Structure-activity relationships and sub-type selectivity in an oxabicyclic estrogen receptor α/β agonist scaffold

Article abstract of DOI:10.1016/j.bmcl.2004.12.077

An oxabicyclic template for estrogen receptor α and β agonists has been identified which can be tuned to provide moderate levels of selectivity for either receptor sub-type. Structure-activity relationships within this phenol-substituted oxabicyclo[3.3.1]

Full text of DOI:10.1016/j.bmcl.2004.12.077

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1463–1466
Structure–activity relationships and sub-type selectivity in an
oxabicyclic estrogen receptor a/b agonist scaffold
Lawrence G. Hamann,a,*, J. Hoyt Meyer,a Daniel A. Ruppar,a Keith B. Marschke,b
Francisco J. Lopez,c Elizabeth A. Allegrettod and Donald S. Karanewskya,*,à
aDepartment of Medicinal Chemistry, Ligand Pharmaceuticals, 10275 Science Centre Dr., San Diego, CA 92121, USA
bDepartment of New Leads Discovery, Ligand Pharmaceuticals, 10275 Science Centre Dr., San Diego, CA 92121, USA
cDepartment of Pharmacology, Ligand Pharmaceuticals, 10275 Science Centre Dr., San Diego, CA 92121, USA
dDepartment of Endocrine Biology, Ligand Pharmaceuticals, 10275 Science Centre Dr., San Diego, CA 92121, USA
Received 18 November 2004; revised 22 December 2004; accepted 29 December 2004
Abstract—An oxabicyclic template for estrogen receptor a and b agonists has been identified which can be tuned to provide mod-
erate levels of selectivity for either receptor sub-type. Structure–activity relationships within this phenol-substituted oxabicy-
clo[3.3.1]nonene series are described. Select compounds from the present series showed activity in vivo after oral dosing in
rodent models of uterine proliferation.
ꢀ 2005 Elsevier Ltd. All rights reserved.
The estrogen receptor (ER) is a member of the nuclear
hormone receptor superfamily of ligand-dependent
transcription factors, which mediate a broad range
of physiological processes in response to binding of
the endogenous ligand 17b-estradiol (1) or exogenous
ligands (Fig. 1).1 Progress in medicinal chemistry pro-
grams focused on the ER has resulted in the identifica-
tion of structurally diverse molecules with unique
biological properties. A number of compounds, which
act through the ER have been approved or are in late
stages of clinical development for female osteoporosis,
breast cancer, and as hormone replacement therapy.2
Most notable among these is raloxifene (2), a selective
estrogen receptor modulator (SERM). Significant ad-
vances have been made to date in elucidating the mech-
anisms by which SERMÕs achieve tissue selective action,
which include the differential recruitment of key co-reg-
ulatory proteins to complete the transcriptional machin-
ery in various cell and gene-specific contexts.3
The relatively recent identification of a second sub-type
of the ER4 may provide an additional opportunity for
selective pharmacology through agents, which preferen-
tially interact with ERb. Although the pharmacology of
ERb is not well understood, its low level of expression in
the uterus relative to ERa has generated interest in the
development of selective modulators of ERb with the
potential for reduced risk of uterine stimulation for
the treatment of vasomotor instability (hot flushes)
and osteoporosis.5 Recent reports6 by workers at Wyeth
demonstrate the utility of an ERb selective agonist in
animal models of chronic intestinal and joint inflamma-
tory disease, suggesting a role for ERb as a modulator of
immune response.
Keywords: Estrogen receptor; Osteoporosis.
Compounds of the general oxabicyclo[3.3.1]nonene
structural class have been reported in the synthetic
chemistry literature as products of the reactions of aro-
matic aldehydes and olefin-containing compounds in the
acidic media of various clays.7 These compounds had
not been shown to possess significant biological activity,
though unrelated bridged bicyclo[3.3.1]nonanes had
been reported by Katzenellenbogen as an ER ligand
*
Corresponding authors. Tel.: +1 609 818 5526; fax: +1 609 818 3550
(L.G.H.); tel.: +1 858 812 1593; fax: +1 858 812 1648 (D.S.K.);
e-mail addresses: lawrence.hamann@bms.com; dkaranew@gnf.org
Present address: Bristol–Myers Squibb, Department of Discovery
Chemistry, Pharmaceutical Research Institute, PO Box 5400, Prince-
ton, NJ 08543-5400, USA.
à Present address: Genomics Institute of the Novartis Research
Foundation, 10675 John Jay Hopkins Dr., San Diego, CA 92121,
USA.
scaffold.8
A recent report from Bayer disclosed
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.12.077

1464
L. G. Hamann et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1463–1466
Figure 1. Estrogen receptor ligands.
compound 3 as a novel ER ligand, and further described
structure–activity relationship (SAR) studies on the
all-carbon isostere of this scaffold.9 In the course of
high-throughput screening efforts we independently
identified compound 3 as a primary screening hit
on ERa/b. Herein we describe the results of our investi-
gations into the SAR in this oxabicyclic series with
respect to both potency and ERa/b sub-type selectivity,
and further describe the oral activity of compounds in
this series in a standard rodent model of estrogenic
action.
Optically pure analogues lacking a hydroxymethyl
group at the bridgehead position were prepared starting
with each respective pure isomer of limonene with gentle
heating in the acidic environment afforded by Montmo-
rillonite K10 clay to produce analogues 31–34 (Scheme
2). By analogy to the compounds derived from
(R)-(+)-limonene (i.e., 31 and 32), the dextrorotatory
isomers 23, 25, 27, 29 isolated by chiral HPLC were
tentatively assigned the absolute stereochemistry as
shown for 31 and 32.
Geminal dimethyl substitutions were introduced at three
separate positions on the oxabicyclo[3.3.1]nonene scaf-
fold by variation of initial Diels–Alder reaction compo-
nents, such as those leading to compounds 35–37, or by
Grignard addition to previously described diester 6 to
afford the appropriate precursor 38 for acid-mediated
cyclization with substituted benzaldehydes to afford
analogues 39–42 (Scheme 3).
Compounds 3 and 8–30 of the present series were pre-
pared by procedures similar to those reported in the
chemical literature (Scheme 1).7 Diels–Alder cycloaddi-
tion of isoprene or 2,3-dimethyl-1,3-butadiene (4a,b)
and diethylmethylene malonate (5a) provided dicarbo-
ethoxycyclohexenes 6, which were then reduced to the
corresponding diols 7 with lithium aluminum hydride.
p-Toluenesulfonic acid mediated cationic cyclization
with various 4-hydroxybenzaldehyes gave the requisite
oxabicyclo[3.3.1]nonene analogues in good to excellent
yields as racemic mixtures of single diastereomers.10
The trans relationship between the hydroxyaryl group
and the hydroxymethylene group of the bicyclic scaffold
was confirmed through 1H NMR (including NOE) anal-
ysis. Compounds 3, and 20–22 were subjected to chiral
preparative HPLC separation to resolve these analogues
into their respective pure enantiomers for the purposes
of assessing absolute stereochemical contributions to
biological activity.
All compounds in the present series, along with the ref-
erence compound 1 were assayed for ERa and ERb
functional activity in cell-based transcriptional assays
in COS-1 cells using methods previously reported (Table
1).11 The initial lead 3 in the series showed moderate
agonist activity on both ER sub-types. The addition of
a fluoro substituent to either the 2 or 3 position of the
aromatic ring lead to a significant enhancement in activ-
ity on both ERa and ERb. Likewise, addition of a
chloro substituent to the 2 position of the aromatic ring
increased potency on both receptors, while introduction
of a chlorine atom to the 3 position increased ERa po-
tency while attenuating activity on ERb. Other aryl sub-
stituents (Me, MeO, HO) tended to decrease potency on
ERa/b. We were surprised to find that most of the ERa
activity present in racemic mixture 3 resided in the dex-
trorotatory optical isomer, while both optical antipodes
Scheme 1. Reagents and conditions: (a) toluene, reflux; (b) LAH,
THF, ꢀ78 ꢁC; (c) (4-HO)ArCHO, p-TsOH, 1,2-DCE, 50 ꢁC.
Scheme 2. Reagents: (a) montmorillonite K10 clay, CH2Cl2.

L. G. Hamann et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1463–1466
1465
were equiactive on ERb (compare 23 and 24). As a
result, the levorotatory isomer 24 is a 22-fold selective
agonist for the ERb versus ERa. In general, it appears
that the levorotatory isomers of these oxabicyclononene
analogues are significantly more ERb selective than their
dextrorotatory counterparts (compare 27 with 28 and 29
with 30). Addition of a methyl group to the bridgehead
carbon as in analogues 19–22 also tended to improve
ERb selectivity relative to their des-methyl counterparts
(e.g., 21 vs 10).
Further optimizations in potency and/or sub-type selec-
tivity were sought by the introduction of geminal
dimethyl substitution at varying positions on the
oxabicyclo[3.3.1]nonene skeleton. It was reasoned that
a steric imposition proximal to a likely point of H-bond-
ing contact within the ligand binding domain of the
receptor (the hydroxymethyl substituent) might discern
differences in the respective binding pockets (Table 2).
Scheme 3. Reagents and conditions: (a) AlCl3, 1,2-DCE, 45 ꢁC;
(b) LAH, THF, ꢀ78 ꢁC; (c) (4-HO)ArCHO, p-TsOH, 1,2-DCE,
50 ꢁC; (d) toluene, reflux; (e) MeMgBr, PhH, rt.
While addition of a gem-dimethyl substituent to the
methylene bridge of 3 (e.g., 37) led to a modest increase
in ERa/b potencies, the regioisomeric gem-dimethyl
analogue 39 displayed dramatically improved potency
on both receptors. The contribution of the bridgehead
hydroxymethyl group to activity of the oxabicyclonon-
ene analogues on ERa/b is best illustrated by comparing
the activity of 39 with its des-hydroxymethyl counter-
parts 31/33. Thus, 31/33 are 129-fold less active on
ERa and 33-fold less active on ERb than the corre-
sponding bridgehead hydroxymethyl analogue 39. The
activity of 39 can be further enhanced by the addition
of a 2 (or 3)-fluoro or 2-chloro substituent. The most po-
tent analogues in this series are single digit nanomolar
on both ER a and b (e.g., 39–42) in the transcriptional
assays.
Table 1. Estrogen receptor a and b agonist potencies of reference
compound 1 and compounds 3 and 8–30 in transiently transfected
COS-1 cells
R
OH
X
1
O
4
HO
Compound Stereo
X
R
ERa
EC50
(nM)
ERb
EC50
(nM)
a,b
a,b
Among the primary endocrinological actions of the na-
tive hormone estradiol (1) is the promotion of the
growth and differentiation of female reproductive tis-
sues. Therefore, the effects of ER agonists can be studied
in rodent models of uterine proliferation in ovariecto-
mized female rats.12 Animals were treated once daily
for 4 days by oral gavage with varying doses (0.1, 0.3,
1
—
Racemic
—
—
—
H
2.4
9.7
3
128
137
743
61
305
2825
25
8
Racemic 3-Me
Racemic 3,5-Di-Me
Racemic 3-F
H
9
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
H
4.6
Racemic 3-Cl
Racemic 2-Cl
Racemic 3-Br
Racemic 2-F
H
36
11
197
H
5.3
H
336
39
485
10
H
Table 2. Estrogen receptor a and b agonist potencies of compounds
31–37 and 39–42 in transiently transfected COS-1 cells
Racemic 2-Me
Racemic 2-OH
Racemic 3-OH
Racemic 3-MeO
H
92
79
H
173
2411
1980
203
43
275
3164
3050
39
Compound
Stereo
Ar
ERa
EC50
(nM)
ERb
EC50
(nM)
H
a,b
a,b
H
Racemic
—
Me
Me
Me
Me
H
31
32
33
34
35
36
37
39
40
41
42
(+)
(+)
4-OH-Ph
3-Me, 4-OH-Ph
4-OH-Ph
203
143
211
218
141
21
418
318
195
318
183
25
Racemic 2-Cl
Racemic 3-F
Racemic 2-F
67
68
50
6
12
(ꢀ)
(ꢀ)
3-Me, 4-OH-Ph
4-OH-Ph
(+)
(ꢀ)
(+)
(ꢀ)
(+)
(ꢀ)
(+)
(ꢀ)
—
—
35
51
38
Racemic
Racemic
Racemic
Racemic
Racemic
Racemic
Racemic
H
835
138
18
2-Cl, 4-OH-Ph
4-OH-Ph
4-OH-Ph
2-Cl
2-Cl
3-F
3-F
2-F
2-F
H
46
59
24
H
6.1
1.6
1.0
1.1
0.6
9.2
6.0
1.4
7.2
H
2.6
40
57
2-Cl, 4-OH-Ph
2-F, 4-OH-Ph
3-F, 4-OH-Ph
H
254
21
H
180
H
19
3.1
a Values are means of three experiments.
b All compounds exhibited at least 75% intrinsic agonist activity.
a Values are means of three experiments.
b All compounds exhibited at least 75% intrinsic agonist activity.

1466
L. G. Hamann et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1463–1466
Table 3. Effects of compound 23, 24, 26, or estrone sulfate on uterine
wet weight in ovariectomized female rats
1996, 37, 6181; (c) Salakhutdinov, N. F.; Volcho, K. P.;
IlÕina, I. V.; Korchagina, D. V.; Tatarova, L. E.;
Barkhash, V. A. Tetrahedron 1998, 54, 15619.
Compound
Uterine weight 95% Confidence
ED50 SEM (mg/kg/day) limits (mg/kg/day)
8. (a) Muthyala, R. S.; Sheng, S.; Carlson, K. E.; Katzen-
ellenbogen, B. S.; Katzenellenbogen, J. A. J. Med. Chem.
2003, 46, 1589; (b) Muthyala, R. S.; Carlson, K. E.;
Katzenellenbogen, J. A. Bioorg. Med. Chem. Lett. 2003,
13, 4485.
Estrone sulfate 0.16 0.02
0.13–0.20
6.69–5.67
—
23
24
26
9.93 2.00
>30
3.44 0.88
2.09–5.67
9. Sibley, R.; Hatoum-Mokdad, H.; Schoenleber, R.; Musza,
L.; Stirtan, W.; Marrero, D.; Carley, W.; Xiao, H.;
Dumas, J. Bioorg. Med. Chem. Lett. 2003, 13, 1919.
10. A representative procedure for the preparation of com-
pound 3 as outlined in Scheme 1 follows: (a) A 250 mL
round-bottom flask was charged with diethyl methylene-
malonate (17 g, 0.099 mol) and isoprene (12 g, 0.176 mol)
and the mixture was heated neat at 100 ꢁC for 3 h. The
mixture was allowed to cool to rt, then concentrated. The
residue was purified by silica gel chromatography (hexanes
followed by 10% EtOAc/hexanes) to afford 17 g (72%) of
the title compound as a colorless oil. 1H NMR (500 MHz,
CDCl3): d 5.34 (m, 1H), 4.15 (m, 4H), 2.51 (m, 2H), 2.12
(m, 2H), 1.98 (m, 2H), 1.62 (s, 3H), 1.23 (t, 6H, J = 7.3).
1.0, 3.0, 10.0 mg/kg) of compound 23, 24, 26, estrone
sulfate (30 mg/kg) as a positive control, or vehicle (Table
3). Compounds 23 and 26 exhibited oral activity in this
animal model, inducing a dose-dependent increase in
uterine wet weight with ED50Õs of 9.93 and 3.44 mg/kg/
day, respectively. In contrast, compound 24 (the optical
antipode of 23) was inactive at 30 mg/kg/day. Since 23
and 24 are approximately equiactive on ERb but differ
significantly in their activity on ERa, the uterotrophic
effects of 23 and 26 are most likely due to the activation
of ERa.
(b) 1,1-Bis(hydroxymethyl)-4-methylcyclohex-3-ene.
A
solution of 4-methylcyclohex-3-ene-1,1-dicarboxylic acid
diethyl ester (1.97 g, 8.21 mmol) in 50 mL of THF was
cooled to ꢀ78 ꢁC. Lithium aluminum hydride (0.494 g,
13.02 mmol) was then added in one portion. The mixture
was allowed to warm to rt and stirred overnight. The
reaction mixture was then poured into aqueous 1 M
NaHSO4 and extracted with EtOAc (2 ·). The organic
layers were combined, washed with brine, dried (MgSO4),
and concentrated. The residue was purified by silica gel
chromatography (hexanes followed by 25% EtOAc/hex-
anes) to afford 0.907 g (71%) of the title compound as a
white solid. 1H NMR (500 MHz, CDCl3): d 5.29 (br s,
1H), 3.61 (s, 4H), 2.21 (br s, 2H), 1.93 (m, 2H), 1.78 (m,
2H), 1.65 (s, 3H), 1.60 (t, 2H, J = 6.5). (c) 4-(40-Hydroxy-
phenyl)-6-methyl-3-oxabicyclo[3.3.1]non-6-ene. To a solu-
tion of 1,1-bis(hydroxymethyl)-4-methylcyclohex-3-ene
(104 mg, 0.66 mmol) in 9 mL of 1,2-dichloroethane was
added p-toluenesulfonic acid monohydrate (39 mg,
0.21 mmol). The reaction mixture was heated to 50 ꢁC,
and 4-hydroxybenzaldehyde (82 mg, 0.67 mmol) was
added. The reaction mixture was stirred for 5 h. The
reaction mixture was allowed to cool to rt, poured into
water, washed with brine, dried (MgSO4), and concen-
trated. Purification by silica gel chromatography (hexanes
followed by 20% EtOAc/hexanes) afforded 70 mg (40%) of
These results serve to validate the utility of the present
template for further optimization to potentially yield
ERb selective pharmacological agents with physical
properties suitable for oral dosing.
References and notes
1. (a) Jordan, V. C. J. Med. Chem. 2003, 46, 883; (b) Jordan,
V. C. J. Med. Chem. 2003, 46, 1081.
2. Sato, M.; Grese, T. A.; Dodge, J. A.; Bryant, H. U.;
Turner, C. H. J. Med. Chem. 1999, 42, 1.
3. (a) Katzenellenbogen, B. S.; Katzenellenbogen, J. A.
Science 2002, 295, 2380; (b) Shang, Y.; Brown, M. Science
2002, 295, 2465.
4. (a) Kuiper, G. G.; Enmark, E.; Pelto-Huikko, M.;
Nilsson, S.; Gustafsson, J. A. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 5925; (b) Mosselman, S.; Polman, J.;
Dijkema, R. FEBS Lett. 1996, 392, 49; (c) Tremblay, G.
B.; Tremblay, A.; Copeland, N. G.; Gilbert, D. J.; Jenkins,
N. A.; Labrie, F.; Giguere, V. Mol. Endocrinol. 1997, 11,
353.
5. Harris, H. A.; Katzenellenbogen, J. A.; Katzenellenbogen,
B. S. Endocrinology 2002, 143, 4172.
1
the title compound as a white solid. H NMR (500 MHz,
6. (a) Harris, H. A.; Albert, L. M.; Leathurby, Y.; Malamas,
M. S.; Mewshaw, R. E.; Miller, C. P.; Kharode, Y. P.;
Marzolf, J.; Komm, B. S.; Winneker, R. C.; Frail, D. E.;
Henderson, R. A.; Zhu, Y.; Keith, J. C. Endocrinology
2003, 144, 4241; (b) Malamas, M. S.; Manas, E. S.;
McDevitt, R. E.; Gunawan, I.; Xu, Z. B.; Collini, M. D.;
Miller, C. P.; Dinh, T.; Henderson, R. A.; Keith, J. C., Jr.;
Harris, H. A. J. Med. Chem. 2004, 47, 5021.
7. (a) Volcho, K. P.; Korchagina, D. V.; Gatilov, Y. V.;
Salakhutdinov, N. F.; Barkhash, V. A. Zh. Org. Khim.
1997, 33, 666; (b) Volcho, K. P.; Korchagina, D. V.;
Salakhutdinov, N. F.; Barkhash, V. A. Tetrahedron Lett.
CDCl3): d 7.17 (d, 2H, J = 8.5), 6.77 (d, 2H, J = 8.5), 5.56
(s, 1H), 4.73 (s, 1H), 4.50 (s, 1H), 3.96 (dd, 1H, J = 11.0,
2.5), 3.63 (d, 1H, J = 11.0), 3.40 (d, 2H, J = 4.0), 2.32 (d,
1H, J = 2.7), 2.21 and 2.08 (ABq, 2H, JAB = 18.9), 1.82
(dd, 1H, J = 11.5, 3.0), 1.67 (ddd, 1H, J = 12.0, 3.0, 3.0),
1.01 (dd, 3H, J = 4.0, 2.5).
11. Zou, A.; Marschke, K. B.; Arnold, K. E.; Berger, E. M.;
Fitzgerald, P.; Mais, D. E.; Allegretto, E. A. Mol.
Endocrinol. 1999, 13, 418.
12. Lundeen, S. G.; Carver, J. M.; McKean, M.-L.; Winneker,
R. C. Endocrinology 1997, 138, 1552.