Development of Selective Estrogen Receptor Modulator (SERM)-Like Activity Through an Indirect Mechanism of Estrogen Receptor Antagonism: Defining the Binding Mode of 7-Oxabicyclo[2.2.1]hept-5-ene Scaffold Core Ligands

Article abstract of DOI:10.1002/cmdc.201200048

Previously, we discovered estrogen receptor (ER) ligands with a novel three-dimensional oxabicyclo[2.2.1]heptene core scaffold and good ER binding affinity act as partial agonists via small alkyl ester substitutions on the bicyclic core that indirectly mo

Full text of DOI:10.1002/cmdc.201200048

MED
DOI: 10.1002/cmdc.201200048
Development of Selective Estrogen Receptor Modulator
(SERM)-Like Activity Through an Indirect Mechanism of
Estrogen Receptor Antagonism: Defining the Binding
Mode of 7-Oxabicyclo[2.2.1]hept-5-ene Scaffold Core
Ligands
Yangfan Zheng,[a] Manghong Zhu,[a] Sathish Srinivasan,[c] Jerome C. Nwachukwu,[c]
Valerie Cavett,[c] Jian Min,[a] Kathryn E. Carlson,[b] Pengcheng Wang,[a] Chune Dong,[a]
John A. Katzenellenbogen,*[b] Kendall W. Nettles,[c] and Hai-Bing Zhou*[a]
Previously, we discovered estrogen receptor (ER) ligands with
a novel three-dimensional oxabicyclo[2.2.1]heptene core scaf-
fold and good ER binding affinity act as partial agonists via
small alkyl ester substitutions on the bicyclic core that indirect-
ly modulate the critical switch helix in the ER ligand binding
domain, helix 12, by interactions with helix 11. This contrasts
with the mechanism of action of tamoxifen, which directly
pushes helix 12 out of the conformation required for gene acti-
vation. We now report that a much larger substitution can be
tolerated at this position of the bicyclic core scaffold, namely
a phenyl sulfonate group, which defines a novel binding epi-
tope for the estrogen receptor. We prepared an array of 14 ox-
abicycloheptene sulfonates, varying the phenyl sulfonate
group. As with the parent compound, 5,6-bis-(4-hydroxyphen-
yl)-7-oxabicyclo[2.2.1]hept-5-ene-2-sulfonic acid phenyl ester
(OBHS), these compounds showed preferential affinity for ERa,
and the disposition and size of the phenyl substituents were
important determinants of the binding affinity and selectivity
of these compounds, with those having ortho substituents
giving the highest, and para substituents the lowest affinities
for ERa. A few analogues exhibit ERa binding affinities that are
comparable to or, in the case of the ortho-chloro analogue,
higher than that of OBHS itself. In cell-based studies, we found
several compounds with activity profiles comparable to tamox-
ifen, but acting entirely as indirect antagonists, allosterically in-
terfering with recruitment of coactivator proteins to the recep-
tor. Thus, the OBHS binding epitope represents a novel ap-
proach to the development of estrogen receptor antagonists
via an indirect mechanism of antagonism.
Introduction
Estrogens, such as the endogenous hormone estradiol (E2), are
important mediators of many physiological functions related
to development, growth, and maintenance of reproductive as
well as non-reproductive tissues in both men and women.[1]
The effects of estrogens are mediated via two estrogen recep-
tor subtypes, ERa and ERb,[2] which have different tissue distri-
butions and significant differences in their ligand binding pref-
erences.[3] Selective estrogen receptor modulators (SERMs),
compounds that show differential levels of agonistic versus an-
tagonistic activity in different estrogen target tissues, as well
as ER subtype-selective ligands, have been investigated inten-
sively in recent years in the search for estrogens that have im-
proved patterns of target-tissue selectivity.[4]
tempts to exploit this unfilled space and the flexibility of the
ER binding pocket as an approach to enhance ligand binding
affinity, SERM behavior, or ER-subtype selectivity. Nevertheless,
a number of ER ligands with diverse three-dimensional chemi-
cal scaffolds have emerged—representative examples include
those based on ferrocene,[8] carboranes,[1a,9] bridged poly-
[a] Y. Zheng, M. Zhu, J. Min, P. Wang, Prof. Dr. C. Dong, Prof. Dr. H.-B. Zhou
State Key Laboratory of Virology
Laboratory of Combinatorial Biosynthesis and Drug Discovery
Ministry of Education, School of Pharmaceutical Sciences
Wuhan University
185 East Lake Road, Wuhan, 430071 (P.R. China)
E-mail: zhouhb@whu.edu.cn
Structural studies of the ligand binding pockets of both ERa
and ERb reveal substantial unoccupied space above and below
the mean plane of the endogenous ligand E2, particularly near
the middle of this molecule (namely, below the B ring and
above the C ring),[5] as well as considerable flexibility in the
shape of the ligand binding pocket.[6] While these characteris-
tics of how different estrogens bind to the ERs have been
known for several years,[5,7] there have been only sporadic at-
[b] K. E. Carlson, Prof. Dr. J. A. Katzenellenbogen
Department of Chemistry, University of Illinois
600 South Mathews Avenue, Urbana, IL 61801 (USA)
E-mail: jkatzene@uiuc.edu
[c] Dr. S. Srinivasan, Dr. J. C. Nwachukwu, V. Cavett, Prof. Dr. K. W. Nettles
Department of Cancer Biology, The Scripps Research Institute
5353 Parkside Drive, Jupiter, FL, 33458 (USA)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201200048.
ChemMedChem 0000, 00, 1 – 8
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

J. A. Katzenellenbogen, H.-B. Zhou et al.
MED
Scheme 1. Structure of estradiol, examples of ER ligands that have three-dimensional elements or core structures, OBHS (1) and title compounds 2.
cyclics,[10] and some other cyclopentadienyl metal tricarbonyl
Results and Discussion
complexes (Scheme 1).[11] We have contributed to this area as
Structure of OBHS-bound ERa
well.[7a]
The formula for making ER antagonists is well established:
Take an agonist ligand and add a bulky side chain to disrupt
helix 12 (Figure 1a,b). Structure–activity relationship (SAR)
studies are then done to optimize for antagonist activity and
potency. We previously discovered that certain small substitu-
tions at different sites on agonist ligands could yield partial ag-
onists, which we called passive or indirect antagonists because
they modulate helix 12 indirectly by inducing small shifts in
helix 11.[12] These partial agonists include a series of oxabicyclic
bridged compounds with small alkyl ester substitutions direct-
ed towards ER helix 11; these compounds exhibit relatively
weak binding affinities, which suggests a limited potential for
expanding SAR.[7a] Intriguingly, however, one member of this
series, OBHS (1) has a much larger, bulky side group, a phenyl
sulfonate, yet binds better than those analogues with smaller
substituents.[7a] This suggested to us that there might actually
be room for much more extensive perturbation of the ligand
binding pocket in this region than we had originally envi-
sioned.
The previously reported ERa co-crystal structures of complexes
of the agonist diethylstilbestrol (DES) and the SERM 4-hydroxy-
tamoxifen (4OHT) illustrate how the bulky side group of ta-
moxifen directly relocates helix 12, thereby destroying the sur-
face-bound coactivator binding site required for gene activa-
tion (Figure 1a,b).[13] In contrast, the bulky side group of OBHS
(1) binds between helices 11 and 8 of the receptor, and in
doing so significantly shifts helix 11 and His524 (Figure 1c).
Since OBHS (1) profiles as a partial agonist (see below), we pre-
pared a series of modifications on the phenyl group to explore
whether this novel epitope can be used to drive antagonist
activity to a level comparable to that of tamoxifen.
Synthesis
The synthesis of the oxabicyclic bridged compounds 2 was ef-
fectively accomplished, as before, by a Diels–Alder reaction of
3,4-diarylfuran 11 with a variety of dienophiles 5. The dieno-
philes (5) were prepared according to a modified literature
procedure by the reaction of 2-chloroethanesulfonyl chloride
(3) with substituted phenols 4a–m in the presence of tri-
ethylamine and were obtained in generally good yields
(Scheme 2).[7a,14]
The structure of OBHS supports this hypothesis and led us
to prepare a series of derivatives targeting this novel binding
epitope. Using a series of cell-based assays in liver and breast
cancer cells, we identify several compounds with higher poten-
cy than OBHS and better activity profiles, matching the level of
antagonism seen with tamoxifen. This work thus demonstrates
a structure-based design approach to blocking ER activity via
a novel binding epitope and an indirect mechanism of antago-
nism.
The synthesis of diarylfuran 11 was accomplished using an
improved procedure, as shown in Scheme 3. a-Bromo-4-me-
thoxyacetophenone (6) reacted with arylacetic acid 7 in the
presence of triethylamine in acetonitrile to give condensation
product 8 in 88% yield.[15] In contrast, our previous procedure
to produce 8 required condensation of the potassium salts of
arylacetic acids 7 with 6 using 18-crown-6 as a catalyst under
refluxing conditions.[7a] Treatment of ester 8 with sodium hy-
dride in anhydrous DMSO gave 2(5H)-furanone 9, which was
&2
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 8
ÝÝ
These are not the final page numbers!

SERM-Like Activity of Oxabicyclo[2.2.1]hept-5-enes
MED
Figure 1. The oxabicycloheptene sulfonate side chain binds in a new epitope between helices 8 and 11. a) Crystal structure of the ERa ligand binding domain
(LBD) in complex with diethylstilbestrol (DES) (PDB ID: 3ERD[13]). The side chains of the helix 3 and helix 11 residues, Glu353 and His524, both of which
engage in hydrogen bonding with DES are shown. b) Crystal structure of the ERa LBD in complex with 4-hydroxytamoxifen (4OHT) (PDB ID: 3ERT[13]). The se-
lective estrogen receptor modulator (SERM) side chain of 4OHT (green) displaces helix 12. c) Crystal structure of the ERa LBD in complex with OBHS (1) at
2.1 ꢁ resolution. The position of His524 in the estradiol-bound structure is shown with faint lines. Notably, the phenylsulfonate moiety (green) of OBHS binds
in a new epitope that significantly shifts helix 11.
demethylated with boron tribromide to afford butenolide 10
in 85% yield,[16] avoiding our prior use of pyridinium chloride
at 2208C to obtain the free phenols.[7a] Diisobutylaluminum hy-
dride reduction of 10 at ꢀ788C gave, after a careful, low-tem-
perature acidic workup, 3,4-diphenol furans 11. Finally, a Diels–
Alder reaction of 11 with dienophiles 5 produced compounds
2a–n. It is noteworthy that, similar to our previous study, high
stereoselectivity was observed in the cycloaddition reaction of
the phenolic furans with the dienophiles, with the exo product
being obtained nearly exclusively; only traces, if any, of the
endo products were observed.[7a] This exo selectivity is pre-
sumed due to the high rate and ready reversibility of the
Diels–Alder reaction of furans, which allows the thermodynam-
ically favored exo product to
Scheme 2. Synthesis of dienophiles 5a–n. Reagents and conditions: a) Et3N,
CH2Cl2, 08C, 1 h.
become dominant.
Binding affinity and SARs
The compounds were assayed
for their binding affinity to ERa
and ERb by a competitive radio-
metric binding assay.[17] These af-
finities are expressed as relative
binding affinity (RBA) values;
RBA values are reported as per-
centages (%) of that of estradiol
(E2), which is set at 100%. All of
the 14 newly synthesized com-
pounds were tested, and the re-
sults are summarized in Table 1.
It is noteworthy that the dis-
position and the size of the sub-
stituent on the phenyl of the sul-
fonates prove to be important
factors in determining the bind-
ing affinity and selectivity of
Scheme 3. Synthesis of oxabicycloheptene sulfonate derivatives 2a–n using an improved procedure. Reagents
and conditions: a) Et3N, CH3CN, RT, 88%; b) NaH, DMSO, 258C, 75%; c) BBr3, CH2Cl2, ꢀ788C!RT, 85%; d) DIBAL-H,
THF, ꢀ788C, 71%; d) 5 (neat), 908C.
these new compounds. Most of
the ligands show moderate-to-
ChemMedChem 0000, 00, 1 – 8
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
3
&
ÞÞ
These are not the final page numbers!

J. A. Katzenellenbogen, H.-B. Zhou et al.
MED
ERa and ERb, respectively, together with an improved selectivi-
ty—now over tenfold—in favor of ERa (Table 1, entry 3). When
the chloro substituent was replaced with a fluoro (2a) or
bromo (2c), a progressive decrease in RBA values was ob-
served (Table 1, entries 2 and 4). For the corresponding meta-
substituted compounds (2d–f), the fluoro (2d) and chloro (2e)
analogues showed lower binding affinities, but the bromo ana-
logue (2 f) also gave good RBA values of 5.84% and 0.707%
for ERa and ERb, respectively (Table 1, entries 5–7). At the para
position, all compounds exhibit rather low binding affinities
(Table 1, entries 8–10). Previously, we prepared the p-hydroxy
and p-methoxy analogues of OBHS (1), and they too had lower
affinities.[7a]
Table 1. Relative binding affinities of compounds 2 for estrogen recep-
tors a and b.[a]
Entry Compd
Ar
ERa
ERb
Ratio b/a
1
2
OBHS (1)
9.28ꢁ0.64 1.71ꢁ0.24
7.28ꢁ0.26 2.24ꢁ0.56
5.5
2a
3.3
10.8
5.4
One can conclude that, in terms of binding affinity, ortho
substitution seems beneficial, while meta substitution has
a more neutral effect, and para substitution is detrimental; at
the ortho position, a chloro substituent is best, but a bromo is
favored at the meta position. Additional SAR investigations fo-
cused on the most promising class, the ortho-substituted
series. Ortho-methyl analogue 2j maintained good affinity, and
while the ortho-ethyl analogue 2k showed a significant drop
in binding affinity, a methoxy group (2l) was well tolerated
(Table 1, entries 11–13). Interestingly, there was a more than
100-fold difference in binding affinity between the 1- and 2-
naphthyl OBHS analogues (2m and 2n) on ERa (Table 1, en-
tries 14 and 15). With respect to a phenyl group, in a way, the
1-naphthyl group combines substitution at the favorable and
neutral ortho and meta positions, while the 2-naphthyl group
begins to occupy the same space as the poorly tolerant para
position.
3
4
5
6
7
2b
2c
2d
2e
2 f
19.0ꢁ4.6
1.76ꢁ0.13
2.29ꢁ0.02 0.419ꢁ0.070
1.76ꢁ0.04 0.39ꢁ0.06
1.91ꢁ0.46 0.36ꢁ0.06
5.84ꢁ0.28 0.707ꢁ0.050
4.5
5.5
8.3
8
2g
2h
2i
1.18ꢁ0.13 0.494ꢁ0.120
0.78ꢁ0.002 0.284ꢁ0.003
0.76ꢁ0.21 0.404ꢁ0.120
2.4
2.8
1.9
9
10
Transcription activation through ERa and b
The ability of these compounds to regulate ERa- and ERb-
mediated transcription was compared to that of tamoxifen in
liver cells, in which tamoxifen displays agonist activity on ERa
but not ERb, due to higher activity of the amino-terminal coac-
tivator binding region in ERa.[18] Cells were transfected with
vectors for full-length human ERa or ERb and a widely used es-
trogen-responsive element (ERE)-driven luciferase reporter
gene.[19] OBHS (1) and the 14 analogues exhibited weaker po-
tency than E2, which activated ERa with an EC50 value of 1–
2 nm, and also lower efficacy (Table 2, agonist mode). It is im-
portant to note that in HepG2 cells, used in the assays report-
ed here, SERMs show higher partial agonist activity than in
HEC-1 cells used previously.[7a,18] In fact, 4OHT has an agonist
activity that is 35% that of estradiol, although fulvestrandt is
still a complete antagonist.[18]
11
12
13
2j
2k
2l
3.49ꢁ0.63 0.294ꢁ0.020
1.03ꢁ0.16 0.096ꢁ0.020
4.39ꢁ0.48 0.210ꢁ0.010
11.9
10.7
20.9
14
15
2m
2n
5.03ꢁ1.50 0.478ꢁ0.120
0.08ꢁ0.02 0.019ꢁ0.005
10.5
2.8
[a] Relative binding affinity (RBA) values were determined by competitive
radiometric binding assays and are expressed as IC50estradiol/IC50 compound ꢂ
100ꢁthe range or standard deviation (n=2–3, RBA, estradiol=100%). In
these assays, the Kd value for estradiol is 0.2 nm on ERa and 0.5 nm on
ERb. Ki values of each compound for each receptor can be obtained from
the RBA values by the equation: Ki =(100/RBA)ꢂKd.
In HepG2 cells, three of the novel compounds, 2a, 2c and
2m, displayed improved EC50 values compared with OBHS (1)
(Table 2). Generally, aryl halides 2a–i showed a similar activity
pattern, a trend where fluoro substitution results in higher effi-
cacy than bromo substitution. A similar pattern is evident with
methyl substitution (compound 2j) inducing greater activity
than ethyl (2k) or methoxy (2l). It is noteworthy that the
much higher affinity 1-naphthyl isomer (2m) has lower activity
on ERa than the 2-naphthyl isomer (2n), which has low affinity
excellent binding affinity, with 2b, a compound that possesses
an ortho-chloro substituent on the phenyl group of the sulfo-
nate, showing the highest RBA values of 19.0% and 1.76% for
&4
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 8
ÝÝ
These are not the final page numbers!

SERM-Like Activity of Oxabicyclo[2.2.1]hept-5-enes
MED
As a final test, a subset of the compounds with the
lowest efficacy or potency were assayed for activa-
tion of 3XERE-luciferase activity in ERa-positive MCF-
7 breast cancer cells, in which tamoxifen acts as an
inverse agonist. While OBHS (1) showed limited ago-
nist activity, we identified several compounds with in-
verse agonist activity. Notably, derivative 2m shows
an improved profile in all assays relative to OBHS (1),
and an antagonist efficacy profile comparable to that
of tamoxifen. While the affinity of derivative 2m is
lower than that of 4OHT, these results establish
proof-of-principle that indirect antagonists can pro-
duce an activity profile comparable to tamoxifen,
through by a different structural mechanism.
Table 2. Estrogen receptor-mediated transcriptional activities.
Entry Compd
Agonist mode[a]
ERa ERb
EC50 [nm][c] Eff. [%] Eff. [%] Eff. [%] IC50 [nm][c] Eff. [%]
Antagonist mode[b]
ERb ERb
ERa
MCF-7
1
2
3
4
5
6
7
8
OBHS (1)
2a
2b
2c
2d
2e
2 f
2g
2h
2i
2j
2k
2l
2m
2n
95
26
–
60ꢁ2
61ꢁ5
29ꢁ5
53ꢁ5
54ꢁ2
34ꢁ1
33ꢁ2
43ꢁ6
36ꢁ2
31ꢁ6
52ꢁ5
33ꢁ2
40ꢁ1
29ꢁ1
48ꢁ2
0ꢁ1
0ꢁ3
1ꢁ1
2ꢁ1
2ꢁ1
1ꢁ2
0ꢁ1
1ꢁ2
1ꢁ5
4ꢁ7
1ꢁ4
0ꢁ1
0ꢁ2
1ꢁ2
3ꢁ1
21ꢁ4
26ꢁ4
ꢀ7ꢁ2
23ꢁ3
n.d.
n.d.
n.d.
n.d.
n.d.
ꢀ2ꢁ2
n.d.
n.d.
n.d.
n.d.
n.d.
ꢀ4ꢁ4
581
–
–
–
453
–
248
214
–
ꢀ16ꢁ2
54ꢁ4
86ꢁ11
11ꢁ14
ꢀ19ꢁ3
ꢀ2ꢁ5
ꢀ26ꢁ4
ꢀ8ꢁ5
ꢀ15ꢁ4
61ꢁ12
ꢀ12ꢁ2
ꢀ4ꢁ4
17ꢁ9
44
140
210
300
200
360
160
220
390
560
46
9
10
11
12
13
14
15
16
17
–
357
3208
1038
–
30ꢁ3
Conclusions
–
–
1.08
–
52ꢁ19
ꢀ23ꢁ0
ꢀ20ꢁ2
Fulvestrandt[d]
4-OHT[d]
1ꢁ1 ꢀ1ꢁ1
1
1
A small library of OBHS (1) analogues, differing in the
substituents on the phenyl ring of the sulfonate,
were prepared and evaluated as ligands for ERa and
ERb. The substitution on the bulky phenyl sulfonate
side group of OBHS (1) is well tolerated and can en-
gender a good binding affinity and antagonist effica-
cy profile. In transcription assays performed in HepG2
cells, some analogues activated ERa only partially
and had little or no activity on ERb, and a few com-
pounds behaved as indirect antagonists exclusively
on ERb. Notably, several of the compounds showed
full antagonism (even inverse agonism) on ERb (e.g.,
2d, 2 f, 2h, and 2j), and most derivatives were partial
35ꢁ3 ꢀ1ꢁ1 ꢀ11 ꢁ1
[a] Luciferase activity was measured in HepG2 cells transfected with 3X-ERE-driven lu-
ciferase reporter and expression vectors encoding ERa or ERb, or where indicated in
ERa+MCF-7 cells with the ERE reporter, and treated in triplicate with increasing
doses (up to 10ꢀ5 m) of test compound. The EC50 values and average efficacy (Eff.),
shown as a percentage of 10ꢀ5 m 17b-estradiol (E2), were determined; data are pre-
sented as the meanꢁSEM. None of the compounds tested activated ERb. Not deter-
mined (n.d.). [b] Average efficacy (Eff.) of the compounds (10ꢀ5 m) in combination with
10ꢀ6 m E2 (E2 only=100%). [c] The EC50 or IC50 value of some compounds was too high
to be determined (–). [d] Fulvestrandt (Faslodex) and 4-hydroxytamoxifen (4OHT)
were used as reference compounds.
and low potency, but higher efficacy. This highlights the fact
that affinity and potency (i.e., EC50) are independent of the allo-
steric conformational signals between ligand and the surface
transcriptional coactivator binding site, which determine the
activity or efficacy (i.e., intrinsic activity) of the compounds.
What was unexpected was the low potency of the ortho-chloro
compound (2b), which had the highest affinity on ERa.
Beyond the factors noted above, cell permeability and metabo-
lism issues might also be contributing to its low potency, as
well as the general difficulty in establishing EC50 values for
compounds with low intrinsic activity. In this context, it is im-
portant to note that the corresponding ortho-fluoro and ortho-
bromo compounds (2a and 2c) have good potency, but only
slightly lower efficacy compared with OBHS (1).
agonist/antagonists on ERa. Importantly, several compounds
also act as inverse agonists in MCF-7 breast cancer cells. Be-
cause the bulky side group binds the receptor between heli-
ces 8 and 11, rather than towards helix 12, compounds such as
2m represent a novel structural class of SERMs. Interestingly,
more than a 100-fold difference in binding affinity was ob-
served between the 1- and 2-naphthyl OBHS derivatives (2m
and 2n) on ERa. Thus, despite the apparent constraints of the
ER ligand binding pocket noted in crystal structures of related
compounds, analogues of OBHS (1) with larger sulfonate aryl
substituents can actually be more potent and active than
OBHS itself. This is, again, a testament to the flexibility of the
ligand binding pocket of the ERs and its ability to adapt to
bind ligands of diverse structure with good affinity.[6,20]
None of the compounds activated ERb as agonists in the
HepG2 luciferase assay, and so, they were profiled only as an-
tagonists in the presence of 1 mm E2. Most of the compounds
are antagonists selectively on ERb (Table 2). Because ERb has
considerable basal activity, especially in HepG2 cells, com-
pounds can be found to have inverse agonist activity, that is,
an intrinsic activity that is less than that of apo-ERb; such com-
pounds are registered as having a negative efficacy value.
While the size of the substituent predicted the activity on ERa,
for ERb, compounds with meta substitution (2d–f) were more
consistently antagonistic. The naphthyl compounds (2m,n)
showed similar activity profiles on both ER subtypes.
Experimental Section
Synthesis
General: Synthetic procedures and characterization data for com-
pounds 5, 8–11 and 2 are given in the Supporting Information. All
compounds assayed were >95% pure as determined by HPLC
analysis. A summary of the HPLC results and HPLC spectra of all
final compounds 2 are given in the Supporting Information.
General procedure for the Diels–Alder reaction: Dienophiles 5
and 3,4-bisphenol-furan 11 were synthesized according to the pre-
viously reported method with modifications (see Supporting Infor-
mation for details).[7a,14] The BBr3 demethylations were performed
ChemMedChem 0000, 00, 1 – 8
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
5
&
ÞÞ
These are not the final page numbers!

J. A. Katzenellenbogen, H.-B. Zhou et al.
MED
according to published procedures.[16] Furan 12 (0.2 mmol) and di-
enophiles 5 (0.24 mmol) were placed in a round flask, and the mix-
ture was stirred under a N2 atmosphere at 1008C for 12–24 h. The
crude product was purified by flash chromatography (10–50%
EtOAc/petroleum ether), preparative thin-layer chromatography, or
recrystallization.
Acknowledgements
The authors are grateful to the National Natural Science Founda-
tion of China (NSFC) (20872116, 20972121, 81172935), the Chi-
nese Ministry of Education, New Century Excellent Talents (NCET)
in University Program (NCET-10-0625), the Chinese National
Mega Project on Major Drug Development (2009ZX09301-014-1),
and the Research Fund for the Doctoral Program of Higher Edu-
cation of China (20100141110021) for support of this research.
Research support from the US National Institutes of Health (PHS
5R37 DK015556 to J.A.K. and R01 DK077085 to K.W.N.) is also
gratefully acknowledged. The authors thank Dr. Teresa Martin
(University of Illinois, Urbana, USA) for help with the binding
assays.
5,6-Bis-(4-hydroxyphenyl)-7-oxabicyclo[2.2.1]hept-5-ene-2-sul-
fonic acid 2-chlorophenyl ester (2b): Flash chromatography (30%
EtOAc/petroleum ether) gave 2b as a light yellow oil (89%):
1H NMR (400 MHz, CDCl3): d=7.56–7.36 (m, 4H), 7.26–6.92 (m, 6H),
6.87–6.64 (m, 4H), 5.79 (d, J=0.9 Hz, 1H), 5.40 (dd, J=12.1, 7.5 Hz,
3H), 3.78 (dd, J=8.4, 4.4 Hz, 1H), 2.64 (dt, J=12.3, 4.4 Hz, 1H),
2.22 ppm (dd, J=12.2, 8.3 Hz, 1H); 13C NMR (101 MHz, CDCl3): d=
155.99, 155.95, 153.7, 149.6, 145.0, 141.5, 137.0, 130.9, 129.5, 128.9,
128.2, 128.1, 126.9, 124.6, 124.2, 123.9, 116.1, 115.9, 115.8, 114.9,
84.4, 83.0, 60.8, 30.7 ppm; HRMS (EI): m/z [M+Na]+ calcd for
C24H19ClO6SNa: 493.04486, found: 493.04831.
Keywords: cycloadditions · estrogen · estrogen receptors ·
hormones · steroids
[1] a) Y. Endo, T. Yoshimi, K. Ohta, T. Suzuki, S. Ohta, J. Med. Chem. 2005, 48,
3941–3944; b) J. A. Gustafsson, Trends Pharmacol. Sci. 2003, 24, 479–
485.
Biological evaluation
[2] G. G. Kuiper, E. Enmark, M. Pelto-Huikko, S. Nilsson, J. A. Gustafsson,
Proc. Natl. Acad. Sci. USA 1996, 93, 5925–5930.
Estrogen receptor binding affinity assays: Relative binding affinities
were determined by a competitive radiometric binding assay, as
previously described,[17b] using 2 nm [3H]estradiol as tracer (Perkin-
Elmer Inc., Waltham, MA, USA) and purified full-length human ERa
and ERb (PanVera/InVitrogen, Carlsbad, CA, USA). Incubations were
18–24 h at 08C. Then, the receptor–ligand complexes were ab-
sorbed onto hydroxyapatite (BioRad, Hercules, CA, USA), and un-
bound ligand was washed away. The binding affinities are ex-
pressed as relative binding affinity (RBA) values, with the RBA value
of estradiol set to 100. The values given are the average ꢁrange
or standard deviation (SD) of two or more independent determina-
tions. Estradiol binds to ERa with a Kd value of 0.2 nm and to ERb
with a Kd value of 0.5 nm.
[3] K. Pettersson, J. A. Gustafsson, Annu. Rev. Physiol. 2001, 63, 165–192.
[4] a) M. De Angelis, F. Stossi, K. A. Carlson, B. S. Katzenellenbogen, J. A. Kat-
zenellenbogen, J. Med. Chem. 2005, 48, 1132–1144; b) C. Gruber, D.
Gruber, Curr. Opin. Investig. Drugs 2004, 5, 1086–1093; c) V. C. Jordan, J.
Med. Chem. 2003, 46, 883–908; d) F. Minutolo, S. Bertini, C. Granchi, T.
Marchitiello, G. Prota, S. Rapposelli, T. Tuccinardi, A. Martinelli, J. R.
Gunther, K. E. Carlson, J. A. Katzenellenbogen, M. Macchia, J. Med.
Chem. 2009, 52, 858–867; e) F. Minutolo, M. Macchia, B. S. Katzenellenb-
ogen, J. A. Katzenellenbogen, Med. Res. Rev. 2011, 31, 364–442; f) A. D.
Palkowitz, A. L. Glasebrook, K. J. Thrasher, K. L. Hauser, L. L. Short, D. L.
Phillips, B. S. Muehl, M. Sato, P. K. Shetler, G. J. Cullinan, T. R. Pell, H. U.
Bryant, J. Med. Chem. 1997, 40, 1407–1416; g) H.-B. Zhou, K. E. Carlson,
F. Stossi, B. S. Katzenellenbogen, J. A. Katzenellenbogen, Bioorg. Med.
Chem. Lett. 2009, 19, 108–110.
Luciferase assay: HepG2 cells cultured in growth media containing
Dulbecco’s minimum essential medium (DMEM) (Cellgro by Medi-
atech, Inc. Manassas, VA, USA) supplemented with 10% fetal
bovine serum (FBS) (Hyclone by Thermo Scientific, South Logan,
UT, USA), 1% nonessential amino acids (Cellgro), penicillin–strepto-
mycin–neomycin antibiotic mixture, and Glutamax (Gibco by Invi-
trogen Corp., Carlsbad, CA, USA), were maintained at 378C and 5%
CO2. Cells in 10 cm dishes were transfected with 10 mg of 3XERE-lu-
ciferase reporter plus 1.6 mg of ERa or ERb expression vector per
dish using FugeneHD reagent (Roche Applied Sciences, Indianapo-
lis, IN, USA). The next day, the cells were re-suspended in phenol-
red-free growth media supplemented with 10% charcoal-dextran
sulfate-stripped FBS, transferred at a density of 20000 cells per
well to 384-well plates, incubated overnight at 378C and 5% CO2,
and treated in triplicate with the various compounds. After 24 h,
luciferase activity was measured using BriteLite reagent (Perkin-
Elmer Inc., Shelton, CT, USA) according to manufacturer’s protocol.
[5] A. M. Brzozowski, A. C. W. Pike, Z. Dauter, R. E. Hubbard, T. Bonn, O. Eng-
strom, L. Ohman, G. L. Greene, J.-A. Gustafsson, M. Carlquist, Nature
1997, 389, 753–758.
[6] K. W. Nettles, J. B. Bruning, G. Gil, E. E. O’Neill, J. Nowak, Y. Guo, Y. Kim,
E. R. DeSombre, R. Dilis, R. N. Hanson, A. Joachimiak, G. L. Greene, EMBO
Rep. 2007, 8, 563–568.
[7] a) H. B. Zhou, J. S. Comninos, F. Stossi, B. S. Katzenellenbogen, J. A. Kat-
zenellenbogen, J. Med. Chem. 2005, 48, 7261–7274; b) Y. L. Wu, X. J.
Yang, Z. Ren, D. P. McDonnell, J. D. Norris, T. M. Willson, G. L. Greene,
Mol. Cell 2005, 18, 413–424; c) A. C. W. Pike, A. M. Brzozowski, R. E. Hub-
bard, T. Bonn, A. G. Thorsell, O. Engstrom, J. Ljunggren, J. K. Gustafsson,
M. Carlquist, EMBO J. 1999, 18, 4608–4618; d) R. W. Hsieh, S. S. Rajan,
S. K. Sharma, Y. Guo, E. R. DeSombre, M. Mrksich, G. L. Greene, J. Biol.
Chem. 2006, 281, 17909–17919.
[8] a) A. Nguyen, S. Top, P. Pigeon, A. Vessiꢃres, E. A. Hillard, M.-A. Plamont,
M. Huchꢄ, C. Rigamonti, G. Jaouen, Chem. Eur. J. 2009, 15, 684–696;
b) D. Plaz?uk, A. Vessie‘res, E. A. Hillard, O. Buriez, E. Labbe‘, P. Pigeon,
M. A. Plamont, C. Amatore, J. Zakrzewski, G. Jaouen, J. Med. Chem.
2009, 52, 4964–4967; c) S. Top, A. Vessieres, G. Leclercq, J. Quivy, J.
Tang, J. Vaissermann, M. Huche, G. Jaouen, Chem. Eur. J. 2003, 9, 5223–
5236; d) G. Gasser, I. Ott, N. Metzler-Nolte, J. Med. Chem. 2011, 54, 3–
25.
[9] a) Y. Endo, T. Iijima, Y. Yamakoshi, H. Fukasawa, C. Miyaura, M. Inada, A.
Kubo, A. Itai, Chem. Biol. 2001, 8, 341–355; b) J. F. Valliant, P. Schaffer,
K. A. Stephenson, J. F. Britten, J. Org. Chem. 2002, 67, 383–387; c) M. L.
Beer, J. Lemon, J. F. Valliant, J. Med. Chem. 2010, 53, 8012–8020; d) P. W.
Causey, T. R. Besanger, J. F. Valliant, J. Med. Chem. 2008, 51, 2833–2844.
[10] a) R. S. Muthyala, S. B. Sheng, K. E. Carlson, B. S. Katzenellenbogen, J. A.
Katzenellenbogen, J. Med. Chem. 2003, 46, 1589–1602; b) K. J. Kieser,
Abbreviations
Estradiol (E2); estrogen receptor (ER); estrogen response element
(ERE); selective estrogen receptor modulators (SERMs); relative
binding affinity (RBA); human liver cancer cells (HepG2); tetra-
hydrofuran (THF).
&6
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 8
ÝÝ
These are not the final page numbers!

SERM-Like Activity of Oxabicyclo[2.2.1]hept-5-enes
MED
D. W. Kim, K. E. Carlson, B. S. Katzenellenbogen, J. A. Katzenellenbogen,
J. Med. Chem. 2010, 53, 3320–3329; c) L. G. Hamann, J. H. Meyer, D. A.
Ruppar, K. B. Marschke, F. J. Lopez, E. A. Allegretto, D. S. Karanewsky,
Bioorg. Med. Chem. Lett. 2005, 15, 1463–1466.
[14] S. J. Liu, B. Zhou, H. Y. Yang, Y. T. He, Z. X. Jiang, S. Kumar, L. Wu, Z. Y.
Zhang, J. Am. Chem. Soc. 2008, 130, 8251–8260.
[15] a) T. Pirali, S. Busacca, L. Beltrami, D. Imovilli, F. Pagliai, G. Miglio, A. Mas-
sarotti, L. Verotta, G. C. Tron, G. Sorba, A. A. Genazzani, J. Med. Chem.
2006, 49, 5372–5376; b) A. Zarghi, P. N. P. Rao, E. E. Knaus, Bioorg. Med.
Chem. 2007, 15, 1056–1061.
[16] D. S. Mortensen, A. L. Rodriguez, K. E. Carlson, J. Sun, B. S. Katzenellenb-
ogen, J. A. Katzenellenbogen, J. Med. Chem. 2001, 44, 3838–3848.
[17] a) J. A. Katzenellenbogen, H. J. Johnson, H. N. Myers, Biochemistry 1973,
12, 4085–4092; b) K. E. Carlson, I. Choi, A. Gee, B. S. Katzenellenbogen,
J. A. Katzenellenbogen, Biochemistry 1997, 36, 14897–14905.
[18] J. C. Gould, L. S. Leonard, S. C. Maness, B. L. Wagner, K. Conner, T. Za-
charewski, S. Safe, D. P. McDonnell, K. W. Gaido, Mol. Cell. Endocrinol.
1998, 142, 203–214.
[11] a) E. S. Mull, V. J. Sattigeri, A. L. Rodriguez, J. A. Katzenellenbogen,
Bioorg. Med. Chem. 2002, 10, 1381–1398; b) C. Ramesh, B. Bryant, T.
Nayak, C. M. Revankar, T. Anderson, K. E. Carlson, J. A. Katzenellenbogen,
L. A. Sklar, J. P. Norenberg, E. R. Prossnitz, J. B. Arterburn, J. Am. Chem.
Soc. 2006, 128, 14476–14477.
[12] a) K. W. Nettles, J. B. Bruning, G. Gil, J. Nowak, S. K. Sharma, J. B. Hahm,
K. Kulp, R. B. Hochberg, H. Zhou, J. A. Katzenellenbogen, B. S. Katzenel-
lenbogen, Y. Kim, A. Joachmiak, G. L. Greene, Nat. Chem. Biol. 2008, 4,
241–247; b) A. K. Shiau, D. Barstad, J. T. Radek, M. J. Meyers, K. W. Net-
tles, B. S. Katzenellenbogen, J. A. Katzenellenbogen, D. A. Agard, G. L.
Greene, Nat. Struct. Biol. 2002, 9, 359–364; c) H. B. Zhou, S. B. Sheng,
D. R. Compton, Y. Kim, A. Joachimiak, S. Sharma, K. E. Carlson, B. S. Kat-
zenellenbogen, K. W. Nettles, G. L. Greene, J. A. Katzenellenbogen, J.
Med. Chem. 2007, 50, 399–403; d) H.-B. Zhou, K. W. Nettles, J. B. Brun-
ing, Y. Kim, A. Joachimiak, S. Sharma, K. E. Carlson, F. Stossi, B. S. Katze-
nellenbogen, G. L. Greene, Chem. Biol. 2007, 14, 659–669.
[19] E. M. McInerney, M. J. Tsai, B. W. O’Malley, B. S. Katzenellenbogen, Proc.
Natl. Acad. Sci. USA 1996, 93, 10069–10073.
[20] J. A. Katzenellenbogen, J. Med. Chem. 2011, 54, 5271–5282.
Received: January 23, 2012
[13] A. K. Shiau, D. Barstad, P. M. Loria, L. Cheng, P. J. Kushner, D. A. Agard,
G. L. Greene, Cell 1998, 95, 927–937.
Published online on && &&, 0000
ChemMedChem 0000, 00, 1 – 8
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
&
7
&
ÞÞ
These are not the final page numbers!

FULL PAPERS
Y. Zheng, M. Zhu, S. Srinivasan,
J. C. Nwachukwu, V. Cavett, J. Min,
K. E. Carlson, P. Wang, C. Dong,
J. A. Katzenellenbogen,* K. W. Nettles,
H.-B. Zhou*
A set of estrogen receptor ligands
with a novel phenyl 2,3-diaryl-7-oxabi-
cyclo[2.2.1]hept-7-ene-5-sulfonate core
structure with various substituents on
the phenyl sulfonate have been pre-
pared (5,6-bis-(4-hydroxyphenyl)-7-oxa-
bicyclo[2.2.1]hept-5-ene-2-sulfonic acid
phenyl ester (OBHS), shown). Evaluation
of these OBHS derivatives showed a dis-
tinctive pattern of affinity and transcrip-
tional activity, and the tolerance for
ortho substituents can be understood
from a new X-ray crystal structure.
&& – &&
Development of Selective Estrogen
Receptor Modulator (SERM)-Like
Activity Through an Indirect
Mechanism of Estrogen Receptor
Antagonism: Defining the Binding
Mode of 7-Oxabicyclo[2.2.1]hept-5-ene
Scaffold Core Ligands
&8
&
www.chemmedchem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 0000, 00, 1 – 8
ÝÝ
These are not the final page numbers!