Peri-substituted hexahydro-indolones as novel, potent and selective human EP3 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2008.12.027

A series of peri-substituted [4.3.0] bicyclic non-aromatic heterocycles have been identified as potent and selective hEP₃ receptor antagonists. These molecules adopt a hair-pin conformation that overlaps with the endogenous ligand PGE₂

Full text of DOI:10.1016/j.bmcl.2008.12.027

Bioorganic & Medicinal Chemistry Letters 19 (2009) 778–782
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Peri-substituted hexahydro-indolones as novel, potent and selective human EP3
receptor antagonists
Matthew O’Connell a, Wayne Zeller a, James Burgeson a, Rama K. Mishra a, Jose Ramirez b, Alex S. Kiselyov a,
Þorkell Andrésson b, Mark E. Gurney a,b, Jasbir Singh a,
*
a deCODE Chemistry, Inc., Medicinal Chemistry, 2501 Davey Road, Woodridge, IL 60517, USA
b deCODE Genetics, Inc., Reykjavik, Iceland
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of peri-substituted [4.3.0] bicyclic non-aromatic heterocycles have been identified as potent and
selective hEP3 receptor antagonists. These molecules adopt a hair-pin conformation that overlaps with
the endogenous ligand PGE2 and fits into an internally generated EP3 pharmacophore model. Optimized
compounds show good metabolic stability and improved solubility over their corresponding bicyclic aro-
matic analogs.
Received 12 October 2008
Revised 4 December 2008
Accepted 4 December 2008
Available online 10 December 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Prostanoid receptor antagonists
EP3 receptor antagonists
Peri-substituted indoles
Acylsulfonamide
Hexahydro-indolones
Prostanoids play an essential role in vascular homeostasis,
including regulation of platelet function. These molecules act
through the specific membrane receptors belonging to the super-
family of G-protein coupled receptors (GPCRs). Among these prosta-
noids, thromboxane A2 (TxA2) is a potent stimulator of platelet
aggregation, whereas prostaglandin I2 (PGI2) inhibits its activation.
Another signaling molecule, prostaglandin E2 (PGE2) binds preferen-
tially to the EP family of receptors, namely EP1–4.1 It has been shown
that EP3 receptor on the platelet is required for PGE2 potentiation of
coagonist signaling. Studies utilizing EP3 KO mice showed that the
stimulatory effects of PGE2 on platelet aggregation are exerted spe-
cifically through EP3 receptor.2 PGE2 has been reported to have bi-
phasic effect on the platelet response. Specifically, it potentiates
platelet aggregation at low concentrations and inhibits this effect
at higher concentrations.3 In addition to its involvement in platelet
function, PGE2 also plays a key role in the regulation of ion trans-
port,4 smooth muscle contraction of the GI tract, acid secretion,
uterine contraction during fertilization and implantation,5 fever
generation, and hyperalgesia.6 Development of specific antagonists
of EP3 receptor are of interest as anti-thrombotic agents.
antagonist activity in both cellular and secondary rat platelet
aggregation assays.7,8 However, these molecules displayed ele-
vated clogP, clogD and possessed low aqueous solubility. Our pre-
liminary structure–activity relationship (SAR) demonstrated that
attempts to modify the solubility of compounds 1 and 2 through
incorporation of either polar or basic functionalities in the bicyclic
template or as Ar1/Ar2 substituents resulted in a significant reduc-
tion in affinity for EP3 receptor (data not shown).9 Another ap-
proach undertaken to improve the solubility of the [4.3.0]
bicyclic framework involved altering the planarity of compounds
1 and 2, and thus reducing the potential of these compounds for
intermolecular
p-stacking. We proposed that the [4.3.0] bicyclic
system of the hexahydro-indolone analog 3 would possess more
favorable physico-chemical parameters than series 1 and 2, and
thus display reduced lipophilicity and improved solubility/formu-
lation properties. Furthermore, this [4.3.0] bicyclic core should pro-
vide for the proper arrangement of the requisite pharmacophoric
features, in a manner analogous to our previously reported bicyclic
heteroaromatic series,7,8 to provide active EP3 analogs (Fig. 1).
Comparison of the energy-minimized10 structure 3 with the previ-
ously reported indole 1 and 2-oxoindole [dihydroindolone] 2
derivatives, showed good spatial and electronic overlap (Fig. 1).
The selection of Ar1 and Ar2 substituent groups was also geared to-
wards improvement of solubility. This report describes the design,
synthesis and SAR for the non-aromatic [4.3.0] bicyclic series 3.
The hexahydro-indolone analogs 3a were readily accessible
from (1-methyl-2-oxo-cyclohexyl)-acetic acid (4a),11 as shown in
Recently, we reported two series of peri-substituted bicyclic
heteroaromatic molecules7,8 which featured sound potency to-
wards the hEP3 receptor and high prostanoid receptor specificity.
Derivatives of indoles 1 and dihydroindolones 2 also showed good
* Corresponding author. Tel.: +1 630 783 4915; fax: +1 630 783 4646.
E-mail address: jsingh@decode.com (J. Singh).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.12.027

M. O’Connell et al. / Bioorg. Med. Chem. Lett. 19 (2009) 778–782
779
CH3
CH3
F
b,c
O
O
O
O
N
N
N
N
N
Ar1
Ar1
Ar1
Ar1
Ar1
O
S
CH3
O
Ar2
(8)
(10)
H
OMe
O
O
N
H
H
H
O
N
S
O
N
S
O
N
S
O
N
O
CH3
(a)
O
O
a
CH3
Ar1
Ar2
(1)
Ar2
Ar2
(2)
b,c
O
(3)
O
O
O
(b)
O
N
N
O
OMe
H
H
Ar1
Ar1
O
S
O
Ar2
(6a)
(9)
O
N
H
(11)
O
OMe
Scheme 2. Reagents and conditions: (a) i—H2, 10% Pd/C, EtOH; ii—silica-gel column
chromatography; (b) NaOH, H2O, MeOH, THF; (c) Ar2SO2NH2, EDCI, DMAP, CH2Cl2.
Similar to the optimized indole 1a and dihydroindolone 2a
derivatives (Ar1 = 2-naphthyl, Ar2 = 4,5-dichloro-thiophenyl), mol-
ecule 3aa showed good activity in the hEP3 receptor binding assay
[hEP3 IC50 = 7.0, 1.5 and 6.8 nM for compounds 1a, 2a and 3aa,
respectively] and low IC50 shift in the presence of human serum
[fold-shift = 7.0, 7.6 and 7.4 for compounds 1a, 2a and 3aa, respec-
tively] indicating favorable plasma protein binding properties.14
Encouraged by these results, we undertook a detailed SAR study
for Ar1 and Ar2 groups in this [4.3.0] bicyclic series (Table 1). A
number of substituted phenyl derivatives at Ar1 (3aa–ah) were
evaluated. With the exception of 3ac, all analogs afforded both
good affinity for hEP3 and low-to-modest plasma protein binding.
4,5-Dichloro substituents on the thiophene sulfonamide portion
(Ar2 substituents) consistently improved activity of the resulting
molecules [see Table 1, 3ac vs 3aj (25-fold)]. The Ar2 group as a
phenyl substituted with two halogens (chloro or fluoro substitu-
ents, 3ay and 3ar) provided good hEP3 affinity. However, these
molecules also displayed large fold shift for hEP3 binding in the
presence of human serum (3al, 3am, 3ao, 3ay, 3az). Detailed data
analysis did not show a direct correlation of their lipophilicity
(clogD7.4) and fold-shift in IC50 values. For example, analogs 3an
versus 3as and 3at versus 3ap showed essentially similar clogD7.4
values while these gave very different IC50 in the presence of plas-
ma protein components.15 This effect of substitution pattern rather
than simple clogD7.4 (or clogP) was further highlighted by compar-
ison of Ar2 isomers which display essentially identical activity in
the binding assay while showing very different PPB effect. For
example 3,4- versus 3,5-difluoro phenyl groups Ar2 displayed sim-
ilar affinities (within 2-fold) for hEP3 receptor but different plasma
protein fold-shifts (7.8/33.7 for 3aq/3am and 4.2/62.4 for 3ap/3ao
pairs). Interestingly, derivative 3az with higher clogD7.4 value ver-
sus 3al afforded 8-fold lower protein binding. Similar to the indole
series,7,8 incorporation of meta-methoxy substituents in Ar1 also
provided very potent analogs, but with varying plasma protein
binding effect (Table 1, 3ah, 3ap, 3ao and 3ax). Fluoro substituted
Ar1 and Ar2 groups (3as and 3aw) exhibited sound hEP3 affinity,
low fold-shift in the presence of plasma proteins and good meta-
bolic stability in a number of species.16
Figure 1. Overlay of AM1 energy minimized10 indole analog (1a, green), indolone
(2a, orange) and hexahydro-indolone (3aa purple), where Ar1 = 2-naphthyl and
Ar2 = 4,5-dichloro-2-thiophene.
Scheme 1. Introduction of the first point of diversity in these ana-
logs involved cyclization of 4a with various benzyl amines using a
modification of the protocol reported by Padwa et al. and afforded
compounds 5a.12 Bromination of hexahydro-indolones 5a with
NBS provided the respective vinyl bromide intermediates. These
were converted to methyl acrylates 6a (R2 = CH3) via standard
Heck coupling protocols. Saponification of 6a provided the key
intermediate acid 7a which was reacted with arylsulfonamide
using EDCI in the presence of DMAP to provide the acylsulfona-
mides 3a.13
Analogs 3b, featuring an ester functionality at the quaternary
bridgehead carbon (R = CO2CH3), were accessed via the same syn-
thetic route (Scheme 1) using 4b for condensation with a benzylic
amine followed by Heck coupling with tert-butyl acrylate. The
orthogonally protected intermediate 6b (R2 = t-Bu), was deprotec-
ted with excess TFA in CH2Cl2 to yield the key mono-acid interme-
diate 7b.
Hydrogenation of the methyl ester intermediate 6a (R2 = CH3) in
the presence of palladium on charcoal provided a mixture of tetra-
substituted olefin and saturated analogs 8 and 9, respectively.
These esters were separated by chromatography, saponified and
reacted with substituted arylsulfonamides to afford compounds
10 and 11, respectively (Scheme 2).
R
R
R
b,c
O
d
a
O
O
OH
N
N
O
Ar1
Ar1
(5)
(4)
(6)
O-R2
(a) R = CH3
(b) R = CO2CH3
O
An additional approach to improve aqueous solubility of these
bicyclic systems was to replace the angular methyl group with a
hydrophylic substituent such as –CH2OH. The neopentyl nature
of this hydroxyl group should have low potential for phase-II
metabolism. Analogs containing a carbomethoxy group as an inter-
mediate to the –CH2OH derivatives were prepared. However, at-
tempted reduction of the angular ester group to corresponding
hydroxyl methyl, under a variety of conditions, was not successful.
The analogs containing the quaternary carbomethoxy group exhib-
ited reduced hEP3 activity in the presence and absence of 10% hu-
man serum in the receptor binding assay relative to the angular
R
R
O
O
N
N
e
Ar1
Ar1
O
S
O
Ar2
O
N
(7)
(3)
O
OH
H
Scheme 1. Reagents and conditions: (a) Ar1CH2NH2, xylenes, 140 °C; (b) NBS,
CH2Cl2; (c) acrylate ester, Pd(OAc)2, (o-tolyl)3P, TEA, DMF; (d) NaOH, H2O, THF,
MeOH or TFA, CH2Cl2; (e) ArSO2NH2, EDCI, DMAP, CH2Cl2.

780
M. O’Connell et al. / Bioorg. Med. Chem. Lett. 19 (2009) 778–782
methyl derivatives; compare pairs 3af versus 3ba and 3av versus
3bc (Tables 1 and 2).
Reduction of the exocyclic double bond in 3aw yielded
10w. The molecule retained affinity for the hEP3 receptor (Ta-
ble 2, compare 10w versus 3aw) but displayed poor metabolic
stability in rat liver microsomal preparations with 4% parent
remaining after 30 min. Further reduction of the endocyclic
olefin in 10w provided octahydro-indolone analog 11w that
was 300-fold less active in hEP3 binding assay compared with
the hexahydro analog 10w. 1H NMR analyses showed the com-
Table 1
Biological activity and metabolic stability of hexahydro-indolones
O
Ar1
N
(3)
H3C
O
N
O
S
O
Ar2
H
Ar1
Ar2
hEP3 IC50 (nM)
hEP3 fold-shiftb
clogD7.4
Metabolic stabilityd
a
c
Compound
Rat
Human
3aa
3ab
3ac
3ad
3ae
3af
3ag
3ah
3ai
2-Naphthyl
4,5-Dichloro thiophen-2-yl
4,5-Dichloro thiophen-2-yl
4,5-Dichloro thiophen-2-yl
4,5-Dichloro thiophen-2-yl
4,5-Dichloro thiophen-2-yl
4,5-Dichloro thiophen-2-yl
4,5-Dichloro thiophen-2-yl
4,5-Dichloro thiophen-2-yl
5-Chloro thiophen-2-yl
2-Thiophenyl
4-Fluoro phenyl
3,5-Difluoro phenyl
3,5-Difluoro phenyl
3,5-Difluoro phenyl
3,5-Difluoro phenyl
3,4-Difluoro phenyl
3,4-Difluoro phenyl
3,4-Difluoro phenyl
6.8
4.4
2.3
2.2
9.5
3.5
3.8
6.4
13.4
57.2
15.2
9.2
3.8
84.1
5.6
12.9
3.4
3.4
9.4
2.5
4.3
3.7
7.7
5.6
9.8
5.6
7.4
6.6
38.4
16.3
8.2
17.7
3.0
4.8
4.9
4.8
4.9
4.3
3.8
3.7
3.8
3.6
4.0
3.0
3.8
4.3
4.2
3.1
3.0
2.8
3.9
4.1
2.9
2.9
3.8
4.0
2.9
2.7
5.3
4.2
54
69
60
ND
18
22
50
64
68
79
64
ND
ND
ND
ND
42
ND
ND
82
ND
50
37
89
84
62
45
64
80
58
ND
67
80
74
62
75
78
70
ND
ND
ND
ND
76
ND
ND
79
ND
49
62
75
76
70
68
2,3-Dichloro phenyl
2,4-Dichloro phenyl
3-Chloro phenyl
3-Fluoro phenyl
3,4-Difluoro phenyl
4-Fluoro phenyl
3-Methoxy phenyl
2,4-Dichloro phenyl
2,4-Dichloro phenyl
2,4-Dichloro phenyl
3,4-Difluoro phenyl
2,3-Dichloro phenyl
4-Fluoro phenyl
3-Methoxy phenyl
3-Methoxy phenyl
2,3-Dichloro phenyl
2,4-Dichloro phenyl
3,4-Difluoro phenyl
4-Fluoro phenyl
2,3-Dichloro phenyl
2,4-Dichloro phenyl
3,4-Difluroro phenyl
3-Methoxy phenyl
2,4-Dichloro phenyl
3,4-Difluoro phenyl
6.9
3aj
3ak
3al
25.2
21.2
217.2
33.7
5.4
62.4
4.2
7.8
18.0
3.2
15.9
5.8
7.0
9.3
3am
3an
3ao
3ap
3aq
3ar
3as
3at
3au
3av
3aw
3ax
3ay
3az
3,4-Difluoro phenyl
3,4-Difluoro phenyl
2,4,5-Trifluoro phenyl
2,4,5-Trifluoro phenyl
2,4,5-Trifluoro phenyl
2,4,5-Trifluoro phenyl
3,5-Dichloro phenyl
6.2
61.2
26.8
3,5-Dichloro phenyl
a
Experimental IC50 values from displacement binding analysis with a minimum of three experiments per value. Displacement binding was assessed with [3H]-PGE2 for
human EP3 receptor membrane preparations in buffer.
b
Fold-shift = (IC50 in the presence of 10% human serum)/(IC50 in buffer (a)).
The values shown are from ACDLabs, version 9.0.
Percent parent compound remaining after 30-min incubation with respective liver microsomal preparations, as determined by LC/MS/MS. ND, not determined.
c
d
Table 2
Bioactivity of hexa- and octahydro-indolones
O
Ar1
N
R
P
Q X
O
N
O
O
S
Y
Ar2
H
Compound
Ar1
Ar2
R
P–Q
X–Y
hEP3 IC50 (nM)
PPB fold-shift
10w
11w
3aw
3ba
3bb
3bc
3,4-Difluoro phenyl
3,4-Difluoro phenyl
3,4-Difluoro phenyl
3,4-Difluoro phenyl
2,4-Dichloro phenyl
2,4-Dichloro phenyl
2,4,5-Trifluoro phenyl
2,4,5-Trifluoro phenyl
2,4,5-Trifluoro phenyl
4,5-Dichloro-thiophen-2-yl
3,4-Difluoro phenyl
Me
Me
Me
CO2Me
CO2Me
CO2Me
C@C
CHCH
C@C
C@C
C@C
C@C
CH2CH2
CH2CH2
CH@CH
CH@CH
CH@CH
CH@CH
1.8
556.9
7.7
29.9
20.6
11.2
33.2
25.7
9.3
308.5
22.0
47.1
2,4,5-Trifluoro phenyl

M. O’Connell et al. / Bioorg. Med. Chem. Lett. 19 (2009) 778–782
781
Table 3
Aqueous solubility, clogD and HPLC retention times for selected compounds
O
O
Ar1
Ar1
Ar1
N
N
N
H3C
O
O
N
O
N
O
O
N S
O
S
O
S
O
O
F
Ar2
Ar2
Ar2
H
H
(3a)
H
(1b-d)
(2b-c)
Compound
Ar1
Ar2
Aqueous solubility (mg/mL)b
clogD7.4
HPLC RT min17
1b
2b
3aw
3,4-Difluoro phenyl
3,4-Difluoro phenyl
3,4-Difluoro phenyl
2,4,5-Trifluoro phenyl
2,4,5-Trifluoro phenyl
2,4,5-Trifluoro phenyl
<0.04
1.58
4.4
3.1a
2.9
27.79
26.32
23.54
>3.0c
1c
2c
3as
3,4-Difluoro phenyl
3,4-Difluoro phenyl
3,4-Difluorophenyl
3,4-Difluoro phenyl
3,4-Difluoro phenyl
3,4-Difluoro phenyl
ND
ND
ND
4.5
3.2a
2.9
27.83
22.09
23.58
1d
3al
3,4-Difluoro phenyl
3,4-Difluoro phenyl
3,5-Difluoro phenyl
3,5-Difluoro phenyl
ND
ND
4.7
2.6
27.93
23.73
1e
3ax
3-Methoxy phenyl
3-Methoxy phenyl
2,4,5-Trifluoro phenyl
2,4,5-Trifluoro phenyl
ND
ND
4.3
2.7
27.56
22.89
1f
3an
4-Fluoro phenyl
4-Fluoro phenyl
3,5-Difluoro phenyl
3,5-Difluoro phenyl
ND
ND
4.8
3.1
27.67
23.23
1g
3ag
4-Fluoro phenyl
4-Fluoro phenyl
4,5-Dichloro thiophen-2-yl
4,5-Dichloro thiophen-2-yl
ND
ND
5.4
3.8
30.25
26.10
a
clogD7.4 for the keto form.
b
The data is for solubility in 50 mM PSB buffer (pH 7.4) and should represent thermodynamic solubility, as the samples were stirred overnight, filtered and analyzed by
HPLC and concd determined using a 5-point standard curve.
c
The sample was homogenous, so the actual solubility is higher than the value shown.
pound 11w to have syn(a):syn(b) orientations, as shown in
Derivatives containing the acryl acylsulfonamide functionality
featured the best activity and sound metabolic stability (Table 1).
Scheme 2.18
A selected series of the more potent analogs were examined
against a panel of prostanoid receptors to assess their EP3 selectiv-
ity. All compounds tested displayed >1000Â selectivity (Table 4) in
radioligand displacement binding assays against other PGE2 (EP1,
EP2 and EP4) and against the FP receptors.
The hexahydro-indolone analogs with good activity in the hEP3
radioligand displacement binding assay and exhibiting low poten-
tial for PPB were shown to exhibit potent and full antagonist activ-
ity in the functional assay.19 This data for selected analogs is shown
in Table 5.
An a,b unsaturated amide is not commonly found in registered ther-
apeutics20 as a structural feature due to its perceived potential to un-
dergo Michael addition. However, the nature of the double bond in
these compounds (e.g., 3aw) suggests that this should not be a liabil-
ity as the acylsulfonamide is a carboxylic acid mimetic.7 Moreover,
the aryl acrylic acylsulfonamide anion present at physiological
pH21 is likely to feature neither chemical nor in vivo reactivity. Rel-
evant synthetic papers22 point out that the
a,b unsaturated amide
functionality is a sub-optimal Michael acceptor.
The hexahydro-indolone series showed reduced retention times
on C18 reverse phase HPLC, compared with their aromatic counter-
parts 1 and 2, suggesting improved hydrophilicity and solubility.
This is corroborated by (a) aqueous solubility data for a series of in-
dole, dihyroindolone and hexahydro-indolone analogs (set-1:
1a ? 2b ? 3aw;) as shown in Table 3 and (b) a Scatter plot of the
clogD7.4 versus HLPC retention time provided a correlation coeffi-
cient of r2 = 0.85 for the entire data shown in Table 3. For example,
compound 3aw displayed >3 mg/mL solubility in PBS buffer at pH
7.4. Thismoleculeshowed goodfunctionalactivityasit blockedinhi-
bition of forskolin-stimulated production of cAMP by PGE2 in CHO-
K1 cells expressing human EP3 with IC50 = 3.5 nM. Several represen-
tative molecules from this class are being further evaluated in the
panel of functional assays in order to select an optimized lead
candidate.
Table 4
Prostanoid receptor activity for potent hexahydro-indolone derivatives
Compound
hEP1
hEP2
hEP4
hFP
3aa
3ag
3ah
3an
3ap
3as
3au
3av
3aw
6.9
IA
IA
IA
1.4
IA
IA
11.6
IA
IA
IA
IA
IA
IA
2.3
IA
IA
13.9
IA
IA
IA
IA
7.5
7.7
IA
20.8
6.2
10.2
27.4
6.1
4.2
6.4
16.4
7.4
7.6
3.8
IC50 (lM) values reported are from receptor binding assays.
IA, inactive; these analogs gave essentially no significant inhibition at 20
lM,
In conclusion, we described both synthesis and SAR of peri-
substituted hexahydro-indolones featuring good overlap with the
previously reported bicyclic series. These series led to the identifi-
cation of analogs exhibiting high affinity for the human EP3 recep-
tor, favorable selectivity across the panel of prostanoid receptors,
full antagonism at the human EP3 receptor in cellular assay and
good metabolic stability across multiple species. In addition, hexa-
hydro-indolones displayed improved aqueous solubility suitable
for further preclinical development.
maximum assay concentration, in this assay.
Table 5
Antagonist activity on CHO-cells expressing human EP3 receptors
Compound
IC50 (nM)
3ab
3af
3ah
3aq
3au
3aw
3az
1.7
5.2
5.8
3.5
4.5
3.5
3.2

782
M. O’Connell et al. / Bioorg. Med. Chem. Lett. 19 (2009) 778–782
phosphine (0.3 equiv). The reaction mixture was heated at 100 °C for 16 h,
References and notes
and then allowed to cool to room temperature. The reaction mixture was
filtered through Celite, the Celite was washed with CH2Cl2, organic layer was
washed with water, brine and dried over anhydrous MgSO4. Organic layer was
concentrated in vacuo. The residue was purified via silica gel chromatography,
using 15% hexanes/CH2Cl2 to obtain the acrylate ester (6a, R2 = CH3).
Hydrolysis of methyl esters with aqueous NaOH (3 equiv) in THF/MeOH
(2:1) provided, after acidification (with 1 N HCl to pH 3) and extractive work-
up with EtOAc, the desired acrylic acid (6a, R2 = H) in 39% yield for the two
steps. Subsequent coupling with 3,4-diflurorbenzenesulfonamide (1.2 equiv)
and DMAP (2.4 equiv) in CH2Cl2 using EDCI (2 equiv) was carried out at room
temperature. The reaction mixture was washed with 1 N HCl, water, brine,
dried over MgSO4 and concentrated in vacuo. The residue was triturated with
CH2Cl2/hexanes to obtain the desired product 3as in 28% yield. 1H NMR
(CDCl3) 1.19 (s, 3H), 1.57 (m, 1H), 1.87 (m, 3H), 2.19 (d, J = 6.8 Hz, 2H), 2.44 (d,
J = 1.2 Hz, 2H), 4.75 (d, J = 16.4 Hz, 1H), 5.17 (d, J = 16.0 Hz, 1H), 5.53 (d,
J = 14.8 Hz, 1H), 7.04 (m, 3H), 7.35 (ddd, J = 16.8, 9.2, 7.6 Hz, 1H), 7.72 (d,
J = 14.8 Hz, 1H), 7.88 (m, 2H), 7.93 (ddd, J = 9.2, 7.2, 2.4 Hz, 1H). LC/MS (95%)
ESI-Calcd 522.5 m/z. Found: 522 m/z.
1. (a) 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. Biochim. Biophys. Acta 2000, 1483, 285; (b)
Kiryiama, A.; Ushukubi, K.; Kobayashi, T.; Hirata, M.; Sugimoto, Y.; Narumiya, S.
Br. J. Pharmacol. 1997, 122, 217.
2. Fabre, J.; Nguyen, M.; Athirakul, K.; Coggins, K.; McNeish, J. D.; Austin, S.; Parise,
L. K.; FitzGerald, G. A.; Coffman, T. M.; Koller, B. H. J. Clin. Invest. 2001, 107, 603.
3. Gross, S.; Tilly, P.; Hentsch, D.; Vonesch, J. L.; Fabre, J. E. J. Exp. Med. 2007, 204,
311.
4. Garcia-erez, A.; Smith, W. L. J. Clin. Invest. 1984, 74, 63.
5. Krall, J. F.; Barrett, J. D.; Jamgotchian, N.; Korenman, S. G. J. Endocrinol. 1984,
102, 329.
6. (a) Xin, L.; Geller, E. B.; Bastepe, M.; Raffa, R. B.; Mao, G.-F.; Ashby, B.; Adler, M.
W. J. Therm. Biol. 2000, 25, 77; (b) Minami, T.; Nishihara, I.; Ito, S.; Hyodo, M.;
Hayaishi, O. Br. J. Pharmacol. 1994, 112, 735; (c) Reinold, H.; Seifollah, A.; Ulrike,
D.; Beate, L.; Heindl, C.; Hamza, M.; Pahl, A.; Brune, K.; Shuh, N.; Muller, U.;
Zeilhofer, H. J. Clin. Invest. 2005, 115, 673.
7. Singh, J.; Zeller, W.; Zhou, N.; Hategen, G.; Mishra, R.; Polozov, A.; Yu, P.; Onua,
E.; Zhang, J.; Zembower, D.; Kiselyov, A.; Ramírez, J.; Sigthorsson, G.; Bjornsson,
J.; Thorsteinnsdottir, M.; Andrésson, T.; Bjarnadottir, M.; Magnusson, O.;
Stefansson, K.; Gurney, M. ACS Chem. Biol., submitted for publication.
8. Zhou, N.; Zeller, W.; Keyvan, M.; Krohn, M.; Anderson, H.; Mishra, R.; Zhang, J.;
Onua, E.; Ramirez, J.; Palsdottir, G.; Halldorsdottir, G.; Andresson, T.; Gurney,
M.; Singh, J. Bioorg. Med. Chem. Lett. 10.1016/j.bmcl.2008.11.007.
14. Based on the data previously reported for the cinnamyl acylsulfonamide series
Juteau, H.; Gareau, Y.; Labelle, M.; Sturino, C. F.; Sawyer, N.; Tremblay, N.;
Lamontagne, S.; Carriere, M. C.; Denis, D.; Metters, K. M. Bioorg. Med. Chem.
2001, 9, 1977. we conducted our in vitro assays in the absence and presence of
human serum to evaluate potential for plasma protein binding (PPB).
15. The shift in plasma protein was measured in the presence of mouse and human
serum and human-serum albumin, and none show a simple correlation with
clogP or clogD.
9. For example, replacement of indole with 5-azaindole core led to over 100-fold
drop in activity versus the corresponding indole analog. Incorporation of (e.g.
2-pyridyl methyl as Ar1 substituents also resulted in analogs with very poor
activity in the hEP3 receptor binding assay.
16. Compound 3aw also showed good metabolic stability with dog and monkey
liver microsomal preparations providing 87% and 54% parent remaining at
30 min, respectively.
17. Waters symmetry C18 column, 4.6 mm  250 mm, 5
lm; flow rate: 1.0 mL/
min; mobile phase A: water (0.05% TFA), linear gradient from 95% A to 95% B
over 35 min.
10. For compounds 1a, 2a and 3aa, 2D–3D structure conversion was performed
using CONCORD 6.0 followed by energy minimization using MMFF94 force
field with conjugate gradient method using SYBYL7.0. Gasteiger–Huckel
charges were assigned and then each minimized structure was subjected to
full conformational search using systematic search (SS) varying all rotatable
bonds by 10ꢀtorsion increment and only two conformation of the olefin (cis and
trans) were allowed. The lowest energy conformer for each molecule thus
obtained was subsequently subjected to AM1 semi-empirical SCF MO energy
minimization using ‘MMOK’ and ‘Precise’ for augmenting convergence criteria.
Each geometry optimized structure with AM1 charges was finally used for
alignment based on the electrostatic charge similarity index principle as
reported by Burt, C.; Richards, W. G.; Huxley, P. The application of molecular
similarity calculations. J. Comp. Chem. 2004, 11, 1139. The resulting overlap of
1a, 2a and 3aa is shown in Figure 1.
18. All four possible isomers, syn(a)–syn(b), syn(a)–anti(b), anti(a)–anti(b) and
anti(a)–syn(b) were energy minimized using semi-empirical, AM1
Hamiltonian. This data indicated that the syn(a)–syn(b) and syn(a)–anti(b) to
have relatively close energy (<1 kcal/mol apart); while the two anti(a)–syn(b)
and anti(a)–anti(b) isomers have much higher energy (ꢀ7–7.5 kcal/mol). The
1D and 2D 1H NMR analyses of 11w supports the syn(a)–syn(b) assignments.
These data support that the isomer isolated to be low energy isomer. Also,
analysis of the two syn(a) isomers with the EP3 antagonist pharmacophore
model is consistent with the observed loss in EP3 binding.
19. CHO-K1 cells stably expressing the hEP3 receptor were treated with
increasing concentrations of a test compound for 10 min at 37 °C in the
presence of
5 lM forskolin and 5 nM PGE2. Control cells treated with a
11. Asselin, A. A.; Humber, L. G.; Dobson, T. A. US Patent 4,057,559, 1975.
12. Rashatasakhon, P.; Ozdemir, A.; Willis, J.; Padwa, A. Org. Lett. 2004, 6, 917.
13. Synthesis of analog 3as, as a representative example: A solution of 4a11 (1 equiv)
and the 3,4-difluorobenzyl amine (1 equiv) in m-xylene was refluxed for 3 h.
The reaction mixture was concentrated in vacuo, and residue was purified via
silica gel chromatography using 10–20% CH2Cl2/hexanes as eluent to obtain
the desired hexahydro-indol-2-ones product (5a) in 85% yield. Compound 5a
was then dissolved in CH2Cl2, cooled to 0 °C and Br2 (1 equiv) was added
drop-wise. The reaction mixture was stirred until bromine color disappeared.
Et3N (3 equiv) was added in one portion, and the reaction mixture was stirred
at room temperature for 10 min. The reaction mixture was washed with
water (3Â), and dried over anhydrous MgSO4. The CH2Cl2 solution was
concentrated in vacuo to provide the desired vinyl bromide in 99% yield. To
solution of the vinyl bromide (1 equiv) and Et3N (10 equiv) in DMF were
added methyl acrylate (1.1 equiv), Pd(OAc)2 (0.1 equiv), and tri-o-tolyl
combination of forskolin and PGE2 showed a 40% inhibition over forskolin-
induced cAMP increase. This inhibition was reversed in a dose-dependent
manner by the test compound.
20. Tranilast containing vinylogous amide functionality has been reported to be
launched in Japan and Korea for the treatment of allergic rhinitis, asthma and
atopic dermatitis by Kessie Pharmaceuticals, Ltd, Data source Iddb3.
21. For analogs listed in Table 1, pKa range from 3.5 to 4.2. Compound 3aw,
pKa = 3.4 (ACDLabs version 9.0).
22. (a) Carroll, F. A. Perspectives on Structure and Mechanism in Organic Chemistry:
Brooks/Cole Publishing Co., 1998, pp 628, report that substituents that are
particularly effective in stabilizing nucleophilic addition to
a,b unsaturated
carbonyl compounds include aldehydes, ketone, esters and other carboxylic
acid derivatives except amide.; (b) Perlmutter, P. Conjugate Addition Reaction in
Organic Synthesis; Pergamon Press: Oxford, England, 1992.