Discovery and preliminary evaluation of 5-(4-phenylbenzyl)oxazole-4-carboxamides as prostacyclin receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2006.12.025

The discovery and evaluation of 5-(4-phenylbenzyl)oxazole-4-carboxamides as prostacyclin (IP) receptor antagonists is described. Analogs disclosed showed high affinity for the IP receptor in human platelet membranes with IC₅₀ values of 0.05-0.50 μM, demonstrated functional antagonism by inhibiting cAMP production in HEL cells with IC₅₀ values of 0.016-0.070 μM, and exhibited significant selectivity versus other prostanoid receptors.

Full text of DOI:10.1016/j.bmcl.2006.12.025

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1211–1215
Discovery and preliminary evaluation of 5-(4-phenylbenzyl)oxazole-
4-carboxamides as prostacyclin receptor antagonists
Marc-Raleigh Brescia,* Laura L. Rokosz, Andrew G. Cole, Tara M. Stauffer,
John M. Lehrach, Douglas S. Auld, Ian Henderson and Maria L. Webb
Pharmacopeia Drug Discovery Inc., PO Box 5350, Princeton, NJ 08543, USA
Received 13 October 2006; revised 6 December 2006; accepted 7 December 2006
Available online 12 December 2006
Abstract—The discovery and evaluation of 5-(4-phenylbenzyl)oxazole-4-carboxamides as prostacyclin (IP) receptor antagonists is
described. Analogs disclosed showed high affinity for the IP receptor in human platelet membranes with IC50 values of 0.05–
0.50 lM, demonstrated functional antagonism by inhibiting cAMP production in HEL cells with IC50 values of 0.016–0.070 lM,
and exhibited significant selectivity versus other prostanoid receptors.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Prostacyclin (PGI2) is an eicosanoid produced from the
cyclooxygenase-mediated metabolism of arachidonic
acid.1 More specifically, it is generated from isomeriza-
tion of endoperoxide H2 (PGH2) via prostacyclin syn-
thase. PGI2 functions as an inhibitor of platelet
aggregation,2 is a potent vasodilator,3 and is a regulator
of blood flow in kidneys.4 PGI2 is also a physiological
antagonist of thromboxane A2 (TXA2).1a PGI2 mediates
its effects through binding and subsequent activation
of the prostacyclin (IP) receptor, a G-protein-coupled
receptor (GPCR) that signals through both Ca2+ and
cAMP.5 A number of studies have shown that PGI2
activity participates in the etiology of pain and inflam-
matory disease.6–11 For example, intraperitonial injec-
tion of PGI2 in mice induces a writhing response,6 and
IP receptor-deficient mice show a distinct reduced re-
sponse to inflammatory pain.11 Previously reported IP
antagonists (RO1138452 and RO3244794, Fig. 1) have
been shown to be orally efficacious in rat acetic-acid
writhing and carrageenan-induced paw hyperalgesia
models, with no effect on bleeding time in the mouse
or cardiovascular effects in the rat.12
N
N
H
N
H
O
RO1138452
F
F
O
H
N
O
(R)
HO
O
O
RO3244794
N
O
H
N
O
HN
O
Compound 6a
Figure 1. IP receptor antagonists.
We sought to identify a structurally distinct and novel
class of IP receptor antagonists by the high-throughput
screening of a diverse 2.1 million-member compound
collection prepared using ECLiPSTM (Encoded Combi-
natorial Libraries on Polymeric Support) technology.13
These compounds were screened using a platelet
membrane binding assay with the radioligand, [3H]-Ilo-
prost.14,15 Of the active structures identified, the 5-(4-
phenylbenzyl)oxazole-4-carboxamide 6a proved to be
Keywords: Prostacyclin; PGI2; Prostanoid receptors; Pain; Inflamma-
tion; ECLiPSTM
.
*
Corresponding author. Tel.: +1 609 452 3799; fax: +1 609 655
4187; e-mail: mbrescia@pcop.com
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.12.025

1212
M.-R. Brescia et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1211–1215
the most interesting, possessing an IC50 value of
0.077 lM (Fig. 1). Interestingly, compound 6a contains
both the aromatic bridging methylene group contained
in RO1138452 and the biaryl motif of RO3244794.
absence of Iloprost and showed no agonist activity
affirming that these analogs are antagonists of the IP
receptor.
The binding and functional activities of the carboxamide
analogs versus the IP receptor are shown in Table 1. In
general, all compounds showed greater potency in the
functional assay compared to the binding assay. Previ-
ously reported IP receptor antagonists have also shown
this phenomenon with up to a sevenfold difference in the
functional assay versus the binding assay.12 The func-
tional assay was validated with RO1138452 which
showed a potency of 0.00083 lM, which is consistent
with the published value of 0.001 lM. The more potent
functional activity was used to define SAR trends.
Hydrophobic substituents led to potent analogs in this
series as seen in benzyl analog 6b and phenethyl analog
6c, which showed potencies in the cAMP assay of less
than 0.15 lM. a-Methyl substitution of the benzylic
methylene of 6b to provide 6d was tolerated, but no in-
crease in potency was observed. N-Methylation of 6c to
form tertiary amide 6e reduced the functional activity
20-fold, thus showing the preference for secondary
amides. Substitution of the phenethyl-based analogs
was also well-tolerated. For example, the incorporation
of a cyclopropyl group to produce 6f generated an IC50
value of 0.36 lM. An increase in potency was also real-
ized by incorporating a-or b-substitution on phenethyl
moieties. (R)-(+)-b-methylphenethyl 6g produced an
IC50 value of 0.051 lM in the functional assay. The R-
configuration at the b-position was preferred over its
enantiomer, as 6h was sixfold less active in the function-
al assay than 6g. Incorporation of a carboxylic acid moi-
ety at the a-position with S-configuration to give
compound 6j resulted in the most potent analog in the
functional assay with an IC50 value of 0.016 lM. The
corresponding R stereoisomer, 6i, was significantly less
potent with an IC50 value of 0.83 lM. It is interesting
to note that the S-configuration is more potent in this
Compound 6a was then supplemented with a series of
analogs designed to expand the SAR around the phen-
ethyl-based carboxamide. Compounds were synthesized
from 4-iodophenylacetic acid (1) in five steps (Scheme
1). Synthesis of acyl azide 2 was accomplished by treat-
ment of 1 with diphenylphosphoryl azide (DPPA),
which was then reacted with the anion of methyl iso-
cyanoacetate to afford oxazole 3. Ester hydrolysis of 3
using lithium hydroxide provided the versatile novel
core 4, which was used in two synthetic pathways to
generate IP receptor antagonists 6. Primarily, com-
pounds 6 were synthesized from 4 via a palladium-
catalyzed biaryl formation under standard Suzuki–
Miyaura16 coupling conditions, to afford biaryl 5,
followed by 1-ethyl-3-(3-dimethylaminopropyl)-carbo-
diimide hydrochloride (EDC)-mediated amide bond
formation. This synthesis provided an efficient way
to examine the SAR at the carboxamide position.
Alternatively, generation of 7 from 4 by performing
the EDC-mediated amide formation first, followed by
Suzuki–Miyaura biaryl formation, was also a viable
synthesis of 6. This latter synthesis also provides greater
flexibility for the future generation of IP receptor antag-
onists aimed at examining the SAR around the biaryl
motif.
IP receptor binding activity was quantified via a filter
binding assay measuring displacement of [3H]-Iloprost
binding to human platelet membranes15 prepared as de-
scribed.17 Functional antagonism was quantified by
measuring the inhibition of cAMP production induced
by Iloprost in human erythroleukemia (HEL) cells using
a Perkin-Elmer LANCETM cAMP detection kit.18 All
analogs were also examined in the cAMP assay in the
O
(i)
HO
(ii)
(iii)
N3
O
I
O
O
N
O
I
I
1
2
3
O
HO
N
O
NH
(iv)
(v)
(v)
5
O
O
N
O
R1
HO
I
O
N
O
NH
4
O
(iv)
6 a-q
R1
O
I
N
O
7
Scheme 1. Reagents and conditions: (i) DPPA, Et3N, THF, 0–25 ꢁC; (ii) NaH, methyl isocyanoacetate, DMF, 0–25 ꢁC; (iii) LiOH, THF/H2O, 25 ꢁC;
(iv) Pd(PPh3)4, K2CO3, 3-acetamidophenylboronic acid, dioxane/H2O, 80 ꢁC; (v) EDC, HOBt, RR0NH, CH2Cl2, 25 ꢁC.

M.-R. Brescia et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1211–1215
Table 1. Biological activity of IP receptor antagonists
1213
N
O
R1
O
HN
O
R1
IPR IC50 (lM)
HEL cAMP IC50 (lM)
a
b
Compound
6ac
0.077 0.012
ND
N
H
N
H
6b
0.561 0.006
0.066 0.016
6c
0.721 0.020
1.23 0.08
0.138 0.000
0.278 0.002
N
H
6dc
N
H
6e
7.16 2.11
2.133 0.435
0.358 0.028
N
6fd
0.331 0.073
N
H
6g
6h
0.053 0.003
0.916 0.148
0.051 0.008
0.296 0.026
(R)
N
H
(S)
N
H
O
OH
6i
4.20 0.56
0.831 0.145
(R)
(S)
N
H
O
OH
6j
0.476 0.193
7.60 2.29
4.95 2.01
4.71 0.21
3.48 0.29
12.2 0.4
0.016 0.001
N
H
6k
6l
1.005 0.157
N
N
H
0.741 0.124
N
N
N
H
6m
6n
6o
0.476 0.173
N
H
0.828 0.072
N
H
3.306 1.026
N
H
(continued on next page)

1214
M.-R. Brescia et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1211–1215
Table 1 (continued)
a
b
Compound
R1
IPR IC50 (lM)
HEL cAMP IC50 (lM)
6p
6q
12.4 4.6
3.584 1.478
3.306 1.026
N
H
N
34.1 6.3
N
H
a Binding IC50 standard deviation.15
b Functional IC50 standard deviation.18
c Racemic.
d trans-Racemic.
series while the R-stereochemistry is presented in
RO3244794. Pyridylethyl analogs, 6k–m, and the phenyl
analog 6n were all less potent, exhibiting potencies of
greater than 0.5 lM. Small alkyl-based analogs (6o–q)
were not tolerated (IC50 >3 lM), showing a significant
reduction in potency relative to the phenethyl-based
antagonists.
Acknowledgments
We thank Weiqing Chen, Lynn Rogers, and Hema
Desai for assistance in ADME profiling and Robert
Swanson and Jingchun Yang for compound analyses
in the EP2 and EP4 antagonist assays.
Of the initial analogs generated, 6b, 6g, and 6j were the
most potent in the cAMP functional assay and were fur-
ther investigated in in vitro ADMET assays to determine
suitability for in vivo evaluation. Metabolic stability was
assessed in both human liver microsomes (HLM) and rat
liver microsomes (RLM) assays.19 After a 30-min incu-
bation with 0.3 lM of liver microsomes, 6j showed sub-
stantial stability in both species (HLM 100% remaining
and RLM 91% remaining), while 6b (HLM 63% remain-
ing and RLM 33% remaining) and 6g (HLM 12%
remaining and RLM 11% remaining) were observed to
be less stable. Since 6j showed good stability in the
microsome assays, it was further examined against a pan-
el of five cytochrome P450 (CYP) enzymes and showed
no significant activity (IC50 >20 lM) against 3A4, 2D6,
1A2, and 2C19, and weak activity (IC50 = 5.1 lM) versus
CYP2C9. Compound 6j also exhibited reasonable per-
meability in a Caco-2 assay,19 with a Papp of 55 nm/s
and no efflux. No significant hERG activity was ob-
served for 6j which showed only 5% inhibition in a rubid-
ium efflux assay at 20 lM. Overall, the in vitro ADMET
properties of 6j are favorable and constitute a lead com-
pound for further optimization against the IP receptor.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2006.12.025.
References and notes
1. (a) Dogne, J. M.; de Leval, X.; Delarge, J.; David, J. L.;
Masereel, B. Curr. Med. Chem. 2000, 7, 609; (b) de Leval,
X.; Hanson, J.; David, J. L.; Maseree, L. B.; Pirotte, B.;
Dogne, J. M. Curr. Med. Chem. 2004, 11, 1243.
2. Whittle, B. J.; Moncada, S.; Vane, J. R. Prostaglandins
1978, 16, 373; Weiss, H. J.; Turitto, V. T. Blood 1979, 53,
244.
3. Moncada, S.; Vane, J. R. Adv. Prostaglandin Thrombox-
ane Res. 1980, 6, 43.
4. Breyer, M. D. Exp. Nephrol. 1998, 6, 180.
5. Murray, R.; Furci, L.; FitzGerald, G. A. FEBS Lett. 1989,
255, 172; Feoktistov, I.; Breyer, R. M.; Biaggioni, I.
Biochem. Pharmacol. 1997, 54, 917.
6. Akarsu, E. S.; Palaoglu, O.; Ayhan, I. H. Meth. Find. Exp.
Clin. Pharmacol. 1989, 11, 273.
7. Schepelmann, K.; Messlinger, K.; Schaible, H. G.;
Schmidt, R. F. Neuroscience 1992, 50, 237.
8. Bley, K. R.; Hunter, J. C.; Eglen, R. M.; Smith, J. A.
Trends Pharmacol. Sci. 1998, 19, 141.
9. Bombardieri, S.; Cattani, P.; Ciabattoni, G.; Di Munno,
O.; Pasero, G.; Patrono, C.; Pinca, E.; Pugliese, F. Br. J.
Pharmacol. 1981, 73, 893.
10. Galvani, M.; Bugiardini, R.; Ferrini, D.; Gridelli, C.;
Mari, L.; Puddu, P.; Lenzi, S. Ric. Clin. Lab. 1985, 15,
145.
11. Murata, T.; Ushikubi, F.; Matsuoka, T.; Hirata, M.;
Yamasaki, A.; Sugimoto, Y.; Ichikawa, A.; Aze, Y.;
Tanaka, T.; Yoshida, N.; Ueno, A.; Oh-ishi, S.; Nar-
umiya, S. Nature 1997, 388, 678.
12. Clark, R. D.; Jahangir, A.; Severance, D.; Salazar, R.;
Chang, T.; Chang, D.; Jett, M. F.; Smith, S.; Bley, K.
Bioorg. Med. Chem. Lett. 2004, 16, 1053; Bley, K. R.;
Bhattacharya, A.; Daniels, D. V.; Gever, J.; Jahangir, A.;
O’yang, C.; Smith, S.; Srinivasan, D.; Ford, A. P.; Jett, M.
F. Br. J. Pharmacol. 2006, 147, 335.
Selectivity of 6j was examined against two other prosta-
noid receptors, EP2 and EP4, using the cloned recombi-
nant human receptors20 expressed in a derivative of
293EBNA cells.21 Significant selectivity over both EP2
and EP4 was observed, with IC50 values of 32 and
24 lM, respectively.
In summary, we have identified novel small-molecule
functional antagonists of the IP receptor. Specifically,
6j was shown to be a potent IP receptor antagonist
with favorable in vitro ADMET properties and is
selective against other prostanoid receptors. This
profile bodes well for further pharmaceutical develop-
ment of compounds in this chemical series. More
detailed selectivity studies, SAR expansion, and
in vivo pharmacokinetic profiling will be reported in
due course.

M.-R. Brescia et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1211–1215
1215
13. Ohlmeyer, M. H. J.; Swanson, R. N.; Dillard, L.; Reader,
J. C.; Asouline, G.; Kobayishi, R.; Wigler, M.; Still, W. C.
Proc. Nat. Acad. Sci. U.S.A. 1993, 90, 10922; Nestler, H.
P.; Bartlett, P. A.; Still, W. C. J. Org. Chem. 1994, 59,
4723.
1· penicillin/streptomycin solution. 5000 HEL cells were
pre-incubated in assay buffer containing 5 mM Hepes,
0.05% BSA, and compound for 15 min at 37 ꢁC in a 384-
well plate (Corning # 3710). 0.01 lM of Iloprost was
added to stimulate cAMP production and the cells were
again incubated for 15 min at 37 ꢁC. Assessment of
agonist activity was measured in the absence of Iloprost.
cAMP signal was detected using a LANCETM cAMP 384
assay kit and the EnVisionTM multilabel counter, from
Perkin-Elmer. Data were normalized relative to maximum
and minimum signal controls and the % maximum data
were fit to the Michaelis–Menten equation to calculate
IC50 values.
14. Grant, S. M.; Goa, K. L. Drugs 1992, 43, 889.
15. IP receptor filter binding assay. 0.01 lM [3H]-Iloprost (GE
Healthcare) was incubated at room temperature with
50 lg of membranes prepared from human platelets17 in
assay buffer containing 20 mM Hepes, 5 mM MgCl2, pH
7.4, in the presence or absence of compound for 1 h.
Compounds were prepared by serially diluting in DMSO,
followed by intermediate dilutions in assay buffer. Final
DMSO concentrations ranged from 1% to 10%, as
indicated. The assay mixture was filtered over a pre-wet
96-well MAFBN0B50 filter plate (Millipore) and washed
twice with 1· PBS (Mediatech). Fifty microliters of
Optiphase Supermix was added to each well and the assay
was counted on a Microbetaꢂ TriLux. Raw data were fit
to the Michaelis–Menten equation to calculate IC50
values. Data are reported as the average of multiple
determinations standard deviation.
16. Miyaura, N.; Yanagi, T.; Suzuki, A. Synth.Comm. 1979,
11, 513; Review: Suzuki, A. Chem. Comm. 2005, 38, 4759.
17. Kaczmarek, D.; Hohlfeld, T.; Wambach, G.; Schror, K.
Eur. J. Clin. Pharmacol. 1993, 45, 451.
18. cAMP assay. HEL 92.1.7 cells were obtained from ATCC
(cat # TIB-180). Cells were cultured at a density of 5 · 104
to 1 · 106 cells/mL in RPMI (Invitrogen) media contain-
ing 10% FBS, 10 mM Hepes, 1 mM sodium pyruvate, and
19. For liver microsome and Caco-2 assay conditions, see:
Merritt, J. R.; Rokosz, L. L.; Nelson, K. H.; Kaiser, B.;
Wang, W.; Stauffer, T. M.; Ozgur, L. E.; Schilling, A.; Li,
G.; Baldwin, J. J.; Taveras, A. G.; Dwyer, M. P.; Chao, J.
Bioorg. Med. Chem. Lett. 2006, 16, 4107.
20. EP2—Regan, J. W.; Bailey, T. J.; Pepperl, D. J.; Pierce,
K. L.; Bogardus, A. M.; Donello, J. E.; Fairbairn, C.
E.; Kedzie, K. M.; Woodward, D. F.; Gil, D. W. Mol.
Pharmacol. 1994, 46, 213; EP4—Honda, A.; Sugimoto,
Y.; Namba, T.; Watabe, A.; Irie, A.; Negishi, M.;
Narumiya, S.; Ichikawa, A. J. Biol. Chem. 1993, 268,
7759.
21. cAMP signaling was quantified in the presence of 560 pM
(EP2) or 170 pM (EP4) PGE2 using a Homogeneous Time
Resolved Fluorescence (HTRF) CIS-US cAMP kit,
cAMP dynamic 2, from CIS-US (Bedford, MA) according
to the manufacturer’s instructions.