Discovery of a new series of potent prostacyclin receptor agonists with in vivo activity in rat

Article abstract of DOI:10.1016/j.bmcl.2015.01.024

The design and synthesis of two closely related series of prostacyclin receptor agonist compounds that showed excellent human IP receptor potency and efficacy is described. Compounds from this series showed in vivo activity after SC dosing in the monocrotaline model of PAH in rat.

Full text of DOI:10.1016/j.bmcl.2015.01.024

Bioorganic & Medicinal Chemistry Letters xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of a new series of potent prostacyclin receptor agonists
with in vivo activity in rat
Thuy-Anh Tran, Young-Jun Shin, Bryan Kramer, Juyi Choi, Ning Zou, Pureza Vallar, Peter Martens,
P. Douglas Boatman, John W. Adams, Juan Ramirez, Yunqing Shi, Michael Morgan, David J. Unett,
⇑
Steve Chang, Hsin-Hui Shu, Shiu-Feng Tung, Graeme Semple
Arena Pharmaceuticals, 6154, Nancy Ridge Drive, San Diego, CA 92121, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The design and synthesis of two closely related series of prostacyclin receptor agonist compounds that
showed excellent human IP receptor potency and efficacy is described. Compounds from this series
showed in vivo activity after SC dosing in the monocrotaline model of PAH in rat.
Ó 2015 Elsevier Ltd. All rights reserved.
Received 5 December 2014
Revised 9 January 2015
Accepted 12 January 2015
Available online xxxx
Keywords:
Prostacyclin
IP receptor agonist
GPCR
PAH
The prostanoids are a family of cyclooxygenase metabolites of
arachidonic acid. These bioactive lipids evoke diverse biological
actions in many tissues and cell types by binding to specific cell
surface G-protein coupled receptors including DP, EP1, EP2, EP3,
EP4, FP, TP and IP.1 Prostacyclin (PGI2) is the major metabolite of
arachidonic acid produced in blood vessels2 and is a potent anti-
platelet, anti-inflammatory vasodilator that also inhibits smooth
muscle cell proliferation.3 The protective effects of PGI2 are medi-
ated by activation of the IP receptor on platelets, smooth muscle
cells and immune cells. Activation of the IP receptor leads primar-
ily to stimulation of adenylate cyclase with a resulting increase in
intracellular cAMP levels.4
Pulmonary arterial hypertension (PAH) is a rare but life-threat-
ening disease. The pathogenesis, which is thought to result, in part,
from an imbalance in the production of PGI2 and thromboxane
A2,5,6 includes pulmonary vasoconstriction, endothelial and
smooth muscle cell proliferation and platelet mediated thrombo-
sis.7 Therapeutic substitution of PGI2 or its analogs in PAH patients
reduces pulmonary artery pressure and slows disease progression.8
However, most compounds of this type have to be dosed intrave-
nously or by inhalation and due to their chemical and metabolic
instability their half-lives in vivo are short and as a result these
drugs have limited and sometimes problematic clinical application.
Thus we sought to design new prostacyclin receptor agonists with
significantly longer half-lives that might be dosed orally.
For our early approaches to new IP receptor agonists, we sur-
veyed multiple series of compounds described previously. In each
case we could discern similar general requirements for activity as
depicted by the three examples shown in Figure 1. Firstly, there
was an absolute requirement for an acidic moiety that we assumed
might mimic the carboxylate function of the endogenous ligand.
There was more variation allowed in what we designated the scaf-
fold portion, with both rigid bicyclic structures and more flexible
linear groups (e.g., MRE269, the active metabolite of the prodrug
NS-3049) being represented. This scaffold was then connected via
a group we termed the linker portion, which again could take the
form of both small and flexible groups through to more rigid het-
erocycles. This linker group then served to orientate the second
key requirement for the molecules which was a pair of aryl rings
suitably arranged in space, either as adjacent substituents in the
case of the aromatic linkers, or both attached directly to a single
atom on the more flexible linkers. With these design principles
in mind, we chose to select both a rigid scaffold and an aromatic
linker as this combination had been only rarely explored. For the
scaffold, we selected a tetrahydro-naphthalene core that was sim-
ilar to that deployed in FK-588, except that it lacked the tertiary
hydroxyl function, and for the aromatic group we surveyed
pyridones and diazinones substituted with two aromatic rings as
novel variants of the pyrazine moiety in NS-304. After the synthe-
sis of a number of possible examples, we quickly focused on the
⇑
Corresponding author.
E-mail address: gsemple@arenapharm.com (G. Semple).
http://dx.doi.org/10.1016/j.bmcl.2015.01.024
0960-894X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Tran, T.-A.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.01.024

2
T.-A. Tran et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
N
O
CO2H
O
ONO-1301
MRE269
FK-788
N
CO2H
CO2H
O
N
N
O
O
N
O
HO
O
CO2H
Biphenyl Group
Scaffold
Acidic Moiety
Figure 1. Design of new prostacyclin receptor antagonist series.
pyridazone series that we will describe here. In each case however,
the synthetic route to the key building block providing the scaffold
moiety (in combination with the acidic group) was identical and its
7-step synthesis is outlined in Scheme 1. The phenol group of
5-hydroxy tetralone (1) was protected as its tert-butyl diphenyl
silyl ether and the resulting ketone 2 converted to the keto-ester
3 by treatment with sodium hydride and diethyl carbonate.10 The
ketone group was then removed by reduction with triethyl silane
to provide the ester intermediate 4. Ester reduction, followed by
deprotection provided 6 which was not isolated but could be selec-
tively alkylated on the phenol with t-butyl bromoacetate to yield 7.
The synthesis of the building blocks 8a or 8b was completed by
bromination or mesylation of the remaining hydroxyl group to
make it available for nucleophilic substitutions. The synthesis of
the first series of this type, the 3,4-diarylpyridazinones is shown
in Scheme 2. To prepare the heterocyclic portion, two routes were
used depending on the whether the 3- or 4-phenyl ring was to be
substituted. In the case of the former (14h–14n), treatment of
phenyl maleic anhydride with benzyl hydrazine hydrochloride in
acetic acid under microwave heating conditions provided the
N-benzyl intermediate 10.11 Conversion to the chloride 11 by treat-
ment with POCl3, was again accomplished using microwave condi-
tions. This intermediate was debenzylated12 and subjected to an
array of Suzuki coupling reactions, albeit in modest yield (using
palladium tetrakis (triphenyl phosphine) as catalyst under micro-
wave conditions), to provide the 3,4-diphenylpyridazinones 13
that were ready to be reacted with the mesylated intermediate
8b that was prepared earlier. This coupling was achieved using
potassium tert-butoxide as base in the presence of a crown ether
at rt overnight. Final removal of the t-butyl ester protecting group
provided the 3-phenyl ring substituted analogues 14h–14n. For
the 4-phenyl substituted series, 3,4-dibromo-5-phenylfuran-
2(5H)-one was treated with hydrazine in ethanol under reflux to
provide the brominated pyridazinone 16 in one step.13 This inter-
mediate too was subjected to an array of Suzuki coupling reactions
then coupled to the scaffold/acid building block 8b using similar
conditions to those described above. Finally, deprotection of the
t-butyl ester under acidic conditions provided the test compounds
14a–14g. All compounds in the series were tested in our cloned
human and rat IP receptor cAMP (HTRF) functional assays.14 The
compound with no substitution on either phenyl ring (14a)
showed only modest activity at the IP receptor. However substitu-
tion on either of the phenyl rings was able to provide compounds
with both improved potency and functional efficacy (as estimated
by measuring intrinsic activity relative to 1 lM iloprost as stan-
dard) comparable to that determined for the literature standard,
MRE269. Potency appeared to be improved by either electron
donating or withdrawing groups in the 3-position on the 4-phenyl
ring and in the 4-position of the 3-phenyl ring, suggesting that the
steric or lipophilic rather than the electronic contribution for the
substituents was of greater importance. A number of significantly
more polar groups were also examined in each of these positions
but none of these were as active as 14a (data not shown) which
we rationalized as being entirely consistent with the highly hydro-
phobic nature of the prostacyclin ligand binding site. Two exam-
ples with substituents on both rings (14o–14p) which combined
our observations above regarding the optimal substitution patterns
were prepared using the 3-substituted ring synthesis. In this case
Please cite this article in press as: Tran, T.-A.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.01.024

T.-A. Tran et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
3
O
O
O
O
(i)
(ii)
OEt
(iii)
OH
OTBDPS
OTBDPS
1
2
3
O
OH
OH
(iv)
(v)
(vi)
OEt
OH
OTBDPS
OTBDPS
4
5
6
R
OH
(vii)
O
O
O
O
O
8a
or (viii)
R = Br
O
8b R = OMs
7
Scheme 1. Synthesis of tetrahydronaphthalene scaffold building blocks. Reagents and conditions: (i) TBDPSCl, imidazole, DMF, 25 °C, 10 h, 90%; (ii) CO(OEt)2, NaH, toluene,
110 °C, 2 h, 80%; (iii) Et3SiH, TFA, 25 °C, 3 h, 86%; (iv) LiAlH4, THF, 90 °C, 60 min, 60%; (v) TBAF, THF, 30 min; (vi) t-butyl bromoacetate, K2CO3, DMF, 60 °C, 2 h, 89%; (vii) CBr4,
PPh3, DCM, rt, o/n, 77%; (viii) MsCl, Et3N, DCM, 94%.
O
O
O
O
(i)
N
N
(ii)
N
N
O
OH
Cl
9
10
11
O
O
NH
N
NH
N
(iii)
(iv)
Cl
R2
12
13
(v)
O
N
N
O
CO2H
R1
R2
14
(viii)
O
O
O
Ph
Br
O
NH
N
(vi)
(vii)
NH
N
Br
Br
R1
17
15
16
Scheme 2. Synthesis of 3,4-diphenylpyridazinones substituted either on the 3- or 4-phenyl ring. Reagents and conditions: (i) PhCH2NHNH2Á2HCl, AcOH,
l
wave, 140 °C, 3 h,
69%; (ii) POCl3, l l
wave, 120 °C, ½ h, 14%; (iii) AlCl3, toluene, 90 °C, 84%; (iv) R2-PhB(OH)2, Pd(PPh3)4, K2CO3, wave, 120 °C, ½ h, 15–70%; (v) (a) tBuOK, 18-crown-6, DMF, 8b;
rt, o/n, (b) 4 M HCl dioxan, 15–50%, two steps; (vi) NH2NH2, EtOH, reflux, 2 h, 99%; (vii) (a) R2-PhB(OH)2, Pd(PPh3)4, K2CO3, wave, 120 °C, ½ h; (viii) (a) tBuOK, 18-crown-6,
l
DMF, 8b; rt, o/n (b) 4 M HCl dioxan, 15–50%, two steps.
Please cite this article in press as: Tran, T.-A.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.01.024

4
T.-A. Tran et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
the synthesis was started from 3-methoxyphenyl maleic anhy-
dride, as 3-methoxy was the 4-phenyl ring substituent that showed
the best activity IP receptor on its own, with this substituent carried
through the whole of the synthesis. Gratifyingly, the SAR was some-
what additive, with the combination of the 3-methoxy-4-phenyl
and 4-methyl-3-phenyl substitutions providing the compound
with the best activity for the IP receptor in this series. It was also
observed that all of the compounds examined appeared to show
diminished activity, by around 3–10 fold, in the rat receptor assay
when compared to the human receptor hence it was likely that
we would need highly potent compounds for the human receptor
to have a significant chance of observing in vivo activity in our rat
model of PAH. Unfortunately, all of this series of compounds
showed significant DP1 receptor agonist activity in a melanophore
dispersion assay.15 Although this counter-screening assay was in a
different platform that can in our experience somewhat over-exag-
gerate the potency and intrinsic efficacy of compounds as a result of
the high (transient) overexpression of the receptor, this was still a
significant cause for concern. It was noted however that essentially
no agonist activity against any of the other receptors in the prosta-
glandin family that we tested, was observed for these series of
compounds in the same melanophore assay platform (see Table 1).
As all the compounds were prepared in racemic form, a number
of the more active examples were separated into their two constit-
uent isomers to see if this would improve selectivity. However,
although one of the isomers in each case was at least 20-times
more potent than its antipode at the human IP receptor, the same
isomer was also responsible for essentially all of the off-target
activity as well.
We therefore decided to examine the alternative 4,5-diphenyl
substitution pattern on the pyridazinone ring to see if this would
also afford potent IP agonists but with improved selectivity over
the DP1 receptor.
Synthesis of the 4,5-substituted heterocyclic derivatives again
required two different approaches depending on which of the phe-
nyl rings were to be substituted, but in this case the variable ring
could be inserted at the penultimate step (Scheme 3). For the 5-
aryl analogues, treatment of 4,5-dichloropyridazin-3(2H)-one
(18) with phenyl magnesium bromide, which reacted only at the
5-chloro position, provided intermediate 19 in good yield. Cou-
pling of 18 with building block 8a provided the key intermediate
20 which was subjected to multiple parallel Suzuki couplings.
Deprotection of the t-butyl ester protecting group provided the test
compounds 21a–21h. Preparation of the 4-aryl analogues again
started from 18, but required four further steps to ensure that
the final Suzuki coupling reaction was directed towards the alter-
nate position. To achieve this, 18 was first protected with benzyl
chloride and then reacted with sodium methoxide to displace the
most active chloride substituent to give intermediate 22. A first
Suzuki coupling was then carried out to install a phenyl ring at
the 4-position. Conversion of the methoxy- group back to chloro-
with POCl3 produced intermediate 24 which was then debenzylat-
ed to provide 25. This intermediate was then set up for the usual
synthesis completion by coupling with intermediate 8a to provide
26, Suzuki coupling to install the required substituted 5-aryl
groups and finally t-butyl ester deprotection to provide test com-
pounds 21i–21m. As can be seen from Table 2, the 4,5-diphenylpy-
ridazinone series was in general somewhat more potent than the
3,4-diphenyl series. The SAR around each of the rings was similar
however, in that 3-substitution on the 5-phenyl ring and 4-substi-
tution on the 4-phenyl ring were favoured. Again, the combination
of two of the best substitution patterns (prepared via the second
route by coupling 5-fluorophenyl boronic acid to intermediate 22
followed by the usual synthesis completion steps) was essentially
additive with the product 21n among the most potent IP agonists
identified in this series. As with the 3,4-diphenyl series most of the
compounds appeared to be close to full agonists of both the rat and
human IP receptors but there was again a significant rightward
shift in activity for the rat receptor. All the compounds tested still
had significant activity in the human DP1 melanophore assay but
with generally lower potency than the 3,4-diphenyl series. This,
in combination with the improved IP activity, resulted in a modest
advantage for these compounds in terms of receptor selectivity.
In addition to its function as a vasodilator, prostacyclin can also
inhibit ADP-induced platelet aggregation in a dose-dependent
fashion. Hence, to confirm activity of a number of our more inter-
esting analogues in an assay with a more physiological readout, we
added a primary human platelet aggregation assay to our testing
repertoire. This assay was carried out as previously described,16
with aggregation induced by of 2.5 lM ADP in the presence or
absence of IP receptor agonists. In addition this assay had the
advantage of providing a read on whether there was any significant
effect of protein binding on the activity of the compounds as
the measurement is made in the presence of plasma proteins.
Table 1
Agonist activity of compounds 14a–14p for the cloned human and rat prostacyclin (IP) receptors
R1
R2
EC50 hIP*
(nM)
IA–
(%)
EC50 rIP*
(nM)
IA–
(%)
EC50 hDP1
(nM)
MRE269
79
51
60
47
88
97
63
77
97
87
88
80
67
89
78
80
81
91
n.d.
—
70
—
21
21
n.d.
17
25
44
61
6
n.d.
n.d.
1.7
n.d.
n.d.
31
3
14a
14b
14c
14d
14e
14f
14g
14h
14i
14j
14k
14l
14m
14n
14o
14p
H
H
H
H
H
H
H
H
969
2670
60
180
380
60
2190
>10,000
310
370
900
4-OMe
3-OMe
3-Me
96
72
97
62
85
—
99
93
99
93
76
75
107
102
3-F
3-Cl
360
350
3-OMe, 2-F
37
H
H
H
H
H
H
H
3-OMe
3-F
4-OMe
4-Me
4-F
4-OMe, 2-F
4-OMe, 3-F
4-Me
4-F
1720
1060
8.5
24
114
20
10
9.5
62
>10,000
2370
12.5
120
1500
112
51
28
620
3-OMe
3-OMe
11
24
*
EC50 for activation of the human or rat IP receptors determined in an HTRF cAMP assay. Values are the mean of at least 3 determinations with SD values <0.3 log units in
all cases.
–
Intrinsic activity (efficacy) relative to 1 lM iloprost as the positive control.
EC50 for the human PGD2 (DP1) receptor determined in a melanophore assay. Values are the mean of at least 3 determinations with SD values <0.3 log units in all cases.
Please cite this article in press as: Tran, T.-A.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.01.024

T.-A. Tran et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
5
O
O
O
(i)
(ii)
Cl
Cl
N
N
NH
N
NH
N
Cl
O
Cl
18
19
20
O
O
(iv)
(iii)
O
MeO
Cl
N
N
R1
O
22
N
N
O
(v)
O
OH
R2
21
O
MeO
N
N
23
(ix)
(vi)
O
O
O
Cl
Cl
N
Cl
NH
N
(viii)
(vii)
N
N
N
O
O
24
O
25
26
Scheme 3. Synthesis of 4,5-diphenylpyridazinones substituted either on the 4- or 5-phenyl ring. Reagents and conditions: (i) THF, PhMgBr (3 equiv, 0–25 °C, 68%); (ii) 8a,
K2CO3, DMF, 5 h, 87%; (iii) (a) R2-PhB(OH)2, Pd(Ph3P)4; (b) 4 M HCl dioxan; (iv) (a) BnCl, K2CO3, 50 °C; (b) NaOMe, MeOH, 88%; (v) PhB(OH)2, Pd(Ph3P)4, K2CO3, 91%; (vi) POCl3,
80 °C, 3 h, 73%; (vii) AlCl3, toluene, reflux, 89%; (viii) 8a, K2CO3, DMF, 40 °C, 5 h, 85%; (ix) (a) R1-PhB(OH)2, Pd(Ph3P)4; (b) HCl.
Table 2
Agonist activity of compounds 21a–n at the cloned human and rat prostacyclin (IP) receptors and activity in the human platelet aggregation assay
R1
R2
EC50 hIP*
(nM)
IA–
(%)
EC50 rIP*
(nM)
IA–
(%)
EC50 hDP1
(nM)
Human platelet IC50 (nM)
MRE269
79
51
75
101
82
93
73
106
87
78
91
100
80
n.d.
260
690
410
260
>3000
420
110
40
370
790
170
11
—
88
81
103
98
—
99
101
98
116
121
104
105
100
103
100
21
180
48
288
325
n.d.
191
n.d.
n.d.
220
87
21a
21b
21c
21d
21e
21f
21g
21h
21i
21j
21k
21l
21m
21n
(R)-21n
H
H
H
H
H
H
H
H
H
100
140
60
3-OMe
3-F
3-Cl
4-OMe
2-F
2,3-F2
2-F, 3-OMe
H
H
H
H
H
88
86
n.d.
n.d.
31
36
40
n.d.
n.d.
n.d.
n.d.
n.d.
42
1060
120
38
10
64
2-F
3-F
4-F
140
260
20
4
85
n.d.
n.d.
230
330
n.d.
62
4-OMe
3-F, 4-OMe
4-F
72
84
94
98
174
266
200
2,3-F2
2,3-F2
33
6
4-F
48
62
*
EC50 for activation of the human or rat IP receptors determined in an HTRF cAMP assay. Values are the mean of at least 3 determinations with SD values <0.3 log units in
all cases.
–
Intrinsic activity (efficacy) relative to 1 lM iloprost as the positive control.
EC50 for the human PGD2 (DP1) receptor determined in a melanophore assay. Values are the mean of at least 3 determinations with SD values <0.3 log units in all cases.
Fortunately for most compounds the shift between the cloned
human receptor and the endogenously-expressed receptor assays
(albeit with different functional readouts) was restricted to around
3–10 fold which we felt to be acceptable for such highly lipophilic
acids.
At this stage of the program we wanted to select a compound
for in vivo testing, partly to validate our in-house testing scheme
but also, as a result of the difference in activity for compounds at
the rat and human receptors, to provide a guideline as to what
level of rat IP receptor activity would be required to provide
Please cite this article in press as: Tran, T.-A.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.01.024

6
T.-A. Tran et al. / Bioorg. Med. Chem. Lett. xxx (2015) xxx–xxx
250
200
150
100
50
constituent enantiomers were prepared18 and only the most active
isomer was tested in the monocrotaline model. This more active
enantiomer (the (R)-configuration) was also modestly more selec-
tive for the IP receptor over DP1 (EC50 in cAMP assay 140 nM
[n = 9; 80% CI 73–151 nM]) and had significantly greater oral bio-
**
availability in rat [F = 76%; i.v. T = 2 h] than of any of the com-
½
**
pounds tested by this route and was thus selected for in vivo
evaluation. Unfortunately, (R)-21n did not show activity in the
monocrotaline model after oral administration at 10 or 30 mg/kg
twice daily despite the fact that it was able to achieve quite high
plasma concentrations. One possible explanation could be that
despite the high bioavailability, the significant right shift in the
agonist potency on going from the human to the rat receptor kept
this compound out of the range where in vivo activity could be
observed. Hence our next round of compound design focused on
reducing both lipophilicity and aromatic atom content to attempt
to improve in vivo potency as well as simplifying the synthesis
to improve cycle times.
(19)
(13)
(9)
(9)
(20)
0
**P<0.001 vs. MCT + Vehicle
Figure 2. Effect of compound 21g on right ventricular weight in the rat model of
In summary we have designed and synthesized two related ser-
ies of prostacyclin receptor agonist compounds that showed excel-
lent receptor potency and efficacy, albeit with moderate selectivity
over the DP1 receptor. Examples from this series showed in vivo
activity after SC but not PO dosing in rats. The subsequent re-
design of further compounds that activate the IP receptor and
had oral activity in vivo, will be presented in due course.
PAH.
in vivo activity. A number of compounds were tested for metabolic
stability in rat and human microsomes and several were found to
be sufficiently stable to allow for sub-cutaneous administration.
None of the compounds containing a methoxy substituent on one
of the aromatic rings however were able to pass this hurdle, as
rather rapid metabolism took place that we later were able to show
was via demethylation to the inactive phenol. Thus we first
selected 21g to test in vivo as a result of its good human potency
and moderate shift in both the rat IP assay and the human platelet
assay.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2015.01.
024.
Monocrotaline (MCT),
a macrocyclic pyrrolizidine alkaloid
derived from Crotalaria spectabilis, induces a syndrome in rats
characterized by a significant increase in pulmonary artery pres-
sure, medial thickness of small pulmonary arteries, and right ven-
tricular hypertrophy, which occurs three weeks after MCT
exposure at a standard dose of 60 mg/kg. Despite uncertainty
about the exact mechanism by which MCT initiates lung injury
in rats, this model continues to be the one most commonly used
to investigate potential therapies for PAH.17
References and notes
1. Narumiya, S.; Sugimoto, Y.; Ushikubi, F. Physiol. Rev. 1999, 79, 1193.
2. Moncada, S.; Gryglewski, R.; Bunting, S.; Vane, J. R. Nature 1976, 263, 663.
3. Vane, J. R.; Botting, R. M. Am. J. Cardiol. 1995, 75, 3A.
4. Wise, H.; Jones, R. L. Trends Pharmacol. Sci. 1996, 17, 17.
5. Rubanyi, G. M. J. Cardiovasc. Pharmacol. 1993, 22, S1.
6. Christman, B. W.; McPherson, C. D.; Newman, J. H.; King, G. A.; Bernard, G. R.;
Groves, B. M.; Loyd, J. E. N. Eng. J. Med. 1992, 327, 70.
7. Rich, S. Chest 1998, 114, 237S.
We determined the half-life of 21g to be around 2.5 h in rat
after sub-cutaneous administration and so we decided to dose
the compound twice daily in the MCT model for 21 days (30, 10
and 1 mg/kg SC). As can be seen in Figure 2, 21 g had a dose-depen-
dent effect on right ventricular weight, showing essentially no
increase over the sham controls at 30 mg/kg SC twice daily. In
addition to a good effect on preventing ventricular weight increase,
measurement of right ventricular systolic pressure in the 10 mg/kg
dose group also showed a clear improvement over the vehicle trea-
ted group.
8. Rubin, L. J.; Groves, B. M.; Reeves, J. T.; Frosolono, M.; Handel, F.; Cato, A. E.
Circulation 1982, 66, 334.
9. Kuwano, K.; Hashino, A.; Asaki, T.; Hamamoto, T.; Yamada, T.; Okubo, K.;
Kuwabara, K. J. Pharmacol. Exp. Ther. 2007, 322, 1181.
10. Hattori, K.; Tanaka, A.; Okitsu, O.; Tabuchi, S.; Taniguchi, K.; Nishio, M.;
Koyama, S.; Higaki, M.; Seki, J.; Sakane, K. Bioorg. Med. Chem. Lett. 2005, 20,
3091.
11. Sun, C.; Huang, Y.; Sitkoff, D. F.; Lee, T.; Ewing, W. U.S. Pat. Appl. Publ. US
20,060,287,324, 2006.
12. Tran, T.-A.; Hayashi, R.; Ibarra, J. B.; Ullman, B.; Zou, N. PCT Int. Appl. WO
2,010,077,275, 2010.
13. Sotelo, E.; Fraiz, N.; Yaez, M.; Terrades, V.; Laguna, R.; Cano, E.; Ravia, E. Bioorg.
Med. Chem. 2002, 10, 2873.
14. Gabriel, D.; Vernier, M.; Pfeifer, M. J.; Dasen, B.; Tenaillon, L.; Bouhelal, R. Assay
Drug Dev. Technol. 2003, 1, 291.
15. Chen, G.; Way, J.; Armour, S.; Watson, C.; Queen, K.; Jayawickreme, C. K.; Chen,
W. J.; Kenakin, T. Mol. Pharmacol. 2000, 57, 125.
16. Adams, J. W.; Ramirez, J.; Ortuno, D.; Shi, Y.; Thomsen, W.; Richman, J. G.;
Morgan, M.; Dosa, P.; Teegarden, B. R.; Al-Shamma, H.; Behan, D. P.; Connolly,
D. T. Eur. J. Pharmacol. 2008, 586, 234.
17. Gomez-Arroyo, J. G.; Farkas, L.; Alhussaini, A. A.; Farkas, D.; Kraskauskas, D.;
Voelkel, N. F.; Bogaard, H. J. Am. J. Physiol. Lung Cell. Mol. Physiol. 2012, 302,
L363.
18. The two enantiomers were prepared as shown in Scheme 3 from the separated
enantiomers of 8b, which in turn were prepared by fractional crystallization of
the (S)-camphanic acid ester of 7 (the S–S isomer crystalized from ethanol as a
white solid whereas the S–R isomer was an oil). The separated esters were then
hydrolysed (KOH, MeOH/water) and the alcohol functionality converted to a
mesylate or tosylate leaving group.
Having shown that an IP agonist discovered using our screening
approach had in vivo activity, we decided to try to demonstrate
activity after oral administration in the MCT model with a com-
pound from this series. However, further testing of 21g at the
hDP1 receptor, this time in a cAMP assay showed the EC50 in this
platform to be 400 nM [n = 9; 80% CI 364–480 nM] giving a selec-
tivity ratio of only around 10-fold over the IP receptor which we
deemed not to be selective enough to progress. Thus we decided
to select another compound to examine whether these analogues
could be efficacious after oral administration. Of the other com-
pounds in hand, 21n was very similar to 21g in most respects with
the added advantage that it did appear to be more selective
over DP1. Before testing this compound in vivo however, the two
Please cite this article in press as: Tran, T.-A.; et al. Bioorg. Med. Chem. Lett. (2015), http://dx.doi.org/10.1016/j.bmcl.2015.01.024