Synthesis and identification of novel oxa-steroids as progesterone receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2006.11.062

A novel series of oxa-steroids 6 derived from (8S, 13S, 14R)-7-oxa-estra-4,9-diene-3,17-dione 1 have been synthesized and identified as potent and selective progesterone receptor antagonists. These novel oxa-steroids showed similar potency to mifepristone

Full text of DOI:10.1016/j.bmcl.2006.11.062

Bioorganic & Medicinal Chemistry Letters 17 (2007) 907–910
Synthesis and identification of novel oxa-steroids as
progesterone receptor antagonists
Fu-An Kang,* George Allan, Jihua Guan, Nareshkumar Jain, Olivia Linton,
Pamela Tannenbaum, Jun Xu, Peifang Zhu, Joseph Gunnet, Xin Chen,
Keith Demarest, Scott Lundeen and Zhihua Sui
Johnson and Johnson Pharmaceutical Research and Development, LLC, 665 Stockton Drive, Exton, PA 19341, USA
Received 4 October 2006; revised 14 November 2006; accepted 20 November 2006
Available online 1 December 2006
Abstract—A novel series of oxa-steroids 6 derived from (8S, 13S, 14R)-7-oxa-estra-4,9-diene-3,17-dione 1 have been synthesized and
identified as potent and selective progesterone receptor antagonists. These novel oxa-steroids showed similar potency to mifepri-
stone. Preliminary SAR study resulted in the most potent 17-phenylethynyl oxa-steroid 6i wih an IC50 of 1.4 nM. In contrast to
the equipotent mifepristone toward the progesterone receptor (PR) and glucocorticoid receptor (GR), compound 6i had over
200-fold selectivity for PR over GR.
ꢀ 2006 Elsevier Ltd. All rights reserved.
The progesterone receptor (PR) is a member of the ste-
roid receptor sub-family of the nuclear hormone receptor
super-family which is a group of ligand-dependent nucle-
ar transcription factors.1 Progesterone is the endogenous
ligand for PR, which regulates ovulation and prepares
uterus to support pregnancy. PR modulators exist clini-
cally as both agonists and antagonists. PR agonists such
as medroxy-progesterone acetate (MPA)2 are mainly
used in contraception and hormone therapy, typically
co-administered with an estrogen. One of the main issues
with PR agonists is that they often bind and modulate
the function of other members of the nuclear hormone
receptor super-family, for example, the androgen (AR),
glucocorticoid (GR), and mineralocorticoid (MR) recep-
tors. In principle, a PR antagonist may have potential
utility as a contraceptive.3 However, current PR antago-
nists, such as mifepristone (RU-486),4 are compromised
as clinically useful contraceptive agents due to overt glu-
cocorticoid receptor antagonism.5
used not only in female contraception, but also poten-
tially for the treatment of various gynecological and
obstetric diseases including hormone-dependent cancers
and non-malignant chronic conditions, such as fibroids6
and endometriosis.7 A variety of steroidal8 and non-ste-
roidal9 PR modulators have been reported in recent
years. As part of our interest in developing novel PR
modulators,10 herein we report the synthesis and
O
Me
O
Me
Me
Me
O
O
Me
H
Me
H
Me
H
H
H
H
O
O
Me
MPA
Progesterone
Me
N
Me
N
With few clinically successful selective PR antagonists
being available, their therapeutic potential has not yet
been fully realized. A selective PR antagonist may be
Me
Me
OH
OH
Me
H
Me
H
Me
R
H
H
O
O
O
Keywords: Oxa-steroid; Progesterone receptor; Antagonist; Modula-
tor; Glucocorticoid receptor; Modeling; Mifepristone; Selective PRM.
Mifepristone
oxa-Steroids
*
Corresponding author. Tel.: +1 610 458 4147; e-mail: fkang@
prdus.jnj.com
Figure 1. Structures of natural and unnatural steroids.
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.062

908
F.-A. Kang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 907–910
identification of a novel series of oxa-steroids (Fig. 1) as
potent and selective PR antagonists.
in Table 1. The ratio of their IC50 values was calculated
as a measure of the separation of PR and GR antagonist
activities. Mifepristone (RU-486) was tested as a con-
trol. It was determined that compound 6a was a potent
PR antagonist with an IC50 of 7.5 nM and it was about
10-fold more selective to PR over GR, which exhibited a
slightly better selective profile to that of mifepristone.
Recently, we have achieved the first enantioselective syn-
thesis of (8S, 13S, 14R)-7-oxa-estra-4,9-diene-3,17-dione
1 with the unambiguous trans-C/D ring junction11
(Scheme 1). We were intrigued in applying this unnatu-
ral oxa-steroidal template to the discovery of novel PR
modulators. Selective mono-ketallation of the 3-carbon-
yl group of compound 1 was achieved in the presence of
pyridinium hydrochloride with azeotropic removal of
water in refluxing benzene. Simultaneously, under the
thermal and acidic reaction condition, the [5,6]-spiro-
ring system induced the 4,9-diene to isomerize to the
thermodynamically more stable 5,11-diene affording
compound 2. A highly regio-/stereo-selective epoxida-
tion with m-CPBA at À30 ꢁC generated predominantly
a-epoxide 3 (a:b = 8:1). It is interesting to note that
epoxidation of the oxa-steroid tended to result in better
stereoselectivity than that of the natural steroid.12 N,N-
Dimethylaniline magnesium bromide was treated with
the homogeneous CuCN–2LiCl complex13 in THF at
0 ꢁC to form the organocuprate which was immediately
added to a-epoxide 3 at 0 ꢁC, stereoselectively producing
the 11-b-substituted oxa-steroid 4. Stereoselective addi-
tion of 1-propynyl magnesium bromide to the carbonyl
group of compound 4 at ambient temperature delivered
compound 5a (R = Me). Finally, deprotection and dehy-
dration under the acidic condition furnished the oxa-ste-
roid 6a (R = Me). Thus, an efficient and stereoselective
synthesis of novel oxa-steroids 6 was achieved in five
steps from compound 1.
Although compound 6a was somewhat less potent than
mifepristone, their similar PR activity was suggested by
our computational study. Shown in Figure 2 is the com-
parison of the possible modes of mifepristone and com-
pound 6a bound in the ligand-binding domain of
progesterone receptor. The model was built based on
the X-ray crystal structures of hPR-norenthindrone16
and hGR-mifepristone17 complexes. In the construction
of these models, helix 12 of the hPR-norenthindrone
crystal structure was removed in order to open the li-
gand-binding site. Mifepristone and compound 6a were
then docked into the ligand-binding site using program
Glide,18 followed by re-packing of helix 12 back to the
antagonism position in reference to the hGR-mifepri-
stone crystal structure. The loop between helices 12
and 11 was re-built using program Prime.18 The final
complex structures were optimized by energy minimiza-
tion using program MacroModel18. In these models,
mifepristone and compound 6a were predicted to bind
in a similar way to PR, with a water molecule bridging
the D-ring hydroxyl group and ASN719 through hydro-
gen bond. At the other end, another hydrogen bond may
also exist between the A-ring carbonyl group and
GLN725. Compared to the 7-methylene group in mife-
pristone, the polar and eletronegative 7-oxygen atom
in oxa-steroid 6a may play a slightly different role in
the ligand–protein interactions with the nearby
MET756 (not shown).
Compound 6a was evaluated for PR antagonist activity
based on its ability to block progesterone induction of
alkaline phosphatase activity in the human breast cancer
cell line T47D.14 It was also tested for GR antagonist
activity based on its ability to inhibit corticoid-induced
transcription from a glucocorticoid response element
(GRE)-linked luciferase reporter gene in the human lung
carcinoma cell line A549.15 The IC50 values of the
compounds from the T47D and A549 assays are listed
As a preliminary SAR study, a number of substituted al-
kyns were added to the carbonyl group of compound 4,
which led to compounds 6b–i (Table 1). Replacement of
the methyl group in compound 6a with the smaller
hydrogen atom in compound 6b lowered the PR activity.
O
O
O
Me
H
Me
Me
H
O
H
H
O
a
b
c
H
H
O
O
O
O
O
O
O
1
2
3
Me
N
Me
N
Me
N
Me
Me
Me
O
OH
OH
Me
H
Me
H
Me
H
d
e
R
R
H
H
H
O
O
O
O
O
O
O
OH
O
OH
6
5
4
Scheme 1. Synthesis of oxa-steroids. Reagents and conditions: (a) (CH2OH)2, pyridine-HCl, benzene, refluxing, 82%; (b) m-CPBA, NaHCO3, DCM,
À30 ꢁC, 70%; (c) Me2NC6H4MgBr, CuCN–2LiCl, THF, 0 ꢁC, 88%; (d) MeCCMgBr (6a, R = Me), THF, rt, 92%; (e) 3 N HCl aq, acetone, 95%.

F.-A. Kang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 907–910
Table 1. SAR study at the 17-substituted-ethynyl groups of compounds 6
909
Compound
R
T47D (PR), IC50 (nM)
A549 (GR), IC50 (nM)
Ratio (GR/PR)
Mifepristone
—
Me
1.4
7.5
1.6
86.5
1
6a
6b
6c
6d
6e
6f
12
11
4
H
19.0
9.6
204.3
33.4
55.5
Cyclopropyl
CF3
9.3
6
CMe3
CMe2(OH)
CH2OMe
CH2NMe2
Ph
185.0
>1000
600
166.9
558.0
315
1
—
0.5
—
217
6g
6h
6i
>1000
1.4
>3000
304.0
Figure 2.
The cyclopropyl group in compound 6c and trifluoro-
methyl group in compound 6d retained the PR activity.
The bulky tert-butyl group in compound 6e significantly
decreased the PR activity. The polar functional groups
such as 2-hydroxy-2-propyl group in compound 6f,
methoxymethyl group in compound 6g, and N,N-di-
methylaminomethyl group in compound 6h resulted in
much less active or literally inactive compounds.
Surprisingly, it was found that change of the methyl
group in compound 6a to the phenyl group led to the
most potent and selective PR antagonist, compound
6i, which had an IC50 of 1.4 nM and was over 200-fold
more selective for PR over GR.
ring junction, we have discovered for the first time that
the oxa-steroids with an oxygen atom in place of the 7-
methylene group have retained the biological properties
of this type of compounds, as it has been demonstrated
by the biological activities of oxa-steroids 6 toward PR
and GR. More significantly, some of the oxa-steroids,
such as compound 6i, have shown outstanding better
selectivity toward PR over GR than that of the natural
steroids such as mifepristone.
In conclusion, a novel series of oxa-steroids 6 derived
from (8S, 13S, 14R)-7-oxa-estra-4,9-diene-3,17-dione 1
have been synthesized. For the first time, the 7-oxa-ste-
roids have been identified as potent and selective proges-
terone receptor antagonists. These novel oxa-steroids
showed similar or comparable potency to that of mife-
pristone. Preliminary SAR study resulted in the discov-
ery of the most potent 17-phenylethynyl oxa-steroid 6i
as a new potent selective PR antagonist. In contrast to
non-selective mifepristone, compound 6i had over 200-
fold selectivity for PR over GR. More SAR study and
their in vivo activity will be reported in due course.
It should be pointed out that oxa-steroids related to
compound 1 were reported before in the literature,19
but the structure and stereochemistry were unfortunate-
ly not clearly characterized and established in the report.
In addition, the biological activities of those oxa-ste-
roids were also evaluated and they were found to be
inactive toward PR. Therefore, it was concluded that
‘the insertion of an oxygen atom into the steroid back-
bone in place of the 7-methylene group has practically
abolished the biological properties of this type of
compounds.’
Acknowledgment
With our first enantioselective synthesis of compound 1
with the unambiguous stereochemistry of the trans-C/D
We thank Dr. Weiqin Jiang, Dr. Ningyi Chen, and Ms.
Heather R. Hufnagel for helpful discussions.

910
F.-A. Kang et al. / Bioorg. Med. Chem. Lett. 17 (2007) 907–910
14. T47D alkaline phosphatase assay. T47D human breast
Supplementary data
cancer cells were plated in 96-well tissue culture plates at
10,000 cells per well in assay medium [RPMI medium
without phenol red (Invitrogen) containing 5% (v/v)
charcoal-treated FBS (Hyclone) and 1% (v/v) penicillin–
streptomycin (Invitrogen)]. Two days later, the medium
was decanted and test compound at various concentra-
tions (0.01–1000 nM), 0.1 nM progesterone, or vehicle
control (dimethyl sulfoxide) at a final concentration of
0.1% (v/v), was added in fresh assay medium. Twenty-four
hours later, an alkaline phosphatase assay was performed
using a SEAP kit (BD Biosciences Clontech, Palo Alto,
CA). Briefly, the medium was decanted and the cells were
fixed for 30 min at room temperature with 5% (v/v)
formalin (Sigma). The cells were washed once at room
temperature with Hanks’ buffered saline solution (Invit-
rogen). Equal volumes (0.05 mL) of 1· dilution buffer,
assay buffer, and 1:20 substrate/enhancer mixture were
then added. After a 1-h incubation at room temperature in
the dark, the lysate was transferred to a white 96-well plate
(Dynex) and luminescence was read using a LuminoSkan
Ascent (Thermo Electron, Woburn, MA).
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2006.11.062.
References and notes
1. Mangelsdorf, D. J.; Thummel, C.; Beato, M.; Herrlich, P.;
Schutz, G.; Umesono, K.; Blumberg, B.; Kastner, P.;
Mark, M.; Chambon, P.; Evans, R. M. Cell 1995, 83, 835.
2. (a) Jeppsson, S. Acta Obstet. Gynecol. Scand. Suppl. 1981,
101, 7; (b) O’Connell, B. J. Curr. Opin. Pediatr. 1995, 7, 371.
3. (a) Ulmann, A.; Peyron, R.; Silvestre, L. Ann. N.Y. Acad.
Sci. 1995, 761, 248; (b) Kekkonen, R.; Lahteenmaki, P.;
Luukkainen, T.; Tuominen, J. Fertil. Steril. 1993, 60, 610;
(c) Chwalisz, K.; Stochkemann, K. World Patent Appli-
cation WO 96/19997, 1996.
4. Brogden, R. N.; Goa, K. L.; Faulds, D. Drugs 1993, 45, 384.
5. Lazar, G.; Lazar, G. Jr.; Husztik, E.; Duda, E.; Agarwal,
K. K. Ann. N.Y. Acad. Sci. 1995, 277, 248.
15. A549 reporter assay. A549 human lung carcinoma cells
were washed with OPTI-MEM I (Gibco). The medium
was removed and lipid–DNA complex solution (1.5 lg/
mL of GRE-luciferase reporter DNA in 8.5 mL OPTI-
MEM I plus 6 lL/mL DMRIE-C reagent in 8.5 mL OPTI-
MEM I, combined and mixed, then incubated at room
temperature for 40 min) was overlayed onto the cells in a
T160 flask. The cells were incubated for 16 h at 37 ꢁC in a
CO2 incubator. The DNA-containing medium was
removed and 30 mL of growth medium containing 5%
(v/v) charcoal-treated fetal bovine serum was added. After
5–6 h, the cells were seeded in 96-well plates and incubated
overnight at 37 ꢁC. To each well were then added test
compounds followed by dexamethasone as a corticoid
challenge. The cells were incubated for 24 h. Luciferase
assay buffer (Promega) was added to each well and the cell
were incubated for 30 min at room temperature. Lucifer-
ase activity was measured in a DYNEX Microlite plate on
a TopCount (Packard).
16. Madauss, K. P.; Deng, J.-S.; Austin, R. J. H.; Lambert,
M. H.; McLay, I.; Pritchard, J.; Short, S. A.; Stewart, E.
L.; Uings, I. J.; Williams, S. P. J. Med. Chem. 2004, 47,
3381.
17. Kauppi, B.; Jakob, C.; Farnegardh, M.; Yang, J.; Ahola,
H.; Alarcon, M.; Calles, K.; Engstrom, O.; Harlan, J.;
Muchmore, S.; Ramqvist, A.-K.; Thorell, S.; Ohman, L.;
Greer, J.; Gustafsson, J.-A.; Carlstedt-Duke, J.; Carlquist,
M. J. Biol. Chem. 2003, 278, 22748.
6. (a) Murphy, A. A.; Kettel, L. M.; Morales, A. J.; Roberts,
V. J.; Yen, S. S. J. Clin. Endocrinol. Metab. 1993, 76, 513;
(b) Murphy, A. A.; Morales, A. J.; Kettel, L. M.; Yen, S.
S. Fertil. Steril. 1995, 64, 187.
7. (a) Kettel, L. M.; Murphy, A. A.; Morales, A. J.; Yen, S.
S. Am. J. Obstet. Gynecol. 1998, 178, 1151; (b) Kettel, L.
M.; Murphy, A. A.; Morales, A. J.; Yen, S. S. Hum.
Reprod. 1994, 9, 116.
8. (a) Teutsch, G.; Philibert, D. Hum. Reprod. 1994, 9, 12; (b)
Chwalisz, K.; Perez, M. C.; DeManno, D.; Winkel, C.;
Schubert, G.; Elger, W. Endo. Rev. 2005, 26, 423; (c)
Chabbert-Buffet, N.; Meduri, G.; Bouchaard, P.; Spitz, I.
M. Hum. Reprod. Update 2005, 11, 293.
9. (a) Allan, G. F.; Sui, Z. Mini-Rev. Med. Chem. 2005, 5,
701; (b) Zhang, P.; Fensome, A.; Winneker, J.; Wrobel,
R.; Zhang, Z. Exp. Opin. 2003, 13, 1839.
10. (a) Jiang, W.; Allan, G.; Fiordeliso, J.; Linton, O.;
Tannenbaum, P.; Xu, J.; Zhu, P.; Gunnet, J.; Demarest,
K.; Lundeen, S.; Sui, Z. Bioorg. Med. Chem. 2006, 14,
6726; (b) Jiang, W.; Allan, G.; Chen, X.; Fiordeliso, J.;
Linton, O.; Tannenbaum, P.; Xu, J.; Zhu, P.; Gunnet, J.;
Demarest, K.; Lundeen, S.; Sui, Z. Steroids 2006, 71, 949.
11. Kang, F.-A.; Jain, N.; Sui, Z. Tetrahedron Lett. 2006,
doi:10.1016/j.tetlet.2006.11.042.
12. (a) Hazra, B.; Basu, S.; Pore, V.; Joshi, P.; Pal, D.;
Chakrabarti, P. Steroids 2000, 65, 157; (b) Prat, D.;
Benedetti, F.; Girard, F.; Bouda, L.; Larkin, J.; Wehrey,
C.; Lenay, J. Org. Process Res. Dev. 2004, 8, 219.
13. The homogeneous CuCN–2LiCl complex in THF was
found to be superior to CuCl12 to ensure a reproducible
and scalable reaction.
18. Glide, Prime and MacroModel, Schrodinger, 1500 S.W.
First Avenue, Suite 1180, Portland, OR 97201, USA.
19. Bucourt, R.; Vignau, M. Bull. Soc. Chim. Fr. 1975, 3–4,
896.