Design, synthesis, and preclinical characterization of novel, highly selective indole estrogens

Full text of DOI:10.1021/jm010086m

1654
J . Med. Chem. 2001, 44, 1654-1657
Letters
Design , Syn th esis, a n d P r eclin ica l
Ch a r a cter iza tion of Novel, High ly
Selective In d ole Estr ogen s
Chris P. Miller,*,† Michael D. Collini,†
Bach D. Tran,‡, Heather A. Harris,‡
Yogendra P. Kharode,‡ J ames T. Marzolf,‡
Robert A. Moran,‡ Ruth A. Henderson,‡
Reinhold H. W. Bender,† Rayomond J . Unwalla,§
Lee M. Greenberger,| J ohn P. Yardley,†
Magid A. Abou-Gharbia,† C. Richard Lyttle,‡ and
Barry S. Komm‡
Wyeth-Ayerst Research, Chemical Sciences,
145 King of Prussia Road, Radnor, Pennsylvania 19087,
Wyeth-Ayerst Research Women’s Health Research Institute,
145 King of Prussia Road, Radnor, Pennsylvania 19087,
Wyeth-Ayerst Research, Biology Chemistry,
145 King of Prussia Road, Radnor, Pennsylvania 19087,
and Wyeth-Ayerst Research, Oncology and
Immunoinflammatory Research,
Pearl River, New York 10965
Received February 27, 2001
In tr od u ction . The onset of menopause has been
associated with a reduction in endogenous estrogen
levels that has been correlated with a variety of meno-
pausal symptoms and an increased risk for osteoporosis
and heart disease.1 Hormone replacement therapy suc-
cessfully alleviates these symptoms, but concerns about
hyperplastic estrogen action on breast and uterine
tissue have been raised.2,3 While the hyperplastic action
of estrogen on uterine tissue can be successfully opposed
by the coadministration of a progestin, the effects of
hormone replacement therapy on breast tissue remain
controversial and is the subject of some debate.4 Largely
as a result of these issues, considerable research has
been dedicated to the discovery of estrogens that act in
a tissue selective fashion.5 The prototypical estrogen
with some tissue selectivity is, paradoxically, the “anti-
estrogen”, tamoxifen 1 (Figure 1). While tamoxifen has
been used for over two decades in the treatment of
hormone-dependent breast cancer, it was not until 1987
that J ordan and co-workers reported that tamoxifen was
able to protect against bone loss in the ovariectomized
rat model.6 J ordan’s group also demonstrated that
raloxifene 2 (raloxifene hydrochloride is currently Evista
for osteoporosis) was also effective as an estrogen
agonist on bone in the ovariectomized rat, while dis-
playing even less uterine stimulation than tamoxifen.
F igu r e 1. Structures of various estrogens.
This discovery has prompted a vigorous search for
compounds with increasingly selective profiles.
Most of the selective estrogens reported to date share
the common structural motif of a core recognition
element consisting of two aryl groups separated by two
atoms (often a stilbene type arrangement). Additionally,
these compounds typically bear a third phenyl group
containing a 4-aminoethoxy substituted phenyl group.
It is believed that selective estrogens function by binding
the ligand binding domain of the estrogen receptor
through their stilbene-like cores and projecting the third
phenyl group that contains the 4-aminoethoxy group
into a region of space which corresponds to the 11
position of an estratriene nucleus (i.e., estrone or
estradiol).7 Indeed, X-ray cocrystallographic studies of
4-OH-tamoxifen (a high-affinity, active metabolite of
tamoxifen) or raloxifene with the ligand binding domain
of the estrogen receptor R (ERR) reveal this to be the
case.8,9 Moreover, when 4-OH tamoxifen or raloxifene
bind the ligand binding domain of ERR, an altered
receptor conformation is adopted where helix 12 is
placed in a hydrophobic groove formed by residues from
helices 3, 4, 5, and the turn connecting helices 3 and 4.
This is in contrast to the situation with DES 3 (di-
ethylstilbesterol, a nonsteroidal estrogen agonist) or
17â-estradiol 4 which both lack the amine-bearing side
chain. When these estrogen agonists bind ER, helix 12
folds over, revealing a groove formed by residues from
helices 3, 4, 5, and 12. This groove provides a binding
* To whom correspondence should be addressed. Tel: 610-341-2168.
Fax: 610-989-4588. E-mail: millerc1@war.wyeth.com.
† Wyeth-Ayerst Research, Chemical Sciences.
‡ Wyeth-Ayerst Research Women’s Health Research Institute.
§ Wyeth-Ayerst Research, Biology Chemistry.
| Wyeth-Ayerst Research, Oncology and Immunoinflammatory Re-
search.
Present Address: Guilford Pharmaceuticals, 6611 Tributary Street,
Baltimore, MD 21224.
10.1021/jm010086m CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/25/2001

Letters
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11 1655
template of the indole can function in a similar fashion
as other selective estrogen stilbene-type cores, we
postulated that the aliphatic, amine-containing side
chain is too flexible to provide for optimal displacement
of helix 12 upon binding. To that end, we decided to
examine the effect of placing a rigidifying linker between
the amine terminus and the indole nitrogen. To do this,
we constructed a phenoxyethyl linker between the
amine of the side chain and the N-1 position of the
indole. In this way, the modified indoles have side
chains that more closely resemble other selective estro-
gens already described. After various explorations around
the core and side chain, we discovered ERA-923 6 and
TSE-424 7, both novel, highly selective estrogens with
particularly non-estrogenic profiles on rat uterine tissue.
F igu r e 2. X-ray structures of the ligand binding domain of
Era/DES (left) with a coactivator fragment (in purple), and
the ligand binding domain of Era/4-OH tamoxifen (right).
Ch em istr y. As shown in Scheme 1, the 3-methyl
indole core 10 was synthesized in a relatively straight-
forward fashion from R-bromopropiophenone 8 and the
aniline hydrochloride 9 via a Bischler-type indole syn-
thesis.12 We found the yields and reproducibility of this
reaction could be increased by conducting the reaction
in two steps but one pot. The side chain 12 was prepared
by alkylation of commercially available 4-OH benzyl
alcohol 11 with ethyl bromoacetate followed by conver-
sion of the alcohol to the benzyl chloride 12 with SOCl2
in THF. Alkylation of the indole 10 with the side chain
12 was accomplished with NaH in DMF. The ester 13
was subsequently reduced with LAH, and the primary
alcohol thus produced was converted to the correspond-
ing bromide by treatment with CBr4 and triphenylphos-
phine. Substitution of this bromide with piperidine
yielded 14, and substitution with hexamethylenimine
furnished 15. Hydrogenation of 14 and 15 and conver-
sion to their respective salts yielded 6 (HCl salt) and 7
(acetate salt).
site for the LXXLL motif of various nuclear receptor
coactivators. The difference in conformation of helix 12
results from the projection of the third phenyl group (in
the case of a selective estrogen) containing the basic
amine into the region of space which is normally
occupied by helix 12 when in an agonist bound confor-
mation (Figure 2).9 The net result of these differences
is that the altered conformation leads to a differential
recognition between the receptor and several coregula-
tory proteins. It is believed, inter alia, that cell-specific
regulation of target gene expression relies on differential
expression or regulation of these various coregulatory
proteins.
Since the stilbene-like cores of raloxifene (X-ray
structure not shown), 4-OH-tamoxifen, and DES all
locate in the ERR ligand binding domain in a similar
fashion, it is reasonable to focus on the location and
structure of the side chain as a fundamental determi-
nant of selective estrogen action. In fact, it has been
demonstrated that even very small changes in the side
chain of raloxifene can result in large changes in its
tissue selectivity profile.10 On the basis of this premise,
we were interested in examining the side chain SAR of
certain estrogenic templates reported in the literature.
Of particular interest was the 2-phenylindole compound,
5.11 Whereas it appears from inspection that the core
Resu lts a n d Discu ssion . Compounds 6 and 7 were
evaluated for their ability to displace [3H]17â-estradiol
from estrogen receptor ligand binding domain constructs
(human) of both the R- and â-receptors (ERR and ERâ).13
The binding results are listed in Table 1 as IC50s; values
for unlabeled 5, 17â-estradiol 4, and 2 are included for
Sch em e 1. Synthesis of Indoles 6 and 7a
a
Reagents and conditions: (a) Et3N, DMF; (b) K2CO3, R-bromoethyl acetate; (c) SOCl2, THF; (d) NaH, DMF; (e) LiAlH4, THF; (f) TPP,
CBr4, THF; (g) piperidine, THF; (h) hexamethylenimine, THF; (i) H2, Pd/C, EtOH/THF.

1656 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11
Letters
F igu r e 3. Uterine wet weight data from immature rats treated for 3 days subcutaneously. Neither 6 nor 7 demonstrate any
increase in uterine weight when administered alone, and both completely antagonize 17â-estradiol stimulated wet weight gains.
Conversely, raloxifene 2 statistically increased uterine wet weight at both 10 and 100 µg and did demonstrate antagonism of
17â-estradiol stimulated weight increase, but it converged with its 100 µg agonist activity. Compound 5 did exhibit agonist activity
at 100 µg, with similar antagonist activity seen with raloxifene converging with its 100 µg effect. Doses administered are per
animal.
Ta ble 1. In Vitro Binding and MCF-7 Transfection Assay
functional estrogenic/antiestrogenic activity in an MCF-7
cell ERE-tk-luciferase (ERR mediated) assay (Table 1).14
radioligand binding assay
IC50(nM)a
transient transfection assay
in MCF-7 cellsa
Since neither 6 nor 7 stimulate transcriptional activity
in these cells, they were evaluated in the antagonist
mode by competition with 10 pM 17â-estradiol. Only 6,
7, and 3 were potent antagonists in this assay; all
having IC50s less than 5 nM. It is interesting to note
that while 5 displays a binding affinity for ERR similar
to that of either 6 or 7, it is not very potent in this
MCF-7 ERE-tk luciferase assay that expresses only
ERR. Perhaps, this is because the binding assay only
utilizes the ligand binding domain of ERR while the
transcriptional assay employs full length receptors with
all domains intact. Alternatively, it may be due to other
factors related to differential cellular uptake, metabo-
lism, cellular transcription, or transactivation.15 The
generally poor correlation between ERR binding affini-
ties and MCF-7 cell proliferation inhibition has been
noted previously for a series of raloxifene analogues.16
ERR
ERâ
tested asb EC50 or IC50 (nM)c
3
2
5
6
7
3 ( 1
4 ( 3
25 ( 13
14 ( 10
23 ( 15
4 ( 2
43 ( 13
242 ( 187
40 ( 24
85 ( 59
agonist
0.007 ( 0.007
0.72 ( 0.2
69 ( 23
antagonist
antagonist
antagonist
antagonist
1.5 ( 0.4
3.7 ( 1.6
a
b
n ) average of at least two determinations ((SD). Agonist
mode run at 10 pM; antagonist mode run with compound against
10 pM 17â-estradiol. c EC50 refers to values in agonist mode; IC50
refers to antagonist mode.
comparison. As evidenced from the data in Table 1, both
6 and 7 show respectable affinity for both forms of the
receptor, albeit with a slight preference for ERR. Con-
sistent with our supposition that the core binding
element generally drives basic binding affinity, com-
pound 5 displays an affinity similar to that of 6 and 7
for the two receptors, while the benzothiophene 2 shows
a high affinity for ERR but a somewhat weaker affinity
for ERâ. Although understanding the distinct physi-
ological role of ERR and ERâ is of considerable interest
to the scientific community, at this time the consequence
of a molecule’s preferential binding to either receptor
is not clear.
Whether being used as a selective estrogen for pre-
vention of osteoporosis or as an antiestrogen for breast
cancer, it is critical that the candidate compound does
not cause stimulation of the uterine endometrium in the
target population, since estrogenic stimulation of the
uterus leads to both an increase in uterine bleeding as
well as an increase in the risk of uterine cancer. Thus,
all compounds were tested in a 3-day immature rat
Compounds 6 and 7 were also evaluated for their

Letters
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11 1657
Su p p or tin g In for m a tion Ava ila ble: Spectral and ana-
lytical data for all new compounds. This information is
available free of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Palacios, S. Current perspectives on the benefits of HRT in
menopausal women. Maturitas 1999, 33 (Suppl. 1), S1-S13.
(2) J acobs, H. S. Hormone replacement therapy and breast cancer.
Endocr.-Relat. Cancer 2000, 7, 53-61.
(3) Elit, L. Endometrial cancer. Prevention, detection, management,
and follow up. Canadian Family Physician 2000, 46, 887-92.
(4) Speroff, L. Postmenopausal hormone therapy and the risk of
breast cancer [editorial]. Maturitas 1999, 32, 123-9.
(5) Grese, T. A.; Dodge, J . A. Selective estrogen receptor modulators
(SERMs). Curr. Pharm. Des. 1998, 4, 71-92.
(6) J ordan, V. C.; Phelps, E.; Lindgren, J . U. Effects of anti-estrogens
on bone in castrated and intact female rats. Breast Cancer Res.
Treat. 1987, 10, 31-5.
F igu r e 4. Total bone mineral density utilizing peripheral
quantitative computerized tomography (pQCT) of the proximal
tibia from ovariectomized rats treated with 7 orally for 6
weeks. TSE-424 demonstrates a significant effect from vehicle
starting at 0.3 mg/kg.
uterine model.13 Figure 3 demonstrates the comparative
uterine stimulation of a subcutaneous administration
of 2, 5, 6, 7, and 4. Both 6 and 7 showed no significant
uterine stimulation when dosed alone. In contrast, 2,
5, and 4 (E2 in Figure 3) all significantly increased the
uterine weights of the treated animals, albeit it to
different degrees. In the antagonist mode, 6 and 7 were
able to completely inhibit the uterine wet weight
increase effected by 4. On the basis of this preclinical
uterine data, we believe that 6 and 7 are among the
most selective of estrogens reported to date.
In a 6-week ovariectomized rat study,17 7 was exam-
ined for its ability to prevent bone loss. Figure 4
demonstrates that 7 protected against bone loss in the
proximal tibia and reached significant efficacy at a dose
of 0.3 mg/kg per day. Additionally, 7 reduced total
cholesterol with maximal efficacy (approximately 50%)
at just 0.1 mg/kg per day while demonstrating no
statistically significant effect on uterine wet weight in
any of the treated animal groups.
(7) Bouhoute, A.; Leclercq, G. Estradiol derivatives bearing the side-
chain of tamoxifen antagonize the association between the
estrogen receptor and calmodulin. Biochem. Pharmacol. 1994,
47, 748-51.
(8) Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard, R. E.;
Bonn, T.; Engstrom, O.; Ohman, L.; Greene, G. L.; Gustaffson,
J .-A.; Carlquist, M. Molecular basis of agonism and antagonism
in the estrogen receptor. Nature (London) 1997, 389, 753-758.
(9) Shiau, A. K.; Barstad, D.; Loria, P. M.; Cheng, L.; Kushner, P.
J .; Agard, D. A.; Greene, G. L. The structural basis of oestrogen
receptor/coactivator recognition and the antagonism of this
interaction by tamoxifen. Cell 1998, 95, 927-937.
(10) Grese, T. A.; Sluka, J . P.; Bryant, H. U.; Cullinan, G. J .;
Glasebrook, A. L.; J ones, C. D.; Matsumoto, K.; Palkowitz, A.
D.; Sato, M.; Termine, J . D.; Winter, M. A.; Yang, N. N.; Dodge,
J . A. Molecular determinants of tissue selectivity in estrogen
receptor modulators. Proc. Natl. Acad. Sci. U.S.A. 1997, 94,
14105-14110.
(11) Nishino, Y.; Schneider, M. R.; Michna, H.; Von Angerer, E.
Pharmacological characterization of a novel estrogen antagonist,
ZK 119010, in rats and mice. J . Endocrinol. 1991, 130, 409-14.
(12) Von Angerer, E.; Prekajac, J .; Strohmeier, J . 2-Phenylindoles.
Relationship between structure, estrogen receptor affinity, and
mammary tumor inhibiting activity in the rat. J . Med. Chem.
1984, 27, 1439-1447.
(13) Miller, C. P.; J irkovsky, I.; Tran, B. D.; Harris, H. A.; Moran, R.
A.; Komm, B. S. Syntheis and estrogenic activities of novel
7-thiosubstituted estratriene derivatives. Bioorg. Med. Chem.
Lett. 2000 10, 147-151.
(14) Bodine, P. V. N.; Harris, H. A.; Komm, B. S. Suppression of
ligand-dependent estrogen receptor activity by bone-resorbing
cytokines in human osteoblasts. Endocrinology 1999, 140, 2439-
2451.
(15) Katzenellenbogen, J . A.; O’Malley, B. W.; Katzenellenbogen, B.
S. Tripartite steroid hormone receptor pharmacology: interac-
tion with multiple effector sites as a basis for the cell- and
promoter-specific action of these hormones. Mol. Endocrinol.
1996, 10, 119-31.
(16) Grese, T. A.; Cho, S.; Finley, D. R.; Godfrey, A. G.; J ones, C. D.;
Lugar, C. W., III.; Martin, M. J .; Matsumoto, K.; Pennington,
L. D.; Winter, M. A.; Adrian, M. D.; Cole, H. W.; Magee, D. E.;
Phillips, D. L.; Rowley, E. R.; Short, L. L.; Glasebrook, A. L.;
Bryant, H. U. Structure-Activity Relationships of Selective
Estrogen Receptor Modulators: Modifications to the 2-Aryl-
benzothiophene Core of Raloxifene. J . Med. Chem. 1997, 40,
146-167.
From the data presented, side chain structure is an
important determinant of selective estrogen action.
Whereas the indole 5 displayed considerably reduced
antiestrogen efficacy in an MCF-7 ERE-tk luciferase
assay as well as some estrogen agonist activity in a
3-day immature rat uterine assay, compounds 6 and 7
behaved as potent antagonists in these assays. While
it has been reported that varying the length of the
aliphatic side chain can also affect the relative degree
of estrogenicity/antiestrogenicity for a given compound
in various cell-based systems, we believe that for the
amine-containing side chains it is optimal to have the
phenoxy group present to rigidify this portion of the
molecule.18
Both 6 and 7 display very attractive selective estrogen
profiles. Due to functional differences between the two
compounds it was decided that 7 should be advanced
for treatment of postmenopausal osteoporosis. Whereas
7 was very efficacious at protecting against bone loss
and reducing total cholesterol in the ovariectomized rat
model of osteopenia, 6 was a very potent and efficacious
antiestrogen in various in vivo MCF-7 xenograft mod-
els.19 Currently, 6 is in phase II clinical trials for the
treatment of hormone-dependent metastatic breast
cancer, and 7 has completed phase II clinical trials for
the prevention and treatment of postmenopausal os-
teoporosis.
(17) Hayward, M. A.; Kharode, Y. P.; Becci, M. M.; Kowal, D. The
effect of conjugated equine estrogens on ovariectomy-induced
osteopenia in the rat. Agents Actions 1990, 31, 152-6.
(18) Biberger, C.; von Angerer, E. 2-Phenylindoles with sulfur
containing side chains. Estrogen receptor affinity, antiestrogenic
potency, and antitumor activity. J . Steroid Biochem. Mol. Biol.
1996, 58, 31-43.
(19) Greenberger, L. M., Komm, B., Miller, C., Annable, T., Lyttle,
R., Frost, P. Comparison of ERA-923, a new selective estrogen
receptor modulator (SERM) for the treatment of estrogen-
receptor positive breast cancer, with other SERMs. NCI-EORTC-
AACR Symposium on New Drugs in Cancer Therapy, Amster-
dam, 2000.
J M010086M