Discovery of highly potent dual EP2and EP3agonists with subtype selectivity

Article abstract of DOI:10.1016/j.bmcl.2015.12.039

The cyclic carbamate derivatives, 2-{[2-((4S)-4-{(1E,3R)-8-fluoro-3-hydroxy-4,4-dimethyl-1-octenyl}-2-oxo-1,3-oxazolidin-3-yl)ethyl]sulfanyl}-1,3-thiazole-4-carboxylic acid (5) and 2-{[2-((4S)-4-{(1E,3R)-3-[1-(4-fluorobutyl)cyclobutyl]-3-hydroxy-1-propenyl}-2-oxo-1,3-oxazolidin-3-yl)ethyl]sulfanyl}-1,3-thiazole-4-carboxylic acid (7) were identified as the first potent dual EP₂and EP₃agonists with selectivity against the EP₁and EP₄subtypes. Compounds 5 and 7 demonstrated highly potent dual EP₂and EP₃agonist activity with EC₅₀values of 10 nM or less. In addition, these compounds possess structural features distinct from natural prostaglandins, such as a cyclic carbamate moiety, a dimethyl or cyclobutyl group and a terminal fluorine atom.

Full text of DOI:10.1016/j.bmcl.2015.12.039

Bioorganic & Medicinal Chemistry Letters 26 (2016) 1016–1019
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of highly potent dual EP2 and EP3 agonists with subtype
selectivity
a
a
a
b
Akihiro Kinoshita a, , Masato Higashino , Yoshiyuki Aratani , Akito Kakuuchi , Hidekazu Matsuya ,
⇑
Kazuyuki Ohmoto a,
⇑
a Medicinal Chemistry Research Laboratories, Ono Pharmaceutical Co., Ltd, 3-1-1 Sakurai, Shimamoto-cho, Mishima-gun, Osaka 618-8585, Japan
b Department of Biology & Pharmacology, Ono Pharmaceutical Co., Ltd, 3-1-1 Sakurai, Shimamoto-cho, Mishima-gun, Osaka 618-8585, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The cyclic carbamate derivatives, 2-{[2-((4S)-4-{(1E,3R)-8-fluoro-3-hydroxy-4,4-dimethyl-1-octenyl}-2-
oxo-1,3-oxazolidin-3-yl)ethyl]sulfanyl}-1,3-thiazole-4-carboxylic acid (5) and 2-{[2-((4S)-4-{(1E,3R)-3-
[1-(4-fluorobutyl)cyclobutyl]-3-hydroxy-1-propenyl}-2-oxo-1,3-oxazolidin-3-yl)ethyl]sulfanyl}-1,3-thi-
azole-4-carboxylic acid (7) were identified as the first potent dual EP2 and EP3 agonists with selectivity
against the EP1 and EP4 subtypes. Compounds 5 and 7 demonstrated highly potent dual EP2 and EP3 ago-
nist activity with EC50 values of 10 nM or less. In addition, these compounds possess structural features
distinct from natural prostaglandins, such as a cyclic carbamate moiety, a dimethyl or cyclobutyl group
and a terminal fluorine atom.
Received 9 October 2015
Revised 28 November 2015
Accepted 11 December 2015
Available online 12 December 2015
Keywords:
Prostaglandin
EP2 receptor
EP3 receptor
Dual agonist
Underactive bladder
Ó 2015 Elsevier Ltd. All rights reserved.
Prostanoid receptors are members of the G-protein coupled
receptor superfamily. Receptors for prostaglandin E2 (PGE2) can
be classified into four subtypes, EP1, EP2, EP3, EP4.1 The diverse bio-
logical activities of PGE2 are considered to be expressed as a hybrid
of the activities mediated by these four EP receptor subtypes.
Among them, the EP2 receptor subtype2,3 induces smooth muscle
relaxation,4 while the EP3 receptor subtype inhibits smooth muscle
relaxation.5
Underactive bladder (UAB) represents dysfunctional conditions
of the bladder where patients are unable to produce an effective
voiding contraction. The most common clinical signs are the eleva-
tion of post-void residual urine volume and the lowering of urine
flow rate. These symptoms have a profoundly negative impact on
quality of life. The primary drugs currently used for UAB are a cho-
linesterase inhibitor, distigmine bromide and a muscarinic recep-
tor agonist, bethanechol chloride. The systemic cholinergic side
effects of these two drugs negatively impact this therapy.
PGE2 is considered to act on both bladder and urethral smooth
muscle. It has been reported that PGE2 prompts contraction of the
isolated bladder and relaxation of the isolated urethra.6 In addition
our pharmacological tests revealed that an EP3 agonist contracts
the bladder and an EP2 agonist relaxes the urethra (American Urol-
ogy Association, 2015).
Our purpose was to develop PGE2 analogs possessing highly
potent dual EP2 and EP3 agonist activity with selectivity against
the other two subtypes because a dual EP2 and EP3 agonist has
the potential as an effective therapeutic addressing unmet medical
needs for UAB.
So far, a potent dual EP2 and EP3 agonist with selectivity against
the EP1 and EP4 receptor subtypes has not been identified. On the
other hand, a dual EP2 and EP4 agonist with selectivity against
CO2H
N
S
O
S
N
CH3
OH
1
EP1
EP2
9.3
90
EP3
540
-
EP4
0.41
0.79
Mouse Binding Assay
>104
-
Ki (nM)
Rat Functional Assay
EC50 (nM)
⇑
Corresponding authors. Tel.: +81 75 961 1151; fax: +81 75 962 9314.
E-mail addresses: ak.kinoshita@ono.co.jp (A. Kinoshita), k.ohmoto@ono.co.jp
(K. Ohmoto).
Figure 1. EP2 and EP4 dual agonist 1.
http://dx.doi.org/10.1016/j.bmcl.2015.12.039
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.

A. Kinoshita et al. / Bioorg. Med. Chem. Lett. 26 (2016) 1016–1019
1017
Table 3
CO2H
N
S
O
Activity profiles of cyclic carbamate derivatives
S
N
CH3
CO2H
N
S
O
S
OH
N
O
R
1
OH
+
O
O
Compd
R
Human functional
assay, EC50 (nM)
Human binding assay
CO2H
CO2Me
CH3
a
Kia (nm)
H3C CH3
OH
EP2
EP3
EP4
EP1
CH3
HO
OH CH3
HO
H3C
*
H3C CH3
CH3
CH3
F
4
5
7.4
50
320
120
Limaprost
Gemeprost
5.7
2.9
4.7
3.9
1220 431
*
6
7
8
9
73
220
CH3
*
*
*
CO2H
N
O
CH3
2 : R =
3 : R =
*
S
S
R
N
CH3
H3C CH3
*
2.9
18
29
10
25
195
705
1080
>10,000
8418
F
CH3
OH
O
CH3
CH3
Figure 2. Molecular design of c-lactam PGE analogs.
2606 745
O
*
*
Table 1
Activity profiles of
c
-lactam derivatives
10
160
27
465
802
CO2H
N
O
N
CH3
2 : R =
3 : R =
*
S
S
R
CH3
11
12
13
60
21
7.4
18
8390 1332
*
H3C CH3
*
CH3
OH
CH3
240
89
a
Compd
Human functional assay, EC50 (nM)
EP3
*
*
EP2
EP4
CH3
CH3
2
3
0.39
0.91
310
8.4
3.0
4.2
3700 6.7
4710 149
CH3
a
EC50 values represent the mean of at least two experiments.
a
EC50 or Ki values represent the mean of at least two experiments.
Table 2
Table 4
Effect of the incorporation of oxygen atom into 5-membered ring
Pharmacokinetics profile of 7 in rats
Iv dosing (0.01 mg/kg)
Oral dosing (1 mg/kg)
CO2H
N
O
S
CL (mL/min/kg)
9.5
T1/2 (h)
AUC (
0.041
lgÁh/mL)
F (%)
S
N
3 : X = CH2
4 : X = O
X
H3C CH3
4.4
2.5
CH3
OH
to its EP2 and EP4 agonist activity. Therefore, the second step was
a
to optimize the 5-membered ring and the
reduction of EP4 agonist activity.
x side chain of 3 toward
Compd
Human functional assay, EC50 (nM)
EP3
EP2
EP4
According to the published data,8 lipophilicity at 5-membered
ring seems to relate to EP4 agonist activity. Therefore, the reduction
of EP4 agonist activity can be expected by the incorporation of
oxygen atom into 5-membered ring. As shown in Table 2, the effect
of modification of the 5-membered ring was investigated. The
functional assay revealed that cyclic carbamate 4 showed a distinct
decrease in EP4 agonist activity versus its lactam counterpart 3 as
expected.
3
4
0.91
7.4
8.4
50
4.2
320
a
EC50 values represent the mean of at least two experiments.
the EP1 and EP3 receptor subtypes was reported (compound 1 in
Fig. 1).7
Our first molecular design for a dual EP2 and EP3 agonist is
described in Figure 2. At first, an increase in affinity for the EP3
x side chain of lima-
prost or gemeprost, which are prostaglandin E1 (PGE1) analogs in
clinical use with high affinity for the EP3 receptor, was introduced
into 1. The activity profiles of the resulting c-lactam derivatives 2
and 3 are shown in Table 1. Of the two resulting compounds, gem-
dimethyl 3 demonstrated potent EP3 agonist activity comparable
Structure–activity relationships (SAR) of the cyclic carbamate
derivatives are shown in Table 3. First, the effect of
receptor was required for compound 1. The
x side chain
was investigated. The functional assays for human EP2–EP4 recep-
tor subtypes and the binding affinity for human EP1 receptor sub-
type were performed to determine subtype selectivity.
Surprisingly, the incorporation of a terminal fluorine atom into 4
enhanced EP3 agonist activity while reducing EP4 agonist activity.

1018
A. Kinoshita et al. / Bioorg. Med. Chem. Lett. 26 (2016) 1016–1019
ClH
NH2
O
O
f), g)
a), b)
c), d), e)
OH
NH
HO
N
O
O
OH
CO2Me
OTBS
14
15
16
CO2Et
N
S
O
CO2Et
N
O
S
O
SAc
h), i)
j), k)
S
N
N
O
H3C
CH3
N
S
O
O
OTBS
R
OH
O
17
18
19 : R = H
20 : R = F
CO2H
N
O
S
N
S
l), m)
O
H3C
CH3
R
OH
4 : R = H
5 : R = F
CH3
CH3
o), p)
H3C
HO
CH3
H3C
CH3
n)
HO
MeO
MeO
F
F
P
O
O
O
O
21
22
23
Scheme 1. Syntheses of 4 and 5. Reagents and conditions: (a) K2CO3, water, triphosgene, toluene, 0 °C; (b) NaBH4, EtOH, rt, 51% (for 2 steps); (c) TBSCl, imidazole, DMF, rt; (d)
bromo ethyl acetate, KOtBu, THF, rt; (e) NaBH4, THF–EtOH, rt, 89% (for 3 steps); (f) MsCl, Et3N, CH2Cl2, 0 °C; (g) KSAc, DMF, rt, 99% (for 2 steps); (h) ethyl 2-bromo-1,3-thiazole-
4-carboxylate, tributylphosphine, K2CO3, EtOH, 50 °C; (i) TBAF, THF, rt, 76% (for 2 steps); (j) SO3-Py, Et3N, DMSO, EtOAc, 10 °C; (k) NaH, dimethyl (3,3-dimethyl-2-oxoheptyl)
phosphonate or 23, THF, 0 °C, 46–59% (for 2 steps); (l) NaBH4, MeOH, AcOH, À78 °C, then separation by silica gel column chromatography; (m) NaOH (aq), MeOH, 0 °C, 26–
44% (for 2 steps); (n) 1-Bromo-4-fluorobutane, n-BuLi, i-Pr2NH, THF, 0 °C; (o) SOCl2, reflux; (p) dimethyl methylphosphonate, n-BuLi, THF, À78 °C, 57% (for 3 steps).
The resulting compound 5 indicated the possibility of potent and
selective dual EP2 and EP3 agonist. Furthermore, compound 7
which possesses a cyclobutyl group in place of the gem-dimethyl
in 5 preserved the feature of a potent dual EP2 and EP3 agonist.
Meanwhile, the activities of other cyclobutyl derivatives 8 and 9
possessing terminal alkoxy group dropped overall. Compound 9,
in particular, suffered a significant loss in EP3 agonist activity. In
contrast, the incorporation of a bulky terminal group like phenyl,
cyclohexyl or tert-butyl led to a selective EP3 agonist. In addition,
3-thiazole-4-carboxylate and removal of TBS group provided
alcohol 18. Horner–Emmons olefination between the aldehyde
derived from 18 and dimethyl (3,3-dimethyl-2-oxoheptyl)phos-
phonate or 23 provided 19 or 20, respectively. Reduction of ketone
19 or 20 resulted in a diastereomeric mixture of each correspond-
ing alcohol. The desired alcohols were separated by silica gel col-
umn chromatography, followed by hydrolysis to provide 4 and 5,
respectively. The configuration of the hydroxyl group was deter-
mined using the Mosher ester method.9 Regarding preparation of
the linear
x
side chain of internal alkyne 12 restored the EP1
phosphonate 23, at first carboxylic acid 21 was a-alkylated with
binding affinity. Regarding the cyclic carbamate derivatives, the
combination of sterically-bulky dimethyl or cyclobutyl group and
the terminal fluorine atom turned to be advantageous potent dual
EP2 and EP3 agonist activity with selectivity against the EP1 and EP4
1-bromo-4-fluorobutane to prepare 22. Carboxylic acid 22 was
converted into the corresponding acid chloride, and then treated
with dimethyl methylphosphonate to give phosphonate 23.
The syntheses of 6–13 are described in Scheme 2. Alkylation of
cyclobutyl carboxylic acid 24 with alkylbromide (R-Br) afforded
carboxylic acids 25a–h. Acid chlorides derived from carboxylic
acids 25a–h were coupled with dimethyl methylphosphonate to
prepare the corresponding phosphonates 26a–h. Oxidation of 18
and Horner–Emmons reaction with 26a–h were performed to pro-
vide alkenes 27a–h. Ketones 27a–h were reduced with NaBH4 and
then purified by silica gel column chromatography. The desired
alcohols were hydrolyzed with aqueous NaOH to give 6–13.
In summary, we have discovered compounds 5 and 7 as the first
highly potent dual EP2 and EP3 agonists with selectivity against the
EP1 and EP4 subtypes. The cyclic carbamate, dimethyl or cyclobutyl
group and terminal fluorine atom were considered key moieties for
dual EP2 and EP3 agonist activity. Further optimization of 5 and 7 to
improve their pharmacokinetic profiles and subtype selectivity will
be reported in due course.
subtypes. Compounds
5 and 7 were expected to be good
candidates for a UAB drug.
Unfortunately, the acceptable in vitro dual EP2 and EP3 agonist
activity of compounds 5 and 7 was offset by pharmacokinetic prob-
lems. Table 4 shows the bioavailability of 7 is only 2.5% in rats. So,
optimization of these compounds toward improvement in pharma-
cokinetic properties must be done to elaborate a clinical candidate.
The syntheses of 4 and 5 are described in Scheme 1. Commer-
cially available D-serine methyl ester hydrochloride 14 was treated
with triphosgene to form cyclic carbamate. Subsequent reduction
of the methyl ester afforded alcohol 15. Protection of the hydroxyl
group of 15 was followed by N-alkylation of cyclic carbamate,
reduction of the ethyl ester, mesylation of the resulting alcohol
16 and substitution by thioacetate to extend the
a side chain.
Treatment of thioacetate 17 with ethyl 2-bromo-1,

A. Kinoshita et al. / Bioorg. Med. Chem. Lett. 26 (2016) 1016–1019
1019
b), c)
a)
HO
MeO
MeO
HO
R
P
R
O
O
O
O
24
25a-h
26a-h
CO2Et
CO2H
N
S
N
S
O
O
S
S
d), e)
f), g)
N
N
O
O
18
R
R
O
OH
27a-h
6-13
R=
*
*
CH3
*
a :
e :
*
*
b :
c :
f :
F
CH3
O
CH3
g :
*
*
CH3
CH3
*
CH3
d :
O
h :
CH3
Scheme 2. Syntheses of 6–13. Reagents and conditions: (a) R-Br, n-BuLi, i-Pr2NH, THF, 0 °C; (b) SOCl2, reflux; (c) dimethyl methylphosphonate, n-BuLi, THF, À78 °C, 54–87%
(for 3 steps); (d) SO3-Py, Et3N, DMSO, EtOAc, 10 °C; (e) NaH, 26a–h, THF, 0 °C, 28–77% (for 2 steps); (f) NaBH4, MeOH, AcOH, À78 °C, then separation by a silica gel column
chromatography; (g) NaOH(aq), MeOH, 0 °C, 22–40% (for 2 steps).
3. Regan, J. W.; Bailey, T. J.; Pepperl, D. J.; Pierce, K. L.; Bogardus, A. M.; Donello, J.
Acknowledgments
E.; Fairbairn, C. E.; Kedzie, K. M.; Woodward, D. F.; Gil, D. W. Mol. Pharmacol.
1994, 46, 213.
We thank Professor Jason J. Chruma (Sichuan University) for
careful reading of this manuscript and helpful suggestions. We also
thank Mr. Masaki Tsujimura for measuring pharmacokinetic pro-
files, Mr. Koji Teraishi and Ms. Hiroko Takano for performing the
biological tests.
4. Coleman, R. A.; Kennedy, I.; Humphrey, P. P. A.; Bunce, K.; Lumley, P. In
Comprehensive Medicinal Chemistry; Emmett, J. C., Ed.; Pergamon: Oxford, 1990;
Vol. 3, pp 643–714.
5. Ushikubi, F.; Hirata, M.; Narumiya, S. J. Lipid Mediators Cell Signalling 1995, 12,
343.
6. Persson, C. G. Acta Physiol. Scand. 1976, 97, 139.
7. Kambe, T.; Maruyama, T.; Nakai, Y.; Oida, H.; Maruyama, T.; Abe, N.; Nishiura,
A.; Nakai, H.; Toda, M. Bioorg. Med. Chem. 2012, 20, 3502.
8. Kambe, T.; Maruyama, T.; Nagase, T.; Ogawa, S.; Minamoto, C.; Sakata, K.;
Maruyama, T.; Nakai, H.; Toda, M. Bioorg. Med. Chem. 2012, 20, 702.
9. Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512.
References and notes
1. Coleman, R. A.; Smith, W. L.; Narumiya, S. Pharmacol. Rev. 1994, 46, 205.
2. Katsuyama, M.; Nishigaki, N.; Sugimoto, Y.; Morimoto, K.; Negishi, M.;
Narumiya, S.; Ichikawa, A. FEBS Lett. 1995, 372, 151.