Structure-activity relationship of triaryl propionic acid analogues on the human EP3 prostanoid receptor

Article abstract of DOI:10.1016/S0960-894X(03)00794-7

Potent and selective ligands for the human EP₃ prostanoid receptor are described. Triaryl compounds bearing an ortho-substituted propionic acid moiety were identified as potent EP₃ antagonists based on the SAR described herein. The binding affinities of key compound on all eight human prostanoid receptors is reported.

Full text of DOI:10.1016/S0960-894X(03)00794-7

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3813–3816
Structure–Activity Relationship of Triaryl Propionic Acid
Analogues on the Human EP3 Prostanoid Receptor
Michel Gallant,* Michel Belley, Marie-Claude Carriere, Anne Chateauneuf,
Danielle Denis, Nicolas Lachance, Sonia Lamontagne, Kathleen M.Metters,
Nicole Sawyer, Deborah Slipetz, Jean Francois Truchon and Marc Labelle
´
Merck Frosst Centre for Therapeutic Research, PO Box 1005, Pointe Claire-Dorval, Quebec, Canada H9R 4P8
Received 18 April 2003; revised 10 July 2003; accepted 28 July 2003
Abstract—Potent and selective ligands for the human EP3 prostanoid receptor are described.Triaryl compounds bearing an ortho-
substituted propionic acid moiety were identified as potent EP3 antagonists based on the SAR described herein.The binding affi-
nities of key compound on all eight human prostanoid receptors is reported.
# 2003 Elsevier Ltd.All rights reserved.
The lipid mediator prostaglandins are generated from
the enzymatic metabolism of free arachidonic acid
through the consecutive action of the cyclo-oxygenases
and distinct synthases.They exert their action in close
proximity to the site of their synthesis by interacting
with specific receptors.To date, eight prostanoid recep-
tors from the growing class of G-protein-coupled
report the identification of a second series of EP3
antagonists.Though structurally similar to the first ser-
ies, the triaryl series differs mainly by the nature of the
acidic moiety essential for potency.In the biaryl series,
affinity for the targeted receptor was achieved by intro-
ducing an acylsulfonamide moiety as the acidic surrogate
(e.g., 1, Fig.1 ).The corresponding carboxylic acid
2
receptors have been cloned and characterized.PGE
being 64-fold less active (Table 1) on the EP3 receptor.10
2
will bind preferentially to the EP1, EP2, EP3, and EP4
receptors, PGD2 to the DP and FP receptors, PGF2a to
the FP and EP3 receptors, PGI2 to the IP receptor and
TXA2 to the TP receptor.1,2 The EP3 receptor has been
found to play a key role in duodenal bicarbonate secre-
tion and mucosal integrity,3 platelet aggregation and
thrombosis,4 fever generation5 and PGE2-mediated
hyperalgesia.6 Thus far, PGE2 or prostanoid-like com-
pounds,7 acting as non-selective agonists and mice
lacking specific receptor (knock-out)8 have been utilized
to study the pharmacological role of the EP3 receptor.
Therefore, the search for selective EP3 antagonists has
become a critical issue in the pharmacological charac-
terization of the EP3 receptor.
Figure 1. Biaryl EP3 antagonists.
Because of the inherent instability of the acylsulfon-
amide moiety in rat plasma, though the introduction of
a double bond as in 1 alleviated part of the problem, we
initiated a search for potent carboxylic acid ligand on
the EP3 receptor.While conducting SAR on the non-
selective EP ligand 3,11 (Scheme 1) we found that the
introduction of a propenoic acid group led to the potent
EP3 selective antagonist 4.Compound 4 was further
investigated but will be the subject of a separate
communication.Introduction of a phenyl group to rigi-
dify the propenyl tether provided us our lead com-
pound, the triaryl propenoic acid 5.
We previously described a series of biaryl acylsulfon-
amides9 which behaves as full antagonists on the human
EP3 receptor.In the present communication, we wish to
*Corresponding author.Te:l+. 51-442-830-54; fax:+51-442-849-00;
e-mail: michel_gallant@merck.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd.All rights reserved.
doi:10.1016/S0960-894X(03)00794-7

3814
M. Gallant et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3813–3816
Table 1. Affinity of lead compounds on prostanoid E2 receptors
Table 3. SAR on the lipophilic residue
Ki (mM)a
Compd
EP1
EP2
EP3
EP4
1
2
3
4
5
8.1
10
0.23
12
72
10
7.5
0.15
8.1
0.83
0.025
1.6
2.3
0.007
0.084
3.7
>10
0.44
Entry
R
Ki (mM)a
EP3
1.9
1.7
EP1
34
9.5
>97
1.1
EP2
EP4
aKi determinations are averages of at least two experiments.
11
12
13
14
H
8.1
2.9
>97
4.6
0.75
6.0
1.0
21
1.1
2-CH2OPh
4-OBn
2-OBn(2,6-Cl2)
0.089
1.5
0.021
aKi determinations are averages of at least two experiments.
The nature of the triaryl scaffold was also explored.
Three disubstituted thiophenes 15, 16, and 17 were pre-
pared and were found to have very distinct affinities
(Table 4) for the EP3 receptor.Interestingly, the 2-5-dis-
ubstituted thiophene 15 was found to be 22-fold less
potent than the corresponding 2,4- (16) and 3,5- (17) dis-
ubstituted analogues.We were unable to identify sig-
nificant geometrical differences between analogues 15, 16,
and 17 as determined by gas-phase energy minimization.12
Several stable conformers were examined and were found
to be superimposable for all three analogues.Differ-
ences in the spatial orientation of the substituents on the
thiophene ring or potential electrostatic interaction with
the sulfur atom could not explain the differences in affi-
nity for these analogues based on our calculations.
Scheme 1. Identification of the lead compound 5.
As exemplified in Table 2, the acid moiety was investi-
gated first.The propanoic acid analogue 6 was found to
be equipotent to the lead compound 5.A two-carbon
tether between the acid moiety and the triaryl scaffold
was, within the limited number of analogues prepared,
optimum for affinity on the EP3 receptor (6 vs 7–9).The
corresponding acylsulfonamide 10, though more potent,
was metabolically unstable as previously disclosed.9
Following optimization of the lipophilic scaffold, we re-
examined the SAR around the acidic moiety following
the observation, in a closely related series, that sub-
stitution at the b position of the propionic chain was
allowed without loss of affinity for the EP3 receptor.
The b-methyl analogue 19 is equipotent to the non-
substituted analogue 17 (Table 5), whereas the a-methyl
18 is a poor ligand of the EP3 receptor.The introduc-
tion of a heteroatom in the linker is also allowed as
Table 2. SAR on the acidic moiety; acylsulfonamide surrogates and
tether length
Entry
R
Ki (mM)a
Table 4. SAR on the triaryl scaffold
EP1
EP2
EP3
EP4
6
7
8
9
10
(CH2)2COOH
CH2COOH
(CH2)3COOH
8.3
9.3
11.9
8.3
3.7
3.6
3.5
3.6
1.6
0.055
1.7
0.59
0.82
0.016
1.8
1.1
2.0
1.2
1.7
CH2CH(CH3)COOH
(CH)2CONHSO2Thb
2.6
aKi determinations are averages of at least two experiments.
Entry
R
Ki (mM)a
EP3
bTh, 2-thienyl.
EP1
27
EP2
EP4
1.1
The presence of an ortho-substituted lipophilic tail on
the triaryl scaffold was found to be essential for affinity
to the EP3 receptor.Removing or displacing the benzyl-
oxy group to the para position, as in 11 and 13, respec-
tively (Table 3), led to dramatic loss in affinity (>14-
fold) compared to compound 6.Increasing the lipo-
philicity by introducing two chlorine atoms as in 14,
improved the affinity for the EP3 receptor by 2-fold
over 6 but with a reduction in selectivity over the EP1
receptor.
15
16
17
4.1
0.52
17.5
17.8
2.9
2.5
0.027
0.023
0.76
0.65
aKi determinations are averages of at least two experiments.

M. Gallant et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3813–3816
3815
exemplified by the amino acetic acid 21.Ultimately, the
2-phenyl cyclopropane carboxylic acid 20 was found to
be the most potent EP3 antagonist discovered within the
scope of the SAR described herein.Consequently, we
identified the antagonist 20, with a Ki of 3 nM on the
EP3 receptor and an acceptable selectivity profile (>40-
fold) against all eight prostanoid receptors (Table 6).
The ligand 20 behaves as a full antagonist with a Kb of
5.8 nM in our EP3 functional assay.13
We have therefore identified a second generation of
ligands for the EP3 human receptor.The new series of
triaryl carboxylic acids originated from integrating
structural features of two separate series which ulti-
mately led to the identification of compound 20.
In general, triaryl propionic acids were prepared in two
consecutive palladium catalyzed cross-coupling reac-
tions.Triphenyl analogues ( Scheme 2) were synthesized
in a straight-forward manner by a Suzuki–Miura cou-
pling between a properly substituted phenylboronic acid
(22) and 1,3-dibromobenzene.The resulting biaryl bro-
mide was converted to the corresponding boronic acid
by lithium–halogen exchange at low temperature fol-
lowed by addition of triisopropyl boronate and acidic
workup.A subsequent coupling between the resulting
boronic acid 23 and 2-bromophenyl propionate 24
afforded, following deprotection, the desired triaryl
propionic acid 26.The latter can be converted to the
corresponding acylsulfonamide 27 using an EDCI/
DMAP coupling7 reaction with a sulfonamide.
Scheme 2. (a) Pd(PPh3)4, Na2CO3, DME, 80 ꢀC; (b) n-BuLi, À78 ꢀC,
B(Oi-Pr)3, À78 ꢀC to rt, then HCl; (c) Pd(PPh3)4, Na2CO3, DME,
80 ꢀC; (d) TFA/CH2Cl2 for t-Bu, LiOH/MeOH/THF for Et; (e) EDCI,
DMAP, thiophene sulfonamine, DMF.
sequence of the two coupling reactions (Scheme 3).
Compound 16, which has the phenyl bearing the acidic
moiety located at the 2 position of the thiophene ring, is
prepared in the following manner.A Suzuki–Miura cou-
pling between 2,4-dibromothiophene and the commer-
cially available 2-hydroxymethylphenyl boronic acid
gave almost exclusively (>10:1) the [2-(4-bromo-thio-
phen-2-yl)-phenyl]-methanol adduct, which was con-
verted to the corresponding benzyl bromide 28.The
regioselectivity of the coupling reaction (2- vs 4-aryl) was
determine by NOE experiments on the benzyl alcohol
intermediate.Addition of the bromide 28 to the enolate of
tert-butyl acetate, generated using a hindered base, affor-
ded the desired ester 29.Subsequent coupling to 2-ben-
zyloxyphenyl boronic acid provided, after hydrolysis, the
desired acid 16.Similarly, compound 17 was synthesized
by a Suzuki–Miura coupling between 2-benzyloxyphenyl
boronic acid and 2,4-dibromothiophene to afford the
The two types of disubstituted thiophene analogues (16
and 17) were prepared independently by inverting the
Table 5. SAR on the acidic moiety
Entry
R
Ki (mM)a
EP1
EP2
EP3
EP4
18
19
20
21
CH2CH(CH3)COOH
CH(CH3)CH2COOH
c-(C3H4)COOH
13
35
48
85
5.5
4.6
2.3
2.5
0.62
0.98
1.1
0.3
0.46
0.035
0.003
0.052
NHCH2COOH
aKi determinations are averages of at least two experiments.
Table 6. Affinity of compounds 21 on all eight prostanoid receptors
Ki (mM)a
EP1
47
EP2
2.3
EP3
EP4
0.30
DP
FP
22
IP
TP
Scheme 3. (a) Pd(PPh3)4, Na2CO3, DME, 80 ꢀC; (b) HBr, AcOH,
110 ꢀC; (c) (i) t-butyl acetate/cyclohexyl-isopropyl amine/n-BuLi,
THF, À78 ꢀC; (ii) add 28, À78 to 0 ꢀC; (d) n-BuLi, À78 ꢀC, B(Oi-Pr)3,
À78 ꢀC to rt, then HCl.
0.003
0.12
0.71
0.40
aKi determinations are averages of at least two experiments.

3816
M. Gallant et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3813–3816
References and Notes
1. Abramovitz, M.; Adam, A.; Boie, Y.; Godbout, C.;
Lamontagne, S.; Rochette, C.; Sawyer, N.; Tremblay, N. M.;
Belley, M.; Gallant, M.; Dufresne, C.; Gareau, Y.; Ruel, R.;
Juteau, H.; Labelle, M.; Ouimet, N.; Metters, K. M. Biochim.
Biophys. Acta 2000, 1483, 285.
2. Kiryiama, A.; Ushikubi, F.; Kobayashi, T.; Hirata, M.;
Sugimoto, Y.; Narumiya, S. Br. J. Pharmacol. 1997, 122, 217.
3. Takeuchi, K.; Ukawa, S.; Kato, S.; Furukawa, O.; Araki,
H.; Sugimoto, Y.; Ichikawa, A.; Ushikubi, F.; Narumiya, S.
Gastroenterology 1999, 117, 1128.
4. Fabre, J. E.; Nguyen, M. T.; Athirakul, K.; Coggins, K.;
McNeish, J. D.; Autin, S.; Parise, L. K.; FitzGerald, G. A.;
Coffman, T.M;. Koller, B.H. J. Clin. Invest. 2001, 107, 603.
5. Ushikubi, F.; Segi, E.; Sugimoto, Y.; Matsuoka, T.;
Kobayashi, T.; Hizaki, H.; Tuboi, K.; Katsuyama, M.; Ichi-
kawa, A.; Tanaka, T.; Yoshida, N.; Narumiya, S. Nature
1998, 395, 281.
6. Minami, T.; Nakano, H.; Kobayashi, T.; Sugimoto, Y.;
Ushikubi, F.; Ichikawa, A.; Narumiya, S.; Ito, S. Br. J. Phar-
macol. 2001, 133, 438.
7. Shimazaki, Y.; Kameo, K.; Tanami, T.; Tanaka, H.; Ono,
N.; Kiuchi, Y.; Okamoto, S.; Sato, F.; Ichikawa, A. Bioorg.
Med. Chem. 2000, 8, 353.
8. Ushikubi, F.; Sugimoto, Y.; Ichikawa, A.; Narumirya, S.
Jpn. J. Pharmacol. 2000, 83, 279.
Scheme 4. (a) (i) cyclohexyl-isopropyl amine /n-BuLi, THF, À78 ꢀC
(ii) MeI-78 ꢀC to rt; (b) CH2N2, Pd(OAc)2, Et2O; (c) (i) Triethyl
phosphonoacetate/NaH, THF, 0 ꢀC (ii) add XX, 0 ꢀC–rt; (d)
PhSO2NHNH2, Tol, reflux; (e) BrCH2COOEt, Net3, DMAP, EtOH,
reflux.
2-(2-benzyloxy-phenyl)-4-bromo-thiophene adduct 30.
The latter was converted to the corresponding boronic
acid 31 using previously described conditions.Finally,
the boronic acid was coupled to methyl 2-bromophenyl-
propionate to produce, after saponification, the desired
analogue 17.
9.Gallant, M;.Carrie re, M. C.; Chateauneuf, D.; Denis, D.;
Gareau, Y.; Godbout, C.; Greig, G.; Juteau, H.; Lachance,
N.; Lacombe, P.; Lamontagne, S.; Metters, K. M.; Rochette,
C.; Ruel, R.; Slipetz, D.; Sawyer, N.; Tremblay, N.; Labelle,
M. Bioorg. Med. Chem. Lett. 2002, 12, 2583.
The syntheses of the functionalized aryl ester inter-
mediates are described in Scheme 4.The a-methyl ester
32 was prepared by addition of MeI to the enolate of
t-butyl 2-bromophenylpropionate generated using an
hindered base at low temperature.The cyclopropane
ester 33 was prepared by a palladium acetate catalyzed
cyclopropanation of 2-bromo cinnamic acid using
diazomethane.The b-methyl ester analogue 35 can be
obtained in two steps from 2-bromo acetophenone.
Horner–Emmons reaction using triethyl phosphonoa-
cetate followed by reduction of the resulting alkene 34
with phenyl sulfonylhydrazide in refluxing toluene gave
the desired ester 35.Finally, the amino acid 36 was
prepared by alkylation of the corresponding aniline
with ethyl bromoacetate.
10.See ref 1 for stable expression of prostanoid receptors in
the human embryonic kidney (HEK) 293 cell line and also for
prostanoid receptor binding assays.The EP -III subtype was
3
used for the binding assay.
11. Breault, G.A.; Oldfield, J.; Tucker, H.; Warner, P. World
Patent WO9606822, 1996; Chem. Abstr. 1996, 125, 86305
12.Conformational analysis using the MMFFs force field and
DFT B3LYP with a G6-31* basis set was performed.
13.EP antagonist assays were performed in human ery-
3
throleukemia cells (ATCC HEL 92.1.7) that endogenously
express the EP3 receptor.cAMP accumulation assays were
performed in a total of 0.2 mL HBSS containing 100 mM RO-
20174 and 15 mM Forskolin. 21 (0–3Â10À6 M) was added to
the incubation mixture in DMSO (0.5% final) and pre-
incubated with the cells (2Â105 cells per incubation) for 10 min
at 37 ꢀC.The reaction was initiated with the EP selective
3
agonist sulprostone (Biomol) (0–3Â10À5 M in 0.5% DMSO),
and incubated at 37 ꢀC for10 min.The reaction was termi-
nated by boiling the samples for 3 min and cAMP was mea-
sured by cAMP SPA assay (Amersham).
Acknowledgements
The authors would like to thank Dr.Renata Oballa for
proofreading this manuscript.