Inhibitors of the inducible microsomal prostaglandin E2 synthase (mPGES-1) derived from MK-886

Article abstract of DOI:10.1016/j.bmcl.2005.05.027

A series of potent and selective inhibitors of the inducible microsomal PGE₂ synthase (mPGES-1) has been developed based on the indole FLAP inhibitor MK-886. Compounds 23 and 30 inhibit mPGES-1 with potencies in the low nanomolar range and with selectivities of at least 100-fold compared to their inhibition of mPGES-2, thromboxane synthase and binding affinity to FLAP. They also block the production of PGE₂ in cell based assays but with a decreased potency and more limited selectivity compared to the enzyme assays.

Full text of DOI:10.1016/j.bmcl.2005.05.027

Bioorganic & Medicinal Chemistry Letters 15 (2005) 3352–3355
Inhibitors of the inducible microsomal prostaglandin E2
synthase (mPGES-1) derived from MK-886
Denis Riendeau,* Renee Aspiotis, Diane Ethier, Yves Gareau, Erich L. Grimm,
´ ´ ` ´
Jocelyne Guay, Sebastien Guiral, Helene Juteau, Joseph A. Mancini, Nathalie Methot,
Joel Rubin and Richard W. Friesen
Merck Frosst Centre for Therapeutic Research, 16711 Trans Canada Hwy, Kirkland, Que., Canada H9H 3L1
Received 18 March 2005; accepted 10 May 2005
Abstract—A series of potent and selective inhibitors of the inducible microsomal PGE2 synthase (mPGES-1) has been developed
based on the indole FLAP inhibitor MK-886. Compounds 23 and 30 inhibit mPGES-1 with potencies in the low nanomolar range
and with selectivities of at least 100-fold compared to their inhibition of mPGES-2, thromboxane synthase and binding affinity to
FLAP. They also block the production of PGE2 in cell based assays but with a decreased potency and more limited selectivity com-
pared to the enzyme assays.
Ó 2005 Elsevier Ltd. All rights reserved.
An inducible form of prostaglandin E2 (PGE2) synthase
(mPGES-1) has been identified as a member of the fam-
ily of membrane-associated proteins in eicosanoid and
glutathione metabolism (MAPEG), which includes cer-
tain glutathione transferases and the 5-lipoxygenase
activating protein (FLAP).1,2 Since its discovery in
1999, mPGES-1 has been shown to be inducible in var-
ious models of pain and inflammation, where it appears
to be the predominant synthase involved in COX-2 med-
iated PGE2 production,3 both at peripheral inflamed
sites4 and in the CNS.5,6 Furthermore, mice deficient
in mPGES-1 show both a reduction in the production
hibit mPGES-1 with
a
relatively low potency
(IC50 = 10–20 lM) while several polyunsaturated fatty
acids and 15-deoxy-PGJ2 are more potent inhibitors
(IC50 = 0.3 lM).9 Since mPGES-1 is a member of the
MAPEG family which includes FLAP, the inhibitory po-
tency of several FLAP inhibitors was analyzed. The po-
tent FLAP inhibitor MK-886 (1) (FLAP IC50 = 26 nM
(n = 2)) functions as a modest inhibitor of human
mPGES-111 with an IC50 of 1.6 lM using our assay con-
ditions (Table 1, entry 1). Using MK-886 as a lead struc-
ture,12 structure–activity relationship (SAR) studies were
then undertaken with the aim of identifying more potent
and selective inhibitors of mPGES-1 (Tables 1–3).
7
of inflammatory PGE2 and a decrease in inflammatory
responses in the collagen-induced arthritis model.8 Thus,
mPGES-1 represents a potential target for the develop-
ment of novel analgesic and anti-inflammatory agents.
Herein, we report the identification and in vitro profiles
of the first potent and selective inhibitors of mPGES-1.
Replacement of the N-p-chlorobenzyl substituent with
sterically smaller (R1 = H, Me, allyl, Table 1, entries
2–4) or larger (R1 = (CH2)3Ph, Table 1, entry 5) substit-
uents resulted in a 2- to 7-fold loss in potency. There did
not appear to be a significant effect on potency by mak-
ing simple changes to the p-chloro substituent (data not
shown) so the remaining SAR was conducted with the
N-p-chlorobenzyl group. The importance of the carbox-
ylic acid function on the side chain at C2 was exempli-
fied by the dramatic loss in potency when it was
modified, either by esterification (R2 = CO2Me, Table
1, entry 6) or by replacement with a primary amide
(R2 = CO2NH2, Table 1, entry 7). The C3 position of
the indole was much more tolerant of modification.
Only a few compounds have been reported to act as
inhibitors of mPGES-1 activity and none of them are
selective mPGES-1 inhibitors. Stable analogs of the
PGH2 substrate9 and the COX-2 inhibitor NS-39810 in-
Keywords: Prostaglandin synthase; FLAP inhibitors; MK-886; Pros-
taglandin E2; Inflammation.
*
Corresponding author. Tel.: +1 514 428 2673; fax: +1 514 428
4900; e-mail: denis_riendeau@merck.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.05.027

D. Riendeau et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3352–3355
Table 1. Structural modification of MK-886
3353
Entry
Compound
R1
R2
R3
mPGES-1 Inhibition IC50, lMa
1
1 (MK-886)
2
CH2(4-Cl–Ph)
H
Me
CO2H
CO2H
CO2H
CO2H
CO2H
CO2Me
CO2NH2
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
CO2H
StBu
1.6 (515)
>10 (1)
>10 (3)
6.7 (5)
2
StBu
3
3
StBu
4
4
CH2(CH@CH2)
(CH2)3Ph
StBu
5
5
StBu
3.2 (2)
7.2 (3)
6
6
CH2(4-Cl–Ph)
CH2(4-Cl–Ph)
CH2(4-Cl–Ph)
CH2(4-Cl–Ph)
CH2(4-Cl–Ph)
CH2(4-Cl–Ph)
CH2(4-Cl–Ph)
CH2(4-Cl–Ph)
CH2(4-Cl–Ph)
StBu
7
7
StBu
Ph
>10 (1)
6.4 (3)
8
8
9
9
OPh
0.65 (2)
0.29 (3)
0.90 (3)
0.26 (2)
0.25 (2)
1.1 (3)
10
11
12
13
14
10
11
12
13
14
CH2(4-tBu–Ph)
CO(2-Me–Ph)
COCH2StBu
t
COCH2 Bu
Me
a Mean of two or more experiments, n values given in parentheses.
Table 2. SAR at C5 and C7
While direct attachment of a phenyl moiety (R3 = Ph,
Table 1, entry 8) resulted in a 4-fold loss in potency,
insertion of an oxygen atom (R3 = OPh, Table 1, entry
9) or CH2 moiety (R3 = CH2(4-tBuPh), Table 1, entry
10) between the indole core and the aromatic group gave
more potent inhibitors. A variety of C3 acyl substituents
were also tolerated (Table 1, entries 11–13), giving inhib-
itors 2- to 6-fold more potent than MK-886. Although
not optimum in terms of increasing potency relative to
the acyl substituents, the simple introduction of a methyl
group at C3 (Table 1, entry 14) provided 14, a 1 lM
inhibitor of mPGES-1. This C3 substituent was used
for further SAR studies (Table 2) since it was found that
modification at C5 had the greatest impact on potency.
Entry
Compound
R5
R7
mPGES-1 Inhibition
IC50, lMa
1
2
3
4
5
6
14
15
16
17
18
19
iPr
H
H
iPr
H
H
H
H
1.1 (3)
4.3 (2)
3.2 (2)
2.6 (3)
0.33 (2)
0.60 (5)
H
F
tBu
Ph
a Mean of two or more experiments, n values given in parentheses.
The relocation of the isopropyl group from C5 to C7
(Table 2, entry 2) or its removal (Table 2, entry 3) re-
sulted in a 3- to 4-fold loss in potency compared to 14.
While introduction of a fluorine atom at C5 (Table 2,
entry 4) was not advantageous, sterically larger substit-
uents (Table 2, entries 5 and 6) led to a moderate po-
tency enhancement.
Table 3. SAR of 5-phenyl indoles
Entry Compound
X
Y
mPGES-1 Inhibition
IC50, lMa
One of the most interesting avenues of SAR that was
pursued, and that lead to a dramatic increase in inhibi-
tor potency, derived from functionalization of the C5
phenyl moiety of inhibitor 19 (Table 3). In particular,
the biphenyls 20 (Table 3, entry 2) and 21 (Table 3, entry
3) exhibited potency enhancements of approximately 4-
and 40-fold, respectively, compared to 19. Extensive
SAR around the biphenyl moiety of 21 provided a num-
ber of potent inhibitors. The key modification appeared
to involve appropriate ortho substitution to the biphe-
nyl bond. For example, introduction of a fluorine atom
at this position provided 23 (Table 3, entry 5), a 7 nM
inhibitor of mPGES-1.
1
19
20
21
22
23
24
25
26
27
28
29
30
31
32
H
H
0.60 (5)
0.16 (3)
2
Ph
H
H
Ph
3
0.016 (6)
0.022 (2)
0.007 (9)
0.032 (2)
0.012 (2)
0.005 (2)
0.004 (2)
0.008 (2)
4
Cl Ph
5
F
F
F
F
F
F
F
F
F
F
Ph
pyrzb
pyrb
6
7
8
2-MeO–Ph
2-Cl–Ph
2-F–Ph
9
10
11
12
13
14
2-MeCO–Ph 0.006 (2)
2-Me–Ph
3-Me–Ph
4-Me–Ph
0.003 (8)
0.033 (2)
0.031 (2)
a Mean of two or more experiments, n values given in parentheses.
b pyrz = 5-(1,3-pyrazinyl); pyr = 3-pyridinyl.
Although the terminal phenyl group could be replaced
with heterocycles (Table 3, entries 6 and 7) with only

3354
D. Riendeau et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3352–3355
moderate loss in potency, additional ortho substitution
to the biphenyl bond on the terminal phenyl group
yielded the best inhibitors in this series (Table 3, entries
8–12). The effect of this double ortho substitution is
exemplified in the last three entries in Table 3. While
the ortho methyl group in 30 (Table 3, entry 12) pro-
vided the most potent inhibitor in the series, 3 nM,
the corresponding meta and para substituted analogs
(Table 3, entries 13 and 14) were 10-fold less potent.
Thus, the biaryl indole 30 is the optimized compound
in this series in terms of in vitro potency and is a
500-fold more potent inhibitor of mPGES-1 than
MK-886.
The syntheses of 23 and 30 (Scheme 1) exemplify the
general route that was used to prepare the biaryl indoles
described in Table 3. The methoxy indole 35 was pre-
pared in 54% by the Fisher indole coupling of N-4-chlo-
robenzyl-N-4-methoxyphenyl hydrazine 33 and 2,2-
dimethyl-4-oxohexanoic acid 34.12 Demethylation, fol-
lowed by triflation converted 35 into the triflate 37.
The biaryl boronates 38a,b were prepared by sequential
palladium-catalyzed coupling reactions of commercially
available 4-bromo-2-fluoro-1-iodobenzene with phen-
ylboronic acid or 2-tolylboronic acid and then bis(pina-
colato)diboron.13 Suzuki–Miyaura palladium-catalyzed
cross-coupling14 of 37 and 38a,b provided the indoles
39a,b. Saponification yielded the desired indole inhibi-
tors 23 and 30.
The inhibitors 23 and 30 were further characterized in
various enzyme assays (Table 4) and were found to be
at least 100-fold less potent inhibitors of recombinant
human mPGES-215 (IC50 > 1 lM) and TXA2 synthase
(IC50s = 0.95–1.4 lM). Interestingly, both inhibitors
showed a similar high degree of selectivity when tested
in the FLAP binding assay (IC50 > 1 lM), being
approximately 40-fold less potent than the original
lead MK-886 (IC50 = 26 nM) in the same assay. The
inhibitors were also tested for their effects on prosta-
noid formation in human A549 epithelial lung carci-
noma cells and primary synoviocytes from
rheumatoid arthritis patients, cells which have been re-
ported to express mPGES-1.10,16,17 When incubated
with the cells at low serum concentration (2% FBS),
23 and 30 inhibited the production of PGE2 by IL-
1b-stimulated human synoviocytes with IC50s of 0.61
and 0.25 lM, respectively. Similar data were obtained
for the inhibition of PGE2 using IL-1b-stimulated
A549 cells (IC50s of 0.3–0.5 lM) (Table 4). In A549
cells, 23 and 30 were 4- to 8-fold less potent at inhib-
iting the production of PGF2a (IC50s of 2.0 and
2.4 lM, respectively) compared to PGE2. At interme-
diate concentrations, a stimulation of PGF2a produc-
tion was observed, which could be due to shunting
of the PGH2 substrate from the PGE2 to the PGF2a
biosynthetic pathway.
Scheme 1. Reagents and conditions: (a) Et3N, PhMe, reflux, 15 h; HCl
t
(52%); (b) BuOH, reflux, 15 h (54%); (c) BBr3, CH2Cl2, À60 to 0 °C;
MeOH (100%); (d) Tf2O, Et3N, CH2Cl2, 0 °C (81%); (e) for 38a:
PdCl2(dppf) (cat.), aq Na2CO3, DMF, 80 °C, 2 h (40%); for 38b:
PdOAc2 (cat.), PPh3 (cat.), aq Na2CO3, DME, reflux, 2 h (99%); (f)
LiOH, THF, MeOH, 60 °C, 4 h (82–97%).
binding. For example, the potency of 30 is shifted
from an IC50 of 3 nM in the enzyme assay to values
of 0.27 and 5.8 lM in the A549 assay in the presence
of 2% and 50% FBS, respectively. In addition, no
inhibitory effects on PGE2 could be detected in LPS-
stimulated human whole blood. Although 23 and 30
are potent and highly selective inhibitors of mPGES-
1 activity in enzyme assays, the large shifts in potency
caused by the presence of plasma protein, and their
limited selectivity in cell-based assays, preclude the
use of these compounds for studying the in vivo effect
of selective mPGES-1 inhibition in models of
inflammation.
The results of the present study demonstrate that the
structure of MK-866 can be modified to yield potent
in vitro inhibitors of mPGES-1, with selectivity >100-
fold compared to FLAP binding affinity and inhibition
of both mPGES-2 and thromboxane synthase. How-
ever, to provide information regarding the effect of selec-
tive mPGES-1 inhibition in vivo, additional studies are
required to identify those analogs with improved po-
tency and selectivity in cell-based assays in the presence
of plasma protein.
As can be seen by the data summarized in Table 4,
this class of inhibitor is also characterized by a shift
in potency from enzyme assays at low protein concen-
trations to whole cell assays in the presence of plasma
proteins, presumably due to a high degree of protein

D. Riendeau et al. / Bioorg. Med. Chem. Lett. 15 (2005) 3352–3355
3355
Table 4. Summary of the potency and selectivity of inhibitors 23 and 30 in various in vitro assays
Enzyme or cell assay
Inhibition potency IC50, lMa
23
30
Human mPGES-1
Human mPGES-2
0.007 (9)
>1 (3)
1.4 (3)
0.003 (8)
>1 (3)
Thromboxane synthase
FLAP binding
0.95 (3)
>1 (3)
>1 (3)
Human synoviocytes (PGE2) (2% FBS)
A549 cells (PGE2) (2% FBS)
A549 cells (PGE2) (50% FBS)
A549 cells (PGF2a) (2% FBS)
0.61 (4)
0.49 (49)
8.0 (49)
2.0 (49)
0.25 (3)
0.27 (3)
5.8 (3)
2.4 (3)
a Mean of three or more experiments, n values given in parentheses.
The cells were then incubated with 10 ng/mL IL-1b for
24 h in the presence of a vehicle or inhibitor from a 100-
fold concentrated DMSO stock solution in a medium
containing 2% or 50% FBS. Levels of PGE2 and PGF2a in
the cell-free medium were measured by EIA (Assay
Designs, MI). IC50 values were derived from eight point
titrations.
Acknowledgments
We acknowledge the contribution of A. Zhao and I.
Smalera (Merck & Co., Inc., Rahway, NJ) for providing
the data on the FLAP binding assay, J. Wang and F.
Masse for the cloning and assay of TXA2 synthase,
and D. Claveau and M. Salem for the human whole
blood assay.
´
11. Mancini, J. A.; Blood, K.; Guay, J.; Gordon, R.;
Claveau, D.; Chan, C. C.; Riendeau, D. J. Biol. Chem.
2001, 276, 4469. The activity of mPGES-1 was tested as
described in this reference using 5–10 lg protein/mL of
References and notes
a
microsomal preparation of recombinant human
mPGES-1 expressed in CHO cells, 1 lM PGH2, and
an incubation time of 0.5 min. The reaction was
terminated with 0.1 N HCl and stannous chloride.
PGE2 accumulation in the media was measured by
EIA (Assay Designs, MI). Inhibitors were pre-incubated
with the recombinant enzyme for 15 min before initiat-
ing the reaction. IC50 values were derived from eight
point titrations.
1. Jakobsson, P. J.; Thoren, S.; Morgenstern, R.; Samuels-
son, B. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 7220.
2. Murakami, M.; Kudo, I. Prog. Lipid Res. 2004, 43, 3.
3. Murakami, M.; Naraba, H.; Tanioka, T.; Semmyo, N.;
Nakatani, Y.; Kojima, F.; Ikeda, T.; Fueki, M.; Ueno, A.;
Oh, S.; Kudo, I. J. Biol. Chem. 2000, 275, 32783.
4. Claveau, D.; Sirinyan, M.; Guay, J.; Gordon, R.; Chan, C.
C.; Bureau, Y.; Riendeau, D.; Mancini, J. A. J. Immunol.
2003, 170, 4738.
5. Inoue, W.; Matsumura, K.; Yamagata, K.; Takemiya, T.;
Shiraki, T.; Kobayashi, S. Neurosci. Res. 2002, 44, 51.
6. Engblom, D.; Ek, M.; Andersson, I. M.; Saha, S.;
Dahlstrom, M.; Jakobsson, P. J.; Ericsson-Dahlstrand,
A.; Blomqvist, A. J. Comp. Neurol. 2002, 452, 205.
7. Uematsu, S.; Matsumoto, M.; Takeda, K.; Akira, S.
J. Immunol. 2002, 168, 5811.
12. (a) Gillard, J. W.; Morton, H. E.; Fortin, R.; Guindon,
Y. U.S. Patent 5,081,138, 1992; (b) Hutchinson, J. H.;
Riendeau, D.; Brideau, C.; Chan, C.; Falgueyret, J.-P.;
´
Guay, J.; Jones, T. R.; Lepine, C.; Macdonald, D.;
McFarlane, C. S.; Piechuta, H.; Scheigetz, J.; Tagari,
´
P.; Therien, M.; Girard, Y. J. Med. Chem. 1994, 37,
1153.
13. Ishiyama, T.; Murata, M.; Miyaura, N. J. Org. Chem.
1995, 60, 7508.
14. (a) Suzuki, A. J. Organomet. Chem. 2002, 653, 83; (b)
Miyaura, N. Top. Curr. Chem. 2002, 219, 5.
15. Human mPGES-2 (AA88-377) was expressed in bacteria
and activity was assayed as described in Tanikawa, N.;
Ohmiya, Y.; Ohkubo, H.; Hashimoto, K.; Kangawa, K.;
Kojima, M.; Ito, S.; Watanabe, K. Biochem. Biophys. Res.
Commun. 2002, 291, 884.
16. Stichtenoth, D. O.; Thoren, S.; Bian, H.; Peters-Golden,
M.; Jakobsson, P. J.; Crofford, L. J. J. Immunol. 2001,
167, 469.
8. Trebino, C. E.; Stock, J. L.; Gibbons, C. P.; Naiman, B.
M.; Wachtmann, T. S.; Umland, J. P.; Pandher, K.;
Lapointe, J. M.; Saha, S.; Roach, M. L.; Carter, D.;
Thomas, N. A.; Durtschi, B. A.; McNeish, J. D.; Hambor,
J. E.; Jakobsson, P. J.; Carty, T. J.; Perez, J. R.; Audoly,
L. P. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 9044.
9. Quraishi, O.; Mancini, J. A.; Riendeau, D. Biochem.
Pharmacol. 2002, 63, 1183.
10. Thoren, S.; Jakobsson, P. J. Eur. J. Biochem. 2000, 267,
6428. A549 cells were seeded in 96-well dishes at a density
of 25,000 cells/well and incubated for 20 h in RPMI 1640
medium supplemented with 2% fetal bovine serum (FBS).
17. Kojima, F.; Naraba, H.; Sasaki, Y.; Okamoto, R.;
Koshino, T.; Kawai, S. J. Rheumatol. 2002, 29, 1836.