Discovery and biological evaluation of a novel class of dual microsomal prostaglandin E2 synthase-1/5-lipoxygenase inhibitors based on 2-[(4,6-diphenethoxypyrimidin-2-yl)thio]hexanoic acid

Article abstract of DOI:10.1021/jm200092b

Various inflammatory diseases are associated with the excessive formation of leukotrienes (LTs) and prostaglandins (PGs). Herein, we present a novel class of dual inhibitors of 5-lipoxygenase (5-LO) and microsomal prostaglandin E ₂ synthase-1 (

Full text of DOI:10.1021/jm200092b

ARTICLE
pubs.acs.org/jmc
Discovery and Biological Evaluation of a Novel Class of Dual
Microsomal Prostaglandin E2 Synthase-1/5-lipoxygenase Inhibitors
Based on 2-[(4,6-Diphenethoxypyrimidin-2-yl)thio]hexanoic Acid
Martina Hieke,†,‡ Christine Greiner,§,‡ Michaela Dittrich,† Felix Reisen,^ Gisbert Schneider,^
Manfred Schubert-Zsilavecz,*,† and Oliver Werz*,||
†Institute of Pharmaceutical Chemistry, ZAFES/LiFF/Goethe-University Frankfurt, Max-von-Laue-Str. 9, D-60438 Frankfurt am Main,
Germany
§Department of Pharmaceutical Analytics, Pharmaceutical Institute, Eberhard-Karls-University Tuebingen, Auf der Morgenstelle 8,
D-72076 Tuebingen, Germany
^Department of Chemistry and Applied Biosciences, Swiss Federal Institute of Technology (ETH), Zurich, Switzerland
Chair of Pharmaceutical/Medicinal Chemistry, Institute of Pharmacy, Friedrich-Schiller-University Jena, Philosophenweg 14,
D-07743 Jena, Germany
S Supporting Information
b
ABSTRACT: Various inflammatory diseases are associated
with the excessive formation of leukotrienes (LTs) and pros-
taglandins (PGs). Herein, we present a novel class of dual
inhibitors of 5-lipoxygenase (5-LO) and microsomal prosta-
glandin E2 synthase-1 (mPGES-1), key enzymes in the forma-
tion of LTs and PGE2, respectively. On the basis of the structure
of 2-[(4,6-diphenethoxypyrimidin-2-yl)thio]hexanoic acid (1),
we performed a detailed SAR analysis, and mechanistic studies
were carried out to elucidate the mode of 5-LO inhibition.
Interestingly, the pyrimidine ring including the thioether of 1
could be replaced by a simple benzyl or a benzylidene moiety
yielding a novel series of bioactive 2-benzylidene- and 2-benzylhexanoic acids exemplified by 2-(2,3-diphenethoxybenzylide-
ne)hexanoic acid, 29 (IC50 5-LO = 0.8 μM; mPGES-1 = 1.1 μM). Importantly, none of the novel bioactive derivatives strongly
inhibited cyclooxygenase activities. Together, we provide novel promising lead compounds for the treatment of inflammatory
diseases valuable for further investigations in vivo.
’ INTRODUCTION
the pro-inflammatory PGE2 by the inducible microsomal pros-
taglandin E2 synthase (mPGES)-1. Nonsteroidal anti-inflamma-
tory drugs (NSAIDs), which inhibit the activity of both COX
isoenzymes, as well as selective COX-2 inhibitors (coxibs), are
widely used in the therapy of pain and inflammatory diseases.
Nevertheless, NSAIDs provoke severe side effects in the gastro-
intestinal tract due to suppression of gastroprotective (mainly
COX-1-derived) PGE2 and elevated LTB4 levels.5 Moreover, the
suppression of prostanoid biosynthesis also causes dysfunction of
the kidney (NSAIDs) and cardiovascular toxicity (coxibs), the
latter due to an imbalance of beneficial prostacyclin and detri-
mental thromboxane A2. Hence, exclusive inhibition of massive
(COX-2-derived) PGE2 formation combined with perpetuation
of homeostatic PGs would be desirable. Moreover, based on the
pro-inflammatory function of LTs and the unwanted effects of
elevated LTs in the gastrointestinal and the cardiovascular
Leukotrienes (LTs) and prostaglandins (PGs) represent
potent lipid mediators that are involved in inflammatory pro-
cesses. They contribute to fever, pain, and allergic diseases and
play a crucial role in cardiovascular diseases (CVD) and various
types of cancer.1ꢀ3 The first step in the biosynthesis of both
groups of eicosanoids is the release of arachidonic acid (AA)
from phospholipids of cellular membranes by cytosolic phos-
pholipase A2 (cPLA2). Subsequently, AA is converted into LTA4
by 5-lipoxygenase (5-LO) or into PGH2 by two different iso-
forms of cyclooxygenase (COX). LTA4 is then metabolized by
LTA4 hydrolase into the chemotactic LTB4 or by LTC4 synthase
into cysteinyl leukotrienes (LTC4, LTD4, and LTE4), which
evoke bronchoconstriction and increase vascular permeability.2,4
COX-1 is constitutively expressed and involved in the bio-
synthesis of physiologically important PGs (e.g., TXA2, PGE2,
PGI2, PGF2R) with house-keeping functions. In contrast, COX-2
is an inducible isoenzyme and up-regulated during inflammatory
processes.4 COX-2 derived PGH2 is efficiently transformed into
Received: January 27, 2011
Published: May 18, 2011
r
2011 American Chemical Society
4490
dx.doi.org/10.1021/jm200092b J. Med. Chem. 2011, 54, 4490–4507
|

Journal of Medicinal Chemistry
ARTICLE
Scheme 1. Synthesis of Thiobarbituric Acid Derivatives 1ꢀ20a
a Reagents and conditions: (I) R-bromo-(R1)-ethyl acetate (1.5 equiv), thiobarbituric acid (1 equiv), TEA (1.5 equiv), DMF, 80 ꢀC, 4 h; (IIa) precursor
resulting from step I (1 equiv), R2-Hal (2.1 equiv), K2CO3 (2.3 equiv), DMF, 80 ꢀC, 3ꢀ28 h or (IIb) precursor resulting from step I (1 equiv), R2-OH
(2 equiv), diethyl azodicarboxylate (DEAD; 2.5 equiv), TPP (2.5 equiv), THF, rt, 1ꢀ3 h; (III) precursor resulting from step II (1 equiv), LiOH H2O
3
(5 equiv), THF/MeOH, H2O, 25ꢀ50 ꢀC, 2ꢀ24 h.
system, dual inhibition of mPGES-1 and 5-LO might be even
more efficient with less side effects than interference with one
single pathway.5
Preparation of 2-[(3,5-diphenethoxyphenyl)thio]hexanoic acid
(21) was carried out in six steps (Scheme 2). The synthesis starts
with the etherification of two hydroxyl groups of phloro-
glucinol.13 The resulting precursor 21a was isolated in low yield
from a mixture of mono-, di-, and trisubstituted phloroglucinol
ethers. The third hydroxyl group was afterward activated with
dimethylthiocarbamoyl chloride to obtain precursor 21b. Heat-
ing at 240 ꢀC led to a NewmanꢀKwart rearrangement yielding
the protected thiol derivative 21c, which was subsequently
deprotected with NaOH to give 21d.14 Finally, a reaction of
21d and R-bromo ethylhexanoate resulted in the ester derivative
21e, which was afterward hydrolyzed to the carboxylic acid 21.
2-(3,5-Diphenethoxyphenoxy)hexanoic acid (22) was synthe-
sized in three steps (Scheme 3). Starting with phloroglucinol,
one hydroxyl group was etherified with R-bromo ethylhexa-
noate.13 Afterward, the two remaining hydroxyl groups were
etherified via Mitsunobu synthesis. Finally, the ester group was
hydrolyzed to yield the product 22.
While there is little known about factors influencing mPGES-1
activity in the cell, the activity of 5-LO is modulated by various
cofactors,6 which may also influence the susceptibility toward
pharmacological inhibitors.7,8 Thus, 5-LO activity is controlled
by phosphorylations at serine residues, Ca2þ, ATP, coactosine-
like protein (CLP), phospholipids, and glycerides.9 Despite
considerable research efforts, zileuton is still the only 5-LO inhi-
bitor that has succeeded clinical trials and reached the market.
The lack of selectivity or mechanism-based side effects were the
main reasons for the failure of 5-LO inhibitors as drug candidates
in clinical trials. Therefore, we included various cell-based and
cell-free assays in our studies to address mechanistic aspects of
5-LO inhibition by one of the lead compounds.
Herein, we present a novel class of dual inhibitors of human
mPGES-1 and 5-LO. In previous studies, we found that a series of
R-substituted pirinixic acid derivatives inhibited mPGES-1 and
5-LO in vitro, and the respective lead compound, 2-(4-chloro-
6-(2,3-dimethylphenylamino)pyrimidin-2-ylthio)octanoic acid
(YS121), exhibited anti-inflammatory efficacy in vivo.10,11 Of
interest, some of these R-substituted pirinixic acid derivatives
also modulated γ-secretase and peroxisome proliferator-acti-
vated receptor γ (PPARγ), such as the 2-[(4,6-diphenethoxy-
pyrimidin-2-yl)thio]hexanoic acid 1.12 Based on the structure of
compound 1, a comprehensive structureꢀactivity relationship
(SAR) study for human mPGES-1 and 5-LO was executed.
Aiming to identify novel, more simplified structural scaffolds,
we performed broad structural variations in the R-position of the
carboxylic acid as well as on the pyrimidine core. To this end, we
established a synthetic method that enabled the development of a
structurally different class of bioactive compounds containing a
trisubstituted benzene scaffold as central part of the molecule.
Disubstituted 2-benzylhexanoic acid and 2-benzylidenehexa-
noic acid derivatives 23ꢀ34 were prepared in four to five steps as
shown in Scheme 4.15 Step one was an Arbuzov reaction between
triethylphosphite and R-bromo ethylhexanoate yielding phos-
phonate precursor A. Second, the respective dihydroxybenzalde-
hyde was etherified in one or two steps under Williamson-like or
Mitsunobu conditions to yield precursor molecules 23b,
25bꢀ28b, and 31bꢀ34b. Subsequently, 23b, 25bꢀ28b, and
31bꢀ34b were converted with phosphonate precursor A to an
ester derivative in a WittigꢀHorner reaction. Afterward, either
the resulting precursors 23c, 25cꢀ28c, and 31cꢀ34c were
directly hydrolyzed to yield 2-benzylidenehexanoic acid deriva-
tives or the double bond was hydrogenated using Pd/C and H2
and the resulting product was finally hydrolyzed to yield 2-ben-
zylhexanoic acid derivatives (compounds 23ꢀ34).
Biological assays. All compounds synthesized within this
study were characterized for their effects on the activity of human
5-LO and mPGES-1. As shown previously, none of the com-
pounds 1ꢀ20 showed cytotoxic effects up to 50 μM,12 providing
a high safety index for the identified lead compounds. Similarly,
analysis of compounds 21ꢀ34 (at 10 μM final concentration) for
acute cytotoxic effects, reflecting the conditions during the
30 min preincubation for 5-LO activity studies in polymorpho-
nuclear leukocytes (PMNL), excluded significant cytotoxicity of
the test compounds.
’ RESULTS AND DISCUSSION
Chemistry. Preparation of thiobarbituric acid derivatives
1ꢀ20 is outlined in Scheme 1 and was described previously.12
First, the respective R-bromo-R1-acetate and thiobarbituric acid
were combined via a nucleophilic substitution. Next, the two
lipophilic moieties were introduced either by a Williamson-like
or a Mitsunobu ether synthesis, and finally, the ester group was
hydrolyzed to give the carboxylic acids 1ꢀ20.
4491
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
Scheme 2. Synthesis of 2-[(3,5-diphenethoxyphenyl)thio]hexanoic acid 21a
a Reagents and conditions: (I) phloroglucinol (1.0 equiv), phenylethyl bromide (2.1 equiv), NaH (3 equiv), DMF (q.s.), rt, 20 h; (II) 21a (1 equiv),
dimethylthiocarbamoyl chloride (1.0 equiv), NaH (1.2 equiv), DMF (q.s.), 60 ꢀC, 5 h; (III) 21b, solvent-free, 240 ꢀC, 5 h; (IV) 21c (1.0 equiv), NaOH
(1 mol/L; 10 mL), THFꢀMeOH, 80 ꢀC, 3 h; (V) 21d (1.0 equiv), R-bromo ethylhexanoate (1.2 equiv), NaH (1.2 equiv), DMF (q.s.), rt, 3 h; (VI) 21e
(1.0 equiv), LiOH H2O (5.0 equiv), THFꢀH2O, 50 ꢀC, 3 h.
3
Scheme 3. Synthesis of 2-(3,5-diphenethoxyphenoxy)hexanoic acid 22a
a Reagents and conditions: (I) phloroglucinol (1.0 equiv), NaH (1.6 equiv), R-bromo ethylhexanoate (0.8 equiv), DMF (q.s.), rt, 8 h; (II) 22a
(1.0 equiv), phenylethanol (2.1 equiv), DEAD (2.5 equiv), TPP (2.5 equiv), THF, rt, 4 h; (III) 22b (1.0 equiv), LiOH H2O (5.0 equiv), THFꢀH2O,
3
50 ꢀC, 24 h.
A detailed description of the methods for analysis of inhibition
of 5-LO and mPGES-1 according to Koeberle et al.10 is summar-
ized in the Materials and Methods. In brief, inhibition of 5-LO
product formation was analyzed in a cell-based assay using
PMNL, as well as in a cell-free assay using purified human
recombinant 5-LO enzyme.16 Because a given test compound
may suppress 5-LO product synthesis in intact cells without
inhibiting 5-LO directly, for example, by interference with
cofactors regulating 5-LO in the cell or with other enzymes
involved in LT synthesis (e.g., FLAP, LTA4H, LTC4S), we
analyzed all compounds for inhibition of isolated 5-LO as well.
On the other hand, since many compounds inhibit 5-LO in cell-
free assays but fail in intact cells for various reasons,7 we
performed the cell-based assay and the cell-free assay side by
side. Inhibition of mPGES-1 activity (transformation of PGH2 to
PGE2) was assessed in a cell-free assay using the mitochondrial
fraction of IL-1β-stimulated A549 lung epithelial adenocarcino-
ma cells and 20 μM PGH2 as substrate.17 In contrast to 5-LO
synthesis inhibitors, for mPGES-1 inhibitors a cell-based test
system that allows selective analysis of interference with endo-
genous mPGES-1 activity in the cell (i.e., the transformation of
PGH2 to PGE2) is not available. Nevertheless, we addressed the
ability of test compounds to interfere with COX enzymes that are
committed in PGE2 biosynthesis from the precursor AA. Inhibi-
tion of COX was analyzed in a cell-free assay using isolated ovine
COX-1 and isolated human recombinant COX-2 enzymes.18 As
reference compounds, the 5-LO inhibitor 35 (BWA4C, (E)-N-
hydroxy-N-(3-(3-phenoxyphenyl)-allyl)acetamide),19 the mPGES-
1 inhibitor 36 (MK-886, 3-(3-(tert-butylthio)-1-(4-chlorobenzyl)-
5-isopropyl-1H-indol-2-yl)-2,2-dimethylpropanoic acid),20 the
4492
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
Scheme 4. Synthesis of Disubstituted 2-Benzylhexanoic Acid and 2-Benzylidenehexanoic Acid Derivatives 23ꢀ34. a
a Reagents and conditions: (I) triethylphosphite (1 equiv), R-bromo ethylhexanoate (1 equiv), 120 ꢀC, 12 h; (IIa) dihydroxybenzaldehyde (1 equiv),
R1-OH (2.1 equiv), TPP (2.5 equiv), DEAD (2.5 equiv), THF, rt, 1.4ꢀ24 h or 2,5-dihydroxybenzaldehyde (1 equiv), 4-(trifluoromethyl)benzylbromide
(1.2 equiv), CsCO3 (1.2 equiv), DMF, 50 ꢀC, 5 h; (IIb) 2,5-dihydroxybenzaldehyde (1 equiv), 4-(trifluoromethyl)benzylbromide (1.2 equiv), CsCO3
(1.2 equiv), DMF, 50 ꢀC, 5 h; (IIc) 34a (1 equiv), 2-cyclohexylethanol (1.1 equiv), TPP (1.2 equiv), DEAD (1.2 equiv), THF, rt. (III) A (1.3 equiv),
23b, 25bꢀ28b, 31bꢀ34b (1 equiv), NaH (1.3 equiv), THF, rt, 2ꢀ22 h; (IV) 23c, 25ꢀ28c, Pd/C (10%), ethanol, rt, 24 h; (V) 24c, 29ꢀ34c or 23d,
25ꢀ28d (1 equiv), LiOH H2O (5 equiv), THFꢀH2O, 50ꢀ60 ꢀC, 6ꢀ40 h.
3
COX-1/2 inhibitor 37 (indomethacin, 2-[1-(4-chlorobenzoyl)-
5-methoxy-2-methyl-1H-indol-3-yl]acetic acid), and the COX-2
inhibitor 38 (celecoxib, 4-(5-(4-methylphenyl)-3-(trifluoromethyl)-
pyrazol-1-yl)benzenesulfonamide) were used.
The carboxylic acid moiety of compound 1 was not altered since
the corresponding ester of 1 did not show significant effects on
PGE2 and 5-LO product formation at a concentration of 10 μM
(remaining activity >80%, respectively; see Koeberle et al10).
The results, summarized in Tables 1ꢀ5, are expressed as IC50
values for all those compounds that inhibit mPGES-1 and 5-LO
activity at 10 μM by more than 50%, for those with lower potency
the remaining enzyme activities at 10 μM inhibitor concentra-
tions are given. Concentrations >10 μM have not been con-
sidered due to solubility inconsistencies in aqueous assay buffers
at higher concentrations of the test compounds. Starting from
the lead 1, the importance of the aliphatic chain in R-position to
the carboxylic acid was investigated by elongation or shortening
of the R-n-butyl residue as well as by its replacement with a
phenyl moiety. These modifications resulted in defined SARs for
5-LO and mPGES-1 (see Table 1). Whereas unsubstituted
StructureꢀActivity Relationships of Dual Inhibitors of
mPGES-1 and 5-LO. We have recently described the screening
of dual mPGES-1/5-LO inhibitors based on pirinixic acid as basic
structure. Within this study, three derivatives containing two
phenethoxy residues at the pyrimidine ring (including com-
pound 1) were synthesized and biologically characterized.10
Encouraged by the efficient inhibition of mPGES-1 (IC50 = 1.2 μM)
and of 5-LO (cell-based IC50 = 0.6 μM, cell-free IC50 = 4.7 μM)
by 2-[(4,6-diphenethoxypyrimidin-2-yl)thio]hexanoic acid, 1,
we designed and synthesized a large set of compounds with
broad modifications of all relevant structural features and per-
formed SAR studies with respect to 5-LO and mPGES-1 inhibition.
4493
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
Table 1. Structural Modification of the Substituents in r-Position of the Carboxylic Group and Resulting Effects on Inhibition of
5-LO and mPGES-1a
a Data are expressed as mean ( SE; n = 3ꢀ4.
compound 2 and ethyl-substituted 3 were less potent on 5-LO in
intact cells (IC50 = 7.6 (2) and 2.8 μM (3)) compared with
n-butyl-substituted 1 (IC50 = 0.6 μM), the elongated R-n-hexyl
derivative 4 and R-phenyl-substituted derivative 5 essentially
retained high potency (IC50 = 0.4 (4) and 0.8 μM (5)).
Inhibition of 5-LO in the cell-free assay followed the same SARs
as in intact cells, with about 3- to 8-fold reduced potency versus
intact cells. The same SARs as for 5-LO are true also for mPGES-
1 inhibition, because compound 4 (IC50 = 2.0) with R-n-hexyl
was similarly potent to R-n-butyl-substituted 1 and the same
potency was obtained by the introduction of R-phenyl moiety
(5). Thus, lipophilic bulky substituents in the R-position of the
carboxylic group seemingly govern potency.
of para-substituents or replacement of phenyl by thiophene
resulted in a subset of potent cellular 5-LO inhibitors (IC50
=
0.3ꢀ0.7 μM). Among this set of compounds, the methoxy-
substituted 13 is the most potent 5-LO inhibitor in intact cells.
Concerning mPGES-1 inhibition, a correlation between the
respective para-substituents and the potency was obvious. Thus,
the introduction of trifluoromethyl (12, IC50 = 1.9 μM) was well
tolerated and especially a methyl moiety (14; IC50 = 0.9 μM)
retained the potency, whereas the corresponding trifluoro-
methoxy (11) or methoxy (13) groups markedly impaired the
activity against mPGES-1. Electron-withdrawing cyano (9) and
nitro (10) groups as well as replacement of phenyl by thiophene
(15) impaired the potency as well.
We used compound 1, as the most potent dual 5-LO and
mPGES-1 inhibitor, as template for further optimization. The
two phenethoxy moieties at the pyrimidine core forming a
lipophilic backbone offer multiple opportunities for structural
optimization. By shortening of the ethylene spacer connecting
the phenyl moiety with the ether bridge via methylene in
compound 6, the potency for 5-LO and mPGES-1 inhibition
was reduced (see Table 2). However, propylene (7) or butylene
(8) spacers were tolerated, visualized by the submicromolar IC50
values of 8 for 5-LO (cell-based) and mPGES-1 (0.7 and 0.9 μM,
respectively).
Next, several electron-withdrawing (cyano (9), nitro (10),
trifluoromethoxy (11), trifluoromethyl (12)) and -donating
(methoxy (13) and methyl (14)) substituents were introduced
in para position of the phenyl moieties of compound 1. Ad-
ditionally, both phenyl moieties of 1 were replaced by bioisos-
teric thiophene (15). As can be seen from Table 2, introduction
In addition, the phenyl moieties of the lipophilic backbone
were replaced by aliphatic rings of various sizes (cycloheptyl
(16), cyclohexyl (17), cyclopentyl (18), and cyclopropyl (19)),
as well as by branched isopropyl (20) moieties. Introduction of
cycloheptyl (16) and cyclopentyl rings (18) was beneficial
neither for 5-LO (IC50 cell-based =2.4 (16) and 4 μM (18))
nor for mPGES-1 (IC50 > 10 μM), whereas the intermediate
sized cyclohexyl derivative 17 was quite potent with IC50 values
for 5-LO in PMNL = 0.8 μM and for mPGES-1 = 2.4 μM.
Interestingly, the smaller isopropyl (20) and cyclopropyl moi-
eties (19) are rather superior over the cyclopentyl-substituted 18.
The second part of the SAR study focused on the impact of the
central core scaffold of the compounds. Replacing the pyrimidine
ring by a benzene moiety (while keeping the thioether of 1) led to
compound 21 that inhibited 5-LO almost equally well (IC50 cell-
based = 0.8 μM) as compound 1 (Table 3) and was somewhat
less active on mPGES-1 (IC50 =4.6 μM). Moreover, replacement of
4494
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
Table 2. Structural Modification of the Lipophilic Backbone and Resulting Effects on Inhibition of 5-LO and mPGES-1a
ARTICLE
a Data are expressed as mean ( SE; n = 3ꢀ4.
4495
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
Table 3. Variation of the Central Core Scaffold and Resulting Effects on Inhibition of 5-LO and mPGES-1a
a Data are expressed as mean ( SE; n = 3ꢀ4.
the thioether of compound 21 by an ether moiety (compound
22) even further (although marginally) improved the potency
in PMNL, mPGES-1 = 2.4 μM). Additionally, we prepared the
corresponding 2-benzylidenehexanoic acid analogues of 25 and
28, that is, the 2-(2,3-diphenethoxybenzylidene)hexanoic acid,
29, and 2-(2,5-diphenethoxybenzylidene)hexanoic acid, 30, re-
spectively. In fact, replacement of the ethylene by an ethenyl
group enhanced the potency for 29 (IC50 5-LO cell-based =
0.8 μM; mPGES-1 = 1.1 μM) against both 5-LO and mPGES-1.
In order to further improve the efficiency, the phenethoxy
residues of compound 30 (IC50 5-LO cell-based = 0.9 μM;
mPGES-1 = 1.5 μM) were modified (Table 5). Replacement of
the phenyl moieties in 30 by cyclohexyl (31) hardly altered the
efficiency. However, substitution of the phenyl moieties by
trifluoromethyl residues in para-position (32) enhanced the
potency against 5-LO in PMNL (IC50 = 0.3 μM). Shortening
the ethylene linker between phenyl and the ether moiety of 32 to
methylene (33) was tolerated with respect to 5-LO but the
potency against mPGES-1 was lost. Also, compound 34 carrying
one p-(trifluoromethyl)benzyloxy substituent and one cyclohex-
ylethoxy residue was not active against mPGES-1 but efficiently
inhibited 5-LO. Data of the reference inhibitors of 5-LO
(compound 35) and mPGES-1 (compound 36) are given at
the end of the table, and onthe basisof the resulting effects of these
compounds that correspond to the values from literature,19,20 we
conclude that our results obtained for the novel compounds 1 to
34 are valid.
against 5-LO in PMNL (IC50 = 0.6 μM) and mPGES-1 (IC50
=
3.4 μM) versus 21. Finally, exchange of the thioether by a
methylene bridge, that is, replacement of all three heteroatoms
of 1 with a pure carbon scaffold (2-(3,5-diphenethoxyben-
zyl)hexanoic acid, 23) was essentially tolerated versus 1 with
respect to mPGES-1 (IC50 = 2.4 μM), whereas for 5-LO in
PMNL a loss of potency was evident (IC50 = 4.7 μM). Accord-
ingly, the presence of a heteroatom (e.g., sulfur or oxygen),
bridging the R-position and the aromatic ring, is not essential but
at least governs bioactivity against 5-LO. Nevertheless, it was
possible to restore the potent 5-LO inhibition of 1 by keeping the
R-carbon atom but forming an ethenyl group (i.e., 2-(3,
5-diphenethoxybenzylidene)hexanoic acid, 24) which could be
prepared within the same synthetic route as 23. We therefore
conclude that the steric configuration in this area in a fixed
position might be favorable.
Based on the modified synthetic procedure of compound 23,
the variation of the two 3,5-diphenethoxy-substitution pattern
was possible, and the respective 2,3- (25), 3,4- (26), 2,4- (27),
and 2,5-(28) diphenethoxy-substituted 2-benzylhexanoic acid
derivatives were prepared (see Table 4). Interestingly, shifting
the two phenethoxy moieties from 3,5- to 2,4-positions (27) led
to inactivity on both targets (remaining activity of 5-LO/
mPGES-1 at 10 μM > 90%). 2,3-Substituted (25) and 3,4-
substituted (26) compounds were about equally active compared
with 3,5-diphenethoxy-substituted 23. The most potent dual
5-LO and mPGES-1 inhibitor from this set turned out to be the
2,5-diphenethoxy-substituted compound 28 (IC50 5-LO = 0.9 μM
Inhibition of COX Enzymes. Inhibition of COX-1/2 by
compounds 1ꢀ20 was examined before;12 compounds 21ꢀ37
were assayed in the COX activity test systems, and the results are
reported in Table 6. Inhibition of isolated ovine COX-1 and
human recombinant COX-2 was determined and expressed as
4496
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
Table 4. Variation of the Substitution Pattern of the Central Benzene Ring and Resulting Effects on Inhibition of 5-LO and mPGES-1a
ARTICLE
a Data are expressed as mean ( SE; n = 3ꢀ4.
remaining activity. In brief, the remaining COX-1 and COX-2
activities are higher than 50% for all compounds tested at a
concentration of 10 μM. Thus, with respect to the high potency
of the identified lead compounds against 5-LO and mPGES-1
(e.g., 1, 4, 8, 14, and 29), the concentrations needed for half-maximal
inhibition of COX enzymes are at least 10-fold higher than those for
dual mPGES-1 and 5-LO inhibition.
Analysis of the Molecular Pharmacology of 2-[(4,6-Diphe-
nethoxypyrimidin-2-yl)thio]octanoic Acid, 4. As shown
above, 5-LO inhibition for all of the 34 test compounds was
4497
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
Table 5. Variation of the Lipophilic Backbone of Compound 30 and Resulting Effects on Inhibition of 5-LO and mPGES-1a
ARTICLE
a Data are expressed as mean ( SE; n = 3ꢀ4.
superior in the cell-based test system compared with the cell-free
assay, and for some compounds, for example, 11, 13, 22, 32, and
34, the IC50 values were even more than 20-fold higher in the
cell-free versus the cell-based assay. Hence, the compounds act as
direct 5-LO inhibitors, but apparently factors in intact cells
govern the potency, or alternatively, the compounds may primarily
(and potently) suppress cellular 5-LO product synthesis by other
additional mechanisms than direct interference with 5-LO.
To further investigate the underlying mechanism of inhibition
of cellular 5-LO product synthesis, we performed mechanistic
studies with compound 4 (one of the most effective 5-LO
inhibitors in cell-free and cell-based assays from our SAR
investigations), applying particular experimental settings. In
order to assess the selectivity of 4 for 5-LO, we investigated
the effects of 4 on the activity of 12/15-LO (15-LO-1), expressed
in eosinophilic granulocytes, and of platelet-type-12-LO derived
from PMNL-adherent platelets. The amount of 15(S)-hydro-
(pero)xy-5,8,11-cis-13-trans-eicosatetraenoic acid (15(S)-H-
(p)ETE), formed by 12/15-LO, and 12(S)-hydro(pero)xy-5,
8-cis-10-trans-14-cis-eicosatetraenoic acid (12(S)-H(p)ETE), formed
by p12-LO, was rather augmented after treatment with 4
(Figure 1A), and a similar pattern could be observed for the
reference 5-LO inhibitor 38 (not shown).
In view of the structural features, no redox- or iron chelating
activities are evident for compound 4 and its analogues, and in a
radical scavenger assay, 4 up to 100 μM failed to reduce 1,1-
diphenyl-2-picrylhydrazyl (DPPH) in contrast to ascorbic acid
and L-cystein (data not shown). Moreover, wash-out experi-
ments (dilution from 3 to 0.3 μM) demonstrated that the
inhibitory effect of 4 on isolated 5-LO can be reverted (not
shown), implying that the compound does not irreversibly
inactivate 5-LO or acts by covalent binding to it.
Previous studies showed that the inhibitory potency of some
5-LO inhibitors can vary depending on the stimulus used to
evoke 5-LO product synthesis in PMNL.21,22 Therefore, besides
induction of 5-LO product formation with A23187, we addressed
5-LO product formation also after stimulation of PMNL with
1 μM N-formyl-methionyl-leucyl-phenylalanine (fMLP) upon
priming with 1 μg/mL lipopolysaccharide (LPS) leading to 5-LO
activation via release of intracellular Ca2þ and phosphorylation
4498
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
Table 6. COX Inhibition of the Compounds 1-34 at a Concentration of 10 μMa
a Values of compounds 1ꢀ20 were reported previously.12 Data are given as mean ( SE, n = 3. Reference inhibitors 37 (indomethacin) and 38
(celecoxib) are listed at the end of the table.
by mitogen-activated protein kinases (MAPK).23 Indeed, the
potency of 4 for inhibition of 5-LO product formation was
somewhat impaired upon stimulation with LPS/fMLP compared
with A23187 (Figure 1B). Activation of 5-LO by phosphoryla-
tion after applying cell stress (300 mM NaCl, together with
40 μM AA) occurs independent of Ca2þ24 and these conditions
led to a slight loss of efficacy of 4 as well (Figure 1C). Con-
clusively, compound 4 is more efficient on 5-LO activated by
Ca2þ rather than under conditions where phosphorylations by
MAPK partially or fully activate 5-LO.
Next, we tested the influence of the substrate concentration on
the 5-LO inhibitory effect of 4. Figure 1D demonstrates that the
potency of 4 is reduced with increasing amounts of AA (20, 40,
and 60 μM), suggesting that 4 may inhibit 5-LO in a competitive
manner. A similar effect on the efficiency of the reference 35 was
obvious, although the loss of potency did not depend on the
various AA concentrations.
A major disadvantage of several non-redox-type 5-LO inhibi-
tors is their dependence on reducing conditions.21 In fact, the
5-LO inhibitory effect of 4 in 100 000 ꢁ g supernatants from
4499
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
Figure 1. Effect of compound 4 on 5-LO product formation in PMNL. (A) Influence of compound 4 on 5-, 12-, and 15-LO product formation in
A23187-stimulated PMNL (5 ꢁ 106/mL) in the presence of 20 μM AA. The amounts of 15(S)-H(p)ETE, 12(S)-H(p)ETE, 5(S)-H(p)ETE, and LTB4
were determined by HPLC. The 100% values correspond to 53.1 ( 13.0 ng/mL 15-H(p)ETE, 150.0 ( 36.4 ng/mL 12-H(p)ETE, 618.3 ( 55.6 ng/mL
5-H(p)ETE, and 85.1 ( 4.5 ng/mL LTB4. Data are means þ SEM, n = 4. (B) Effect of compound 4 on LTB4 formation in A23187- and LPS/fMLP-
stimulated PMNL (5 ꢁ 106/mL). Cells were preincubated with 4 for 15 min at 37 ꢀC, and 2.5 μM A23187 was added. LTB4 was determined by HPLC
after 10 min incubation at 37 ꢀC. Alternatively, cells (2 ꢁ 107/mL) were preincubated with 4 for 15 min at 37 ꢀC, primed with 1 μg/mL LPS, and then
stimulated with 1 μM fMLP. After 5 min, the amount of released LTB4 was determined by ELISA. The 100% values correspond to 189.2 ( 57.7 pg/mL
LTB4. Data are means þ SEM, n = 4; **p < 0.01 vs controls that were stimulated with A23187, ANOVA þ Tukey post-hoc test. (C) Efficiency of
compound 4 under conditions that induce 5-LO product synthesis by A23187 or cell stress. PMNL (5 ꢁ 106/ml) were preincubated with 4 for 15 min at
37 ꢀC. NaCl (0.3 M) was supplemented 3 min before addition of 40 μM AA; 2.5 μM A23187 was added together with 40 μM AA. After 10 min at 37 ꢀC,
the reaction was stopped and 5-LO products were measured. The 100% values correspond to 529.4 ( 61.2 and 335.8 ( 50.0 ng/mL 5-LO products for
cells stimulated by A23187 and by NaCl, respectively. Data are means þ SEM, n = 4. (D) Efficiency of compounds 4 and 35 at variant substrate
concentrations. PMNL (5 ꢁ 106/ml) were preincubated with 4 or 35 for 15 min at 37 ꢀC, the indicated amounts of AA and 2.5 μM A23187 were added,
and the reaction was stopped after 10 min to analyze 5-LO product formation: 199.2 ( 37.9, 1136.7 ( 299.7, 529.4 ( 61.2 and 498.4 ( 87.7 ng/mL
5-LO products at 0, 20, 40, and 60 μM AA, respectively. Data are means þ SEM, n = 3.
PMNL homogenates (S100, IC50 = 12.5 μM) or on isolated
5-LO (IC50 = 2.8 μM) was significantly attenuated versus intact
cells (IC50 = 0.4 μM), which was true also for the non-redox-type
5-LO inhibitors 1-ethyl-6-((3-fluoro-5-(4-methoxy-3,4,5,6-tetra-
hydro-2H-pyran-4-yl)phenoxy)methyl)-2-quinolone (ZM230487)25
or 2-cyano-4-(3-furyl)-7-[[6-[3-(3-hydroxy-6,8-dioxabicyclo-
[3.2.1]octanyl)]-2-pyridyl]methoxy]-naphthalene (L-739.010).21
In contrast to these non-redox-type inhibitors, addition of 1 mM
dithiothreitol (DTT, to reconstitute GPx activity and thus to
create reducing conditions) did not restore potent 5-LO inhibition
4500
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
Figure 2. Influence of the redox tone on 5-LO inhibition by compound 4 in cell-free assays. (A) Influence of DTT. The S100 from homogenates of
PMNL (5 ꢁ 106/mL) was preincubated with compound 4 or with vehicle (0.3% DMSO) in the presence or absence of 1 mM DTT for 5ꢀ10 min at
4 ꢀC. Then, samples were prewarmed at 37 ꢀC for 30 s and 2 mM CaCl2 and 20 μM AA were added, and after 10 min at 37 ꢀC, 5-LO activity was analyzed
by HPLC. The 100% values correspond to 765.2 ( 261.5 and 1044.8 ( 263.0 ng/mL 5-LO products in homogenates and in homogenates plus DTT,
respectively. Data are means þ SEM, n = 3. (B) Influence of 13(S)-HpODE. Purified 5-LO (0.5 μg/mL, human recombinant) was preincubated with 4
or with vehicle (0.3% DMSO) for 5ꢀ10 min at 4 ꢀC. Samples were prewarmed at 37 ꢀC for 30 s, and 5-LO product formation was started by addition of
2 mM CaCl2 and 20 μM AA in the presence or absence of 1 μM 13(S)-HpODE. After another 10 min, 5-LO products were determined. The 100% values
correspond to 923.4 ( 196.1 and 1373.9 ( 197.4 ng 5-LO products per 0.5 μg of enzyme in the absence and presence of 13(S)-HpODE, respectively.
Data are means þ SEM, n = 3.
Figure 3. Computer-assisted protein binding site analysis. Potential binding sites for dual mPGES-1 and 5-LO inhibitors were predicted by PoLiMorph,
as highlighted: (A) 5-LO (PDB-ID 3o8y) and (B) mPGES-1 (PDB-ID 3dww). Reasonable binding poses were obtained by automated ligand-docking,
exemplarily shown for compound 1 in the pockets of (C) 5-LO and (D) mPGES-1.
in S100 (Figure 2A), suggesting that the potency of 4 is inde-
pendent of the redox tone. Along these lines, inclusion of 1 μM
13(S)-HpODE in the cell-free assay using isolated 5-LO did not
alter the potency of compound 4 (Figure 2B).
4501
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
Computational Analysis of Potential Ligand Binding Sites.
In a preliminary attempt to identify potential ligand binding sites
for dual 5-LO and mPGES-1 inhibitors, cavities on the surface of
the two proteins were compared. First, potential binding sites of
Protein Data Bank (PDB)26 entries 3o8y27 (5-LO, chain A) and
3dww28 (mPGES-1) were extracted using our software Pocket-
Picker.29 For 5-LO, we found nine and for mPGES-1 six surface
cavities exceeding a volume of 110 Å3 (less than 5% of all protein
binding sites accommodating a ligand have a smaller volume;
study performed on scPDB30 v.2009, data not published). Then,
pairwise comparisons between all identified cavities from mPGES-1
and 5-LO were performed. This was done with PoLiMorph,31 a
software tool that is able to reveal related receptorꢀligand
interaction potentials even in the absence of protein sequence
homology. Only one pocket pair from 5-LO and mPGES-1
achieved a similarity score greater than PoLiMorph’s reliability
threshold of 0.15.31 In 5-LO, this predicted binding site lies
distant from the active site and the C2-like domain (Figure 3A).
risk of side effects. Here, we describe the discovery of a novel class
of potent dual inhibitors of 5-LO and mPGES-1 based on the
structure of 2-[(4,6-diphenethoxypyrimidin-2-yl)thio]hexanoic
acid, 1, with minor influence on the activity of COX-1/2.
Compound 14 is one of the most potent leads with a well-
balanced activity against 5-LO (IC50 cell-based = 0.5 μM) and
mPGES-1 (IC50 = 0.9 μM). Of interest, we established a new
synthetic procedure allowing the replacement and broad mod-
ification of the central pyrimidine core of 1 in order to vary the
phenethoxy positions at a central benzene ring with large space
for structural optimization. Note that these 2-(diphenethoxy-
benzylidene)hexanoic acids constitute a novel structural class of
dual mPGES-1/5-LO inhibitors with 2-(2,3-diphenethoxyben-
zylidene)hexanoic acid 29 (IC50 5-LO in PMNL = 0.8 μM;
mPGES-1 = 1.1 μM) as the most potent representative from this
set of compounds.
Evaluation of previous 5-LO inhibitors showed that complex
regulation of 5-LO by various cofactors6 influences the suscept-
ibility toward inhibitors and the efficiency of the compounds may
strongly depend on experimental settings.7,8 Characterization in
various cell-based and cell-free assays indicate that compound 4
shares characteristics with classical non-redox-type 5-LO inhibi-
tors, including impaired potency when 5-LO is activated by
phosphorylation events and the loss of efficiency at increased
substrate concentrations.33 However, 4 markedly inhibits 5-LO
product formation also under nonreducing conditions, which is
not the case for non-redox-type 5-LO inhibitors and thus en-
courages further studies. Taken together, we present different
novel classes of compounds that dually act on 5-LO and mPGES-
1 with minor activity on COX enzymes. Based on the overall
suitable pharmacological profile, one may expect a synergistic
anti-inflammatory activity along with unwanted side effects of
traditional NSAIDs and coxibs. These data encourage for further
pharmacological and pharmacokinetic studies in order to reveal
the in vivo characteristics of the compounds in animal models.
It is flanked by residues Arg221, His225, Met230, Tyr234, Leu237
,
Lys319, Tyr467, and Glu656. In mPGES-1, the respective predicted
binding site is formed by residues His71, Arg72, and Met76 from
each of the three chains. Here, the pocket is located at the center
of the protein structure where these residues come into close
proximity (Figure 3B). Notably, both pockets contain an arginineꢀ
histidineꢀmethionine triangle with similar edge lengths. To
further elaborate the hypothesis that dual mPGES1/5-LO in-
hibitors might bind to these two pockets proposed by PoLi-
Morph, automated ligand docking studies were performed for
compounds 1, 2, 5, 13, 14, 29, and 32 using the GOLD, v. 5.0.1,
software in combination with the GOLDscore function.32 Favor-
able average docking score values of 81 (stddev = 7) for the
predicted mPGES-1, and 79 (stddev = 3) for the predicted 5-LO
binding site were obtained. These scores are statistically indis-
tinguishable (MannꢀWhitney U-test, p = 0.46), which points to
actual pocket similarity. In comparison, docking of the com-
pounds into the active site of 5-LO yielded significantly lower
scores (mean = 61; stddev = 4; MannꢀWhitney U-test: p <
0.001). Visual inspection of docking poses revealed reasonable
conformations in both potential binding sites (Figure 3C,D). In
the cavity of 5-LO, an aromatic face-to-face interaction between
His225 and one of the two phenyl-residues of compound 1 is
predicted, and the carboxylic group of compound 1 might act as a
hydrogen-bridge acceptor for Lys319 and Gln659 side chains. In
the binding pocket of mPGES-1, aromatic face-to-face interac-
tions seem feasible between both phenyl residues and His72 in
chain A and His72 in chain B, respectively. Here, the carboxylic
group acts as a potential hydrogen-bridge acceptor for Arg73 in
chain A. As a result of this computational study, we suggest
that the two identified pockets might accommodate dual
inhibitor 1. The proposed interacting amino acid residues
are candidates for mutation studies, which will help assess the
validity of our hypothesis.
’ MATERIALS AND METHODS
Compounds and Chemistry. The structures of previously re-
ported compounds 1ꢀ20 were confirmed by 1H and 13C NMR as well as
by mass spectrometry (ESI); the purity (>95%) was determined by
combustion analysis as described.12 Compounds 21ꢀ34 were synthe-
sized as described in Schemes 2ꢀ4. All commercial chemicals and
solvents are of reagent grade and were used without further purification,
unless otherwise specified. 1H and 13C NMR spectra were measured in
DMSO-d6 or CDCl3 on a Bruker ARX 300 or AV 300 spectrometer.
Chemical shifts are reported in parts per million (ppm) using tetra-
methylsilane (TMS) as an internal standard. Mass spectra were obtained
on a Fissous Instruments VG Platform 2 spectrometer measuring in the
positive- or negative-ion mode (ESI-MS system). The purities of the
final compounds were determined by combustion analysis, which has
been performed by the Microanalytical Laboratory of the Institute of
Organic Chemistry and Chemical Biology, Goethe-University Frankfurt,
on a Foss Heraeus CHN-O-rapid elemental analyzer or an Elementar
Vario Micro Cube. All compounds described here have a purity of 95%
or higher. The synthetic procedure is described representatively for
compounds 21, 23, and 24. Detailed synthetic and analytical data of all
compounds are provided in the Supporting Information.
’ CONCLUSIONS
The therapeutic use of NSAIDs and coxibs against inflamma-
tory diseases has become highly prevalent, although these classes
of compounds possess critical target (i.e., COX)-related side
effects resulting from the suppression of homeostatically relevant
or protective PGs.4 Agents that dually inhibit the PGE2 and the
LT synthetic pathway are thought to be more efficient than
traditional NSAIDs and, in particular, may also exhibit reduced
Synthesis of 21 (2-[(3,5-Diphenethoxyphenyl)thio]hexanoic
Acid) (Following Scheme 2). In method A (synthetic procedure of
steps I, II, and V), sodium hydride (95%) was suspended in anhydrous
DMF (q.s.) and stirred under argon atmosphere and ice bath cooling.
Subsequently, E1 was slowly added to the solution via syringe. After the
4502
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
mixture was stirred for 30ꢀ60 min, E2 was added, and the suspension
was stirred at room temperature (steps I and V) or 60 ꢀC (step II) until
completed (TLC control). The mixture was then diluted with water and
extracted two times with ethyl acetate. The organic fractions were dried
over MgSO4 and evaporated to yield the crude product. Purification was
done by silica gel column chromatography using n-hexane/ethyl acetate
to yield the pure products.
O-CH2), 4.12ꢀ4.17 (t, 4H, J = 6.8 Hz, Ph-O-CH2), 6.38ꢀ6.39 (m, 1H,
Ph-H), 6.51 (d, 2H, J = 2.1 Hz, Ph-H), 7.17ꢀ7.30 (m, 10H, Ph0-H). 13C
NMR (75.44 MHz, (CD3)2SO): δ = 13.67 (Bu-CH3), 13.84 (Et-CH3),
21.67 (Bu-CH2), 28.63 (Bu-CH2), 30.80 (Bu-CH2), 34.78 (2C, Ph0-
CH2), 48.83 (S-CH), 60.64 (O-CH2), 68.27 (2C, Ph-O-CH2), 100.29
(Ph-C4), 108.98 (2C, Ph-C2/6), 126.23 (2C, Ph0-C4), 128.24 (4C, Ph0-
C2/6), 128.89 (4C, Ph0-C3/5), 135.48 (Ph-C1), 138.23 (2C, Ph0-C1),
159.58 (2C, Ph-C3/5), 171.62 (COO-). MS (ESIþ) = m/e = 493.3
[M þ 1]þ
Step I, Synthesis of 21a (3,5-Diphenethoxyphenol):13. Method A:
NaH (95%, 2.4 g, 90 mmol); E1, phloroglucinol (3.78 g, 30 mmol); E2,
phenethyl bromide (11.7 g, 63 mmol); duration of synthesis, 20 h;
Step VI, Synthesis of Final Compound 21 (2-[(3,5-Diphenethox-
yphenyl)thio]hexanoic Acid): The corresponding ester 21e (0.13 g, 0.26
mmol) was dissolved in a mixture of 5 mL of THF/10 mL of MeOH and
1
product, orange solid; yield, 18.4% (1.84 g). H NMR (300.13 MHz,
(CD3)2SO): δ = 2.94ꢀ2.99 (t, 4H, J = 6.7 Hz, Ph0-CH2), 4.05ꢀ4.13 (m,
4H, Ph-O-CH2), 5.90ꢀ5.92 (m, 3H, Ph-H), 7.17ꢀ7.31 (m, 10H, Ph0-
H), 9.37 (s, 1H, Ph-OH). 13C NMR (75.44 MHz, (CD3)2SO): δ =
34.86 (Ph0-CH2), 67.38 (2C, Ph-O-CH2), 92.30 (Ph-C4), 94.65 (2C,
Ph-C2/6), 126.46 (2C, Ph0-C1), 128.10 (4C, Ph0-C2/6), 128.87 (4C, Ph0-
C3/5), 138.40 (2C, Ph0-C4), 159.04 (Ph-C1), 160.22 (2C, Ph-C3/5). MS
(ESIþ) = m/e = 335.1 [M þ 1]þ.
a solution of LiOH H2O (0.05 g, 1.31 mmol) in 3 mL of H2O was
3
added. After stirring at 50 ꢀC for 3 h, the solvent was removed, and the
residue was dissolved in water (under heating; if necessary, low amounts
of MeOH were added). The solution was acidified with diluted hydro-
chloric acid. The formed precipitate was filtered, washed to neutrality
with water, and then washed with n-hexane. Purification of the product
was done using silica gel column chromatography and n-hexane/ethyl
acetate to obtain the pure product as light yellow oil. Yield: 65.6%
(0.12 g). 1H NMR (300.13 MHz, (CD3)2SO): δ = 0.82ꢀ0.87 (t, 3H, J =
7.0 Hz, Bu-CH3), 1.24ꢀ1.41 (m, 4H, Bu-CH2), 1.58ꢀ1.84 (m, 2H, Bu-
CH2), 2.98ꢀ3.04 (t, 4H, J = 6.5 Hz, Ph0-CH2), 3.77ꢀ3.83 (t, 1H, J =
7.0 Hz, S-CH), 4.14ꢀ4.19 (t, 4H, J = 6.7 Hz, Ph-O-CH2), 6.37ꢀ6.41
(m, 1H, Ph-H), 6.53ꢀ6.56 (m, 2H, Ph-H), 7.19ꢀ7.33 (m, 10H, Ph0-H),
12.75 (s/br, 1H, COOH). 13C NMR (75.44 MHz, (CD3)2SO): δ =
13.70 (Bu-CH3), 21.72 (Bu-CH2), 28.71 (Bu-CH2), 31.01 (Bu-CH2),
34.82 (2C, Ph0-CH2), 49.09 (S-CH), 68.26 (2C, Ph-O-CH2), 99.96 (Ph-
C4), 108.46 (2C, Ph-C2/6), 126.24 (2C, Ph0-C4), 128.26 (4C, Ph0-C2/6),
128.91 (4C, Ph0-C3/5), 136.30 (Ph-C1), 138.24 (2C, Ph0-C1), 159.61
(2C, Ph-C3/5), 173.06 (COO). MS (ESIþ) = m/e = 465.5 [M þ 1]þ
Synthesis of Compounds 23 and 24 (Exemplary of All
Compounds with Central Carbon Scaffold (23ꢀ37)):15. Step
I, Synthesis of Precursor A (Ethyl 2-(Diethoxyphosphoryl)hexanoate):
Triethylphosphite and R-bromo ethylhexanoate were solved together.
After stirring overnight at 120 ꢀC, the mixture was distilled. First, the
resulting ethylbromide was removed at 40 ꢀC. Next, the product was
distilled under vacuum (0.5 mbar) and 100 ꢀC. 1H NMR (300.13 MHz,
(CD3)2SO): δ = 0.81ꢀ0.85 (t, 3H, J = 7.1 Hz, Bu-CH3), 1.15ꢀ1.24 (m,
13H, Et-CH3 þ Bu-CH2), 1.61ꢀ1.82 (m, 2H, Bu-CH2), 2.91ꢀ3.04 (m,
1H, CH-COO-), 3.94ꢀ4.16 (m, 6H, O-CH2). MS (ESIþ) = m/e =
281.1 [M þ 1]þ.
Step II, Synthesis of 21b (O-(3,5-Diphenethoxyphenyl)dimethyl-
carbamothioate):14. Method A: NaH (95%, 0.16 g, 6.3 mmol), E1, 21a
(1.7 g, 5.1 mmol); E2, dimethylthiocarbamoyl chloride (0.62 g, 5.1
mmol); temperature, 60 ꢀC; duration of synthesis, 5 h; product,
transparent oil; yield, 65.4% (1.4 g). 1H NMR (300.13 MHz, CDCl3):
δ = 3.03ꢀ3.11 (t, 4H, J = 6.8 Hz, Ph0-CH2), 3.31 þ 3.39 (s, 6H, N-CH3),
4.11ꢀ4.23 (m, 4H, Ph-O-CH2), 5.97ꢀ6.00 (m, 1H, Ph-H), 6.24ꢀ6.35
(m, 3H, Ph-H), 7.24ꢀ7.37 (m, 10H, Ph0-H). 13C NMR (75.44 MHz,
CDCl3): δ = 35.35 (2C, Ph0-CH2), 38.70 (2C, N-CH3), 68.82 (2C, Ph-
O-CH2), 98.85 (Ph-C4), 101.02 (2C, Ph-C2/6), 126.47 (2C, Ph0-C4),
128.46 (4C, Ph0-C2/6), 128.97 (4C, Ph0-C3/5), 138.12 (2C, Ph0-C1),
153.00 (Ph-C1), 160.01 (2C, Ph-C3/5), 187.42 (O-C-S). MS (ESIþ) =
m/e = 422.3 [M þ 1]þ.
Step III, Synthesis of 21c (S-(3,5-Diphenethoxyphenyl)dimethyl-
carbamothioate): In this step, a NewmanꢀKwart rearrangement was
done to introduce the thiol group.14 Precursor 21b (1.4 g, 3.3 mmol) was
heated to 240 ꢀC (solvent-free) and stirred for 5 h. The transparent oil
turns into a dark brown solution. Purification of the crude product (silica
gel column chromatography; solvents n-hexane and ethyl acetate)
1
yielded the pure compound as yellow oil. Yield: 43.6% (0.61 g). H
NMR (300.13 MHz, (CD3)2SO): δ = 2.87ꢀ3.01 (m, 10H, N-CH3 þ
Ph0-CH2), 4.07ꢀ4.18 (m, 4H, Ph-O-CH2), 6.23 (d, 1H, Ph-H), 6.52ꢀ
6.57 (m, 2H, Ph-H), 7.17ꢀ7.30 (m, 10H, Ph0-H). 13C NMR (75.44
MHz, (CD3)2SO): δ = 34.78 (2C, Ph0-CH3), 36.41 (N-CH3), 67.89
(2C, Ph-O-CH2), 102.08 (Ph-C4), 113.63 (2C, Ph-C2/6), 126.23 (2C,
Ph0-C4), 128.25 (4C, Ph0-C2/6), 128.90 (4C, Ph0-C3/5), 138.24 (2C,
Ph0-C1), 159.31 (2C, Ph-C3/5), 159.50 (Ph-C1), 164.63 (CdS). MS
(ESIþ) = m/e = 422.3 [M þ 1]þ
Step IIa, Synthesis of 23b (3,5-Diphenethoxybenzaldehyde): 3,
5-Dihydroxybenzaldehyde (0.88 g, 6.34 mmol, 1 equiv), 2-phenyletha-
nol (1.55 g, 12.7 mmol, 2.1 equiv), and triphenylphosphine (TPP;
3.66 g, 13.9 mmol, 2.5 equiv) were dissolved in anhydrous THF and
stirred under argon atmosphere with ice bath cooling. 1,10-(Azodicar-
bonyl)dipiperidine) (ADDP; 3.36 g, 13.3 mmol, 2.5 equiv), diluted in
5 mL of THF, was added dropwise via a syringe, and the solution was
stirred for 36 h. Subsequently, THF was evaporated, and the remaining
residue was purified by silica gel column chromatography using
n-hexane/ethyl acetate to yield the pure product as transparent oil.
Yield: 54.7% (1.2 g). 1H NMR (300.13 MHz, (CD3)2SO): δ =
3.42ꢀ3.47 (t, 3H, J = 6.8 Hz, Ph0-CH2), 4.64ꢀ4.68 (t, 3H, J = 6.8
Hz, Ph-O-CH2), 7.20ꢀ7.21 (d, 1H, J = 2.2 Hz, Ph-H), 7.45ꢀ7.46
(d, 2H, J = 5.3 Hz, Ph-H), 7.62ꢀ7.65 (m, 2H, Ph0-H), 7.66ꢀ7.75 (m, 8H,
Ph0-H), 10.29 (s, 1H, CHO). 13C NMR (75.44 MHz, (CD3)2SO): δ =
35.19 (2C, Ph0-CH2), 68.98 (2C, Ph-O-CH2), 107.74 (Ph-C4), 108.01
(2C, Ph-C2/6), 126.70 (2C, Ph0-C4), 128.71 (4C, Ph0-C2/6), 129.35 (4C,
Ph0-C3/5), 138.60 (2C, Ph0-C1), 138.64 (Ph-C1), 160.49 (2C, Ph-C3/5),
193.18 (CHO). MS (ESIþ) = m/e = 361.2 [M þ OHþ]þ.
Step IV, Synthesis of 21d (3,5-Diphenethoxybenzenethiol): For
deprotection of the thiol,14 21c (0.6 g, 1.42 mmol) was dissolved in a
mixture of THF and MeOH. After the mixture was heated at 80 ꢀC,
NaOH (1 mol/L, 10 mL) was added, and the solution was stirred for 3 h.
Purification of the product was done with silica gel column chromatog-
raphy yielding the product as milky oil in 50.2% yield (0.25 g). 1H NMR
(300.13 MHz, (CD3)2SO): δ = 3.02ꢀ3.07 (t, 4H, J = 6.7 Hz, Ph0-CH2),
4.15ꢀ4.21 (t, 4H, J = 6.8 Hz, Ph-O-CH2), 5.40 (s, 2H, -SH), 6.28ꢀ6.30
(m, 1H, Ph-H), 6.50ꢀ6.51(m, 2H, Ph-H), 7.23ꢀ7.40 (m, 10H, Ph0-
CH2). MS (ESIꢀ) = m/e = 349.1 [M ꢀ 1]ꢀ
Step V, Synthesis of 21e (Ethyl 2-[(3,5-Diphenethoxyphenyl)thio]-
hexanoate): Method A: NaH (95%, 0.02 g, 0.85 mmol); E1, 21d (0.25 g,
0.71 mmol); E2, R-bromo ethylhexanoate (0.19 g, 0.85 mmol); duration
1
of synthesis, 3 h; product, transparent oil; yield, 51.3% (0.18 g). H
NMR (300.13 MHz, (CD3)2SO): δ = 0.80ꢀ0.84 (t, 3H, J = 7.0 Hz, Bu-
CH3), 1.02ꢀ1.06 (t, 3H, J = 7.1 Hz, Et-CH3), 1.23ꢀ1.37 (m, 4H, Bu-
CH2), 1.62ꢀ1.79 (m, 2H, Bu-CH2), 2.96ꢀ3.00 (t, 4H, J = 6.8 Hz, Ph0-
CH2) 3.86ꢀ3.90 (t, 1H, J = 6.5 Hz, S-CH), 3.98ꢀ4.05 (q, 2H, J = 6.3 Hz,
Step III, Synthesis of 23c (Ethyl 2-(3,5-Diphenethoxybenzylidene)-
hexanoate): Sodium hydride (95%; 0.11 g, 4.3 mmol, 1.3 equiv) was
suspended in anhydrous THF and stirred under argon atmosphere with
4503
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
ice bath cooling. The precursor A (1.21 g, 4.3 mmol, 1.3 equiv; in 5 mL of
THF) was added to the suspension via a syringe leading to a clear
solution. After 1 h, 23b (1 g, 2.9 mmol, 1 equiv; in 5 mL of THF) was
added, and the solution was stirred at room temperature for 3 h until the
aldehyde was completely converted to 23c (TLC control). After
completion of the reaction, THF was evaporated, and the remaining
solid was diluted with ethyl acetate. The organic phase was washed with
water two times, dried over MgSO4, and evaporated. The crude product
was purified by column chromatography using n-hexane/ethyl acetate to
yield 23c as transparent oil. Yield: 89% (1.2 g).1H NMR (300.13 MHz,
(CD3)2SO): δ = 0.76ꢀ0.81 (t, 3H, J = 7.1 Hz, Bu-CH3), 1.20ꢀ1.40
(m, 5H, Et-CH3 þ Bu-CH2), 1.43ꢀ1.50 (m, 2H, Bu-CH2), 2.37ꢀ2.43
(m, 2H, Bu-CH2), 2.97ꢀ3.02 (t, 4H, J = 6.8 Hz, Ph0-CH2), 4.10ꢀ4.19
(m, 6H, Ph-O-CH2 þ O-CH2), 6.48ꢀ6.52 (dd, 3H, J = 1.7; 9.2 Hz,
Ph-H), 7.16ꢀ7.29 (m, 10H, Ph0-H), 7.45 (s, 1H, C=CH). 13C NMR
(75.44 MHz, (CD3)2SO): δ = 13.56 (Bu-CH3), 13.91 (Et-CH3), 22.01
(Bu-CH2), 26.94 (Bu-CH2), 30.91 (Bu-CH2), 34.15 (2C, Ph0-CH2),
60.40 (O-CH2), 68.26 (2C, Ph-O-CH2), 100.47 (Ph-C4), 107.61 (2C,
Ph-C2/6), 126.22 (2C, Ph0-C4), 128.22 (4C, Ph0-C2/6), 128.89 (4C, Ph0-
C3/5), 133.47 (Ph-C1), 137.49 (COO-C=CH), 138.22 (Ph-CH),
138.31 (2C, Ph0-C1), 159.54 (2C, Ph-C3/5), 167.33 (COO-). MS
(ESIþ) = m/e = 473.1 [M þ 1]þ.
(Prop-C2), 68.01 (2C, Ph-O-CH2), 98.74 (Ph-C4), 107.48 (2C, Ph-
C2/6), 126.20 (2C, Ph0-C4), 128.24 (4C, Ph0-C2/6), 128.90 (4C, Ph0-C3/5),
138.34 (2C, Ph0-C4), 141.95 (Ph-C1), 159.60 (2C, Ph-C3/5), 176.29
(COOH). MS (ESIþ) = m/e = 447.5 [M þ 1]þ.
Compound 24 (2-(3,5-Diphenethoxybenzylidene)hexanoic Acid):
1H NMR (300.13 MHz, (CD3)2SO): δ = 0.77ꢀ0.82 (t, 3H, J = 7.0 Hz,
Bu-CH3), 1.24ꢀ1.27 (m, 2H, Bu-CH2), 1.29ꢀ1.40 (m, 2H, Bu-CH2),
2.35ꢀ2.40 (m, 2H, Bu-CH2), 2.98ꢀ3.03 (t, 4H, J = 6.8 Hz, Ph0-CH2),
4.15ꢀ4.19 (t, 4H, J = 6.8 Hz, Ph-O-CH2), 6.48ꢀ6.53 (dd, 3H, J = 1.7;
9.2 Hz, Ph-H), 7.20ꢀ7.28 (m, 10H, Ph0-H), 7.44 (s, 1H, C=CH), 12.0
(s/br, 1H, COOH). 13C NMR (75.44 MHz, (CD3)2SO): δ = 13.64 (Bu-
CH3), 22.38 (Bu-CH2), 27.03 (Bu-CH2), 30.91 (Bu-CH2), 34.85 (2C,
Ph0-CH2), 68.26 (2C, Ph-O-CH2), 101.41 (Ph-C4), 107.61 (2C, Ph-C2/6),
126.25 (2C, Ph0-C4), 128.26 (4C, Ph0-C2/6), 128.92 (4C, Ph0-C3/5),
134.19 (Ph-C1), 137.25 (CH-COO), 137.47 (Ph-CdCH), 138.26 (2C,
Ph0-C4), 159.53 (2C, Ph-C3/5), 169.04 (COOH). MS (ESIꢀ) = m/e =
443.5 [M ꢀ 1]ꢀ.
Assay Systems. Materials. Pirinixic acid, arachidonic acid, calcium
ionophore A23187, DTT, fMLP, LPS, and all other fine chemicals were
from Sigma (Deisenhofen, Germany), unless stated otherwise. 13(S)-
HpODE was from Cayman Chemical (Ann Arbor, Michigan, USA);
adenosine deaminase (Ada) and HPLC solvents were from Merck
(Darmstadt, Germany).
Cells and Cell Viability Assay. Human PMNL were freshly isolated
from leukocyte concentrates obtained at the Blood Center of the
University Hospital Tuebingen (Germany). In brief, venous blood
was taken from healthy adult donors, and leukocyte concentrates were
prepared by centrifugation at 4000 ꢁ g for 20 min at 20 ꢀC. PMNL were
immediately isolated by dextran sedimentation, centrifugation on Ny-
coprep cushions (PAA Laboratories, Linz, Austria), and hypotonic lysis
of erythrocytes as described previously.34 Cells were finally resuspended
in phosphate-buffered saline pH 7.4 (PBS) containing 1 mg/mL glucose
and 1 mM CaCl2 (PGC buffer).
Human A549 cells were cultured in DMEM/high glucose (4.5 g/L)
medium supplemented with heat-inactivated fetal calf serum (FCS,
10% v/v), penicillin (100 U/mL) and streptomycin (100 μg/mL) at
37 ꢀC and 5% CO2. After 20 h, cells were transferred into DMEM/high
glucose with FCS (2% v/v), penicillin (100 U/mL), and streptomycin
(100 μg/mL), stimulated with 2 ng/mL interleukin-1β (IL-1β), and
cultured for another 24 h at 37 ꢀC. Then, confluent cells were detached
using 1 ꢁ trypsin/EDTA solution; cell viability was measured using the
colorimetric thiazolyl blue tetrazolium bromide (MTT) dye reduction
assay as described.35 In brief, A549 cells (1 ꢁ 104 cells/100 μL) in
medium supplemented with FCS (10% v/v) and IL-1β (2 ng/mL) were
plated into a 96-well microplate and incubated at 37 ꢀC and 5% CO2 with
test compounds, or solvent (DMSO) for 24 h. MTT (20 μL, 5 mg/mL) was
added, and the incubations were continued for 1 h. The formazan
product was solubilized with sodium dodecylsulfate (10%, w/v in
20 mM HCl) overnight shaking. The absorbance of each sample was
measured at 595 nm relative to that of vehicle (DMSO)-treated control
cells using a multiwell scanning spectrophotometer (Victor3 plate read-
er, PerkinElmer, Rodgau-Juegesheim, Germany).
Determination of Product Formation by 5-LO, 15-LO, and
12-LO in Cell-Based Assays. For assays of intact cells stimulated
with calcium ionophore A23187 or NaCl, 5 ꢁ 106 freshly isolated
PMNL were resuspended in 1 mL of PGC buffer. After preincubation
with the compounds for 15 min at 37 ꢀC, 5-LO product formation was
started by addition of 2.5 μM A23187 and exogenous AA as indicated or
0.3 M NaCl was supplemented 3 min before addition of AA. After 10 min
at 37 ꢀC, the reaction was stopped with 1 mL of methanol, and 30 μL of
1 N HCl, 200 ng of PGB1, and 500 μL of PBS were added. Formed 5-LO
metabolites were extracted and analyzed by HPLC as described.36
5-LO product formation is expressed as nanograms of 5-LO products
per 106 cells, which includes LTB4 and its all-trans isomers, 5(S),
Step IV, Synthesis of Compound 23d (Ethyl 2-(3,5-Diphenethoxy-
benzyl)hexanoate): For synthesis of compound 23, the exocyclic double
bond was hydrogenated by solid-phase catalysis. The precursor 23c was
diluted in ethanol, and 10% (weight) Pd/C was added to the solution.
The mixture was placed in an autoclave, gassed with H2, and stirred at
5 bar. After 24 h, hydrogenation was completed, and the product was
filtered over Celite. The solvent was evaporated, and the product 23d
1
was used without further purification (quantitative yield). H NMR
(300.13 MHz, (CD3)2SO): δ = 0.78ꢀ0.83 (t, 3H, J = 7.1 Hz, Bu-CH3),
1.01ꢀ1.06 (t, 3H, J = 7.0 Hz, Et-CH3), 1.08ꢀ1.21 (m, 4H, Bu-CH2),
1.30ꢀ1.48 (m, 2H, Bu-CH2), 2.58ꢀ2.69 (m, 3H, Ph-CH2 þ CH-
COO), 2.95ꢀ3.00 (t, 4H, J = 6.8 Hz, Ph0-CH2), 3.91ꢀ3.99 (m, 2H,
O-CH2), 4.08ꢀ4.13 (t, 4H, J = 6.8 Hz, Ph-O-CH2), 6.29 (s, 3H, Ph-H),
7.17ꢀ7.30 (m, 10H, Ph0-H). 13C NMR (75.44 MHz, (CD3)2SO): δ =
13.97 (Bu-CH3), 14.00 (Et-CH3), 21.92 (Bu-CH2), 28.92 (Bu-CH2),
31.36 (Bu-CH2), 32.65 (Prop-C3), 34.88 (Ph0-CH2), 46.54 (Prop-C2),
59.51 (O-CH2), 68.03 (2C, Ph-O-CH2), 98.97 (Ph-C4), 107.42 (2C,
Ph-C2/6), 126.18 (2C, Ph0-C4), 128.21 (4C, Ph0-C2/6), 128.89 (4C, Ph0-
C3/5), 138.35 (2C, Ph0-C1), 141.55 (Ph-C1), 159.35 (2C, Ph-C3/5),
174.66 (COO-). MS (ESIþ) = m/e = 475.3 [M þ 1]þ
Step V, Synthesis of Final Compounds 23 and 24. The correspond-
ing ester 23d (0.4 g, 0.8 mmol) or 23c (0.4 g, 0.8 mmol) was dissolved in
6 mL of THF/4 mL of MeOH and a solution of LiOH H2O (0.18 g,
3
4.2 mmol) in 3 mL of H2O was added. The mixture was stirred at 50 ꢀC
until the hydrolysis was completed (TLC control). Next, the solvent was
removed, and the residue was dissolved in water (under heating; if
necessary, low amounts of MeOH were added). The solution was
acidified with diluted hydrochloric acid. The resulting precipitate was
filtered, washed to neutrality with water, and afterward washed with
n-hexane. Purification of 23 was done by silica gel column chromatog-
raphy using n-hexane and ethyl acetate to give the pure product (23) as
transparent oil in 76.3% (0.28 g) yield. Compound 24 was purified by
recrystallization using methanol and H2O to give a white solid. Yield:
93.1% (0.35 g).
1
Compound 23 (2-(3,5-Diphenethoxybenzyl)hexanoic Acid): H
NMR (300.13 MHz, (CD3)2SO): δ = 0.79ꢀ0.84 (t, 3H, J = 7.1 Hz, Bu-
CH3), 1.14ꢀ1.23 (m, 4H, Bu-CH2), 1.37ꢀ1.48 (m, 2H, Bu-CH2),
2.48ꢀ2.73 (m, 3H, Ph-CH2 þ CH-COO), 2.96ꢀ3.01 (t, 4H, J = 6.8 Hz,
Ph0-CH2), 4.09ꢀ4.13 (t, 4H, J = 6.8 Hz, Ph-O-CH2), 6.29ꢀ6.31 (m, 3H,
Ph-H), 7.17ꢀ7.30 (m, 10H, Ph0-H). 13C NMR (75.44 MHz,
(CD3)2SO): δ = 13.79 (Bu-CH3), 22.02 (Bu-CH2), 28.86 (Bu-CH2),
31.33 (Bu-CH2), 32.67 (Prop-C3), 34.91 (2C, Ph0-CH2), 46.56
4504
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
12(S)-dihydroxy-6,10-trans-8,14-cis-eicosatetraenoic acid (5(S),12(S)-
DiHETE), and 5(S)-hydro(pero)xy-6-trans-8,11,14-cis-eicosatetraenoic
acid (5-H(p)ETE). Cysteinyl LTsC4, D4, and E4 were not detected, and
oxidation products of LTB4 were not determined. 15(S)-H(p)ETE was
analyzed as product of 15-LO and 12(S)-H(p)ETE as product of 12-LO.
To assess 5-LO product formation in cells stimulated with fMLP,
2 ꢁ 107 PMNL were resuspended in 1 mL of PGC buffer, primed with
1 μg/mL LPS for 10 min at 37 ꢀC, and 0.3 U/mL Ada was added. After
another 10 min, cells were treated with the compounds for 10 min and
1 μM fMLP was added. The reaction was stopped on ice after 5 min, and
cells were centrifuged (800 ꢁ g, 10 min). Formed LTB4 was determined
by ELISA according to the manufacturer’s protocol (Assay Designs, Ann
Arbor, MI).
Expression and Purification of Human Recombinant 5-LO
from Escherichia coli and Determination of 5-LO Activity in
Cell-Free Systems. E. coli BL21 was transformed with pT3-5LO
plasmid, recombinant 5-LO protein was expressed at 37 ꢀC, and 5-LO
was purified as described.33 Purified 5-LO was immediately used for
5-LO activity assays. For determination of 5-LO activity in corres-
ponding supernatants of PMNL homogenates, 5 ꢁ 106 freshly iso-
lated PMNL were resuspended in 1 mL of PBS containing 1 mM
EDTA and sonicated (3 ꢁ 10 s). Cell homogenates were centrifuged at
100 000 ꢁ g for 1 h at 4 ꢀC to obtain 100 000 ꢁ g supernatants (S100).
For determination of the activity of recombinant 5-LO, 0.5 μg of
partially purified 5-LO was diluted with PBS/EDTA. Aliquots of S100
or purified 5-LO (0.5 μg) were diluted with PBS/EDTA plus 1 mM
ATP, with or without 1 mM DTT, to 1 mL and preincubated with the
test compounds. After 5ꢀ10 min at 4 ꢀC, samples were prewarmed for
30 s at 37 ꢀC, and 2 mM CaCl2 and 20 μM AA, with or without 13(S)-
HpODE, were added to start 5-LO product formation. The reaction was
stopped after 10 min at 37 ꢀC by addition of 1 mL of ice-cold methanol,
and the formed metabolites were analyzed by HPLC as described for
intact cells.
Induction of mPGES-1 in A549 Cells, Isolation of Micro-
somes, and Determination of PGE2 Synthase Activity in
Microsomes of A549 Cells. Preparation of human A549 cells was
performed as described.37ꢀ39 In brief, cells (2 ꢁ 106/20 mL DMEM/
high glucose (4.5 g/L) medium containing FCS (2%, v/v)) were
incubated for 16 h at 37 ꢀC and 5% CO2. Subsequently, the culture
medium was replaced by fresh medium, IL-1β (1 ng/mL) was added,
and cells were incubated for another 72 h. Thereafter, cells were
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, 1 mM phenylmethylsulfonyl fluoride, 60 μg/mL soybean
trypsin inhibitor, 1 μg/mL leupeptin, 2.5 mM glutathione, and 250 mM
sucrose) was added, and after 15 min, cells were resuspended and
sonicated on ice (3 ꢁ 20 s). The homogenate was subjected to
differential centrifugation at 10 000 ꢁ g for 10 min and at 174 000 ꢁ
g for 1 h at 4 ꢀC. The pellet (microsomal fraction) was resuspended in
1 mL of homogenization buffer, and the protein concentration was
determined by the Coomassie protein assay. The microsomal mem-
branes were then diluted in potassium phosphate buffer (0.1 M, pH 7.4)
containing 2.5 mM glutathione (100 μL total volume), and test
compounds or vehicle (DMSO) was added. After 15 min, PGE2
formation was initiated by addition of PGH2 (20 μM, final concen-
tration). After 1 min at 4 ꢀC, the reaction was terminated with 100 μL of
stop solution (40 mM FeCl2, 80 mM citric acid, and 10 μM of
11β-PGE2), PGE2 was separated by solid-phase extraction on re-
versed-phase (RP)-C18 material using acetonitrile (200 μL) as eluent
and analyzed by RP-HPLC (30% acetonitrile aqueous þ 0.007% TFA
(v/v), Nova-Pak C18 column, 5 ꢁ 100 mm2, 4 μm particle size, flow
rate 1 mL/min) with UV detection at 195 nm. 11β-PGE2 was used as
internal standard to quantify PGE2 product formation by integration
of the area under the peaks.
Determination of COX Inhibition. Inhibition of the activities of
isolated ovine COX-1 and human recombinant COX-2 was performed
as described.10 Briefly, purified COX-1 (ovine, 50 units) or COX-2
(human recombinant, 20 units) was diluted in 1 mL of reaction mixture
containing 100 mM Tris buffer, pH 8, 5 mM glutathione, 5 μM
hemoglobin, and 100 μM EDTA at 4 ꢀC and preincubated with the
test compounds for 5 min. Samples were prewarmed for 60 s at 37 ꢀC,
and AA (5 μM for COX-1, 2 μM for COX-2) was added to start the
reaction. After 5 min at 37 ꢀC, the COX product 12(S)-hydroxy-5-cis-
8,10-trans-heptadecatrienoic acid (12-HHT) was extracted and then
analyzed by HPLC as described.25,40
Statistics. Data are expressed as mean ( SE. IC50 values were
calculated by nonlinear regression using SigmaPlot 9.0 (Systat Software
Inc., San Jose, USA) one site binding competition. Statistical evaluation
of the data was performed by one-way ANOVA followed by a Bonferroni
or TukeyꢀKramer post-hoc test for multiple comparisons, respectively.
A p value < 0.05 (/) was considered significant.
’ ASSOCIATED CONTENT
1
S
Supporting Information. Chemical syntheses, H and
b
13C NMR values of intermediates and final products, mass
spectrometry, and combustion analysis. This material is available
free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*M.S.-Z: phone þ49-069-79829339; fax þ49-69-79829332;
e-mail schubert-zsilavecz@pharmchem.uni-frankfurt.de. O.W.:
phone þ49-03641-949801; fax þ49-03641-949802; e-mail
oliver.werz@uni-jena.de.
Author Contributions
‡These authors contributed equally to this study.
’ ACKNOWLEDGMENT
The authors thank Daniela M€uller for expert technical assis-
tance and Aureliasan GmbH/Tuebingen, Germany and the
Doktor-Robert-Pfleger-Stiftung for financial support.
’ ABBREVIATIONS
AA, arachidonic acid; CLP, coactosine-like protein; COX, cyclo-
oxygenase; cPLA2, cytosolic phospholipase A2; CVD, cardiovas-
cular disease; DPPH, 1,1-diphenyl-2-picrylhydrazyl; fMLP,
N-formyl-methionyl-leucyl-phenylalanine; FLAP, 5-lipoxy-
genase activating protein; LPS, lipopolysaccharide; 5-LO, 5-
lipoxygenase; LT, leukotriene; mPGES-1, microsomal prostaglandin
E2 synthase-1; MAPK, mitogen-activated protein kinase; NSAID,
nonsteroidal anti-inflammatory drug; PG, prostaglandin; PGC
buffer, phosphate-buffered saline, pH 7.4, containing 1 mg/mL
glucose and 1 mM CaCl2; PMNL, polymorphonuclear leuko-
cyte; SAR, structureꢀactivity relationship; q.s., quantum satis,
meaning the amount that is needed
’ REFERENCES
(1) Radmark, O.; Samuelsson, B. Microsomal prostaglandin E
synthase-1 and 5-lipoxygenase: Potential drug targets in cancer. J. Intern.
Med. 2010, 268, 5–14.
(2) Peters-Golden, M.; Henderson, W. R., Jr. Leukotrienes. N. Engl J.
Med. 2007, 357, 1841–1854.
4505
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
(3) Wang, D.; DuBois, R. Eicosanoids and cancer. Nat. Rev. Cancer
2010, 10, 181–193.
(4) Rainsford, K. D. Anti-inflammatory drugs in the 21st century.
Subcell. Biochem. 2007, 42, 3–27.
(5) Celotti, F.; Laufer, S. Anti-inflammatory drugs: New multitarget
compounds to face an old problem. The dual inhibition concept.
Pharmacol. Res. 2001, 43, 429–436.
during cyclooxygenase-2-dependent prostaglandin E2 production
in the rat adjuvant-induced arthritis model. J. Immunol. 2003, 170,
4738–4744.
(21) Werz, O.; Szellas, D.; Henseler, M.; Steinhilber, D. Nonredox
5-lipoxygenase inhibitors require glutathione peroxidase for efficient
inhibition of 5-lipoxygenase activity. Mol. Pharmacol. 1998, 54,
445–451.
(6) Radmark, O.; Werz, O.; Steinhilber, D.; Samuelsson, B. 5-
Lipoxygenase: Regulation of expression and enzyme activity. Trends
Biochem. Sci. 2007, 32, 332–341.
(7) Werz, O.; Steinhilber, D. Development of 5-lipoxygenase in-
hibitors--lessons from cellular enzyme regulation. Biochem. Pharmacol.
2005, 70, 327–333.
(8) Feisst, C.; Pergola, C.; Rakonjac, M.; Rossi, A.; Koeberle, A.;
Dodt, G.; Hoffmann, M.; Hoernig, C.; Fischer, L.; Steinhilber, D.;
Franke, L.; Schneider, G.; Radmark, O.; Sautebin, L.; Werz, O. Hyper-
forin is a novel type of 5-lipoxygenase inhibitor with high efficacy in vivo.
Cell. Mol. Life Sci. 2009, 66, 2759–2771.
(9) Werz, O. 5-Lipoxygenase: Cellular biology and molecular phar-
macology. Curr. Drug Targets: Inflammation Allergy 2002, 1, 23–44.
(10) Koeberle, A.; Zettl, H.; Greiner, C.; Wurglics, M.; Schubert-
Zsilavecz, M.; Werz, O. Pirinixic acid derivatives as novel dual inhibitors
of microsomal prostaglandin E2 synthase-1 and 5-lipoxygenase. J. Med.
Chem. 2008, 51, 8068–8076.
(11) Koeberle, A.; Rossi, A.; Zettl, H.; Pergola, C.; Dehm, F.; Bauer,
J.; Greiner, C.; Reckel, S.; Hoernig, C.; Northoff, H.; Bernhard, F.;
Dotsch, V.; Sautebin, L.; Schubert-Zsilavecz, M.; Werz, O. The
molecular pharmacology and in vivo activity of 2-(4-chloro-6-(2,
3-dimethylphenylamino)pyrimidin-2-ylthio)octanoic acid (YS121), a
dual inhibitor of microsomal prostaglandin E2 synthase-1 and 5-lipox-
ygenase. J. Pharmacol. Exp. Ther. 2010, 332, 840–848.
(12) Hieke, M.; Ness, J.; Steri, R.; Dittrich, M.; Greiner, C.; Werz,
O.; Baumann, K.; Schubert-Zsilavecz, M.; Weggen, S.; Zettl, H. Design,
synthesis, and biological evaluation of a novel class of γ-secretase
modulators with PPARγ activity. J. Med. Chem. 2010, 53, 4691–4700.
(13) Bharate, S. B.; Khan, S. I.; Yunus, N. A.; Chauthe, S. K.; Jacob,
M. R.; Tekwani, B. L.; Khan, I. A.; Singh, I. P. Antiprotozoal and
antimicrobial activities of O-alkylated and formylated acylphlorogluci-
nols. Bioorg. Med. Chem. 2007, 15, 87–96.
(22) Werz, O.; Burkert, E.; Fischer, L.; Szellas, D.; Dishart, D.;
Samuelsson, B.; Radmark, O.; Steinhilber, D. 5-Lipoxygenase activa-
tion by MAPKAPK-2 and ERKs. Adv. Exp. Med. Biol. 2003, 525,
129–132.
(23) Pergola, C.; Dodt, G.; Rossi, A.; Neunhoeffer, E.; Lawrenz, B.;
Northoff, H.; Samuelsson, B.; Radmark, O.; Sautebin, L.; Werz, O. ERK-
mediated regulation of leukotriene biosynthesis by androgens: A
molecular basis for gender differences in inflammation and asthma. Proc.
Natl. Acad. Sci. U.S.A. 2008, 105, 19881–19886.
(24) Werz, O.; Burkert, E.; Samuelsson, B.; Radmark, O.; Steinhilber,
D. Activation of 5-lipoxygenase by cell stress is calcium independent in
human polymorphonuclear leukocytes. Blood 2002, 99, 1044–1052.
(25) Teixeira, M. M.; Hellewell, P. G. Effect of a 5-lipoxygenase
inhibitor, ZM 230487, on cutaneous allergic inflammation in the guinea-
pig. Br. J. Pharmacol. 1994, 111, 1205–1211.
(26) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat,
T. N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E. The Protein Data
Bank. Nucleic Acids Res. 2000, 28, 235–242.
(27) Gilbert, N. C.; Bartlett, S. G.; Waight, M. T.; Neau, D. B.;
Boeglin, W. E.; Brash, A. R.; Newcomer, M. E. The structure of human
5-lipoxygenase. Science 2011, 331, 217–219.
(28) Jegerschold, C.; Pawelzik, S. C.; Purhonen, P.; Bhakat, P.;
Gheorghe, K. R.; Gyobu, N.; Mitsuoka, K.; Morgenstern, R.; Jakobsson,
P. J.; Hebert, H. Structural basis for induced formation of the inflam-
matory mediator prostaglandin E2. Proc. Natl. Acad. Sci. U.S.A. 2008,
105, 11110–11115.
(29) Weisel, M.; Proschak, E.; Schneider, G. PocketPicker: Analysis
of ligand binding-sites with shape descriptors. Chem. Cent. J. 2007, 1
No. 7.
(30) Paul, N.; Kellenberger, E.; Bret, G.; Muller, P.; Rognan, D.
Recovering the true targets of specific ligands by virtual screening of the
protein data bank. Proteins 2004, 54, 671–680.
(14) Yoshida, Y.; Barrett, D.; Azami, H.; Morinaga, C.; Matsumoto,
S.; Matsumoto, Y.; Takasugi, H. Studies on anti-Helicobacter pylori
agents. Part 1: Benzyloxyisoquinoline derivatives. Bioorg. Med. Chem.
1999, 7, 2647–2666.
(15) Zettl, H.; Steri, R.; L€ammerhofer, M.; Schubert-Zsilavecz, M.
Discovery of a novel class of 2-mercaptohexanoic acid derivatives as
highly active PPARalpha agonists. Bioorg. Med. Chem. Lett. 2009,
19, 4421–4426.
(16) Karg, E. M.; Luderer, S.; Pergola, C.; Buhring, U.; Rossi, A.;
Northoff, H.; Sautebin, L.; Troschutz, R.; Werz, O. Structural optimiza-
tion and biological evaluation of 2-substituted 5-hydroxyindole-3-car-
boxylates as potent inhibitors of human 5-lipoxygenase. J. Med. Chem.
2009, 52, 3474–3483.
(17) Koeberle, A.; Siemoneit, U.; Buhring, U.; Northoff, H.; Laufer,
S.; Albrecht, W.; Werz, O. Licofelone suppresses prostaglandin E2
formation by interference with the inducible microsomal prostaglandin
E2 synthase-1. J. Pharmacol. Exp. Ther. 2008, 326, 975–982.
(18) Siemoneit, U.; Hofmann, B.; Kather, N.; Lamkemeyer, T.;
Madlung, J.; Franke, L.; Schneider, G.; Jauch, J.; Poeckel, D.; Werz, O.
Identification and functional analysis of cyclooxygenase-1 as a molecular
target of boswellic acids. Biochem. Pharmacol. 2008, 75, 503–513.
(19) Tateson, J. E.; Randall, R. W.; Reynolds, C. H.; Jackson, W. P.;
Bhattacherjee, P.; Salmon, J. A.; Garland, L. G. Selective inhibition of
arachidonate 5-lipoxygenase by novel acetohydroxamic acids: bio-
chemical assessment in vitro and ex vivo. Br. J. Pharmacol. 1988, 94,
528–539.
(31) Reisen, F.; Weisel, M.; Kriegl, J. M.; Schneider, G. Self-
organizing fuzzy graphs for structure-based comparison of protein
pockets. J. Proteome Res. 2010, 9, 6498–6510.
(32) Jones, G.; Willett, P.; Glen, R. C. Molecular recognition of
receptor sites using a genetic algorithm with a description of desolvation.
J. Mol. Biol. 1995, 245, 43–53.
(33) Fischer, L.; Szellas, D.; Radmark, O.; Steinhilber, D.; Werz, O.
Phosphorylation- and stimulus-dependent inhibition of cellular 5-lipox-
ygenase activity by nonredox-type inhibitors. FASEB J. 2003, 17,
949–951.
(34) Albert, D.; Zundorf, I.; Dingermann, T.; Muller, W. E.; Steinhilber,
D.; Werz, O. Hyperforin is a dual inhibitor of cyclooxygenase-1 and
5-lipoxygenase. Biochem. Pharmacol. 2002, 64, 1767–1775.
(35) Tretiakova, I.; Blaesius, D.; Maxia, L.; Wesselborg, S.; Schulze-
Osthoff, K.; Cinatl, J., Jr.; Michaelis, M.; Werz, O. Myrtucommulone
from Myrtus communis induces apoptosis in cancer cells via the
mitochondrial pathway involving caspase-9. Apoptosis 2008, 13,
119–131.
(36) Werz, O.; Szellas, D.; Steinhilber, D.; Radmark, O. Arachidonic
acid promotes phosphorylation of 5-lipoxygenase at Ser-271 by MAPK-
activated protein kinase 2 (MK2). J. Biol. Chem. 2002, 277, 14793–14800.
(37) Jakobsson, P. J.; Thoren, S.; Morgenstern, R.; Samuelsson, B.
Identification of human prostaglandin E synthase: A microsomal,
glutathione-dependent, inducible enzyme, constituting a potential novel
drug target. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 7220–7225.
(38) Werz, O.; Greiner, C.; Koeberle, A.; Hoernig, C.; George, S.;
Popescu, L.; Syha, I.; Schubert-Zsilavecz, M.; Steinhilber, D. Novel and
potent inhibitors of 5-lipoxygenase product synthesis based on the
structure of pirinixic acid. J. Med. Chem. 2008, 51, 5449–5453.
(20) Claveau, D.; Sirinyan, M.; Guay, J.; Gordon, R.; Chan, C. C.;
Bureau, Y.; Riendeau, D.; Mancini, J. A. Microsomal prostaglandin E
synthase-1 is a major terminal synthase that is selectively up-regulated
4506
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507

Journal of Medicinal Chemistry
ARTICLE
(39) Popescu, L.; Rau, O.; Bottcher, J.; Syha, Y.; Schubert-Zsilavecz, M.
Quinoline-based derivatives of pirinixic acid as dual PPAR alpha/gamma
agonists. Arch. Pharm. (Weinheim, Ger.) 2007, 340, 367–371.
(40) Kato, M.; Nishida, S.; Kitasato, H.; Sakata, N.; Kawai, S.
Cyclooxygenase-1 and cyclooxygenase-2 selectivity of non-steroidal
anti-inflammatory drugs: investigation using human peripheral mono-
cytes. J. Pharm. Pharmacol. 2001, 53, 1679–1685.
4507
dx.doi.org/10.1021/jm200092b |J. Med. Chem. 2011, 54, 4490–4507