Synthesis, pharmacophore modeling and in vitro activity of 10,11-dihydrodibenzo[b,f]oxepine-4-carboxamide derivatives as novel and potent antagonists of the prostaglandin EP4 receptor

Article abstract of DOI:10.1016/j.bmcl.2011.08.102

The construction of a EP₄ antagonists pharmacophore model and the discovery of a highly potent oxepinic series of EP₄ antagonists is discussed. Compound 1a exhibits an excellent selectivity profile toward EP₂ receptor subtype and low cytochrome P450 inhibition potential.

Full text of DOI:10.1016/j.bmcl.2011.08.102

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6336–6340
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis, pharmacophore modeling and in vitro activity
of 10,11-dihydrodibenzo[b,f]oxepine-4-carboxamide derivatives as novel
and potent antagonists of the prostaglandin EP4 receptor
Luigi Piero Stasi a, , Kanji Bhimani d, Manuela Borriello a, Luca Canciani c, Gianfranco Caselli b,
⇑
Fabrizio Colace a, Cristian Ferioli b, Mehul Kaswala d, Laura Mennuni b, Tiziana Piepoli b, Sabrina Pucci a,
Matteo Salvi a, Vikas Shirsath d, Tiziano Zanelli b, Silvia Zerbi b
a Rottapharm Madaus, Medicinal Chemistry Department, via Valosa di Sopra 9, Monza 20900, Italy
b Rottapharm Madaus, Pharmacology & Toxicology Department, via Valosa di Sopra 9, Monza 20900, Italy
c Rottapharm Madaus, Translational Sciences & Pharmacokinetics Department, via Valosa di Sopra 9, Monza 20900, Italy
d Oxygen Bio Research Pvt Ltd, Economic Zone, Pharmez, Matoda, Sanand, Ahmedabad 382 110, Gujarat, India
a r t i c l e i n f o
a b s t r a c t
Article history:
The construction of a EP4 antagonists pharmacophore model and the discovery of a highly potent oxepinic
series of EP4 antagonists is discussed. Compound 1a exhibits an excellent selectivity profile toward EP2
receptor subtype and low cytochrome P450 inhibition potential.
Received 28 July 2011
Revised 24 August 2011
Accepted 25 August 2011
Available online 1 September 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
EP4 antagonists
3D-QSAR model
Pharmacophore modeling
Prostaglandins
PGE2
Prostaglandin E2 (PGE2), an important pro-inflammatory mes-
senger, is a lipid derived from the arachidonic acid metabolism
and elicits its biological functions through four EP subtypes recep-
tors, EP1–4. Inhibition of PGE2 production by NSAIDs and COX-2
inhibitors (Coxibs) relieves arthritis pain symptoms and thus is
the basis of widespread uses of these drugs as analgesics.1 How-
ever, the therapeutic use of these drugs is limited by their potential
to cause either gastro-intestinal (GI) toxicity2 (NSAIDs) or cardio-
vascular (CV) side effects3 (both NSAIDs and Coxibs). The CV ad-
verse events associated with these drugs are probably linked to
the inhibition of prostacyclin (PGI2) and thromboxane A2 (TxA2)
biosynthesis which may cause prothrombotic and hypertensive
effects.4
Several pieces of evidence have been accumulated in literature
regarding the potential utility of antagonizing the PGE2 pro-
inflammatory effect by selectively blocking the EP4 receptors.
McCoy et al. reported, in a mouse model of collagen-antibody in-
duced arthritis (CIA), that the EP4À/À mice showed a good resis-
tance in both the incidences and symptom scores (paw swelling/
redness, ankylosis) compared to the wild type control, while the
EP1–3À/À mice had no effect, suggesting that the effect of PGE2
in inflammation was mediated predominantly by the EP4 recep-
tor.5 Lin et al. demonstrated that EP4, not EP1–3, contributed to
inflammatory pain hypersensitivity in rats.6 Recently a research
group of Merck Frosst laboratories,7 using highly selective EP1,
EP3 and EP4 antagonists, demonstrated that EP4, not EP1 or EP3,
was the primary receptor involved in joint inflammation and pain
in rodent models of rheumatoid arthritis, further supporting EP4
antagonism as a valid strategy for treating inflammatory pain. In
addition, there is evidence that a highly selective EP4 antagonist
can be a safer alternative in relieving arthritis symptoms without
causing potential GI side effects observed with NSAIDs and COX-
2 inhibitors. Takeuchi et al. reported that the highly selective EP4
antagonist CJ-42794 does not cause any damage in the normal
rat gastrointestinal mucosa and in the arthritic rat stomach.8 Also
a safer CV safety profile could be envisaged with the use of selec-
tive EP4 antagonists since they do not interfere with biosynthesis
of any prostanoids including prostacyclin and thromboxanes. In
addition to its role in inflammation, the EP4 receptor has also been
implicated in destabilizing atherosclerotic plaques9 in human, in
developing tumor metastasis,10 in migraine headache11 and in Alz-
heimer’s disease.12 Therefore, EP4 antagonism represents a poten-
tial promising new therapeutic approach for several diseases and
⇑
Corresponding author. Tel.: +39 039 7390 630; fax: +39 039 7390 673.
E-mail address: luigi.stasi@rottapharm.com (L.P. Stasi).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.08.102

L. P. Stasi et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6336–6340
6337
O
N
O
O
Cl
N
H
N
H
CO2H
CO2H
F
CF3
MF-766
CJ-42794
O
O
O
O
O
NH
H
N
N
S
N
O
O
O
O
OH
ONO-AE3-208
MF-498
Figure 1. Representative examples of EP4 antagonists.
Figure 2. EP4 pharmacophore points: green spheres are hydrophobic features (H),
pink sphere is H-bond acceptor (A), red sphere is an anionic feature (N), golden
rings are aromatic moieties (R), purple dotted lines are distances within features
and are expressed in Å.
has been the subject of recent research interests. Examples of po-
tent EP4 antagonists belonging to different chemical classes have
been recently reported in the literature as the phenoxy derivative
CJ-42794,13 the phenylic derivative ONO-AE3-208,14 the indole
derivative MF-76615 and the acylsulphonamide derivative MF-
49816 (Fig. 1).
Table 1
Statistical 3D-QSAR parameters: SD is standard deviation of regression, R2 is
coefficient of determination for training set, F is Fisher coefficient and express the
overall significance of the model, RMSE is root mean square error in prediction, Q2
coefficient of determination for test set and P is Pearson correlation coefficient
In order to screen in silico our ideas and prioritize synthetic ef-
fort to quickly identify a suitable chemical starting point for our
exploration we decided to set up an EP4 ligands pharmacophore.
For this purpose a dataset of 179 EP4 antagonists was assembled
from literature data17 with measured binding affinities (Ki values).
All dataset compounds were drawn with ISIS/Draw v 2.5 SP418 and
their 3D structures were minimized with the LIGPREP v2.3 program19
using the OPLS_2005 force field at pH 7.0 with the rest of the
parameters as default. All molecular descriptors of dataset com-
pounds were calculated with Dragon 5.5.20 A multivariate analysis
by Principal Component Analysis (PCA) was calculated with Simca-
P+ v12.0.1.21 An onion design with evenly occupied layers of equal
thickness was therefore chosen as an appropriate method to select
a diverse and representative subset out of the whole compounds
pool. We used D-optimal onion design22 as implemented in the
MODDE software.21 Only the first two PCA components were cho-
sen as the basis design, with 10 layers plus the central points, to
identify an appropriate representative set to establish the 3DQSAR
model. A set of 33 compounds was selected with this methods. The
PHASE v3.1211 program19 was employed to carry out the calcula-
tions using the Atom-Based 3D-QSAR panel.23 For each ligand a
conformational analysis was carried out using OPLS_2005 force
field at pH 7.0 with a distance dependent dielectrics model as sol-
vation treatment, setting at 200 the maximum number of confor-
mations per structure, constraining the amide bound to trans and
using a maximum atom deviation of 2 Å to eliminate redundant
conformers. We used an activity threshold cut off of pKi >9 for ac-
tive compound and pKi <5 for inactive compounds to find the a
common pharmacophores variant among actives versus inactives.
The best model obtained is composed of six pharmacophoric fea-
tures (Fig. 2). Two hydrophobic interactions, together with an H
bond acceptor and an acid moiety, were found to be the main
contributors of the binding energy.
Hypothesis
AHHNRR
Factors
SD
R2
F
RMSE
Q2
P
1
2
3
0.632
0.349
0.250
0.802
0.944
0.973
60.6
0.881
0.813
0.789
0.524
0.595
0.619
0.733
0.811
0.825
117.2
156.1
Satisfactory statistical parameters were obtained for the best
hypothesis identified as summarized in Table 1.
Figure 3. Fitting of compounds CJ-42794 and MF-498 on EP4 pharmacophore.
CJ-42794 and MF-498 fitting in our EP4 receptor pharmacophore
model is reported in Figure 3.
point to design novel EP4 antagonists series by constraining the
aromatic rings A and B, highlighted in pink circles in Figure 4, to
screen in our pharmacophore model.
Considering its good in vitro and in vivo activity Pfizer’s phen-
oxy derivative CJ-4279413 was selected as an interesting starting

6338
L. P. Stasi et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6336–6340
O
O
Compound 1d, as an example of a novel 10,11-dihydrodibenzo
[b,f]oxepine-4-carboxamide series, emerged as a potential hit to
prepare after the pharmacophore screening of the CJ-41794 con-
strained structures (Fig. 5).
N
H
A
C
CO2H
B
This new oxepinc scaffold fitted very well into our pharmaco-
phore key features as exemplified by the overlay of compound
1d (Fig. 6). Interestingly compound 1d overlays on the EP4 pharma-
cophore in a slightly different way with respect to CJ-42794, show-
ing that the key interactions were due to the hydrophobic aromatic
ring A and to the substituent position of ring B.
CF3
Figure 5. Example of 10,11-dihydrodibenzo[b,f]oxepine-4-carboxamide series,
compound 1d.
Recently an interesting paper published from a Merck Frosst
group on benzyl indoline series24 showed that the intrinsic activity
at the h-EP4 receptor (agonism vs antagonism) was easily modu-
lated by minor modifications, as the relative CF3 group position
on the benzylic group (Fig. 7). Based on this finding, considering
the non-convergent synthetic approach of our oxepinic left hand
side, we decided to modulate the substitution pattern on oxepinic
ring B since the beginning to quickly gain crucial information on
the functional activity of this new series. We also focused our right
hand side exploration by modulating the benzyl amine substitu-
tion pattern (methyl vs cyclopropyl) while maintaining unaltered
the aromatic ring C moiety.
Synthesis of the key intermediate 10,11-dihydrodibenzo
[b,f]oxepine-4-carboxylic acids 12a,b with CF3 group both on the
positions 2 (meta vs the oxygen linker) and 3 (para vs the oxygen
linker) of ring B of oxepinic core was accomplished as shown in
Scheme 1. Starting from the commercially available acid 2 we ob-
tained its ester derivatives 3 which were then converted to the
phenoxy ethers 4a,b by using the corresponding meta (a) and para
(b) CF3-phenylboronic acid by mean of copper catalyzed coupling
reactions. Compounds 4a,b were then reduced to the correspond-
ing alcohol derivatives 5a,b which were eventually transformed
in their bromine derivatives 6a,b. Cyanide derivatives 7a,b, ob-
tained upon nucleophilic displacement of the corresponding ben-
zylic bromines with NaCN, were hydrolyzed to carboxylic acids
8a,b. The tricyclic oxepinones 9a,b were obtained after intramolec-
ular Friedel–Crafts reactions of 8a,b. Reduction of benzylic ketone
moieties from intermediates 9a,b yielded the oxepines 10a,b
which were selectively oxidized in their 6-methyl position to the
desired acids 12a,b25 by using a two steps protocol. Although sev-
eral oxidizing protocols were tried to transform 10a,b to 12a,b only
the reported two steps procedure gave the desired products even if
with very low yields; considering that we were principally inter-
ested in testing our hypothesis the overall chemistry pathway
was not further optimized.
Figure 6. Fitting of compound 1d on EP4 pharmacophore.
O
O
N
N
H
H
CO2H
N
CO2H
N
F3C
CF3
1f
EP4 agonist
1e
EP4 antagonist
Figure 7. Example of EP4 antagonism-agonism switch in indoline series.
in the antagonist mode, by inhibition of PGE2-induced cAMP eleva-
tion assay and in the agonist mode, by potentiating the cAMP
28
elevation in the absence of PGE2 (Table 2).
Standard amide coupling of acids 12a,b with commercially
available (S)-methyl 4-(1-aminoethyl)benzoate hydrochloride 13
or cyclopropyl analog 1424 in presence of HCTU and DIPEA, fol-
lowed by hydrolysis of the resultant methyl esters, gave the final
oxepinic compounds 1a–d26 (Scheme 2).
As shown in Table 2, compounds 1a–d were found to be very po-
tent ligands at the human recombinant EP4 receptor (h-EP4) within
excellent IC50s values. In particular compound 1a achieved a five-
fold increase of affinity versus the h-EP4 receptor with respect to
the competitor’s molecule CJ-42794. They also behave as full antag-
onists in the EP4 functional assay ranging from moderate to excel-
lent antagonistic potency. Compounds 1a and 1b, sharing meta
substitution on the bottom phenyl ring, showed the highest affinity
versus the EP4 receptor behaving however differently at the func-
tional assay: compound 1a was found to be an excellent EP4 antag-
onist, with a similar f-IC50 with respect to the phenoxy derivative
CJ-42794, while compound 1b showed only a modest antagonistic
activity against this receptor. This functional data shows that the
meta-CF3 sub-series is very sensitive to the variation of the benzylic
substitution pattern of the RHS aromatic ring. In particular it seems
that the cyclopropyl group alters the molecular features necessary
for a good antagonistic activity. Surprisingly compounds 1c–d (bot-
tom para-CF3 phenyl sub-series) showed a good antagonistic activ-
ity at the EP4 receptor despite the substitution pattern of the benzyl
amine RHS moiety. Our pharmacophore model, being constructed
Affinity at EP4 and EP2 human receptors of compounds 1a–d
27
was determined using a radioligand binding assay [3H]-PGE2 on
membrane cell of HEK293 cell lines over-expressing the human
prostanoid receptors EP4 or EP2; intrinsic activity was evaluated,
O
Cl
N
A
C
H
O
CO2H
constrain
introduction
B
F
CJ-42794
Figure 4. Design of novel constrained series from CJ-42794.

L. P. Stasi et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6336–6340
6339
HO
COOC2H5
OH
OH
R1
R2
R1
R2
O
O
COOH
COOC2H5
i)
ii)
iii)
2
3
4a,b
5a,b
iv)
Br
HOOC
NC
O
R1
R2
O
R
R1
R2
O
1
vi)
v)
R2
8a,b
vii)
7a,b
6a,b
OHC
O
R1
R2
O
R1
R2
R1
R2
O
viii)
ix)
O
9a,b
10a,b
11a,b
x)
HOOC
O
R1
R2
12a,b
Scheme 1. Synthesis of the key 10,11-dihydrodibenzo[b,f]oxepine-4-carboxamides 12a (R1 = H, R2 = CF3) and 12b (R1 = CF3, R2 = H). Reagents and conditions: (i) DCM, SOCl2,
DMF, reflux, 73% yield; (iia) (3-(trifluoromethyl)phenyl)boronic acid, TEA, DCM, Cu(OAc)2, rt, 26% yield (4a); (iib) (4-(trifluoromethyl)phenyl)boronic acid, TEA, DCM,
Cu(OAc)2, rt, 34% yield (4b); (iii) THF, LiAlH4, 0 °C, 99% yield (5a), 89% yield (5b); (iv) DCM, PBr3, À10 °C to 10 °C, 83%yield (6a), 72% yield (6b); (v) DMSO/H2O, NaCN, DABCO,
80 °C, 84% yield (7a), 87% yield (7b); (vi) H2O, KOH, 100 °C, 99% yield (8a), 100% yield (8b); (vii) DCM, triflic acid, trifluoro acetic anhydride, 10 °C, 32% yield (9a), 38% yield
(9b); (viii) TFA, triethyl silane, DCM, from À10 °C to rt, 58% yield (10a), 70% yield (10b); (ix) AcOH 50%, ACN, 80 °C, 24% yield (11a), 31% yield (11b); (x) H2O, KMnO4, 70 °C,
10% yield (12a), 13% yield (12b).
O
O
Table 2
(a) IC50 (nM) values in [3H]-PGE2 competition assays (b) and inhibition of PGE2-
induced cAMP accumulation (c) are means from at least three independent
experiments
N
H
H2N
HCl
CO2H
CO2Me
13
Entry
R1
R2
Binding affinitya
Functional assaya,c
EP4
O
O
R1
i) ii)
b
b
EP4
EP2
R2
1a, c
OH
CJ-42794
10
2
4
63
27
>1000
>1000
>1000
>1000
>1000
13
20
1000
42
34
1a
1b
1c
1d
CF3
CF3
H
H
H
CF3
CF3
i) ii)
R1
O
O
R2
H2N
HCl
H
N
H
CO2Me
12a-b
14
CO2H
R1
could be the origin of this subtle difference. Furthermore com-
pounds 1a–d highlighted a good selectivity versus the EP2 receptor
when tested in recombinant human EP2 binding assay (Table 2)
thus proving that our constrained oxepinic core had no negative im-
pact on the selectivity against this promiscuous PGE2 receptor
target.
Compounds 1a–d were further characterized in terms of preli-
minary ADME characteristic showing a low inhibition potential
against the major human Cytochrome P450s (CYP450) as reported
in Table 3, thus placing this novel chemical class in a low drug to
drug interactions risk area.
R2
1b, d
Scheme 2. Synthesis of the 10,11-dihydrodibenzo[b,f]oxepine-4-amides 1a–d: 1a
(R1 = CF3, R2 = H), 1b (R1 = CF3, R2 = H), 1c (R1 = H, R2 = CF3), 1d (R1 = H, R2 = CF3).
Reagents and conditions: (i) HCTU, DIPEA; DMF, 28–94% yield; (ii) LiOH, H2O,
dioxane, 33–86% yield.
on affinity and not activity data, however was not able to predict
these results. Interestingly both sub-series did not show any ago-
nistic activity at the h-EP4 receptor regardless of the CF3 group po-
sition on the oxepine B ring thus behaving differently from the
Merck’s indoline compounds;24 probably a different binding mode
of our ligands with respect to the molecules reported in literature
In conclusion, starting from the phenoxy compound CJ-42794
and by using a pharmacophore model approach, we were able,
with a limited chemical exploration, to discover a novel series of
potent EP4 antagonists. All compounds prepared did not show

6340
L. P. Stasi et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6336–6340
11. Maubach, K. A.; Davis, R. J.; Clark, D. E.; Fenton, G.; Lockey, P. M.; Clark, K. L.;
Table 3
Oxford, A. W.; Hagan, R. M.; Routledge, C.; Coleman, R. A. Br. J. Pharmacol. 2009,
156, 316.
Gentest™ microplated-based fluorimetric assay with recombinant human P450
enzymes: data reported as % inhibition at 3
lM compound’s concentration against
12. Hoshino, T.; Namba, T.; Takehara, M.; Nakaya, T.; Sugimoto, Y.; Araki, W.;
Narumiya, S.; Suzuki, T.; Mizushima, T. J. Biol. Chem. 2009, 284, 18493.
13. Murase, A.; Okumura, T.; Sakakibara, A.; Tonai-Kachi, H.; Nakao, K.; Takada, J.
Eur. J. Pharmacol. 2008, 580, 116.
14. Kabashima, K.; Saji, T.; Murata, T.; Nagamachi, M.; Matsuoka, T.; Segi, E.;
Tsuboi, K.; Sugimoto, Y.; Kobayashi, T.; Miyachi, Y.; Ichikawa, A.; Narumiya, S. J.
Clin. Invest. 2002, 109, 883.
15. Colucci, J.; Boyd, M.; Berthelette, C.; Chiasson, J. F.; Wang, Z.; Ducharme, Y.;
Friesen, R.; Wrona, M.; Levesque, J. F.; Denis, D.; Mathieu, M. C.; Stocco, R.;
Therien, A. G.; Clarke, P.; Rowland, S.; Xu, D.; Han, Y. Bioorg. Med. Chem. Lett.
2010, 20, 3760.
CYP450 3A4, 2D6, 1A2, 2C19 and 2C9 isoforms
Entry
1A2
2C9
2C19
2D6
3A4dbf
3A47bfc
1a
1b
1c
1d
NI
NI
NI
NI
21
11
17
26
NI
NI
11
13
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
NI
DBF = Dibenzyl fluorescein. BFC = 7-benzyloxy-4-trifluoromethyl coumarin. NI = no
inhibition.
16. Burch, J. D.; Belley, M.; Fortin, R.; Deschenes, D.; Girard, M.; Colucci, J.; Farand,
J.; Therien, A. G.; Mathieu, M.; Denis, D.; Vigneault, E.; Lévesque, J.; Gagne´, S.;
Wrona, M.; Xu, D.; Clark, P.; Rowland, S.; Han, Y. Bioorg. Med. Chem. Lett. 2008,
18, 2048.
any in vitro agonistic activity in the EP4 cAMP functional assay.
Compound 1a had an excellent selectivity profile toward prosta-
glandin EP2 receptor and low CYP450 inhibition potential and
therefore represents a new useful tool for further investigate the
role of EP4 antagonism both in vitro and in vivo.
17. (a) Hattori, K.; Tanaka, A.; Fujii, N.; Takasugi, H.; Tenda, Y.; Tomita, M.;
Nakazato, S.; Nakano, K.; Kato, Y.; Kono, Y.; Murai, H.; Sakane, K. J. Med. Chem.
2005, 48, 3103; (b) Clark, D. E. et al. WO2004067524.; (c) Blouin, M. et al.
WO2008017164.; (d) Murase, A.; Taniguchi, Y.; Tonai-Kachi, H.; Nakao, K.;
Takada, J. Life Sci. 2008, 82, 226; (e) Machwate, M.; Harada, S.; Leu, C. T.; Seedor,
G.; Labelle, M.; Gallant, M.; Hutchins, S.; Lachance, N.; Sawyer, N.; Slipetz, D.;
Metters, K. M.; Rodan, S. B.; Young, R.; Rodan, G. A. Mol. Pharm. 2001, 60, 36; (f)
Clark, D. E. et al. WO2005037812.; (g) Nozawa, E. et al. WO2009005076.; (h)
Yamagishi, T. et al. WO2005021508.; (i) Nozawa, E. et al. WO2009139373.
18. MDL Information System Inc. Symyx MDL, Accelrys Inc., San Diego, CA, USA.
http://www.accelrys.com.
19. Schrodinger LLC, Portland, USA. http://www.schrodinger.com.
20. Talete srl, Milan, Italy. http://www.talete.mi.it.
21. Umetrics AB, Umeå, Sweden. http://www.umetrics.com.
22. Eriksson, L.; Arnhold, T.; Beck, B.; Fox, T.; Johansson, E.; Kriegl, J. M. J. Chemom.
2004, 18, 188.
23. Dixon, S. L.; Smondyrev, A. M.; Knoll, E. H.; Rao, S. N.; Shaw, D. E.; Friesner, R. A.
J. Comput. Aided Mol. Des. 2006, 20, 647.
Acknowledgments
The authors thank Mrs. Anna Stucchi and Mr. Mario Montagna
for their skillful technical assistance in binding and functional as-
says, Mrs. Silvia Giambuzzi for CYP450 data generation and the
members of the Rottapharm Madaus R&D Analytical Chemistry
group for their assistance in the structural confirmation of the
compounds described in this manuscript.
24. Boyd, M. J.; Berthelette, C.; Chiasson, J. F.; Clark, P.; Colucci, J.; Denis, D.; Han,
Y.; Lévesque, J. F.; Mathieu, M. C.; Stocco, R.; Therien, A.; Rowland, S.; Wrona,
M.; Xu, D. Bioorg. Med. Chem. Lett. 2011, 21, 484.
References and notes
1. Flower, R. J. Nat. Rev. Drug Disc. 2003, 2, 179.
25. Compound 12a analytical data (C16H11F3O3, Molecular Weight = 308.25): 1H
NMR (400 MHz, DMSO-d6) d ppm, 7.57 (m, 3H), 7.48 (d, 1H), 7.41 (d, 1H), 7.21
(t, 1H), 3.20 (m, 4H); HPLC-MS [MH]+ = 309.1.
2. Bombardier, C.; Laine, L.; Reicin, A.; Shapiro, D.; Burgos-Vargas, R.; Davis, B.;
Day, R.; Bosi Ferraz, R.; Hawkey, C. J.; Hochberg, M. C.; Kvien, T. K.; Schnitzer, T.
J. N. Eng. J. Med. 2000, 343, 1520.
3. (a) Solomon, S. D.; McMurray, J. J. V.; Pfeiffer, M. A.; Wittes, J.; Fowler, R.; Finn,
P.; Anderson, W. F.; Zauber, A.; Hawk, E.; Bertagnolli, M. N. Eng. J. Med. 2005,
352, 1071; (b) Nussmeier, N. A.; Whelton, A. A.; Brown, M. T.; Langford, R. M.;
Hoeft, A.; Parlow, J. L.; Boyce, S. W.; Verbug, K. M. N. Eng. J. Med. 2005, 352,
1081; (c) Bresalier, R. S.; Sandler, R. S.; Quan, H.; Bolognese, J. A.; Oxenius, B.;
Horgan, K.; Lines, C.; Ridell, R.; Morton, D.; Lanas, A.; Konstam, M. A.; Baron, J.
A. N. Eng. J. Med. 2005, 352, 1092.
4. (a) McAdam, B. F.; Catella-Lawson, F.; Mardini, I. A.; Kapoor, S.; Lawson, J. A.;
FitzGerald, G. A. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 272; (b) FitzGerald, G. A.;
Patrono, C. N. Eng. J. Med. 2001, 345, 433; For a review see: (c) Funk, C. D.;
FitzGerald, G. A. J. Cardiovasc. Pharmacol. 2007, 50, 470.
5. McCoy, J. M.; Wicks, J. R.; Audoly, L. P. J. Clin. Invest. 2002, 110, 651.
6. Lin, C. R.; Amaya, F.; Barrett, L.; Wang, H.; Takada, J.; Samad, T. A.; Woolf, C. J. J.
Pharmacol. Exp. Ther. 2006, 319, 1096.
7. Clark, P.; Rowland, S. E.; Denis, D.; Mathieu, M.-C.; Stocco, R.; Poirier, H.; Burch,
J.; Han, Y.; Audoly, L.; Therien, A. G.; Xu, D. J. Pharmacol. Exp. Ther. 2008, 325, 425.
8. Takeuchi, K.; Tanaka, A.; Shinichi, K.; Aihara, E.; Amagase, K. J. Pharmacol. Exp.
Ther. 2007, 322, 903.
9. Cipollone, F.; Fazia, M. L.; Iezzi, A.; Cuccurollo, C.; De Cesare, D.; Ucchino, S.;
Spigonardo, F.; Marchetti, A.; Buttitta, F.; Paloscia, L.; Mascellanti, M.;
Cuccurollo, F.; Mezzetti, A. Arterioscler. Thromb. Vasc. Biol. 2005, 25, 1925.
10. (a) Ma, X.; Kundu, N.; Rifat, S.; Walser, R.; Fulton, A. M. Cancer Res. 2006, 66,
2923; (b) Chell, S. D.; Witherden, I. R.; Dobson, R. R.; Moorghen, M.; Herman, A.
A.; Qualtrough, D.; Williams, A. C.; Paraskeva, C. Cancer Res. 2006, 66, 3106; (c)
Yang, L.; Huang, Y.; Porta, R.; Yanagisawa, K.; Gonzalez, A.; Segi, E.; Johnson, D.
J.; Narumiya, S.; Carbone, D. P. Cancer Res. 2006, 66, 9665.
Compound 12b analytical data (C16H11F3O3, Molecular Weight = 308.25): 1H
NMR (400 MHz, DMSO-d6) d ppm, 7.55 (m, 2H), 7.49 (d, 1H), 7.42 (s, 2H), 7.21
(t, 1H), 3.15 (m, 4H); HPLC-MS [MH]+ = 309.1.
26. Compound 1a analytical data (C25H20F3NO4, Molecular Weight = 455.43): 1H
NMR (400 MHz, MeOD) d ppm, 9.03 (d, J = 7.8 Hz, 1H), 8.04 (d, J = 8.3 Hz, 2H),
7.60 (d, J = 8.3 Hz, 2H), 7.42–7.27 (m, 4H), 7.23–7.09 (m, 2H), 5.45–5.29 (m,
1H), 3.30–3.10 (m, 4H), 1.62 (d, J = 6.8 Hz, 3H); UPLC-MS [MH]+ = 456.3.
Compound 1b analytical data (C26H20F3NO4, Molecular Weight = 467.44): 1H
NMR (400 MHz, MeOD) d ppm, 9.29 (s, 1H), 8.01 (d, J = 8.3 Hz, 2H), 7.52 (d,
J = 8.8 Hz, 3H), 7.41–7.30 (m, 3H), 7.22–7.13 (m, 1H), 7.00 (d, J = 8.8 Hz, 1H),
3.27–3.15 (m, 4H), 1.56–1.42 (m, 4H); UPLC-MS [MH]+ = 468.3.
Compound 1c analytical data (C25H20F3NO4, Molecular Weight = 455.43): 1H
NMR (400 MHz, MeOD) d ppm, 8.06 (d, J = 8.3 Hz, 2H), 7.60 (d, J = 8.3 Hz, 2H),
7.46 (s, 1H), 7.37–7.30 (m, 2), 7.27 (dd, J = 2.0, 8.3 Hz, 1H), 7.21–7.12 (m, 1H),
6.75 (d, J = 8.3 Hz, 1H), 5.37 (d, J = 6.8 Hz, 1H), 3.28–3.11 (m, 4H), 1.62 (d,
J = 7.3 Hz, 3H); UPLC-MS [MH]+ = 456.3.
Compound 1d analytical data (C26H20F3NO4, Molecular Weight = 467.44): 1H
NMR (400 MHz, DMSO-d6) d ppm = 9.21–9.16 (m, 1H), 7.91–7.85 (m, 2H),
7.49–7.39 (m, 3H), 7.38–7.29 (m, 1H), 7.27–7.22 (m, 1H), 7.21–7.15 (m, 1H),
3.22–3.15 (m, 4H), 1.44–1.34 (m, 4H); UPLC-MS [MH]+ = 468.3.
27. Abramovitz, M.; Adam, M.; Boie, Y.; Carrière, M.; Denis, D.; 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. Biochem. Biophys. Acta 2000, 1483, 285.
28. The cAMP assay, a bead-based proximity assay Alphascreen (PerkinElmer Inc.,
MA, USA), is performed in cell membranes from HEK293 stably expressing
human EP4 receptors, according to supplier indications.