type.
Fragment
Target
mPGES-1
Occupancy
5-LO
Occupancy
Interaction Type
Carboxylate
ARG52
ARG52
SER127
98%
20%
22%
ASN425
ALA424
TYR181
TRP599
TYR181
LEU414
ALA603
Membrane*
98%
73%
72%
57%
n.c.
Salt Bridge
Hydrophilic P.
Water Bridge
p-chlorophenyl
LEU39
n.c.
vdW Interaction
Hydrophobic-1P.
TYR130
15%
n.c.
π
–
π
Stacking
Isoxazole-5-phenyl
Membrane
GLN363
LEU607
PHE421
THR364
LEU368
ALA410
HIS372
HIS367
HIS372
ASN407
ALA410
ARG411
n.c.
vdW Interaction
Hydrophobic-2P.
66%
49%
n.c.
π – π Stacking
Hydrogen Bond
vdW Interaction
Heteroaryl
ILE32
n.c.
Hydrophobic-3P.
GLN134
TYR130
31%
27%
10%
24%
42%
74%
π – π Stacking
Orthogonal Multipolar Interaction
Hydrogen Bond
Fig. 8. A) Time-base scale of carrageenan-induced paw edema in 90, 180, 270, and 360 min. Values are expressed as mean ± S.E.M. B) Effect of 18 on the
carrageenan-induced paw edema of mice at 270th min. Values are expressed as mean ± S.E.M. Differences from control, ***P < 0.001; SF, serum physio-
logic solution.
used without further purification. H and 13C NMR spectra were recor-
1
and have the potential to provide multitarget inhibitors, exemplified by
compound 18, with potent in vitro and in vivo efficacy. As future
perspective, further investigations on the nature of the heteroaryl frag-
ment can be fundamental to increase the interaction efficiency with the
hydrophobic binding pocket of the active site in three target proteins to
identify multitarget inhibitors. In conclusion, our results provide the
efficiency of the heteroaryl side arm in the design of promising candi-
dates as anti-inflammatory agents, and shed light on the chemical dec-
orations functional for the design of further members belonging to this
inhibitor class.
3 6
ded in CDCl or DMSO‑d on a Varian Mercury 400 MHz spectrometer
using tetramethylsilane as the internal standard. All chemical shifts were
recorded as δ (ppm). All coupling constants are reported as Hertz. High
resolution mass spectra data (HRMS) were collected using Waters LCT
Premier XE Mass Spectrometer (high sensitivity orthogonal acceleration
time-of-flight instrument) operating in ESI (+) or ESI (-) method, also
coupled to an AQUITY Ultra Performance Liquid Chromatography sys-
tem (Waters Corporation) using a UV detector monitoring at 254 nm.
Purity for all final compounds were > 95%, according to the UPLC-MS
method using (A) water + 0.1% Formic Acid and (B) acetonitrile +
0.1% Formic Acid; flow rate = 0.3 ml/min, Column: Aquity BEH C18
column (2.1 × 100 mm, 1.7 mm). The reactions were followed by TLC on
precoated Merck silica gel plates purchased from Merck. The developed
plates were visualized using 254 nm or 365 nm UV. Flash column
chromatography on silica gel was performed on Interchim prepacked
4
. Experimental
4
.1. Chemistry
Starting materials were purchased from commercial suppliers, and
8

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
+
disposable silica gel columns using Interchim Puriflash 4250 (Mon-
tluçon, France). Melting points of the synthesized compounds were
determined by an SMP50 automatic melting point apparatus (Stuart,
Staffordshire, ST15 OSA, UK) and were uncorrected. Compounds 1–6
were synthesized in the same manner as reported previously [24].
160.8, 166.4; HRMS m/z [M + H] calcd for C22
3 4
H19ClN O 424.1064;
found 424.1074.
4.1.1.6. 4-(4-Chlorophenyl)-5-{4-[(1,3-dimethyl-1H-pyrazol-5-yl)
methoxy]phenyl}isoxazole-3-carboxylic acid (22). Purified by recrystal-
◦
1
lization from methanol. Yield 72%; mp 190.7–192.3 C. H NMR
(DMSO‑d ): δ 2.08 (3H, s), 3.70 (3H, s), 5.11 (2H, s), 6.12 (1H, s), 7.08
4
.1.1. Synthesis of title isoxazol-3-carboxylic acid derivatives 17–26
6
Ester intermediates 7–16 (0.004 mol) and lithium hydroxide (0.008
(2H, d, J = 8.8 Hz), 7.36 (2H, d, J = 8.8 Hz), 7.38 (2H, d, J = 8.8 Hz),
◦
3
mol) in THF/water mixture (1:1) were stirred at at 60 C for 2 h. After
the reaction is complete, the reaction mixture was allowed to cool to
ambient temperature, diluted with water, acidified with 2 N HCl to pH 6,
and the resulting precipitate was filtered and recrystallized from the
appropriate solvent or purified by automated flash chromatography.
7.48 (2H, d, J = 8.8 Hz); C NMR (DMSO‑d ): δ 13.1, 36.0, 60.1, 106.5,
6
113.9, 115.4, 119.1, 128.1, 128.4, 128.6, 132.0, 133.1, 137.4, 145.6,
+
155.9, 159.5, 160.8, 166.3; HRMS m/z [M
+
H] calcd for
C22H19ClN O 424.1064; found 424.1061.
3 4
4
.1.1.7. 4-(4-Chlorophenyl)-5-(4-{[1-(propan-2-yl)-1H-imidazol-2-yl]
4
.1.1.1. 4-(4-Chlorophenyl)-5-{4-[(6-methylpyridin-2-yl)methoxy]
methoxy}phenyl) isoxazole-3-carboxylic acid (23). Purified by recrystal-
◦
1
phenyl}isoxazole-3-carboxylic acid (17). Purified by recrystallization
lization from methanol. Yield 95%; mp 126.4–127.7 C. H NMR
◦
1
from ethyl acetate. Yield 34%; mp 193.7–195.1 C. H NMR (DMSO‑d
6
):
(DMSO‑d ): δ 1.37 (6H, d, J = 6.4 Hz), 4.52–4.59 (1H, m), 5.27 (2H, s),
6
δ 2.46 (3H, s), 5.12 (2H, s), 6.90 (2H, d, J = 8.4 Hz), 7.09 (2H, d, J = 8.4
Hz), 7.17 (1H, d, J = 7.6 Hz), 7.29 (1H, d, J = 7.6 Hz), 7.51 (2H, d, J =
7.11 (1H, s), 7.14 (2H, d, J = 8.8 Hz), 7.37 (2H, d, J = 8.4 Hz), 7.39 (2H,
1
3
d, J = 8.8 Hz), 7.48 (2H, d, J = 8.4 Hz), 7.53 (1H, s); C NMR
1
3
8
.0 Hz), 7.69 (1H, t, J = 7.6 Hz), 7.98 (2H, d, J = 8.0 Hz); C NMR
(DMSO‑d ): δ 23.1, 48.1, 61.1, 113.8, 115.4, 118.2, 119.5, 125.7, 128.2,
6
(
DMSO‑d
6
): δ 23.9, 70.2, 113.1, 115.2, 118.5, 122.1, 123.3, 124.4,
128.4, 128.6, 132.0, 133.0, 141.0, 156.6, 159.3, 161.0, 165.9; HRMS m/
z [M + H]+ calcd for C23H21ClN O 438.1221; found 438.1200.
1
1
4
26.9, 128.2, 129.1, 131.4, 137.1, 137.9, 139.8, 156.1, 157.4, 157.8,
3
4
+
80.1; HRMS m/z [M + H] calcd for C23
2 4
H18ClN O 421.0955; found
21.0965.
4
.1.1.8. 4-(4-Chlorophenyl)-5-{4-[(2-methylpyridin-3-yl)methoxy]
phenyl}isoxazole-3-carboxylic acid (24). Purified by recrystallization
◦
1
4
.1.1.2. 4-(4-Chlorophenyl)-5-(4-((5-(trifluoromethyl)furan-2-yl)
from methanol. Yield 45%; mp 154.3–155.9 C. H NMR (DMSO‑d ): δ
6
methoxy)phenyl)isoxazole-3-carboxylic acid (18). Purified by successive
2.50 (3H, s), 5.16 (2H, s), 7.01 (2H, d, J = 8.4 Hz), 7.18 (2H, d, J = 8.4
◦
1
washing with hot hexane. Yield 73%; mp 133.8–135.8 C. H NMR
DMSO‑d
): δ 5.18 (2H, s), 6.82 (1H, d, J = 3.6 Hz), 7.09 (2H, d, J = 8.8
Hz), 7.19–7.20 (1H, m), 7.34–7.39 (4H, m), 7.44 (2H, d, J = 8.8 Hz);
Hz), 7.23–7.27 (1H, m), 7.56 (2H, d, J = 7.8 Hz), 7.79–7.82 (1H, m),
1
3
(
6
7.94 (2H, d, J = 7.8 Hz), 8.40–8.42 (1H, m); C NMR (DMSO‑d ): δ
6
1
3
C
22.2, 67.4, 113.9, 115.1, 121.7, 124.7, 124.9, 125.6, 127.7, 129.9,
3
NMR (DMSO‑d
6
, NaH): δ 61.5, 73.4, 112.0, 113.6, 114.3 ( JC-F = 2.3
Hz), 119.4 ( JC-F = 265.2 Hz), 124.3, 124.6, 125.1, 127.5, 129.1, 129.6,
130.8, 131.8, 136.2, 137.0, 139.1, 148.6, 157.2, 158.6, 179.9; HRMS m/
z [M + H]+ calcd for C23H18ClN O 421.0955; found 421.0938.
1
2
4
2
1
38.5, 139.7, 140.6 ( JC-F = 41.2 Hz), 154.2, 157.7, 182.4; HRMS (m/z):
[
M + H]+ calcd for C22
3 5
H14ClF NO 464.0513; found 464.0504.
4
.1.1.9. 5-{4-[(2H-1,3-Benzodioxol-5-yl)methoxy]phenyl}-4-(4-chlor-
ophenyl)-1,2-oxazole-3-carboxylic acid (25). Purified by recrystalliza-
◦
1
4
.1.1.3. 5-{4-[(1H-1,2,3-Benzotriazol-1-yl)methoxy]phenyl}-4-(4-chlor-
tion from methanol. Yield 69%; mp 167.7–168.2 C. H NMR
ophenyl)isoxazole-3-carboxylic acid (19). Purified by recrystallization
(DMSO‑d ): δ 4.99 (2H, s), 6.01 (2H, s), 6.88–6.93 (2H, m), 6.99 (1H, s),
6
◦
1
from methanol. Yield 88%; mp 173.6–174.9 C. H NMR (DMSO‑d
6
6
): δ
7.05 (2H, d, J = 8.4 Hz), 7.37 (4H, d, J = 8.4 Hz), 7.48 (2H, d, J = 8.4
Hz); 13C NMR (DMSO‑d ): δ 69.2, 100.9, 108.1, 108.5, 113.5, 115.3,
.82 (2H, s), 7.24 (2H, d, J = 8.8 Hz), 7.36–7.46 (5H, m), 7.48 (2H, d, J
6
=
8.8 Hz), 7.62 (1H, t, J = 7.6 Hz), 7.95 (1H, d, J = 8.4 Hz), 8.09 (1H, d,
118.8, 121.7, 122.1, 128.3, 128.5, 130.0, 131.9, 132.9, 146.9, 147.3,
1
3
+
J = 8.4 Hz); C NMR (DMSO‑d
6
): δ 73.5, 110.6, 114.3, 116.5, 119.4,
156.6, 159.9, 160.9, 165.9; HRMS m/z [M + H] calcd for C24H17ClNO6
1
1
20.5, 124.7, 127.9, 128.3, 128.5, 128.6, 132.0, 132.6, 133.1, 145.3,
450.0744; found 450.0741.
+
55.9, 157.5, 160.7, 166.0; HRMS m/z [M
+
H] calcd for
23 4 4
C H16ClN O 447.0860; found 447.0860.
4
.1.1.10. 5-{4-[(5-Chloro-1-benzothiophen-3-yl)methoxy]phenyl}-4-(4-
chlorophenyl)-1,2-oxazole-3-carboxylic acid (26). Purified by recrystal-
◦
1
4
.1.1.4. 4-(4-Chlorophenyl)-5-[4-({1-propyl-1H-imidazo[4,5-c]pyridin-2-
lization from methanol. Yield 44%; mp 170.8–171.7 C. H NMR
yl}methoxy)phenyl] isoxazol-3-carboxylic acid (20). Purified by recrys-
(DMSO‑d ): δ 5.38 (2H, s), 7.15 (2H, d, J = 8.8 Hz), 7.37–7.44 (5H, m),
6
◦
1
tallization from methanol. Yield 83%; mp 148.7–150.3 C. H NMR
DMSO‑d
): δ 0.86 (3H, t, J = 7.2 Hz), 1.77–1.82 (2H, m), 4.29 (2H, t, J
7.2 Hz), 5.51 (2H, s), 7.18 (2H, d, J = 8.4 Hz), 7.38 (2H, d, J = 8.0 Hz),
.42 (2H, d, J = 8.4 Hz), 7.48 (2H, d, J = 8.0 Hz), 7.76 (1H, d, J = 4.8
7.49 (2H, d, J = 8.4 Hz), 7.94 (2H, d, J = 2.0 Hz), 7.97 (1H, s), 8.04 (1H,
1
3
(
6
d, J = 8.4 Hz); C NMR (DMSO‑d ): δ 63.8, 113.8, 115.4, 118.9, 121.6,
6
=
124.5, 124.6, 128.1, 128.3, 128.6, 129.0, 129.5, 130.8, 132.0, 133.0,
+
7
138.3, 138.9, 155.9, 159.8, 160.7, 166.3. HRMS m/z [M + H] calcd for
1
3
Hz), 8.39 (1H, d, J = 4.8 Hz), 9.0 (1H, s); C NMR (DMSO‑d
6
): δ 10.9,
25 16 2 4
C H Cl NO S 496.0177; found 496.0170.
2
1
2.6, 45.3, 62.3, 106.6, 113.9, 115.4, 119.6, 128.2, 128.4, 128.5, 131.9,
33.0, 139.0, 140.5, 140.7, 141.1, 151.3, 156.3, 159.2, 160.8, 166.0;
+
4.2. Computational methods
HRMS m/z [M + H] calcd for C26
4 4
H22ClN O 489.1330; found
4
89.1308.
4
.2.1. Molecular docking studies
The probable binding orientations of the active compounds were
4
.1.1.5. 4-(4-Chlorophenyl)-5-{4-[(1-ethyl-1H-pyrazol-3-yl)methoxy]
identified by conducting molecular docking studies against mPGES-1, 5-
LO and FLAP by following the previously applied procedures [24,42-
phenyl}isoxazole-3-carboxylic acid (21). Purified by recrystallization
◦
1
from hexane-ethyl acetate. Yield 45%; mp 189.6–190.8 C. H NMR
DMSO‑d
4
4]. The PDB codes of the crystal structures used in this study are 5TL9
36], 3O8Y [37] and 2Q7M [38], respectively. The docking results of 17,
8, 19, 25, 26 were visualized at the mPGES-1 active site, 17, 18, 24,
25, 26 were visualized at the 5-LO active site, and 17, 18, 21, 22, 24, 25,
(
6
): δ 1.33 (3H, t, J = 7.2 Hz), 4.08 (2H, q, J = 7.2 Hz), 5.00 (2H,
[
s), 6.27 (1H, d, J = 2.0 Hz), 7.06 (2H, d, J = 8.8 Hz), 7.36 (2H, d, J = 8.8
1
Hz), 7.37 (2H, d, J = 8.8 Hz), 7.48 (2H, d, J = 8.8 Hz), 7.67 (1H, d, J =
.0 Hz); 1 C NMR (DMSO‑d
3
2
1
6
): δ 15.4, 46.1, 63.8, 105.1, 113.8, 115.3,
2
6 were visualized at FLAP active site.
18.7, 128.2, 128.3, 128.6, 130.2, 132.0, 133.1, 146.4, 155.9, 160.0,
Crystal selection was made (i) from the crystals that contain
9

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
glutathione (GSH) and a potent mPGES-1 inhibitor in the active site of
the complex, (ii) having a better crystal resolution, (iii) being co-
crystallized with potent mPGES-1 inhibitors, (iv) analysis of electron
density maps and (v) reproducibility of the ligand binding mode of the
co-crystallized ligand by the applied docking protocol for mPGES-1.
Regarding crystal selection for 5-LO, the selected structure is found
suitable for docking 4,5-diarylisoxazol-3-carboxylic acid derivatives.
The other ones either (i) include a mutated residue to mimic 15-LO’s or a
natural product which is identified (ii) a redox inhibitor or (iii) an
allosteric inhibitor. Selected FLAP crystal structure is obtained in low-
resolution, however we used a refined model of the original structure
by applying a relaxation protocol. The readers are invited to read further
our previous publication for further details [23]. The relaxed FLAP
model is compared with recently published crystals and low RMSD value
is obtained (RMSD < 0.5) by aligning the displayed active site residues
in Fig. 5. A brief report of each published crystal structure is given in
Table S4, S5 and S6.
MD trajectories and energy calculation results for each trajectory. The
simulation system relaxation protocol is given in as stages before start-
ing the main MD simulation. There were 9335 water molecules for
mPGES-1 simulation and 22,988 water molecules for 5-LO. Long-range
Coulombic interactions cutoff radius value was set to 9.0 Å. All the
simulation systems were prepared with OPLS2005 forcefield. The sys-
tems were neutralized, and the salt concentrations were defined as 0.15
+
-
M, and calculated numbers of Na and Cl ions were added. The simu-
lations were started to run with the relaxation protocol of a series of
standard minimization and short MD simulations to relax the systems.
The relaxation protocol is applied as described followingly. Initial sys-
tem relaxations were conducted with up to 2000 minimization steps, a
convergence criterion of 50 kcal/mol/Å, and the presence of harmonic
2
restraints on the solute atoms (force constant = 50.0 kcal/mol/Å ). In
the second step, minimization routines were conducted without re-
straints. The third stage includes 12 ps at 10 K with harmonic restraints
2
on the heavy atoms of the solute (force constant = 50.0 kcal/mol/Å ), by
Ligands were drawn by using Maestro 11.1 [46] Interface. The atom
types and the protonation states of the proteins and the ligands were
identified with OPLS2005 forcefield. Protein Preparation Wizard [47]
was used to parameterize the atom types, to add predicted positions of
the missing side chains, and to relax the active sites of the proteins.
LigPrep [48] routine was used at pH 7.0 ± 2.0 to prepare the ligands.
The predicted protein-ligand interactions by docking studies were
generated with Glide 7.4 [49-51]. Firstly, the docking boxes were
generated by identifying binding sites’ centroids with the amino acid
residues that are present in docking figures. These centroids’ co-
ordinates were used to identify binding site enclosing boxes and to limit
each docked pose’s distance to this point by 10 Å. Van der Waals radius
scaling factor and partial charge cutoff values are kept with their default
values (1.0 and 0.25, respectively). The docking simulations were done
in standard precision mode (GlideScore SP) [49,51,52]. The most
representative binding mode of 4,5-diarylisoxazol-3-carboxylic acid
derivatives for each target were identified by (i) keeping top 5 ranked
poses by their docking scores and (ii) selecting the lowest RMSD value
one for each ligand within all docked poses. Image generation was done
with PyMOL 2.3 [40] after identifying protein-ligand contacts with PLIP
using NVT ensemble and Berendsen thermostat. The fourth step 12 ps at
10 K, by retaining the harmonic restraints and using NPT ensemble and
Berendsen thermostat and barostat. The fifth step was done to heat the
system for 24 ps at 323 K (above the transition temperature (314.4 K) of
DPPC from gel to liquid [58]), by retaining the harmonic restraints and
using NPT ensemble and Berendsen thermostat and barostat. At the final
stage of the relaxation of the system was done for 24 ps at 323 K without
harmonic restraints, by using NPT ensemble and Nose-Hoover thermo-
stat and Martyna-Tobias-Klein barostat. Subsequently, 100 ns long MD
simulations were carried out with four copies for both simulation sys-
tems, to have statistically more reliable results. All the simulations were
run with an NPT ensemble at 323 K. The simulation trajectories were
visualized with Maestro 11.1 [46] and VMD 1.9.3 [59]. The pro-
tein–ligand interaction occupancy values and RMSD values during the
simulations were evaluated with Simulation Interactions Diagram module,
and further interaction analyses of trifluoromethyl moiety of 18 were
done with Simulation Event Analysis module of Schrodinger Suite 2017-1.
Conformational clustering was done with gmx cluster module of GRO-
MACS 2018.3 [60] with a cut-off value of 0.5 Å to obtain the most
representative snapshot from the whole 100 ns simulation. Later, the
pose generations on the highest populated pose for each protein was
done with PLIP 1.4.4 [39] and PyMOL 2.3 [40].
1
.4.4 [39]. Directional and non-directional interactions are identified by
′
PLIP 1.4.4 s [39] rule-based detection algorithm [53] with default pa-
rameters [54] to visualize the interactions. Hydrophilic interactions are
identified based on their distance and angle cutoffs [54]. However,
hydrophobic interactions are a result from their entropic effects, there-
fore there are no clear geometries for their association. We stored these
occurrences in static pictures by using the generous cutoffs of PLIP 1.4.4
4.3. Biological assays
4.3.1. Cells
Neutrophils were isolated from human blood as reported before [25].
Briefly, human peripheral blood was obtained from healthy adult
(18–65 years) male and female donors with consent that had not taken
antiinflammatory drugs during the last 10 days, by venipuncture in
heparinized tubes (16 IE heparin/ml blood; University Hospital Jena,
[
39]. Salt bridges are identified by the identification of opposite charged
atoms and then controlling the distances between to be below 5.5 Å
55]. interactions are identified by computationally placing dummy
atoms to the middle of the appropriate rings’ centers and checking
formation of parallel or T shaped -stackings by their distances or angles
56]. Van der Waals interactions were identified similarly. Firstly, car-
[
π-π
◦
π
Germany). The blood was centrifuged (4000 × g for 20 min at 20 C) for
[
preparation of leukocyte concentrates. Leukocyte concentrates were
then subjected to dextran sedimentation and centrifugation on Nyco-
prep cushions. Contaminating erythrocytes of pelleted neutrophils were
lysed by hypotonic lysis using water. Neutrophils were washed twice in
ice-cold PBS (purity > 96–97%) and finally resuspended in PBS pH 7.4
bon atoms that are bound to other carbons or hydrogens are classified as
hydrophobic atom, then distances between such atoms are analyzed and
close ones are identified as interacting atoms. Only the closest ones
within these residues are shown as interaction for a clear visualization.
Further information can be found in the source code [53,54].
containing 1 mg/ml glucose and 1 mM CaCl (PGC buffer). For analysis
2
of acute cytotoxicity of the compounds during preincubation periods
◦
4
.2.2. Molecular dynamic (MD) studies
(30 min at 37 C), cellular integrity of neutrophils was analyzed by
The best ranking docking result of compound 18 with mPGES-1 and
-LO were used to evaluate the predicted binding modes further. The
trypan blue exclusion with a Vi-cell counter (Beckmann Coulter GmbH,
Krefeld). None of the compounds caused significant loss of neutrophil
viability within 30 min (studied by trypan blue staining and light mi-
croscopy, data not shown).
5
simulation systems were generated with System Builder utility, to run MD
simulations with Desmond 4.9 [57]. mPGES-1 complex including 18 was
embedded in dipalmitoyl phosphatidylcholine (DPPC) bilayers to
include membrane structure in the simulation system. The predicted
location of the lipid bilayer was obtained from Orientations of Proteins
in Membrane (OPM) [41] database. SPC model was used for water
molecules. Recording interval is set to 10 picoseconds while saving the
4.3.2. Determination of 5-LO products in intact cells
For determination of LO products in intact cells, neutrophils (5 ×
6
10 ) were resuspended in 1 ml PGC buffer, preincubated for 15 min at
◦
2+
37 C with test compounds or vehicle (0.1% DMSO) and Ca -ionophore
1
0

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
◦
A23187 (2.5 mM) was added. After 10 min at 37 C, the reaction was
original draft. Ulrich S. Schubert: Resources. Oliver Werz: Resources,
Supervision. Erden Banoglu: Conceptualization, Resources, Writing -
review & editing, Funding acquisition.
stopped on ice by addition of 1 ml of methanol. 30 µl 1 N HCl and 500 µl
PBS, and 200 ng prostaglandin (PG)B were added and the samples were
1
subjected to solid phase extraction on C18-columns (100 mg, UCT,
Bristol, PA, USA). 5-LO products (LTB and its trans-isomers, and 5-H(P)
ETE) were analyzed by RP-HPLC and quantities calculated on the basis
of the internal standard PGB . Cys-LTs C , D and E were not detected
amounts were below detection limit), and oxidation products of LTB
4
Declaration of Competing Interest
1
4
4
4
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
(
4
were not determined.
4
.3.3. Determination of the activity of isolated 5-LO in cell-free assays
Acknowledgements
Escherichia coli BL21 cells were transformed with pT3-5-LO plasmid,
◦
and recombinant 5-LO protein was expressed at 27 C as described [25].
Cells, resuspendend in 50 mM triethanolamine/HCl pH 8.0, 5 mM EDTA,
soybean trypsin inhibitor (60 µg/ml), 1 mM phenylmethanesulphonyl
We acknowledge the Turkish Academy of Sciences (TUBA) for partial
financial support. The GPUs used in this study were donated by NVIDIA
Corporation. This work was also supported by the Deutsche For-
schungsgemeinschaft, SFB1278/1 Polytarget 316213987 project A04.
fluoride, and lysozyme (500 µg/ml), were homogenized by sonication
◦
(
3 × 15 s). After centrifugation at 40,000g for 20 min at 4 C, the su-
pernatant was applied to an ATP-agarose column to partially purify 5-LO
as described previously [21]. Aliquots of semi-purified 5-LO were
diluted with ice-cold PBS containing 1 mM EDTA, and 1 mM ATP was
Appendix A. Supplementary material
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bioorg.2021.104861.
added. Samples were pre-incubated with the test compounds or vehicle
◦
(
0.1% DMSO) as indicated. After 15 min at 4 C, samples were pre-
◦
warmed for 30 s at 37 C, and 2 mM CaCl
2
plus 20 µM AA was added
to initiate 5-LO product formation. After 10 min at 37 C, the reaction
was stopped by addition of 1 ml ice-cold methanol, and the formed
metabolites were analyzed by RP-HPLC as described [21]. 5-LO products
References
◦
[1] M.T. Donnelly, C.J. Hawkey, Review article: COX-II inhibitors–a new generation of
safer NSAIDs? Aliment. Pharmacol. Ther. 11 (2) (1997) 227–236.
[
2] S. Tacconelli, A. Bruno, R. Grande, P. Ballerini, P. Patrignani, Nonsteroidal anti-
inflammatory drugs and cardiovascular safety - translating pharmacological data
into clinical readouts, Expert Opin. Drug Saf. 16 (7) (2017) 791–807.
3] H.H. Chang, E.J. Meuillet, Identification and development of mPGES-1 inhibitors:
where we are at? Future Med. Chem. 3 (15) (2011) 1909–1934.
4
include the all-trans isomers of LTB and 5-H(P)ETE.
[
[
4
.3.4. Determination of mPGES-1 activity
Microsomal preparations of A549 cells expressing mPGES-1 were
prepared as previously described [26]. In brief, A549 cells were culti-
4] A. Koeberle, O. Werz, Perspective of microsomal prostaglandin E2 synthase-1 as
drug target in inflammation-related disorders, Biochem. Pharmacol. 98 (1) (2015)
1–15.
ꢀ
1
vated in DMEM containing FCS (2%) and IL-1β (2 ng mL ) for 72 h
[
[
5] A. Koeberle, O. Werz, Inhibitors of the microsomal prostaglandin E(2) synthase-1
as alternative to non steroidal anti-inflammatory drugs (NSAIDs)-a critical review,
Curr. Med. Chem. 16 (32) (2009) 4274–4296.
◦
(
37 C, 5% CO
2
). Cells were harvested and resuspended in homogeni-
zation buffer consisting of potassium phosphate (0.1 M, pH 7.4), phe-
6] D. Xu, S.E. Rowland, P. Clark, A. Giroux, B. Cote, S. Guiral, M. Salem, Y. Ducharme,
R.W. Friesen, N. Methot, J. Mancini, L. Audoly, D. Riendeau, MF63 [2-(6-chloro-
nylmethanesulfonyl fluoride (1 mM), soybean trypsin inhibitor (60 µg
mL ), leupeptin (1 g mL ), glutathione (2.5 mM), and sucrose (250
ꢀ 1
ꢀ 1
μ
1
H-phenanthro[9,10-d]imidazol-2-yl)-isophthalonitrile], a selective microsomal
mM). After shock-freezing of the cells in liquid nitrogen, sonication (3 ×
prostaglandin E synthase-1 inhibitor, relieves pyresis and pain in preclinical
models of inflammation, J. Pharmacol. Exp. Ther. 326 (3) (2008) 754–763.
◦
2
0 s), and differential centrifugation at 10000 × g (10 min, 4 C) and
◦
[7] Y. Jin, C.L. Smith, L. Hu, K.M. Campanale, R. Stoltz, L.G. Huffman Jr., T.
A. McNearney, X.Y. Yang, B.L. Ackermann, R. Dean, A. Regev, W. Landschulz,
Pharmacodynamic comparison of LY3023703, a novel microsomal prostaglandin e
synthase 1 inhibitor, with celecoxib, Clin. Pharmacol. Ther. 99 (3) (2016)
1
74000g (60 min, 4 C), the pellets were resuspended in homogenization
buffer. Microsomes were diluted in potassium phosphate buffer (0.1 m,
pH 7.4) with glutathione (2 mM) and pre-incubated with the test com-
2
74–284.
pounds or vehicle (0.1% DMSO) on ice for 15 min. After stimulation for
[
8] B.H. Norman, M.J. Fisher, M.A. Schiffler, S.L. Kuklish, N.E. Hughes, B.A. Czeskis, K.
C. Cassidy, T.L. Abraham, J.J. Alberts, D. Luffer-Atlas, Identification and mitigation
of reactive metabolites of 2-aminoimidazole-containing microsomal prostaglandin
E synthase-1 inhibitors terminated due to clinical drug-induced liver injury, J. Med.
Chem. 61 (5) (2018) 2041–2051.
◦
1
min at 4 C with 20
μ
M PGH
2
as substrate, the reaction was terminated
(40 mM), citric acid
M, as internal standard) and analyzed for
by RP-HPLC as reported before [26].
by the addition of stop solution containing FeCl
80 mM), and 11β -PGE (10
PGE
3
(
2
μ
2
[
9] S. Sant, M. Tandon, V. Menon, G. Gudi, V. Kattige, N.K. Joshi, K. Korukonda,
O. Levine-Dolberg, Grc 27864, novel, microsomal prostaglandin E synthase-1
enzyme inhibitor: Phase 1 study to evaluate safety, Pk and Biomarkers in Healthy,
Adult Subjects, Osteoarthr Cartilage 26 (2018) S351–S352.
4
.4. Statistics
[
[
[
10] Z.T. Gur, B. Caliskan, E. Banoglu, Drug discovery approaches targeting 5-lipoxy-
genase-activating protein (FLAP) for inhibition of cellular leukotriene biosynthesis,
Eur. J. Med. Chem. 153 (2018) 34–48.
Results are presented as mean ± standard error of the mean (SEM)
out of n independent experiments, where n represents the number of
performed experiments on different days or with different donors. IC50
values were calculated from at least 5 different concentrations using a
nonlinear regression interpolation of semi-logarithmic graphs using
GraphPad Prism (Graphpad Software Inc., San Diego, CA). Statistical
evaluation was performed by one-way ANOVA using GraphPad InStat
11] O. Radmark, O. Werz, D. Steinhilber, B. Samuelsson, 5-Lipoxygenase, a key enzyme
for leukotriene biosynthesis in health and disease, Biochim Biophys Acta 1851 (4)
(2015) 331–339.
12] M. Lemurell, J. Ulander, S. Winiwarter, A. Dahlen, O. Davidsson, H. Emtenas,
J. Broddefalk, M. Swanson, D. Hovdal, A.T. Plowright, A. Pettersen, M. Ryden-
Landergren, J. Barlind, A. Llinas, M. Herslof, T. Drmota, K. Sigfridsson, S. Moses,
C. Whatling, Discovery of AZD6642, an inhibitor of 5-lipoxygenase activating
protein (FLAP) for the treatment of inflammatory diseases, J. Med. Chem. 58 (2)
(
Graphpad Software Inc., San Diego, CA) followed by a Bonferroni post-
(
2015) 897–911.
hoc test for multiple or student t-test for single comparisons, respec-
tively. Pvalues < 0.05 were considered as significant.
[
13] D. Pettersen, J. Broddefalk, H. Emtenas, M.A. Hayes, M. Lemurell, M. Swanson,
J. Ulander, C. Whatling, C. Amilon, H. Ericsson, A. Westin Eriksson, K. Granberg,
A.T. Plowright, I. Shamovsky, A. Dellsen, M. Sundqvist, M. Nagard, E.L. Lindstedt,
Discovery and early clinical development of an inhibitor of 5-lipoxygenase
activating protein (AZD5718) for treatment of coronary artery disease, J. Med.
Chem. 62 (9) (2019) 4312–4324.
CRediT authorship contribution statement
Tu g˘ ba Gürses: Investigation. Abdurrahman Ol g˘ aç: Investigation,
Visualization. Ulrike Garscha: Methodology, Investigation. Tu g˘ çe Gür
Maz: Methodology, Formal analysis. Nur Banu Bal: Investigation.
Orhan Uluda g˘ : Methodology. Burcu Çalıs¸kan: Supervision, Writing -
[14] H. Takahashi, D. Riether, A. Bartolozzi, T. Bosanac, V. Berger, R. Binetti,
J. Broadwater, Z. Chen, R. Crux, S. De Lombaert, R. Dave, J.A. Dines, T. Fadra-
Khan, A. Flegg, M. Garrigou, M.H. Hao, J. Huber, J.M. Hutzler, S. Kerr, A. Kotey,
W. Liu, H.Y. Lo, P.L. Loke, P.E. Mahaney, T.M. Morwick, S. Napier, A. Olague,
E. Pack, A.K. Padyana, D.S. Thomson, H. Tye, L. Wu, R.M. Zindell,
1
1

T. Gürses et al.
Bioorganic Chemistry 112 (2021) 104861
A. Abeywardane, T. Simpson, Synthesis, SAR, and series evolution of novel
oxadiazole-containing 5-lipoxygenase activating protein inhibitors: discovery of 2-
4-(3-{(r)-1-[4-(2-amino-pyrimidin-5-yl)-phenyl]-1-cyclopropyl-ethyl}-[1,2,4]oxa-
diazol-5-yl)-pyrazol-1-yl]-N, N-dimethyl-acetamide (BI 665915), J. Med. Chem. 58
[34] O. Werz, 5-lipoxygenase: cellular biology and molecular pharmacology, Curr. Drug
Targets Inflamm. Allergy 1 (1) (2002) 23–44.
[
[35] D. Riendeau, R. Aspiotis, D. Ethier, Y. Gareau, E.L. Grimm, J. Guay, S. Guiral,
H. Juteau, J.A. Mancini, N. Methot, J. Rubin, R.W. Friesen, Inhibitors of the
inducible microsomal prostaglandin E2 synthase (mPGES-1) derived from MK-886,
Bioorg. Med. Chem. Lett. 15 (14) (2005) 3352–3355.
(
4) (2015) 1669–1690.
[
15] F. Celotti, S. Laufer, Anti-inflammatory drugs: New multitarget compounds to face
an old problem. The dual inhibition concept, Pharmacol. Res. 43 (5) (2001)
[36] K.M. Partridge, S. Antonysamy, S.N. Bhattachar, S. Chandrasekhar, M.J. Fisher,
A. Fretland, K. Gooding, A. Harvey, N.E. Hughes, S.L. Kuklish, J.G. Luz, P.
R. Manninen, J.E. McGee, D.R. Mudra, A. Navarro, B.H. Norman, S.J. Quimby, M.
A. Schiffler, A.V. Sloan, A.M. Warshawsky, J.M. Weller, J.S. York, X.P. Yu,
Discovery and characterization of [(cyclopentyl)ethyl]benzoic acid inhibitors of
microsomal prostaglandin E synthase-1, Bioorg. Med. Chem. Lett. 27 (6) (2017)
1478–1483.
4
29–436.
[
[
16] S.K. Kulkarni, V.P. Singh, Licofelone: the answer to unmet needs in osteoarthritis
therapy? Curr. Rheumatol. Rep. 10 (1) (2008) 43–48.
17] E. Banoglu, B. Caliskan, S. Luderer, G. Eren, Y. Ozkan, W. Altenhofen, C. Weinigel,
D. Barz, J. Gerstmeier, C. Pergola, O. Werz, Identification of novel benzimidazole
derivatives as inhibitors of leukotriene biosynthesis by virtual screening targeting
5
3
-lipoxygenase-activating protein (FLAP), Bioorg. Med. Chem. 20 (12) (2012)
728–3741.
[37] N.C. Gilbert, W.E. Boeglin, A.R. Brash, M.E. Newcomer, The Structure of Human 5-
Lipoxygenase, 217(2011) (2011).
[
[
[
18] B. Caliskan, E. Banoglu, Overview of recent drug discovery approaches for new
generation leukotriene A4 hydrolase inhibitors, Expert Opin. Drug Discov. 8 (1)
[38] A.D. Ferguson, B.M. McKeever, S. Xu, D. Wisniewski, D.K. Miller, T.-T. Yamin, R.
H. Spencer, L. Chu, F. Ujjainwalla, B.R. Cunningham, J.F. Evans, J.W. Becker,
Crystal structure of inhibitor-bound human 5-lipoxygenase-activating protein 317
(2007) 510–512.
(
2013) 49–63.
19] B. Caliskan, S. Luderer, Y. Ozkan, O. Werz, E. Banoglu, Pyrazol-3-propanoic acid
derivatives as novel inhibitors of leukotriene biosynthesis in human neutrophils,
Eur. J. Med. Chem. 46 (10) (2011) 5021–5033.
[39] S. Salentin, S. Schreiber, V.J. Haupt, M.F. Adasme, M. Schroeder, PLIP: fully
automated protein-ligand interaction profiler, Nucleic Acids Res. 43 (W1) (2015)
W443–W447.
20] S. Levent, J. Gerstmeier, A. Olgac, F. Nikels, U. Garscha, A. Carotti, A. Macchiarulo,
O. Werz, E. Banoglu, B. Caliskan, Synthesis and biological evaluation of C(5)-
substituted derivatives of leukotriene biosynthesis inhibitor BRP-7, Eur. J. Med.
Chem. 122 (2016) 510–519.
[40] The PyMOL Molecular Graphics System, Schr o¨ dinger, LLC.
[41] M.A. Lomize, A.L. Lomize, I.D. Pogozheva, H.I. Mosberg, OPM: orientations of
proteins in membranes database, Bioinformatics 22 (5) (2006) 623–625.
[42] M. Boccellino, M. Donniacuo, F. Bruno, B. Rinaldi, L. Quagliuolo, M. Ambruosi,
S. Pace, M. De Rosa, A. Olgac, E. Banoglu, N. Alessio, A. Massa, H. Kahn, O. Werz,
A. Fiorentino, R. Filosa, Protective effect of piceatannol and bioactive stilbene
derivatives against hypoxia-induced toxicity in H9c2 cardiomyocytes and
structural elucidation as 5-LOX inhibitors, Eur. J. Med. Chem. 180 (2019)
637–647.
[
21] C. Pergola, J. Gerstmeier, B. Monch, B. Caliskan, S. Luderer, C. Weinigel, D. Barz,
J. Maczewsky, S. Pace, A. Rossi, L. Sautebin, E. Banoglu, O. Werz, The novel
benzimidazole derivative BRP-7 inhibits leukotriene biosynthesis in vitro and in
vivo by targeting 5-lipoxygenase-activating protein (FLAP), Br. J. Pharmacol. 171
(
12) (2014) 3051–3064.
[
[
[
22] S. Shekfeh, B. Caliskan, K. Fischer, T. Yalcin, U. Garscha, O. Werz, E. Banoglu,
A Multi-step Virtual Screening Protocol for the Identification of Novel Non-acidic
Microsomal Prostaglandin E2 Synthase-1 (mPGES-1) Inhibitors, ChemMedChem 14
[43] F. Bruno, S. Errico, S. Pace, M.B. Nawrozkij, A.S. Mkrtchyan, F. Guida, R. Maisto,
A. Olgac, M. D’Amico, S. Maione, M. De Rosa, E. Banoglu, O. Werz, A. Fiorentino,
R. Filosa, Structural insight into the optimization of ethyl 5-hydroxybenzo[g]indol-
3-carboxylates and their bioisosteric analogues as 5-LO/m-PGES-1 dual inhibitors
able to suppress inflammation, Eur. J. Med. Chem. 155 (2018) 946–960.
[44] Z.T. Gur, B. Caliskan, U. Garscha, A. Olgac, U.S. Schubert, J. Gerstmeier, O. Werz,
E. Banoglu, Identification of multi-target inhibitors of leukotriene and
prostaglandin E2 biosynthesis by structural tuning of the FLAP inhibitor BRP-7,
Eur. J. Med. Chem. 150 (2018) 876–899.
(
2) (2019) 273–281.
23] A. Olgac, A. Carotti, C. Kretzer, S. Zergiebel, A. Seeling, U. Garscha, O. Werz,
A. Macchiarulo, E. Banoglu, Discovery of novel 5-lipoxygenase-activating protein
(
FLAP) inhibitors by exploiting a multistep virtual screening protocol, J. Chem. Inf.
Model. 60 (3) (2020) 1737–1748.
24] E. Banoglu, E. Celikoglu, S. Volker, A. Olgac, J. Gerstmeier, U. Garscha,
B. Caliskan, U.S. Schubert, A. Carotti, A. Macchiarulo, O. Werz, 4,5-Diarylisoxazol-
3
-carboxylic acids: A new class of leukotriene biosynthesis inhibitors potentially
targeting 5-lipoxygenase-activating protein (FLAP), Eur. J. Med. Chem. 113 (2016)
–10.
[45] C.A. Winter, E.A. Risley, G.W. Nuss, Carrageenin-induced edema in hind paw of the
rat as an assay for antiiflammatory drugs, Proc. Soc. Exp. Biol. Med. 111 (1962)
544–547.
1
[
25] U. Garscha, S. Voelker, S. Pace, J. Gerstmeier, B. Emini, S. Liening, A. Rossi,
C. Weinigel, S. Rummler, U.S. Schubert, G.K. Scriba, E. Celikoglu, B. Caliskan,
E. Banoglu, L. Sautebin, O. Werz, BRP-187: A potent inhibitor of leukotriene
biosynthesis that acts through impeding the dynamic 5-lipoxygenase/5-lipoxyge-
nase-activating protein (FLAP) complex assembly, Biochem. Pharmacol. 119
[46] Schr o¨ dinger Release 2017-1: Maestro, Schr o¨ dinger, LLC, New York, NY, 2017.
[47] Schr o¨ dinger Release 2017-1: Schr o¨ dinger Suite 2017-1 Protein Preparation
Wizard; Epik, Schr o¨ dinger, LLC, New York, NY, 2017; Impact, Schr o¨ dinger, LLC,
New York, NY, 2017; Prime, Schr o¨ dinger, LLC, New York, NY, 2017.
[48] Schr o¨ dinger Release 2017-1: LigPrep, Schr o¨ dinger, LLC, New York, NY, 2017.
[49] Schr o¨ dinger Release 2017-1: Glide, Schr o¨ dinger, LLC, New York, NY, 2017.
[50] R.A. Friesner, J.L. Banks, R.B. Murphy, T.A. Halgren, J.J. Klicic, D.T. Mainz, M.P.
Repasky, E.H. Knoll, M. Shelley, J.K. Perry, D.E. Shaw, P. Francis, P.S. Shenkin,
Glide: A new approach for rapid, accurate docking and scoring. 1. Method and
assessment of docking accuracy, J. Med. Chem. 47(7) (2004) 1739–1749.
[51] T.A. Halgren, R.B. Murphy, R.A. Friesner, H.S. Beard, L.L. Frye, W.T. Pollard, J.
L. Banks, Glide: a new approach for rapid, accurate docking and scoring. 2.
Enrichment factors in database screening, J. Med. Chem. 47 (7) (2004) 1750–1759.
[52] T.A. Halgren, R.B. Murphy, R.A. Friesner, H.S. Beard, L.L. Frye, W.T. Pollard, J.
L. Banks, Glide: A new approach for rapid, accurate docking and scoring. 1. Method
and assessment of docking accuracy, J. Med. Chem. 47 (7) (2004) 1750–1759.
[53] https://github.com/pharmai/plip/blob/v1.4.4/DOCUMENTATION.md (Accessed
on 1 December 2020).
(
2016) 17–26.
[
[
[
26] A. Koeberle, U. Siemoneit, U. Buhring, H. Northoff, S. Laufer, W. Albrecht, O. Werz,
Licofelone suppresses prostaglandin E2 formation by interference with the
inducible microsomal prostaglandin E2 synthase-1, J. Pharmacol. Exp. Ther. 326
(
3) (2008) 975–982.
27] M.G. Chini, R. De Simone, I. Bruno, R. Riccio, F. Dehm, C. Weinigel, D. Barz,
O. Werz, G. Bifulco, Design and synthesis of a second series of triazole-based
compounds as potent dual mPGES-1 and 5-lipoxygenase inhibitors, Eur. J. Med.
Chem. 54 (2012) 311–323.
28] R. De Simone, M.G. Chini, I. Bruno, R. Riccio, D. Mueller, O. Werz, G. Bifulco,
Structure-based discovery of inhibitors of microsomal prostaglandin E2 synthase-1,
5
-lipoxygenase and 5-lipoxygenase-activating protein: promising hits for the
development of new anti-inflammatory agents, J. Med. Chem. 54 (6) (2011)
1
565–1575.
[54] https://github.com/pharmai/plip/blob/v1.4.4/plip/modules/config.py (Accessed
on 1 December 2020).
[
29] C. Greiner, H. Zettl, A. Koeberle, C. Pergola, H. Northoff, M. Schubert-Zsilavecz,
O. Werz, Identification of 2-mercaptohexanoic acids as dual inhibitors of 5-lipox-
ygenase and microsomal prostaglandin E(2) synthase-1, Bioorg. Med. Chem. 19
[55] D.J. Barlow, J.M. Thornton, Ion-pairs in proteins, J. Mol. Biol. 168 (4) (1983)
867–885.
(
11) (2011) 3394–3401.
[56] G.B. McGaughey, M. Gagne, A.K. Rappe, pi-Stacking interactions. Alive and well in
proteins, J. Biol. Chem. 273 (25) (1998) 15458–15463.
[
[
[
[
30] M. Hieke, C. Greiner, T.M. Thieme, M. Schubert-Zsilavecz, O. Werz, H. Zettl,
A novel class of dual mPGES-1/5-LO inhibitors based on the alpha-naphthyl
pirinixic acid scaffold, Bioorg. Med. Chem. Lett. 21 (5) (2011) 1329–1333.
31] K. Meirer, D. Steinhilber, E. Proschak, Inhibitors of the arachidonic acid cascade:
interfering with multiple pathways, Basic Clin. Pharmacol. Toxicol. 114 (1) (2014)
[57] Schr o¨ dinger Release 2017-1: Desmond Molecular Dynamics System, New York, NY,
2017. Maestro-Desmond Interoperability Tools, Schr o¨ dinger, New York, NY, 2017.
[58] C.H. Huang, J.R. Lapides, I.W. Levin, Phase-transition behavior of saturated,
symmetric chain phospholipid bilayer dispersions determined by Raman
spectroscopy: correlation between spectral and thermodynamic parameters, J. Am.
Chem. Soc. 104 (22) (1982) 5926–5930.
8
3–91.
32] O. Radmark, B. Samuelsson, Regulation of the activity of 5-lipoxygenase, a key
enzyme in leukotriene biosynthesis, Biochem. Biophys. Res. Commun. 396 (1)
[59] W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics, J. Mol.
Graph 14(1) (1996) 33-8, 27-8.
[60] M.J. Abraham, T. Murtola, R. Schulz, S. Pall, J.C. Smith, B. Hess, E. Lindahl,
GROMACS: High performance molecular simulations through multi-level
parallelism from laptops to supercomputers, SoftwareX 1–2 (2015) 19–25.
(
2010) 105–110.
33] J. Gerstmeier, C. Weinigel, S. Rummler, O. Radmark, O. Werz, U. Garscha, Time-
resolved in situ assembly of the leukotriene-synthetic 5-lipoxygenase/5-lipoxyge-
nase-activating protein complex in blood leukocytes, FASEB J. 30 (1) (2016)
2
76–285.
1
2