Discovery of a benzenesulfonamide-based dual inhibitor of microsomal prostaglandin E2 synthase-1 and 5-lipoxygenase that favorably modulates lipid mediator biosynthesis in inflammation

Article abstract of DOI:10.1016/j.ejmech.2018.07.031

Leukotrienes (LTs) and prostaglandin (PG)E₂, produced by 5-lipoxygenase (5-LO) and microsomal prostaglandin E₂ synthase-1 (mPGES-1), respectively, are key players in inflammation, and pharmacological suppression of these lipid mediators (LM) represents a strategy to intervene with inflammatory disorders. Previous studies revealed that the benzenesulfonamide scaffold displays efficient 5-LO-inhibitory properties. Here, we structurally optimized benzenesulfonamides which led to an N-phenylbenzenesulfonamide derivative (compound 47) with potent inhibitory activities (IC₅₀ = 2.3 and 0.4 μM for isolated 5-LO and 5-LO in intact cells, respectively). Compound 47 prevented the interaction of 5-LO with its activating protein (FLAP) at the nuclear envelope in transfected HEK293 cells as shown by in situ proximity ligation assay. Comprehensive assessment of the LM profile produced by human macrophages revealed the ability of 47 to selectively down-regulate pro-inflammatory LMs (i.e. LTs and PGE₂) in M1 but to enhance the formation of pro-resolving LMs (i.e. resolvins and maresins) in M2 macrophages. Moreover, 47 strongly inhibited LT formation and cell infiltration in two in vivo models of acute inflammation (i.e., peritonitis and air pouch sterile inflammation in mice). Together, 47 represents a novel LT biosynthesis inhibitor with an attractive pharmacological profile as anti-inflammatory drug that also promotes the biosynthesis of pro-resolving LM.

Full text of DOI:10.1016/j.ejmech.2018.07.031

European Journal of Medicinal Chemistry 156 (2018) 815e830
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Discovery of a benzenesulfonamide-based dual inhibitor of
microsomal prostaglandin E2 synthase-1 and 5-lipoxygenase that
favorably modulates lipid mediator biosynthesis in inflammation
,
,
Sun-Yee Cheung a 1, Markus Werner b 1, Lucia Esposito b, Fabiana Troisi b,
b
b
b
b
€
Vincenza Cantone , Stefanie Liening , Stefanie Konig , Jana Gerstmeier ,
Andreas Koeberle b, Rossella Bilancia c, Roberta Rizza c, Antonietta Rossi c,
Fiorentina Roviezzo c, Veronika Temml d, Daniela Schuster e, Hermann Stuppner d,
Manfred Schubert-Zsilavecz a, Oliver Werz b *, Thomas Hanke a **, Simona Pace b
,
,
, 1
a Institute of Pharmaceutical Chemistry, Goethe-University Frankfurt, Max-von-Laue-Str. 9, Frankfurt am Main, D-60438, Germany
b Department of Pharmaceutical/Medicinal Chemistry, Institute of Pharmacy, Friedrich-Schiller-University Jena, Philosophenweg 14, Jena, D-07743,
Germany
c Department of Pharmacy, School of Medicine, University of Naples Federico II, Via Domenico Montesano 49, Naples, 80131, Italy
d Institute of Pharmacy/Pharmacognosy and Center for Molecular Biosciences Innsbruck, University of Innsbruck, Innrain 80/82, Innsbruck, 6020, Austria
e Paracelsus Medical University Salzburg, Institute of Pharmacy, Department of Pharmaceutical and Medicinal Chemistry, Strubergasse 21, Salzburg, 5020,
Austria
a r t i c l e i n f o
a b s t r a c t
Article history:
Leukotrienes (LTs) and prostaglandin (PG)E2, produced by 5-lipoxygenase (5-LO) and microsomal pros-
taglandin E2 synthase-1 (mPGES-1), respectively, are key players in inflammation, and pharmacological
suppression of these lipid mediators (LM) represents a strategy to intervene with inflammatory disor-
ders. Previous studies revealed that the benzenesulfonamide scaffold displays efficient 5-LO-inhibitory
properties. Here, we structurally optimized benzenesulfonamides which led to an N-phenyl-
Received 26 May 2018
Received in revised form
5 July 2018
Accepted 11 July 2018
benzenesulfonamide derivative (compound 47) with potent inhibitory activities (IC50 ¼ 2.3 and 0.4
mM
for isolated 5-LO and 5-LO in intact cells, respectively). Compound 47 prevented the interaction of 5-LO
with its activating protein (FLAP) at the nuclear envelope in transfected HEK293 cells as shown by in situ
proximity ligation assay. Comprehensive assessment of the LM profile produced by human macrophages
revealed the ability of 47 to selectively down-regulate pro-inflammatory LMs (i.e. LTs and PGE2) in M1
but to enhance the formation of pro-resolving LMs (i.e. resolvins and maresins) in M2 macrophages.
Moreover, 47 strongly inhibited LT formation and cell infiltration in two in vivo models of acute
inflammation (i.e., peritonitis and air pouch sterile inflammation in mice). Together, 47 represents a
novel LT biosynthesis inhibitor with an attractive pharmacological profile as anti-inflammatory drug that
also promotes the biosynthesis of pro-resolving LM.
Keywords:
5-Lipoxygenase
Microsomal prostaglandin E2 synthase-1
Lipid mediators
Specialized pro-resolving mediators
Inflammation
© 2018 Elsevier Masson SAS. All rights reserved.
Abbreviations: 5-LO, 5-lipoxygenase; AA, arachidonic acid; COX, cyclooxygenase; cPLA2, cytosolic phospholipase A2; FLAP, 5-lipoxygenase-activating protein; fMLP, N-
formyl-methionyl-leucyl-phenylalanine; H(P)ETE, hydro(pero)xyeicosatetraenoic acid; HHT, hydroxyheptadecatrienoic acid; IL, interleukin; LM, lipid mediator; LPS, lipo-
polysaccharide; LT, leukotriene; mPGES-1, microsomal prostaglandin E2 synthase-1; PG, phosphate-buffered saline plus 1 mg/ml glucose; PGC, phosphate-buffered saline
plus 1 mg/ml glucose and 1 mM CaCl2; SPM, specialized pro-resolving mediators.
* Corresponding author. Chair of Pharmaceutical/Medicinal Chemistry, Institute of Pharmacy, Friedrich-Schiller-University Jena, Philosophenweg 14, Jena, D-07743,
Germany.
** Corresponding author.
E-mail addresses: oliver.werz@uni-jena.de (O. Werz), hanke@pharmchem.uni-frankfurt.de (T. Hanke).
Authors contributed equally to this work.
1
https://doi.org/10.1016/j.ejmech.2018.07.031
0223-5234/© 2018 Elsevier Masson SAS. All rights reserved.

816
S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
1. Introduction
in position R1 (Fig. 1), we synthesized the secondary amines by
nucleophilic aromatic substitution (Scheme 1). Here, we used
ethyl-4-fluorobenzoate and primary aliphatic amines to avoid side
reactions, which might occur under reductive amination condi-
tions, by using enolizable aldehydes. Next, the secondary amines
were coupled with different sulfonyl chlorides (Scheme 1, step II;
Scheme 2, step II; Scheme 3, step II and Scheme 6, step I). Because of
the electron withdrawing group in para position to the aniline
derivate, the nucleophilicity is strongly reduced, so different con-
ditions were evaluated for the sulfonamide coupling, yielding the
best results by using pyridine and 4-DMAP in DCM (Supporting
Information Table S1). Then, the ethyl ester was cleaved by using
LiOH in THF/water to obtain the corresponding benzoic acid de-
rivatives (Scheme 1, step III; Scheme 2, step III; Scheme 3, step III;
Scheme 5, step II; and Scheme 6, step III). In case of having an
alkyne derivative in position R2 (compound 43), Sonogashira
coupling (Scheme 6, step II) was performed prior to ester cleavage.
For the bioisosteric replacement of the benzoic acid to an N-acyl-
sulfonamide derivative the benzoic acid derivatives were coupled
with the corresponding sulfonamide moiety by using EDC and a
catalytical amount of 4-DMAP (Scheme 4, step I). Likewise, for
bioisosteric replacement of the central sulfonamide moiety by a
tertiary amide the secondary amine (13a) was coupled with
biphenyl-4-carbonyl chloride by using dry pyridine in dry DCM
(Scheme 5, Step I).
5-Lipoxygenase (5-LO) and microsomal prostaglandin E2
synthase-1 (mPGES-1) are crucial enzymes in the biosynthesis of
pro-inflammatory eicosanoids, namely leukotrienes (LT) and
prostaglandin (PG)E2 [1,2]. 5-LO catalyzes the initial transformation
of arachidonic acid (AA) to 5-hydroperoxyeicosatetraenoic acid (5-
HPETE) and its subsequent conversion to LTA4. Distinct enzymatic
reactions lead then from LTA4 to LTB4 or to the cysteinyl-LTs C4, D4
and E4 that play prominent roles in inflammatory and allergic re-
actions [1]. Thus, anti-LT therapy has been indicated for asthma,
allergic rhinitis, arthritis, neuroinflammatory disorders, cardiovas-
cular disease and cancer [1,3]. In the biosynthesis of PGE2, AA is first
converted by cyclooxygenases (COX)-1 and ꢀ2 to PGH2 which is
substrate for various prostanoid synthases that generate bioactive
PGs and thromboxane A2, but particularly, also for three terminal
PGES that convert PGH2 to produce PGE2 [2]. The mPGES-1 is an
inducible, membrane-bound enzyme that is often co-expressed
with the inducible COX-2 and is suggested to be responsible for
the massive PGE2 formation at sites of inflammation due to the
coupled catalysis of COX-2 and mPGES-1 [4]. Therefore, selective
targeting of mPGES-1-derived PGE2 is considered as superior
pharmacological strategy over COX-1/2 inhibition by non-steroid-
anti-inflammatory drugs (NSAIDs) and coxibs, because adverse
on-target effects of these drugs such as gastric, renal and cardio-
vascular toxicity might be avoidable [4,5].
The dual interference with PGE2 and LT biosynthesis by target-
ing both mPGES-1 and 5-LO or its helper protein 5-LO-activating
protein (FLAP) is considered to be superior over single interference
in terms of both efficacy and adverse effects [4,6,7]. Thus, both
types of pro-inflammatory eicosanoids are blocked within the COX/
5-LO pathway and shunting phenomena that can occur by single
interference with one pathway [8,9] can be avoided. Although
several dual mPGES-1/5-LO inhibitors have been identified from
synthetic and natural sources [4,7], their pharmacological assess-
ment and their efficacy in vivo have often been incompletely
investigated. Along these lines, the modulation of the biosynthesis
of the plethora of oxylipins within the complex lipid mediator (LM)
network has often been neglected [8,9]. Here, we focus on a lead
structure that was recently identified by a virtual screening
approach [10]. The hit compound (1) with a benzenesulfonamide
scaffold showed moderate potencies against human mPGES-1 and
2.2. Biological assays
Inhibition of 5-LO and mPGES-1 was analyzed in cell-free assays
using partially purified human recombinant 5-LO and microsomes
from IL-1b-treated human A549 cells that abundantly express
mPGES-1. Under these assay conditions, inhibition of enzymatic
activity is most likely due to direct interference of the test com-
pounds with the target enzymes. As reference drugs, the 5-LO in-
hibitor zileuton (N-[1-(1-benzothien-2-yl)ethyl]-N-hydroxyurea)
and the mPGES-1 inhibitor MK886 (3-[3-(tert-butylsulfanyl)-1-[(4-
chlorophenxyl)methyl]-5-(propan-2-yl)-1H-indol-2-yl]-2,2-
dimethylpropanoic acid) were used that inhibited the formation of
5-LO products and PGE2 in the assays with IC50 ¼ 0.6 and 2.4
mM,
respectively, as expected. In addition, the test compounds were
assessed for inhibition of 5-LO activity in intact cells, using neu-
trophils stimulated with 2.5
20 M AA as substrate, and the 5-LO products LTB4 and its trans-
m
M Ca2þ-ionophore A23187 plus
5-LO with IC50 values of 3.7 and 5.7
mM, respectively (Fig. 1). Thus,
m
we developed a simple straightforward synthetic approach to
design and achieve various derivatives for determining structure-
activity relationships (SAR), eventually leading to the optimized
compound 47 that inhibits both targets with IC50 values in the
submicromolar range. Compound 47 is highly efficient in sup-
pressing 5-LO product formation in intact leukocytes and displays
significant in vivo activity in two animal models of acute inflam-
mation. Moreover, we present metabololipidomics data for 47,
addressing the modulation of the biosynthesis of multiple eicosa-
noids and docosanoids, and we provide a comprehensive phar-
macological assessment of 47 in cell-free and cell-based assays.
isomers, and 5-H(P)ETE were determined. In intact cells, the test
compounds may affect cellular co-factors (ATP, Ca2þ, lipids etc.) or
other proteins involved in LT biosynthesis such as FLAP, LTA4 hy-
drolase, or LTC4 synthase. Analysis of direct interference of a given
compound with mPGES-1 in the cell (i.e., the transformation of
exogenously added PGH2 to PGE2) is not feasible. Descriptions of
these assays are summarized in the Experimental Part.
2.3. SAR for 5-LO and mPGES-1
Starting from 1 as lead, we maintained the benzenesulfonamide
scaffold and focused on modifications of three different residues: (i)
the n-octyl (R1), (ii) the methyl (R2) of the tosyl moiety and (iii) the
carboxylic acid residue (R3). Shortening the n-octyl (R1) to n-hexyl
(2) or to n-butyl (3) as well as replacement by cyclohexylmethyl (4),
benzyl (5), p-tolyl-methyl (6), or 4-, 3-, or 2-chlorophenyl-methyl
(7e9) strongly reduced the potency against both 5-LO and mPGES-
2. Results and discussion
2.1. Chemistry
The benzensulfonamide derivatives were synthesized as
described in Schemes 1e4 and 6. In general, the syntheses of the
secondary amines were performed under reductive amination
conditions using ethyl-4-aminobenzoate with different aldehydes
and sodium triacetoxyborhydride as reducing agent (Scheme 2,
step I, and Scheme 3, step I). However, for insertion of alkyl chains
1 (IC50 > 10
the 2-chloro-6-fluorobenzyl derivative 11 were inactive up to
10 M in all investigated assay systems. However, dichlorination to
the 3,4-dichlorobenzyl derivative 12 restored activity and inhibited
5-LO with IC50 of 2.4 M in neutrophils and mPGES-1 with
mM; Table 1). Also, the 3,4-dimethoxybenzyl (10) and
m
m

S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
817
Fig. 1. From the lead structure to the SAR of the N-phenylbenzenesulfonamide derivatives.
a
Scheme 1. Synthesis of compounds 2; 3; 28e30. Reagents and conditions: Step I) n-alkylamine (1.0 equiv), ethyl-4-fluorobenzoate (1.0 equiv), Cs2CO3 (1.5 equiv), DMF, 110 ꢁC,
20 h; Step II) sulfonyl chloride (1.0 equiv), pyridine (1.0 equiv), 4-DMAP (catalytically), DCM, 40 ꢁC, 18 h; Step III) LiOH*H2O (5.0 equiv), THF/water, 60 ꢁC, 18 h.
IC50 ¼ 6.3
m
M. These results encouraged us to go on with more
showed potent mPGES-1 inhibition (IC50 ¼ 2.7
mM) but 5-LO ac-
bulky residues in this position such as 2-naphthyl-methly (13),
which however, was less potent in neutrophils versus 12, and
without activity against 5-LO and mPGES-1 in cell-free assays. Of
interest, introduction of 4-phenoxy-benzyl (14) led to activity
against 5-LO in neutrophils and mPGES-1, but still lacked sub-
stantial activity against 5-LO in the cell-free assay. Smaller aromatic
moieties such as thiophene-methyl (15, 16) were again inactive.
Recently, Elkady et al. [11] have shown that bioisosteric
replacement of the carboxylic acid group of common NSAIDs by an
N-acylsulfonamide moiety yields dual 5-LO/mPGES-1 inhibitors.
Therefore, we replaced the carboxylic acid group in 5 by an N-
acylsulfonamide moiety resulting in compound 17, which indeed
tivity was not affected.
Next, we maintained the cyclohexylmethyl or the 2-
thiophenylmethyl residues of inactive 4 and 15 and modified the
methyl group (R2) of the tosyl moiety (Table 2). First, the methyl
residue of 15 was replaced by small halogens like fluorine (18) and
chlorine (19) and also by more polar groups like a carboxylic acid
moiety (20), but without any effect on the two targets
(IC50 > 10 mM). Also, conversion of the benzoic acid of 15 to 2-
phenylacetic acid (21) failed in this respect. Increasing the lip-
ophilicity by replacement of the methyl moiety (R3) by 2-naphthyl
(22), failed in this respect as well. However, incorporation of a
phenyl residue (23) instead of methyl restored the activities, except

818
S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
Scheme 2. Synthesis of compounds 2e16; 18e20; 22e27; 31e34; 37e39; 41; 42; 44e47. a Reagents and conditions: Step I) Ethyl-4-aminobenzoate (1.0 equiv), carboxaldehyde (1.1
equiv), AcOH (2.0 equiv), DCE, RT, 2 h, NaBH(OAc)3 (1.4 equiv), RT, 18 h; Step II) sulfonyl chloride (1.0 equiv), pyridine (1.0 equiv), 4-DMAP (catalytically), DCM, 40 ꢁC, 18 h; Step III)
LiOH*H2O (5.0 equiv), THF/water, 60 ꢁC, 18 h.
against isolated 5-LO. Comparable SARs were obtained for the
cyclohexylmethyl series. Thus, replacement of the methyl by fluo-
rine in 24 was again ineffective while a phenyl residue (25) led to
comparable efficiency similar as 23. Since the 4-phenoxybenzyl
group (R1) in 14 was beneficial, we introduced this moiety also in
R3 of 24 yielding 26 with activity against both mPGES-1 and 5-LO in
neutrophils. Of interest, when also the cyclohexylmethyl in 26 was
replaced by phenoxyphenyl (27) a marked increase in potency was
R2 and the 2-naphthylmethyl moiety in R1 we tried elongation of
the benzoic acid derivative to a phenylacetic acid derivative (35)
and a phenylpropionic acid derivative (36) which however, resulted
in impaired potency versus 34.
Next, we replaced also the sulfonamide moiety of 34 by a car-
boxamide group, yielding 40, which essentially retained suppres-
sion of 5-LO product formation but was detrimental for mPGES-1.
However, when the naphthylmethyl moiety in 34 was changed to a
4-tert-butylbenzyl residue (41) the potency against 5-LO in intact
obtained against mPGES-1 (IC50 ¼ 0.8
mM) and 5-LO in neutrophils
(IC50 ¼ 1.0
mM).
cells (IC50 ¼ 0.3
mM) and in the cell-free assay further increased,
In a third series we mainly maintained the biphenyl system, that
is, the phenyl residue as R2 (Table 3) and focused on position R1.
Compound 28 with an n-heptyl substituent was 10-fold more
although potency against mPGES-1 was reduced. Replacing the
biphenyl residue by a tert-butyl-phenyl moiety in 42 or elongation
of the tert-butyl scaffold with a rigid alkyne moiety in 43 slightly
decreased the potencies against 5-LO, although 43 was somewhat
more active against mPGES-1.
potent against 5-LO in neutrophils than 1, with an IC50 of 0.5 mM
with minor effectiveness against isolated 5-LO and mPGES-1.
Shortening to the n-hexyl (29) or to n-butyl (30) derivatives
clearly reduced the potency. Replacement of n-octyl of 28 by benzyl
(31) slightly improved mPGES-1 inhibition without effects against
5-LO. While the smaller heterocyclic furan-3-yl residue in 33 was
again less potent, insertion of the more bulky 2-naphthylmethyl
moiety (34) strongly improved the activity against mPGES-1
Finally, we replaced the naphthylmethyl by aromatic ether de-
rivatives such as 3-methoxyphenyl (44), that was similar potent to
34 against 5-LO in neutrophils but was 20-fold less potent against
mPGES-1. When a 3-phenoxyphenyl group was introduced instead
(45), then 5-LO inhibition in neutrophils was improved
(IC50 ¼ 0.3
mM) with minor impact on inhibition of isolated 5-LO
(IC50 ¼ 0.3
m
M) and against 5-LO product formation in neutrophils
and mPGES-1. Insertion of a methylene spacer yielding the 4-
benzyloxyphenyl derivative 46 was similar efficient in neutrophils
and on isolated 5-LO as compared to 45, but less potent against
mPGES-1. Shifting the benzyloxy group from 4- to 3-position led to
47 that was similar potent than 46 in neutrophils with
(IC50 ¼ 1.1
mM). Replacement of the 2-naphthyl moiety by quino-
lone-3-yl-methyl (37), by 2,3-dihydro-1,4-benzodioxin-6-yl-
methyl (38) or by 2H-1,3-benzodioxol-5-yl-methyl (39) groups all
failed to improve potency versus 34. Maintaining the biphenyl as

S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
819
a
Scheme 3. Synthesis of compounds 21; 35; 36. Reagents and conditions: Step I) Aniline derivative (1.0 equiv), carboxaldehyde (1.1 equiv), AcOH (2.0 equiv), DCE, RT, 2 h,
NaBH(OAc)3 (1.4 equiv), RT, 18 h; Step II) sulfonyl chloride (1.0 equiv), pyridine (1.0 equiv), 4-DMAP (catalytically), DCM, 40 ꢁC, 18 h; Step III) LiOH*H2O (5.0 equiv), THF/water, 60 ꢁC,
18 h.
a
Scheme 4. Synthesis of compound 17. Reagents and conditions: Step I) benzensulfonamide (1.2 equiv), EDC (1.2 equiv), 4-DMAP (catalytically), DCM, 0 ꢁC to RT, 2 h.
a
Scheme 5. Synthesis of compound 40. Reagents and conditions: Step I) Biphenyl-4-carbonyl chloride (1.0 equiv), pyridine (1.0 equiv), DCM, 40 ꢁC, 18 h; Step II) LiOH*H2O (5.0
equiv), THF/water, 60 ꢁC, 18 h.
(IC50 ¼ 0.4
(IC50 ¼ 0.7
m
M), but about 9-fold more potent against mPGES-1
(Fig. 2, Supporting Information, Fig. S1). The poses were analyzed
by taking into account three factors: the presence of key hydrogen
bonds, the docking score, and the similarity of the outcome poses
(see Experimental Procedures).
mM). Also, 47 was one of the most potent derivatives
against isolated 5-LO (IC50 ¼ 2.3
mM).
The series displayed a clear binding pattern within mPGES-1.
The binding site (Fig. 2C) is formed by a large hydrophobic cavity,
contains an acid binding site, formed by two Arg residues (38 and
126), and offers three potential hydrogen bond partners for the
sulfonamide, i.e., Gln36, Thr131 and Ser127.
2.4. Molecular docking on mPGES-1 and 5-LO
The original compound and the new derivatives were
1
computationally docked to the binding sites of the two target en-
zymes 5-LO and mPGES-1 to gain more insights into their SAR (pdb
entry 4bpm [12] for mPGES-1 and pdb entry 3o8y [13] for 5-LO)
The docking simulations showed that the size of R2 and R3 were

820
S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
Scheme 6. Synthesis of compound 43. a Reagents and conditions: Step I) 4-iodobenzenesulfonyl chloride (1.0 equiv), pyridine (1.0 equiv), 4-DMAP (catalytically), DCM, 40 ꢁC, 18 h;
Step II) tert-butylacetylene (2.0 equiv), PdCl2(PPh3)2 (0.02 equiv), CuI (0.05 equiv), TEA, RT, 2 h; Step III) LiOH*H2O (5.0 equiv), THF/water, 60 ꢁC, 18 h.
Table 1
Inhibition of 5-LO and mPGES-1 by benzenesulfonamide derivatives. The residual enzyme activity at 10
m
M compound concentration is given in %; for compounds that
inhibited by > 50% at 10 mM the IC50 values were determined and are given in m
M. Data are expressed as means SEM; n ¼ 3e4.
Compound
R1
5-LO activity (cell-based)
residual (%) IC50 M)
5-LO activity (cell-free)
residual (%) IC50 M)
mPGES-1 (cell-free)
residual (%) IC50 M)
(
m
(
m
(m
R2
R3
1
2
3
4
5
6
7
8
n-heptyl
n-pentyl
n-propyl
cyclohexyl
phenyl
p-tolyl
4-Cl-phenyl
3-Cl-phenyl
2-Cl-phenyl
3,4-dimethoxyphenyl
2-Cl-6-F-phenyl
3-Cl-4-Cl-phenyl
2-naphthyl
4-phenoxyphenyl
thiophen-2-yl
thiophen-3-yl
phenyl
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-CH3
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-CONHSO2Ph
4.5
m
M
5.7
m
M
3.7 mM
65.2 10.4%
111.8 10.3%
76.9 1.3%
75.7 25.9%
96.7 4.6%
83.2 6.5%
84.7 5.1%
95.2 3.5%
110.3 12.4%
75.4 5.9%
2.4 0.4
4.8 1.0
3.5 0.5
103.8 5.5%
98.2 7.2%
80.0 6.0%
86.1 8.3%
99.1 0.1%
91.4 12.1%
81.1 11.9%
93.4 6.6%
100.5 7.8%
106.9 12.6%
115.1 12.9%
93.2 8.5%
86.7 6.0%
97.6 5.9%
96.7 11.0%
61.4 7.8%
92.7 2.2%
106.1 11.6%
88.94 14.1%
79.9 2.6%
79.9 2.6%
91.6 4.3%
91.3 5.1%
82.8 1.6%
71.2 5.1%
74.1 þ 8.1%
71.7 8.0%
95.7 7.1%
85.7 1.1%
6.3 0.6
52.8 1.7%
5.3 0.9 mM
92.5 9.4%
90.7 5.9%
9
10
11
12
13
14
15
16
17
m
m
m
M
M
M
mM
2.7 0.3 mM
crucial for positioning the ligands in such manner that hydrogen
bonds between the sulfonamide group and the protein could be
formed. While smaller groups in R1, e.g. 15, 16 allowed for more
freedom within the binding pocket in the simulation (and conse-
quently very diverse poses), larger groups, as seen in 47 (Fig. 2A,E)
led to ideal positioning for hydrogen bonding and a more conclu-
sive positioning. Compound 47 binds to the acid binding site, and
the sulfonamide group acts as a hydrogen bond acceptor for Gln36.
In addition, a hydrogen bond between Thr131 and the ether func-
tion of the 3-benzyloxyphenyl residue is formed.
and membrane binding domain of 5-LO, described in Ref. [13]. The
anatomy of this binding site (Fig. 2D) in some way resembles the
one found in mPGES-1, explaining why many dual ligands for those
two proteins have been identified [7]. One such common feature is
the (fatty) acid binding site, combined with hydrogen bond donors
within a similar distance in both enzymes. There is an acid binding
site formed by Tyr383, Thr403 and Arg401, a large hydrophobic
cavity and putative hydrogen bonding partners for the sulfonamide
group on one side of the binding pocket, Tyr81, and Trp102.
A similar SAR pattern as for mPGES-1 was observed for 5-LO. A
good fit in the binding pocket, with both the sulfonamide and the
carboxyl group being positioned within range to form interactions
with the protein, mostly requires a large group in R1 position. Many
Since 47 did not inhibit 5-LO in a competitive manner (see
below), it is very likely that the compound inhibits the enzyme by
binding to an allosteric binding site, located between the catalytic

S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
821
Table 2
Inhibition of 5-LO and mPGES-1 by benzenesulfonamide derivatives. The residual enzyme activity at 10
mM compound concentration is given in %; for compounds that
inhibited by > 50% at 10 mM the IC50 values were determined and are given in m
M. Data are expressed as means SEM; n ¼ 3e4.
Compound
R1
5-LO activity (cell-based)
5-LO activity (cell-free)
mPGES-1 (cell-free)
residual (%) IC50 (mM)
residual (%) IC50
(m
M)
residual (%) IC50
(
m
M)
R2
R3
18
19
20
21
22
23
24
25
26
27
thiophen-2-yl
-F
-Cl
-COOH
-COOH
-COOH
-CH2-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
94.6 6.6%
97.2 4.6%
111.5 3.2%
101.6 0.5%
69.7 4.3%
86.7 13.9%
86.1 17.5%
93.0 11.8%
97.0 6.0%
71.9 12.9%
66.4 6.4%
109.3 5.8%
57.5 7.4%
72.0 3.6%
96.5 5.3%
96.7 9.7%
94.6 7.7%
96.4 7.3%
76.7 2.8%
thiophen-2-yl
thiophen-2-yl
thiophen-2-yl
thiophen-2-yl
thiophen-2-yl
cyclohexyl
cyclohexyl
cyclohexyl
3-phenoxyphenyl
-COOH
-CH3
-2-naphthyl
-Ph
-F
-Ph
4.4 1.0
mM
4.7 1.7
mM
98.7 4.5%
88.9 2.9%
3.2 0.6
2.6 0.6
0.8 0.2
mM
mM
mM
3.4 0.8
5.4 0.3
1.0 0.1
m
m
m
M
M
M
-phenoxy
-phenoxy
6.0 0.2 mM
Table 3
Inhibition of 5-LO and mPGES-1 by benzenesulfonamide derivatives. The residual enzyme activity at 10
mM compound concentration is given in %; for compounds that
inhibited by > 50% at 10 mM the IC50 values were determined and are given in m
M. Data are expressed as means SEM; n ¼ 3e4.
Compound
R1
5-LO activity (cell-based)
residual (%) IC50 M)
5-LO activity (cell-free)
residual (%) IC50 M)
mPGES-1 (cell-free)
residual (%) IC50 M)
(
m
(m
(m
R2
R3
28
29
30
31
32
33
34
35
36
37
38
39
40
n-heptyl
n-pentyl
n-propyl
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-Ph
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-CH2-COOH
-(CH2)2-COOH
-COOH
-COOH
-COOH
0.5 0.2
1.5 0.3
4.0 0.4
3.3 0.2
1.0 0.2
8.4 0.9
1.1 0.1
3.4 1.4
1.8 0.7
2.1 0.5
1.1 0.3
3.2 1.2
1.7 0.6
m
m
m
m
m
m
m
m
m
m
m
m
m
M
M
M
M
M
M
M
M
M
M
M
M
M
5.2 0.1
m
M
5.0 0.3
1.9 1.1
52.6 3.5%
4.0 0.6
5.2 1.0
56.8 1.9%
0.3 0.04 mM
1.1 0.1
0.7 0.1
4.7 0.8
1.6 1.4
1.9 1.4
2.9 0.4
m
m
M
M
105.0 10.9%
114.7 8.6%
96.2 7.2%
65.6 4.9%
109.7 7.5%
6.4 0.5
8.7 3.3
5.5 0.5
75.0 13.0%
116.9 5.1%
88.4 4.3%
78.0 11.2%
m
m
M
M
-p-tolyl
-furan-3-yl
-2-naphthyl
-2-naphthyl
-2-naphthyl
quinoline-3-yl
m
m
m
M
M
M
m
M
M
M
M
M
M
m
m
m
m
m
2,3-dihydro-1,4-benzodioxin-6-yl
2H-1,3-benzodioxol-5-yl
41
42
43
44
45
46
47
4-tert-butylphenyl
4-tert-butylphenyl
4-tert-butylphenyl
3-methoxyphenyl
3-phenoxyphenyl
4-benzyloxyphenyl
3-benzyloxyphenyl
-Ph
-4-tert-butyl
3,3-dimethylbutyn-1-yl
-Ph
-Ph
-Ph
-Ph
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
-COOH
0.3 0.1
0.7 0.4
0.9 0.4
1.3 0.4
0.3 0.1
0.5 0.1
0.4 0.1
m
m
m
m
m
m
m
M
M
M
M
M
M
M
5.6 1.5
50.2 5.3%
2.3 0.6
78.6 4.5%
m
M
1.2 0.5
3.2 0.3
2.7 0.5
6.2 0.2
3.8 0.5
6.5 0.7
m
m
m
m
m
m
M
M
M
M
M
M
mM
7.2 0.5
6.6 0.5
2.3 0.4
m
m
m
M
M
M
0.7 0.04 mM
of the derivatives with thiophenyl in R1 position were not placed in
the binding site at all by the docking program or were placed with
very diverse poses, while e.g. 27 containing a 3-phenoxyphenyl or
47 with the 3-benzyloxyphenyl group in R1 were placed steadily,
forming all key interactions (see Fig. 2B,F). The carboxyl group of 47
was predicted to interact with Tyr383 and the sulfonamide in-
teracts with Tyr81. Together, the docking simulations underscore
the similarities between the two binding modes, explaining why so
many of the derivatives also showed activities against both, 5-LO
and mPGES-1.
2.5. Mode of action of compound 47 to inhibit 5-LO product
formation
Compound 47 emerged as potent dual 5-LO/mPGES-1 inhibitor
that exhibits 8-fold superior activity against 5-LO in intact cells
(IC50 ¼ 0.4
m
M) as compared to a cell-free assay (IC50 ¼ 2.3
mM,
Fig. 3A). First, this strong efficiency in intact cells renders 47 an
interesting candidate with potentially high efficacy in a biological
relevant environment. Second, the higher potency in the cell-based
assay versus isolated 5-LO implies potential interference with other

822
S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
Fig. 2. Binding sites and putative binding modes for 47 in mPGES-1 (A, C, E) and 5-LO (B, D, F). A) Docking pose of 47 in mPGES-1 (pdb entry 4bpm [12]). The binding site is
shown in detail in C) and composed of a large hydrophobic cavity, an acid binding site, formed by Arg38 and Arg126, and three potential hydrogen bond interaction partners Gln36,
Thr131 and Ser127. E) Compound 47 binds to the acid binding site (red star) and forms hydrogen bonds (red arrows) with Gln36 via the sulfonamide group and with Thr131 via the
ether moiety. B) Docking pose of 47 in 5-LO (based on pdb entry 3o8y [13]). D) The binding site is structured in a manner that partly corresponds to the binding site of mPGES-1,
there is an acid binding site formed by Tyr383, Thr403 and Arg401, a large hydrophobic cavity and hydrogen bond partners for the sulfonamide group on one side of the binding
pocket (Tyr81 and Trp102). F) Compound 47 is predicted to form an ionic interaction with Tyr383 and a hydrogen bond with Tyr 81, while the hydrophobic cavity is filled with the
ring system. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
key players in the biosynthesis of 5-LO products such as cPLA2 or
FLAP. Thus, we continued with further pharmacological in-
vestigations on 47. Neutrophils were supplemented with exoge-
significantly compromised (data not shown). Moreover, the effi-
ciency of 47 to inhibit 5-LO in the cell-free assay was independent
of the substrate concentration (Fig. 4), excluding a substrate-
competitive mechanism of 47 for interaction with 5-LO. Together,
47 is a direct 5-LO inhibitor with increased potency in intact cells
that is not primarily related to modulation of substrate release by
cPLA2.
nous AA (20 mM) and stimulated with A23187 in order to
circumvent cPLA2-mediated AA supply for 5-LO product formation.
Interestingly, addition of exogenous AA did not impair the potency
of 47 (Fig. 3A). Nevertheless, 47 slightly impaired the release of AA
in human neutrophils at concentrations (10
m
M) that clearly
Besides the absolute supply of endogenous substrate for 5-LO,
the accessibility of AA is another crucial event in 5-LO product
formation of intact cells and depends on the transfer of AA from the
nuclear membrane-bound FLAP to 5-LO [14]. Thus, FLAP inhibitors
efficiently block 5-LO product formation in intact cells by impeding
5-LO/FLAP complex formation at the nuclear envelope and thereby
AA transfer to 5-LO [15,16]. We investigated whether 47 could
inhibited 5-LO product biosynthesis in these cells (Fig. 3B). Since
also the cPLA2 inhibitor pyrrolidine was much more efficient, we
conclude that inhibition of cPLA2 is not the primary cause of potent
suppression of 5-LO product formation by 47. Note that unspecific
detrimental effects of 47 on intact cells such as acute cytotoxicity
can be excluded since cell viability (at 10 mM 47) was not

S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
823
Fig. 3. Effect of compound 47 on 5-LO product formation. A) Inhibition of 5-LO activity in cell-based (human neutrophils) and cell-free (purified 5-LO) assays. Purified 5-LO
(0.5
g/ml, at 4 ꢁC) or human neutrophils (at 37 ꢁC) were pre-incubated with 47 or vehicle (0.1% DMSO) for 15 min. For the cell-free assay, samples were pre-warmed for 30 s at
37 ꢁC, and 2 mM CaCl2 and 20
M AA were added. For the intact cell assay, neutrophils were activated with 2.5 M A23187 with or without 20 M AA. After 10 min 5-LO product
formation was determined. B) Analysis of [3H]AA in supernatants of [3H]AA-labeled neutrophils after pre-incubation with 47 (10
M) or cPLA2 inhibitor pyrrolidine (1 M) and
stimulation with 2.5
M A23187. Data are expressed as means þ S.E.M.; n ¼ 3e4 independent experiments. **P < 0.01; ***P < 0.001 versus 100% control.
m
m
m
m
m
m
m
prevent 5-LO nuclear translocation and/or the interaction between
5-LO and FLAP at the nuclear membrane in 5-LO/FLAP-transfected
HEK293 cells [17], making use of an in situ proximity ligation assay
(PLA) [14]. Although 47 failed to block 5-LO translocation (Fig. 5A),
it clearly diminished the fluorescent signal of 5-LO/FLAP interaction
(Fig. 5B), a typical feature of FLAP inhibitors, such as MK886 (not
shown), as reported by us before [14,15].
2.6. Effect of compound 47 on the lipid mediator profile of activated
human macrophages
In neutrophil incubations with A23187 and AA, other LOs (such
as 12-LO and 15-LO) are activated and contribute to transformation
of AA. Therefore, we evaluated whether 47 affects formation of
other LMs by these related LOs. Interestingly, while 5-LO activity
was strongly suppressed by 47, the production of 12-HETE and 15-
HETE (formed by 12-LO and 15-LO, respectively) were increased
(Fig. 6A), prompting for a more detailed investigation of LM profiles
produced by various LOs and COXs. Hence, human monocyte-
derived macrophages were polarized towards pro-inflammatory
M1-like cells (producing mainly LTs and PGs) and towards pro-
resolving M2-like macrophages (forming substantial amounts of
SPMs), pre-treated with 47 (30 min) and further challenged with
pathogenic E. coli as described recently [18]. In M1 macrophages, 47
inhibited the formation of pro-inflammatory LMs derived from 5-
LO. Of interest, in M2 macrophages 47 increased the production
of several SPM, particularly RvD2, RvD5 and MaR1, including their
monohydroxylated precursors 17-HDHA and 14-HDHA as well as
other 12/15-LO-derived LMs (i.e., 12-HEPE, 12-HETE, 15-HETE and
15-HEPE) (Fig. 6B). These data suggest that 47 might beneficially
affect the resolution of inflammation. As expected, PGE2 formation
was suppressed by 47 due to its mPGES-1-inhibitory effect, while
other COX-derived products were neither reduced in M1 nor in M2
Fig. 4. Effects of compound 47 on the activity of human recombinant 5-LO at
varying AA substrate concentrations. Purified 5-LO was incubated with 47 or vehicle
(DMSO, 0.1%) for 15 min at 4 ꢁC and stimulated with the indicated concentrations of AA
for 10 min at 37 ꢁC. Data are expressed as percentage of control (100%), means þ S.E.M.,
n ¼ 3.
MAPEG family member related to mPGES-1 and FLAP (data not
shown).
2.7. Efficiency of compound 47 in in vivo models of acute
inflammation related to LTs
Based on the favorable activity of 47 against 5-LO product for-
mation along with increased production of SPMs, we next evalu-
ated the effectiveness of this compound in two in vivo models of
acute inflammation that are both critically related to LTs [19e21].
(Fig. 6B). Accordingly, 47 (10
in cell-free assays (data not shown). Additionally, 47 displayed
inhibitory effects against LTC4 synthase (IC50 ¼ 1.8 M) another
mM) failed in blocking COX-1/2 activity
m

824
S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
Fig. 5. Influence of compound 47 on 5-LO subcellular localization and 5-LO/FLAP
interaction. A) Subcellular localization of 5-LO and FLAP. HEK293 cells (stably trans-
fected with 5-LO and FLAP) were pre-incubated with 47 (10
mM) or vehicle for 10 min
and then stimulated with 2.5
m
M A23187 for 20 min at 37 ꢁC. Images show single
staining for FLAP (green), 5-LO (red), and overlay. Results are representative for 100
individual cells of three independent experiments. (B) 5-LO/FLAP interaction. In situ
PLA, using proximity probes against 5-LO and FLAP, was performed in HEK293 cells
that were pre-incubated and stimulated as described above. DAPI (blue) was used to
stain the nucleus and PLA signals (magenta dots) visualize 5-LO/FLAP interaction.
Results are representative for 100 individual cells analyzed in three independent ex-
periments. (For interpretation of the references to colour in this figure legend, the
reader is referred to the Web version of this article.)
Fig. 6. Modulation of the lipid mediator profile in human neutrophils and mac-
rophages by compound 47. A) Effect of 47 on the formation of 5-LO products, 12-HETE
and 15-HETE in human neutrophils stimulated with A23187 (2.5 mM) plus AA (20 mM)
for 10 min at 37 ꢁC. B) Modulation of LM in M1 and M2 macrophages by 47. Human
monocytes were differentiated with GM-CSF or M-CSF (20 ng/ml, each) for 6 days. Cells
were either polarized with 100 ng/ml LPS plus 20 ng/ml IFN-g (GM-CSF) to obtain M1,
or with 20 ng/ml IL-4 (M-CSF) to obtain M2 for 48 h. M1 and M2 were pre-incubated
with 47 for 15 min at 37 ꢁC and then challenged with E. coli (O6:K2:H1; ratio ¼ 1:50)
for another 90 min. LM were analyzed via LC-MS/MS. Data are shown as heat map
representation of the changes in the percentage of LM produced by M1 and M2
macrophages to the vehicle (100%) control.
We first analyzed 47 in zymosan-induced peritonitis in mice
(10 mg/kg, intraperitoneally (i.p.)) against the production of LTC4.
Thus, pre-treatment of mice with 47 (30 min before zymosan in-
jection), significantly (P < 0.001) reduced LTC4 formation with
comparable efficiency as the well-known LT biosynthesis inhibitor
zileuton (Fig. 7A). As consequence of LT inhibition, cell recruitment
into the peritoneal cavity (4 h after peritonitis induction) was
impaired after treatment with 47 (Fig. 7B). Similar beneficial effects
were evident in the air pouch model in mice, where i. p. injection of
47 (10 mg/kg) significantly reduced LTB4 formation (P < 0.01) as
well as leukocyte recruitment (P < 0.05) in the pouches upon
zymosan challenge (Fig. 7C and D). Again, the efficiency of 47 in this
respect was comparable to that of zileuton (10 mg/kg). Together,

S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
825
Fig. 7. In vivo pharmacology of compound 47. A,B) Effect of 47 in zymosaneinduced peritonitis in mice. A) Effect of 47 and zileuton (10 mg/kg, each) on LTC4 production 30 min
after i. p. zymosan (0.5 mL of 2 mg/mL suspension) injection. B) Effect of 47 and zileuton (10 mg/kg, each) on cell recruitment in the peritoneal cavity 4 h after zymosan injection.
C,D) Effect of 47 and zileuton on air pouch model of acute inflammation in mice. Air pouches were developed by a subcutaneous injection (2.5 mL) of sterile air into the back of mice.
Compound 47 and zileuton were (10 mg/kg, each) given i. p., 30 min before zymosan injection into the pouch. After 4 h mice were killed and LTB4 (C) as well as infiltrated cells (D)
were measured in the inflammatory exudate. Data are expressed as means þ S.E.M., n ¼ 6 animals per group. Statistical analysis was performed by using ANOVA two-tailed test
followed by Bonferroni post hoc test or by using student t-test. *P < 0.05; **P < 0.01; ***P < 0.001 versus vehicle control.
compound 47 strongly inhibits LT biosynthesis in vivo and in vitro
seemingly due to direct interference with 5-LO and due to pre-
vention of 5-LO/FLAP interaction, along with anti-inflammatory
effectiveness in mice.
moiety by a biphenyl moiety were critical structural modifications
leading to improved IC50 values of 47 (2.3 M and 0.7 M for 5-LO
and mPGES-1). It is remarkable, however, that 47 inhibits 5-LO
quite potently in intact cells (IC50 ¼ 0.4 M), possibly due to pre-
m
m
m
vention of the 5-LO/FLAP interaction which is an essential process
for 5-LO activity in the cell. Another attractive characteristics of 47
is the redirection of LM formation in M2 macrophages towards
SPMs (i.e., RvD2, RvD5, and MaR1) and their 12/15-LO-derived
monohydroxylated precursors. Upregulation of 12- and 15-HETE by
dual 5-LO/mPGES-1 inhibitors [25] was observed before and was
discussed as class specific feature of dual 5-LO/mPGES-1 inhibitors
[8,9]. However, an upregulation of SPMs has not been demon-
strated before for a dual 5-LO/mPGES-1 or eicosanoid multi-
targeting inhibitor. We conclude that 47 may combine anti-
inflammatory features related to suppression of LTs and PGE2
with pro-resolving properties related to elevating SPM formation.
In fact, the strong effectiveness of 47 in vivo in two simple models of
acute inflammation hold promise for potential anti-inflammatory
therapy. Future studies addressing the ability of 47 to promote
resolution and tissue repair along with SPM formation in vivo are
challenging and hold promise for more insights that allow better
3. Conclusions
Series of sulfonamides that either inhibit mPGES-1 or 5-LO were
reported before [22e24], and by exploiting an initial virtual
screening approach we recently identified the benzenesulfonamide
core as scaffold of the balanced dual mPGES-1/5-LO inhibitor 1 [10].
Here, we successfully explored the benzenesulfonamide scaffold in
a comprehensive SAR study using 46 newly synthesized de-
rivatives.
We
eventually
discovered
the
phenylbenzenesulfonamide-based 47 as multi-targeting drug that
inhibits 5-LO, mPGES-1, LTC4 synthase and presumably also FLAP.
Intriguingly, 47 favorably modulates LM biosynthesis suppressing
the production of pro-inflammatory LTs and PGE2 and consequently
elevating the 12/15-LO-mediated SPM formation with anti-
inflammatory and pro-resolving properties. Replacement of the
n-octyl chain in 1 by a 3-(benzyloxy)phenyl residue and of the tolyl

826
S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
appraisal of its therapeutic potential to intervene with inflamma-
tory disorders.
compound 2b, 3b, 28be30b in Scheme 1, step II; compound
4be16b, 18be20b, 22be27b, 31be34b, 37be39b, 41b, 42b,
44be47b in Scheme 2, step II; compound 21b, 35b and 36 in
Scheme 3, step II; compound 43b in Scheme 6, step I): The sec-
ondary amine (E1, 1.0 eq) and dry pyridine (E2, 1.0 eq) were dis-
solved in 20 ml dry dichloromethane. The benzenesulfonylchloride
(E3,1.0 eq) and a catalytically amount of 4-DMAP were added to the
mixture. The reaction was stirred under argon-atmosphere and
refluxed for 18 h. The solvent was evaporated under reduced
pressure, and the residue was suspended in water. The product was
extracted three times with ethyl acetate and the organic layer was
dried over MgSO4. Subsequently the solvent was evaporated again.
Recrystallization or column chromatography gave the pure product.
D) General procedure for ester hydrolysis (synthesis of com-
pound 2, 3, and 28e30 in Scheme 1, step III; compound 4e16,
18e20, 22e27, 31e34, 37e39, 41, 42, and 44e47 in Scheme 2, step
III; compound 21 and 35 in Scheme 3, step III; compound 40 in
Scheme 5, step II; compound 43 in Scheme 6, step III): The ester (E1,
1.0 eq) was dissolved in 20 ml THF. LiOH∙H2O (E2, 5.0 eq) was
dissolved in 2 ml H2O and added dropwise to the mixture. The re-
action was stirred at 60 ꢁC for 18 h. The solvent was evaporated
under reduced pressure and water was added to the residue. The
water layer was acidified by 2 N HCl. The resulting precipitate was
filtered over a filter (Por. 4) washed with n-hexane and water and
dried on a high vacuum pump. The crude product was recrystal-
lized with n-hexane/ethyl acetate to obtain pure product.
4. Experimental section
4.1. Compounds and chemistry
The structures of the presented compounds were verified by 1H,
13C NMR, and mass spectrometry (ESI); the purity (>95%) was
determined by combustion analysis, or by HPLC plus additional
HRMS for compound 6, 25 and 37. All commercial chemicals and
solvents were of reagent grade and were used without further
purification. 1H and 13C NMR spectra were measured in DMSO‑d6 on
a Bruker DPX 250, AV 300, AV 400 or AV500 spectrometer. Chemical
shifts are reported in parts per million (ppm) in the scale relative to
DMSO‑d6, 2.50 ppm for 1H NMR and 39.52 for 13C NMR. Mass
spectra were obtained on two Thermo Fisher Surveyor MSQ's, (one
MSQ coupled to a Camag TLC-MS interface 2 for direct measure-
ments of TLC spots) measuring in the positive- and/or negative-ion
mode. The combustion analysis has been performed by the
Microanalytical Laboratory of the Institute of Organic Chemistry
and Chemical Biology, Goethe-University Frankfurt, on a Foss Her-
aeus CHN-O-rapid elemental analyzer. All compounds described
have a purity of 95% or higher. High resolution mass spectra were
recorded on a MALDI LTQ ORBITRAP XL instrument (Thermo Fisher
Scientific). Compound purity was analyzed on a Varian ProStar
HPLC equipped with a MultoHigh100 RP18-5
m
240 mm þ 4 mm
E) General procedure for amide coupling (synthesis of com-
pound 40b in Scheme 5, step I): The secondary amine (13a, 1.0 eq)
and dry pyridine (E2, 1.0 eq) were dissolved in 20 mL dry
dichloromethane. Biphenyl-4-carbonyl chloride (1.0 eq) was added
and the reaction stirred under argon-atmosphere at 40 ꢁC for 18 h.
The solvent was removed under reduced pressure and the residue
was suspended in water. The product was extracted three times
with ethyl acetate and the organic layer was dried over MgSO4.
Subsequently the solvent was evaporated again, and the crude
product was further purified by column chromatography (n-hex-
ane/ethyl acetate).
column (CS-Chromatographie Service GmbH, Langerwehe, Ger-
many) using a gradient (H2O þ 0.1% formic acid/MeOH 80:20 iso-
cratic for 5 min to MeOH after additional 45 min and MeOH for
additional 10 min) at a flow rate of 1 mL/min and UV detection at
245 and 280 nm.
The synthesis of the presented compounds was performed in
accordance to Schemes 1e6 and is described here in a general
procedure. For more detailed information and analytical data of the
compounds see Supporting Information.
4.2. General procedures
F) General procedure for N-acylsulfonamide coupling (synthesis
of compound 17 in Scheme 4, step I): Benzoic acid derivative (5, 1.0
eq), benzenesulfonamide (E2, 1.2 eq) and a catalytically amount of
4-DMAP were dissolved in dichloromethane. EDC (E3, 1.2 eq) were
also dissolved in dichloromethane and added to the reaction
mixture dropwise at 0 ꢁC. Reaction was stirred 2 h at 0 ꢁC. The
product was washed with 1 M hydrochloric acid. The organic layer
was dried over MgSO4 and the solvent was evaporated.
G) General procedure for Sonogashira coupling (synthesis of
compound 43c in Scheme 6, step II): Alkyne (E1, 2.0 eq), aryl halides
(43b, 1.0 eq), and copper (I) iodide (E2, 0.02 eq) were dissolved in
25 ml trimethylamine. The flask was filled with argon and bis(-
triphenylphosphine)palladium (II) dichloride (E3, 0.02 eq) was
added to the reaction mixture. The reaction was stirred at room
temperature for 2 h. After reaction, the product was diluted with
dichloromethane and saturated ammonium chloride. The product
was extracted with dichloromethane and washed with distilled
water. Column chromatography (n-hexane/ethyl acetate) gave the
pure product.
A) General procedure for nucleophilic aromatic substitution
(synthesis of compound 2a, 3a and 28a in Scheme 1, step I): Amine
(E1, 1.0 eq) and ethyl-4-fluorobenzoate (E2, 1.0 eq) were dissolved
in 20 ml dry N,N-dimethylformamide. Caesium carbonate (E3, 1.5
eq) was added to the mixture. The reaction was stirred at 110 ꢁC
under argon-atmosphere for 20 h. The mixture was quenched with
30 ml distilled water. The product was extracted with ethyl acetate
and the organic layer was dried over MgSO4. Subsequently the
solvent was evaporated again. Recrystallization or column chro-
matography will give the pure product.
B) General procedure for reductive amination (synthesis of
compound 4ae16a, 27a, 33a, 37ae39a, 41a, 44a, 46a and 47a in
Scheme 2, step I; compound 21a, 35a and 36a in Scheme 3, step I):
Amine (E1, 1.0 eq), the carbonyl compound (E2, 1.1 eq) and glacial
acetic acid (E3, 2.0 eq) were dissolved in 20 ml 1,2-dichloroethane.
The mixture was stirred at room temperature under argon-
atmosphere for 2 h. After imine formation, the reaction was
treated with sodium triacetoxyborohydride (E4, 1.4 eq) at room
temperature and argon-atmosphere for 18 h until the reactants
were consumed as determined by TLC analysis. The reaction
mixture was quenched by adding 1 N NaOH (stirred for 30 min) and
the product was extracted with diethyl ether. The organic layer was
dried over MgSO4 and the solvent was evaporated under reduced
pressure to give the crude product. Pure product was obtained by
recrystallization or column chromatography.
4.3. Molecular docking
Ligands were drawn in Chemdraw (vs 14) and underwent con-
version from 2D to 3D.sdf format in Pipeline Pilot (vs 9.1 Accelrys
Inc, 3.0; San Diego, CA, 2015) They were then energetically mini-
mized in omega 2.5.1.4. (OpenEye Scientific Software, Santa Fe, NM.
http://www.eyesopen.com) [26] prior to docking. Docking simu-
lations were performed on a windows 7 machine with an i5-4670 S
C) General procedure for sulfonamide formation (synthesis of

S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
827
CPU using the CCDC's Gold software (version 5.2) [27]. Gold was set
4.5. Induction of mPGES-1 in A549 cells, isolation of microsomes,
and determination of mPGES-1 activity
to calculate ten poses per molecule. If the genetic algorithm leads to
very similar poses, this is a sign of a stable energetic minimum.
For mPGES-1 the structures were docked to the crystal structure
4bpm [12]. In this structure the inhibitor ligand, BI1 (Ki ¼ 2.4 nM) is
co-crystallized with glutathione. BI1 was extracted from the
structure and its binding site was used to define the docking site in
an 8 Å radius. CHEMPLP was selected as a scoring function. The
original ligand BI1 was prepared in the same workflow as the
candidate molecules and redocked into the binding site, with an
RMSD of 1.002.
Human A549 cells were treated and prepared as described
previously [30]. In brief, cells (2 ꢂ 106/20 ml DMEM/High glucose
(4.5 g/l) medium containing FCS (2%, v/v)) were incubated for
16 h at 37 ꢁC, 5% CO2. The culture medium was replaced by fresh
medium, interleukin-1
b (1 ng/ml) was added, and cells were
incubated for another 72 h. Cells were then detached with trypsin/
EDTA, washed with PBS and frozen in liquid nitrogen. Ice-cold
homogenization buffer (0.1 M potassium phosphate buffer pH 7.4,
For docking to 5-LO the crystal structure 3o8y [13] was prepared
by inserting four virtual mutations (mutated residues Glu13, His14,
Gly75, and Ser76 in the stable 5-LO to Trp13, Phe14, Trp75, and
Leu76 in the wild-type enzyme) to represent the wild-type enzyme
instead of the crystallized “Stable-5-LOX”. The protein structure
was then minimized in Discovery Studio (version 3.5; Discovery
Studio, Accelrys Inc, 3.0; San Diego, CA, 2014). The binding site was
selected in a 10 Å radius around Ile167 and GOldPLP was used as a
scoring function. This binding site and workflow was validated in a
previous project [28]. The minimization was performed with the
Smart Minimizer algorithm, provided by Discovery Studio. The
maximum minimization steps were 200 and the RMSD threshold
was 0.1.
1 mM phenylmethanesulfonyl fluoride, 60
m
g/ml soybean trypsin
inhibitor, 1
m
g/ml leupeptin, 2.5 mM glutathione, and 250 mM su-
crose) was added. After 15 min at 4 ꢁC, cells were re-suspended and
sonicated (3 ꢂ 20 s). The homogenate was subjected to differential
centrifugation (10,000ꢂg for 10 min and at 174,000ꢂg for 1 h at
4
ꢁC). The pellet (microsomal fraction) was re-suspended in 1 ml
homogenization buffer, and protein concentration was determined
by the Coomassie protein assay. The microsomal membranes were
then diluted in potassium phosphate buffer (0.1 M, pH 7.4) con-
taining 2.5 mM glutathione (100
pounds or vehicle (DMSO) were added. After 15 min, enzymatic
PGE2 formation was initiated by addition of 20 M PGH2 (final
concentration). After 1 min at 4 ꢁC, the reaction was terminated
with 100 l of stop solution (40 mM FeCl2, 80 mM citric acid and
10 M of 11 -PGE2), PGE2 was separated by solid phase extraction
on reversed phase (RP)-C18 material using acetonitrile (200 l) as
acetonitrile
ml total volume) and test com-
m
m
m
b
m
4.4. Isolation of cells and cell culture
eluent,
and
analyzed
by RP-HPLC
(30%
aqueous þ 0.007% TFA (v/v), Nova-Pak® C18 column, 5 ꢂ 100 mm,
Venous blood (University Hospital Jena, Germany) was collected
from fasted, adult, healthy donors that were informed about the
aim of the study and that gave written consent. The protocols for
experiments with neutrophils or monocytes were approved by the
ethical commission of the Friedrich-Schiller-University Jena
(approval number 4025e02/14). All methods were performed in
accordance with the relevant guidelines and regulations. Leukocyte
concentrates were obtained by centrifugation (4000ꢂg, 20 min,
20 ꢁC) of freshly withdrawn heparinized blood. Neutrophils and
monocytes were immediately isolated as described before [29]. In
brief, leukocyte concentrates were subjected to dextran sedimen-
tation and centrifuged on lymphocyte separation medium (LSM
1077, PAA, Coelbe, Germany). For isolation of pelleted neutrophils,
remaining erythrocytes were removed by hypotonic lysis, washed
twice with ice-cold PBS and resuspended in respective buffers.
Monocytes were separated from peripheral blood mononuclear
cells (PBMC) by adherence to cell culture flasks (Greiner Bio-one,
Nuertingen, Germany) for 1.5 h (37 ꢁC, 5% CO2) in RPMI 1640 con-
4 mm particle size, flow rate 1 ml/min) with UV detection at 195 nm.
11b-PGE2 was used as internal standard to quantify PGE2 formation.
4.6. Expression, purification and activity assay of human
recombinant 5-LO
5-LO was expressed in E. coli B121 (DE3) transformed with pT3-
5LO plasmid, and purification of 5-LO was performed as described
previously [31]. Briefly, E. coli were lysed in 50 mM triethanol-
amine/HCl, pH 8.0, 5 mM EDTA, 60 mg/ml soybean trypsin inhibitor,
1 mM phenylmethanesulfonyl fluoride, 1 mM dithiothreitol and
1 mg/ml lysozyme and then sonified (3 ꢂ 15 s). The homogenate
was then centrifuged at 40,000ꢂg for 20 min at 4 ꢁC. 5-LO in the
supernatant was partially purified by affinity chromatography on
an ATP-agarose column (SigmaeAldrich). Semi-purified 5-LO was
diluted in PBS containing EDTA (1 mM) and ATP (1 mM) and
immediately used for 5-LO activity assays. Samples were pre-
incubated with the test compounds or vehicle (0.1% DMSO) for
10 min at 4 ꢁC. Samples were pre-warmed for 30 s at 37 ꢁC, and
2 mM CaCl2 plus the indicated concentrations of AA were added to
start 5-LO product formation. The reaction was stopped after
10 min by addition of one volume of ice-cold methanol, and the
formed metabolites were analyzed by RP-HPLC as described [32]. 5-
LO products include the all-trans isomers of LTB4 as well as 5-
HPETE and its corresponding alcohol 5(S)-hydroxy-6-trans-
8,11,14-cis-eicosatetraenoic acid (5-HETE).
taining L-glutamine (1 mM), heat-inactivated FCS (10%), penicillin
(100 U/mL) and streptomycin (100
scraping and resuspension in PBS.
mg/mL), followed by cell-
M1 and M2 macrophages were obtained as described before
[18]. Briefly, for M1, freshly isolated monocytes were differentiated
for 6 days by GM-CSF (20 ng/ml) in RPMI containing 10% of fetal calf
serum and subsequently polarized by LPS (100 ng/ml) plus INF-
g
(20 ng/ml) treatment for 48 h. In order to get M2, the differentiation
procedure was performed in presence of M-CSF (20 ng/ml) for 6
days and the polarization by addition of IL-4 (20 ng/ml) for 48 h.
HEK293 cells were maintained in monolayers (37 ꢁC, 5% CO2) in
DMEM containing FCS (10%), penicillin (100 U/mL) and strepto-
4.7. Determination of 5-LO product formation in cell-based assays
Freshly isolated human neutrophils (5 ꢂ 106) were resuspended
in 1 ml PBS buffer containing 1 mg/ml glucose (PG buffer). After
pre-incubation with the compounds (15 min, 37 ꢁC), LO product
mycin (100
FLAP were selected using geneticin (400
(200 g/mL), respectively, as previously described [14]. Trans-
mg/mL). HEK293 cell lines stably expressing 5-LO and
m
g/mL) and hygromycin
m
formation was started by addition of 1 mM CaCl2 and 2.5
mM
fection of HEK293 was performed using pcDNA3.1 plasmids and
lipofectamine according to the manufacturer's instructions (Invi-
trogen, Darmstadt, Germany).
A23187 with or without AA (20
m
M). After 10 min at 37 ꢁC, the re-
action was stopped with 1 ml of methanol and products formed and
released in the supernatants were extracted and analyzed by HPLC.

828
S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
4.8. Determination of [3H]-AA-release in neutrophils
The reaction was stopped by adding methanol containing the
deuterium-labeled internal standards d8-5S-HETE, d4-LTB4, d5-
LXA4, d5-RvD2, and d4-PGE2 (500 pg, each) to facilitate quantifica-
tion and sample recovery.
Neutrophils (1 ꢂ 107) were incubated in 1 ml of RPMI 1640
medium with 5 nM [3H]-AA (corresponding to 0.5
mCi/mL, specific
activity 200 Ci/mmol) for 120 min at 37 ꢁC in 5% CO2 atmosphere.
Cells were washed twice with PG buffer containing 2 mg/ml acid-
free bovine albumin (BSA, fatty acid-free to remove unincorpo-
rated [3H]-AA) and pre-incubated with 0.1% DMSO or test com-
pounds (15 min, 37 ꢁC). Neutrophils were then stimulated with
4.12. Incubations of macrophages and LM metabololipidomics
M1 and M2 macrophages (2 ꢂ 106/ml) were incubated in PBS
containing 1 mM CaCl2 with 47 or vehicle control (0.1% DMSO) for
15 min at 37 ꢁC. Then, E. coli (serotype O6:K2:H1) at a ratio of 1:50
(M1/M2:E. coli) were added and after 180 min the supernatants
2.5
m
M A23187. After 5 min at 37 ꢁC the reaction was stopped for
L) was collected after centrifu-
10 min on ice. Supernatant (300
m
gation (500ꢂg, 15 min, 20 ꢁC) and added to 2 ml LSC fluid (Rotiszint
Eco Plus, Carl Roth GmbH & Co. KG, Karlsruhe, Germany). The
released [3H]-AA was measured using a scintillation counter (Per-
kinElmer Life Sciences, Waltham, MA).
were transferred to 2 ml of ice-cold methanol containing 10
deuterium-labeled internal standards (200 nM d8-5S-HETE, d4-
LTB4, d5-LXA4, d5-RvD2, d4-PGE2 and 10 M d8-AA). Samples
ml of
m
were kept at ꢀ20 ꢁC for 60 min to allow protein precipitation. After
centrifugation (1200 g, 4 ꢁC, 10 min) 8 ml acidified water was added
(final pH ¼ 3.5) and subjected to solid phase extraction. Solid-phase
cartridges (Sep-Pak® Vac 6 cc 500 mg/6 ml C18; Waters, Milford,
MA) were equilibrated with 6 ml methanol and 2 ml water before
samples were loaded onto columns. After washing with 6 ml water
and additional 6 ml hexane, LM were eluted with 6 ml methyl
formate, the samples were brought to dryness using an evaporation
system (TurboVap LV, Biotage, Uppsala, Sweden), and resuspended
4.9. Analysis of subcellular localization of 5-LO by
immunofluorescence microscopy
Subcellular localization of 5-LO and FLAP was analyzed by
immunofluorescence (IF) microscopy as previously described [17].
In brief, HEK cells expressing 5-LO/FLAP were seeded onto acid-
washed (50% sulfuric acid) and poly-
coverslips and cultured for 48 h at 37 ꢁC until ~60% confluency. Cells
were pre-incubated with 10 M compound 47 or vehicle (0.1%
DMSO) for 10 min before stimulation with 2.5 M A23187 for
L-lysine (0.01%)-coated glass
in 100 ml methanol-water (50/50, v/v) for UPLC-MS/MS automated
m
injections. LM were analyzed with an Acquity™ UPLC system
(Waters, Milford, MA, USA) and a QTRAP 5500 Mass Spectrometer
(ABSciex, Darmstadt, Germany) equipped with a Turbo V™ Source
and electrospray ionization. LM were separated on an ACQUITY
m
20 min at 37 ꢁC. After fixation with 4% paraformaldehyde solution,
followed by permeabilization by using 50 mM ammonium chloride
and acetone (3 min, 4 ꢁC), cells were incubated with 10% non-
immune goat serum blocking solution for 1 h. Then, HEK cells
were incubated over night with mouse monoclonal anti-5-LO
UPLC® BEH C18 column (1.7
mm, 2.1 ꢂ 100 mm; Waters, Eschborn,
Germany) at 50 ꢁC with a flow rate of 0.3 ml/min and a mobile
phase consisting of methanol-water-acetic acid of 42:58:0.01 (v/v/
v) that was ramped to 86:14:0.01 (v/v/v) over 12.5 min and then to
98:2:0.01 (v/v/v) for 3 min. The QTRAP 5500 was operated in
negative ionization mode using scheduled multiple reaction
monitoring coupled with information-dependent acquisition. The
scheduled MRM window was 60 s, optimized LM parameters were
adopted (CE, EP, DP, CXP), and the curtain gas pressure was set to
35 psi. The retention time and at least six diagnostic ions for each
LM were confirmed by means of external standards (Cayman
Chemicals). Quantification was achieved by calibration curves for
each LM. Linear calibration curves were obtained for each LM and
gave r2 values of 0.998 or higher (for fatty acids 0.95 or higher).
Additionally, the limit of detection for each targeted LM was
determined.
antibody (1:100) and rabbit polyclonal anti-FLAP antibody (5 mg/
mL). The coverslips were intensively washed with PBS before
staining with the fluorophore-labeled secondary antibodies Alexa
Fluor 488 goat anti-rabbit (1:500) and Alexa Fluor 555 goat anti-
mouse (1:500) for 20 min in the dark. The coverslips were moun-
ted on glass slides with mowiol containing 2.5% n-propyl gallate
and 40,6-diamidin-2-phenylindol (DAPI; Invitrogen, Carlsbad, CA), a
staining for nuclear DNA. Samples were analyzed by a Zeiss Axio-
vert 200 M microscope, and a Plan Neofluar ꢂ 100/1.30 Oil (DIC III)
objective (Carl Zeiss, Jena, Germany). An AxioCam MR camera (Carl
Zeiss) was used for image acquisition.
4.10. Analysis of 5-LO/FLAP interaction by in situ proximity ligation
assay (PLA)
4.13. Animals
To detect in situ interaction of 5-LO with FLAP in HEK cells, a PLA
was performed as reported elsewhere [14]. Cells were treated with
test compounds and stimuli, fixed, and incubated with primary
antibody as mentioned for IF microscopy. Secondary antibodies
conjugated to oligonucleotides (PLA probe anti-mouse MINUS/anti-
rabbit PLUS or anti-goat MINUS/anti-rabbit PLUS; SigmaeAldrich,
Taufkirchen, Germany) were applied for 1 h at 37 ꢁC to the samples
prior to the ligation step (30 min, 37 ꢁC). Generated DNA circle was
amplified by rolling-circle amplification and hybridization with
fluorescently labeled nucleotides for 100 min at 37 ꢁC. Nuclear DNA
was stained with DAPI. PLA signals were detected using the same
microscope and camera as described for IF microscopy above.
All animal care and experimental procedures were approved by
the Italian Ministry according to International and National law and
policies (EU Directive, 2010/63/EU for animal experiments, ARRIVE
guidelines and the Basel declaration including the 3 R concept).
Animal studies are reported in compliance with the ARRIVE
guidelines [33,34].
4.14. Murine zymosan peritonitis
CD-1 male mice (6e8 weeks, Charles River, Italy) were housed in
a controlled environment with free access to water and food under
controlled temperature, humidity and light/dark cycle. Mice (6 each
group) received 47 (10 mg/kg) intraperitoneally (i.p.), zileuton
(10 mg/kg) or vehicle (2% DMSO in saline) for 30 min before peri-
tonitis induction. Then, a suspension of 2 mg/mL of zymosan
4.11. Determination of LM profile in activated human macrophages
M1 and M2 (2 ꢂ 106) were re-suspended in PBS buffer con-
taining 1 mM of CaCl2 and pre-incubated with 3
m
M compound 47
(500 mL each mouse, Sigma, Milan) was administered i. p. in order
to induce inflammation. At selected time points (30 min for LTC4
for 15 min at 37 ꢁC. LM production was started by challenging the
cells with pathogenic E. coli (O6:K2:H1; ratio ¼ 1:50) for 90 min.
measurement or 4 h for determination of cell recruitment), mice

S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
829
lipoxygenase: potential drug targets in cancer, J. Intern. Med. 268 (2010)
5e14.
[7] A. Koeberle, O. Werz, Natural products as inhibitors of prostaglandin E2 and
pro- inflammatory 5-lipoxygenase-derived lipid mediator biosynthesis, Bio-
technol. Adv. (2018), https://doi.org/10.1016/j.biotechadv.2018.02.010 pii:
S0734-9750(18)30026-0.
[8] C. He, Y. Wu, Y. Lai, Z. Cai, Y. Liu, L. Lai, Dynamic eicosanoid responses upon
different inhibitor and combination treatments on the arachidonic acid
metabolic network, Mol. Biosyst. 8 (2012) 1585e1594.
were sacrificed by inhalation of CO2 and the peritoneal cavity was
washed with 3 ml of PBS. Cells in the inflammatory exudates were
immediately counted (dilution 1:10) by using a Burker's chamber
and vital trypan blue staining, and then samples were centrifuged
(18,000 ꢂ g, 5 min, 4 ꢁC). LTC4 measurement in the supernatant was
performed by ELISA according to the manufacturer's instructions.
Data are expressed as ng/ml.
[9] L.L. Mazaleuskaya, J.A. Lawson, X. Li, G. Grant, C. Mesaros, T. Grosser, I.A. Blair,
E. Ricciotti, G.A. FitzGerald, A broad-spectrum lipidomics screen of antiin-
flammatory drug combinations in human blood, JCI Insight 1 (2016).
[10] B. Waltenberger, K. Wiechmann, J. Bauer, P. Markt, S.M. Noha, G. Wolber,
J.M. Rollinger, O. Werz, D. Schuster, H. Stuppner, Pharmacophore modeling
and virtual screening for novel acidic inhibitors of microsomal prostaglandin
E(2) synthase-1 (mPGES-1), J. Med. Chem. 54 (2011) 3163e3174.
[11] M. Elkady, R. Niess, A.M. Schaible, J. Bauer, S. Luderer, G. Ambrosi, O. Werz,
S.A. Laufer, Modified acidic nonsteroidal anti-inflammatory drugs as dual in-
hibitors of mPGES-1 and 5-LOX, J. Med. Chem. 55 (2012) 8958e8962.
[12] D. Li, N. Howe, A. Dukkipati, S.T. Shah, B.D. Bax, C. Edge, A. Bridges,
P. Hardwicke, O.M. Singh, G. Giblin, A. Pautsch, R. Pfau, G. Schnapp, M. Wang,
V. Olieric, M. Caffrey, Crystallizing membrane proteins in the lipidic meso-
phase. Experience with human prostaglandin E2 synthase 1 and an evolving
strategy, Cryst. Growth Des. 14 (2014) 2034e2047.
[13] N.C. Gilbert, S.G. Bartlett, M.T. Waight, D.B. Neau, W.E. Boeglin, A.R. Brash,
M.E. Newcomer, The structure of human 5-lipoxygenase, Science 331 (2011)
217e219.
[14] J. Gerstmeier, C. Weinigel, S. Rummler, O. Radmark, O. Werz, U. Garscha, Time-
resolved in situ assembly of the leukotriene-synthetic 5-lipoxygenase/5-
lipoxygenase- activating protein complex in blood leukocytes, Faseb. J. 30
(2016) 276e285.
[15] 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-
lipoxygenase-activating protein (FLAP) complex assembly, Biochem. Phar-
macol. 119 (2016) 17e26.
[16] O. Werz, J. Gerstmeier, U. Garscha, Novel leukotriene biosynthesis inhibitors
(2012-2016) as anti-inflammatory agents, Expert Opin. Ther. Pat. 27 (2017)
607e620.
[17] J. Gerstmeier, C. Weinigel, D. Barz, O. Werz, U. Garscha, An experimental cell-
based model for studying the cell biology and molecular pharmacology of 5-
lipoxygenase- activating protein in leukotriene biosynthesis, Biochim. Bio-
phys. Acta 1840 (2014) 2961e2969.
[18] O. Werz, J. Gerstmeier, S. Libreros, X. De la Rosa, M. Werner, P.C. Norris,
N. Chiang, C.N. Serhan, Human macrophages differentially produce specific
resolvin or leukotriene signals that depend on bacterial pathogenicity, Nat.
Commun. 9 (2018) 59.
4.15. Mouse air pouch
Air pouches were developed by subcutaneous injection of 2.5 ml
sterile air into the back of CD-1 male mice (6e8 weeks, Charles
River, Italy), followed by a re-injection of 2.5 ml sterile air into the
same cavity after 3 days. On the day 6 from the first injection of
sterile air, mice received 47 (10 mg/kg), zileuton (10 mg/kg) or
vehicle (2% DMSO in saline) i. p. 30 min. before 1 ml zymosan 1% w/
v injection into the air pouch. After 4 h of zymosan injection, mice
were sacrificed by CO2 exposure and exudates in the pouch were
collected with 1 ml of saline and centrifuged at 125 ꢂ g for 10 min at
4 ꢁC. The total leukocyte amount was counted by Burker chamber
and LTB4 in the supernatant was measured by ELISA according to
the manufacturer's instructions. Data are expressed as ng/mL.
4.16. Statistical analysis
Data are expressed as mean S.E.M. IC50 values were calculated
from averaged measurements at 4e5 different concentrations of
the compounds by nonlinear regression using GraphPad Prism
software (San Diego, CA) one site binding competition. Data ob-
tained in animal experiments are given as mean SEM obtained
from 6 mice in total. Statistical evaluation of the data was per-
formed by one-way ANOVA followed by a Bonferroni post hoc test
for multiple comparisons respectively.
considered significant.
A p value < 0.05 was
Funding sources
[19] A.M. Schaible, R. Filosa, V. Krauth, V. Temml, S. Pace, U. Garscha, S. Liening,
C. Weinigel, S. Rummler, S. Schieferdecker, M. Nett, A. Peduto, S. Collarile,
M. Scuotto, F. Roviezzo, G. Spaziano, M. de Rosa, H. Stuppner, D. Schuster,
B. D'Agostino, O. Werz, The 5-lipoxygenase inhibitor RF-22c potently sup-
presses leukotriene biosynthesis in cellulo and blocks bronchoconstriction
and inflammation in vivo, Biochem. Pharmacol. 112 (2016) 60e71.
[20] U. Garscha, E. Romp, S. Pace, A. Rossi, V. Temml, D. Schuster, S. Konig,
J. Gerstmeier, S. Liening, M. Werner, H. Atze, S. Wittmann, C. Weinigel,
S. Rummler, G.K. Scriba, L. Sautebin, O. Werz, Pharmacological profile and
efficiency in vivo of diflapolin, the first dual inhibitor of 5-lipoxygenase-
activating protein and soluble epoxide hydrolase, Sci. Rep. 7 (2017) 9398.
[21] F. Maione, V. Cantone, S. Pace, M.G. Chini, A. Bisio, G. Romussi, S. Pieretti,
O. Werz, A. Koeberle, N. Mascolo, G. Bifulco, Anti-inflammatory and analgesic
activity of carnosol and carnosic acid in vivo and in vitro and in silico analysis
of their target interactions, Br. J. Pharmacol. 174 (2017) 1497e1508.
[22] G. Lauro, P. Tortorella, A. Bertamino, C. Ostacolo, A. Koeberle, K. Fischer,
I. Bruno, S. Terracciano, I.M. Gomez-Monterrey, M. Tauro, F. Loiodice,
E. Novellino, R. Riccio, O. Werz, P. Campiglia, G. Bifulco, Structure-based
design of microsomal prostaglandin E2 Synthase-1 (mPGES-1) inhibitors us-
ing a virtual fragment growing optimization scheme, ChemMedChem 11
(2016) 612e619.
[23] T. Hanke, F. Rorsch, T.M. Thieme, N. Ferreiros, G. Schneider, G. Geisslinger,
E. Proschak, S. Grosch, M. Schubert-Zsilavecz, Synthesis and pharmacological
characterization of benzenesulfonamides as dual species inhibitors of human
and murine mPGES-1, Bioorg. Med. Chem. 21 (2013) 7874e7883.
[24] G. Mustafa, I.U. Khan, M. Ashraf, I. Afzal, S.A. Shahzad, M. Shafiq, Synthesis of
new sulfonamides as lipoxygenase inhibitors, Bioorg. Med. Chem. 20 (2012)
2535e2539.
[25] T. Hanke, F. Dehm, S. Liening, S.D. Popella, J. Maczewsky, M. Pillong, J. Kunze,
C. Weinigel, D. Barz, A. Kaiser, M. Wurglics, M. Lammerhofer, G. Schneider,
L. Sautebin, M. Schubert-Zsilavecz, O. Werz, Aminothiazole-featured pirinixic
acid derivatives as dual 5-lipoxygenase and microsomal prostaglandin E2
synthase-1 inhibitors with improved potency and efficiency in vivo, J. Med.
Chem. 56 (2013) 9031e9044.
We thank the Deutsche Forschgungsgemeinschaft SFB1127
(ChemBioSys) and SFB1278 (Polytarget) for financial support. J.G.
received a Carl Zeiss stipend.
Author contributions
The manuscript was written through contributions of all au-
thors. All authors have given approval to the final version of the
manuscript.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.ejmech.2018.07.031.
References
[1] O. Radmark, O. Werz, D. Steinhilber, B. Samuelsson, 5-Lipoxygenase, a key
enzyme for leukotriene biosynthesis in health and disease, Biochim. Biophys.
Acta 1851 (2015) 331e339.
[2] A. Koeberle, S.A. Laufer, O. Werz, Design and development of microsomal
prostaglandin E2 Synthase-1 inhibitors: challenges and future directions,
J. Med. Chem. 59 (2016) 5970e5986.
[3] M. Peters-Golden, W.R. Henderson Jr., Leukotrienes, N. Engl. J. Med. 357
(2007) 1841e1854.
[4] A. Koeberle, O. Werz, Perspective of microsomal prostaglandin E2 synthase-1
as drug target in inflammation-related disorders, Biochem. Pharmacol. 98
(2015) 1e15.
[26] P.C. Hawkins, A.G. Skillman, G.L. Warren, B.A. Ellingson, M.T. Stahl, Conformer
generation with OMEGA: algorithm and validation using high quality struc-
tures from the Protein Databank and Cambridge Structural Database, J. Chem.
[5] K.D. Rainsford, Anti-inflammatory drugs in the 21st century, Subcell. Biochem.
42 (2007) 3e27.
[6] O. Radmark, B. Samuelsson, Microsomal prostaglandin E synthase-1 and 5-

830
S.-Y. Cheung et al. / European Journal of Medicinal Chemistry 156 (2018) 815e830
Inf. Model. 50 (2010) 572e584.
Pirinixic acid derivatives as novel dual inhibitors of microsomal prostaglandin
E2 synthase-1 and 5-lipoxygenase, J. Med. Chem. 51 (2008) 8068e8076.
[31] L. Fischer, D. Szellas, O. Radmark, D. Steinhilber, O. Werz, Phosphorylation-
and stimulus-dependent inhibition of cellular 5-lipoxygenase activity by
nonredox-type inhibitors, Faseb. J. 17 (2003) 949e951.
[32] D. Steinhilber, T. Herrmann, H.J. Roth, Separation of lipoxins and leukotrienes
from human granulocytes by high-performance liquid chromatography with a
Radial-Pak cartridge after extraction with an octadecyl reversed-phase col-
umn, J. Chromatogr. A 493 (1989) 361e366.
[27] G. Jones, P. Willett, R.C. Glen, A.R. Leach, R. Taylor, Development and valida-
tion of a genetic algorithm for flexible docking, J. Mol. Biol. 267 (1997)
727e748.
[28] A. Peduto, M. Scuotto, V. Krauth, F. Roviezzo, A. Rossi, V. Temml, V. Esposito,
H. Stuppner, D. Schuster, B. D'Agostino, C. Schiraldi, M. de Rosa, O. Werz,
R. Filosa, Optimization of benzoquinone and hydroquinone derivatives as
potent inhibitors of human 5-lipoxygenase, Eur. J. Med. Chem. 127 (2017)
715e726.
[29] S. Pace, C. Pergola, F. Dehm, A. Rossi, J. Gerstmeier, F. Troisi, H. Pein,
A.M. Schaible, C. Weinigel, S. Rummler, H. Northoff, S. Laufer, T.J. Maier,
O. Radmark, B. Samuelsson, A. Koeberle, L. Sautebin, O. Werz, Androgen-
mediated sex bias impairs efficiency of leukotriene biosynthesis inhibitors in
males, J. Clin. Invest. 127 (2017) 3167e3176.
[33] C. Kilkenny, W. Browne, I.C. Cuthill, M. Emerson, D.G. Altman,
N.C.R.R.G.W. Group, Animal research: reporting in vivo experiments: the
ARRIVE guidelines, Br. J. Pharmacol. 160 (2010) 1577e1579.
[34] J.C. McGrath, G.B. Drummond, E.M. McLachlan, C. Kilkenny, C.L. Wainwright,
Guidelines for reporting experiments involving animals: the ARRIVE guide-
lines, Br. J. Pharmacol. 160 (2010) 1573e1576.
[30] A. Koeberle, H. Zettl, C. Greiner, M. Wurglics, M. Schubert-Zsilavecz, O. Werz,