Synthesis and evaluation of (17α,20Z)-21-(4-substituted-phenyl)-19- norpregna-1,3,5(10),20-tetraene-3,17β-diols as ligands for the estrogen receptor-α hormone binding domain: Comparison with 20E-isomers

Article abstract of DOI:10.1021/jm040157s

As part of our ongoing program to develop probes for the hormone binding domain of the estrogen receptor-α (ERα), we prepared and evaluated a series of 17α,Z-(4-substituted-phenyl)-vinyl estradiol derivatives. The results indicated that the relative bindi

Full text of DOI:10.1021/jm040157s

4300
J. Med. Chem. 2005, 48, 4300-4311
Synthesis and Evaluation of
(17r,20Z)-21-(4-Substituted-phenyl)-19-norpregna-1,3,5(10),20-tetraene-3,17â-diols
as Ligands for the Estrogen Receptor-r Hormone Binding Domain: Comparison
with 20E-Isomers
Robert N. Hanson,*,†,‡ Carolyn J. Friel,‡ Robert Dilis,† Alun Hughes,§ and Eugene R. DeSombre§
Departments of Chemistry and Pharmaceutical Sciences, Northeastern University, 360 Huntington Avenue, Boston,
Massachusetts 02115-5000, and The Ben May Institute for Cancer Research, The University of Chicago, 5485 South Maryland
Avenue, Chicago, Illinois 60637
Received August 16, 2004
As part of our ongoing program to develop probes for the hormone binding domain of the
estrogen receptor-R (ERR), we prepared and evaluated a series of 17R,Z-(4-substituted-phenyl)-
vinyl estradiol derivatives. The results indicated that the relative binding affinities (RBAs) at
25 °C for the new compounds were significant (RBA ) 9-57) although less than that of estradiol
(RBA ) 100) or of the parent unsubstituted phenylvinyl estradiol (RBA ) 66). All of the
Z-compounds were full agonists in the uterotrophic assay, indicating that the ligands formed
estrogen-like complexes with the estrogen receptor-R hormone binding domain (ERR-HBD).
Comparison of corresponding Z- and E-4-substituted phenylvinyl ligands complexed with the
ERR-HBD indicated small but significant differences in binding modes that may account for
the differing trends seen in the structure-activity relationships for the two series.
Introduction
the estrogen receptor hormone binding domain
(ERR/â-HBD) and as potential therapeutic agents for
the treatment of hormone-responsive disorders, includ-
ing breast cancer. Observations that ERR is overex-
pressed in most breast cancers (>75%) and that acti-
vation of that receptor stimulates breast cancer cell
proliferation strongly suggested that agents that selec-
tively block hormone-initiated processes would have
significant therapeutic potential.11-13 Other investiga-
tors have described efforts to develop selective estrogen
receptor modulators (SERMs) as potential chemothera-
peutic agents.14-22 Additional studies23,24 highlighting
the linkage between the estrogen receptor and tumor
progression have made identification of new estrogen
receptor ligands that clarify the activation process an
important research target.
Our previous publications in the field described the
identification of the 17R-phenyl-X-vinyl or halovinyl
estradiols as potential ER ligands and synthetic meth-
ods for their preparation.25-29 Subsequent studies sug-
gested that modifications on structurally similar hor-
mone receptor ligands may elicit subtle changes in
receptor conformations that modulate the biological
responses of these compounds.30-34 Our recent evalua-
tion of the series of 17R,E-(2-, 3-, 4-trifluoromethylphe-
nylvinyl) estradiols in the uterotrophic assay indicated
a range of estrogenic potencies that was only partially
accounted for by the relative binding affinities (RBAs)35
(Figure 1). On the basis of these observations, we
prepared a series of new ER ligands encompassing a
number of 17R,E-(4-substituted-phenyl)vinyl estradiols
as probes for the interactions between the ligand and
binding site of the ERR-HBD (Figure 2).36 Understand-
ing those interactions would allow us to design and
prepare compounds having higher affinity for the recep-
tor, possessing selectivity for specific subtype and
The estrogen receptor (ER), a member of the nuclear
receptor (NR) superfamily, regulates a wide range of
developmental and physiological responses including
reproductive functions, cardiovascular modulation, and
bone density.1-4 This receptor, similar to the other NRs,
is characterized by a distinctive structural homology
consisting of six domains (A-F) that include the
N-terminal transcription activation function-1 (AF-1)
domain, a DNA-binding domain (DBD), a hinge region,
a hormone-binding domain (HBD), and a transcription
activation function-2 (AF-2) domain. Sequencing and
cloning studies revealed that the DBD utilized two zinc
fingers to recognize the cognate portion of the ER-
response element (ERE), while selective components of
the HBD provided the binding site for the hormone.
Subsequent studies have generated an increasingly
complex picture of how the interactions of hormone
agonists and antagonists with the ER initiate intracel-
lular events that eventually express the pharmacological
response.5-8 Differential expression of ERR and ERâ
subtypes, coupled with variable coactivator or corepres-
sor proteins, may contribute to the explanation of how
the same hormonal ligand can elicit a variety of effects
in different tissues. These newer models of estrogen
action help to understand the roles of the hormone in
pathological states, such as carcinoma of the breast, in
which cellular growth and differentiation have been
altered.9,10
Our research program has focused on the develop-
ment of novel estradiol derivatives, both as probes for
* To whom correspondence should be addressed. Phone:
(617) 373-3313. Fax: (617) 373-8795. E-mail: r.hanson@neu.edu.
† Department of Chemistry, Northeastern University.
‡ Department of Pharmaceutical Sciences, Northeastern University.
§ The University of Chicago.
10.1021/jm040157s CCC: $30.25 © 2005 American Chemical Society
Published on Web 06/02/2005

Diols for Hormone Binding Domain
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4301
temperature, and when the reaction time was relatively
short (<4 h). The isolated yields for 2 were 30-40%,
with the E-isomer 3 being present in 5-15% yields. The
remaining material under these conditions was mostly
unreacted ethynyl estradiol 3-acetate 1 accompanied by
small amounts of the proteodestannylated 4. Efforts to
increase the conversion to the Z-isomer by modifying
this method gave higher yields of the E-isomer. Al-
though there are other literature methods for preparing
Z-vinylstannanes,37,38 we did not investigate them
because we could obtain sufficient quantities of pure
intermediate for our immediate studies.
Figure 1. Phenyl ring isomers.
The Z-stannylvinyl estradiol 3-acetate 2 was then
coupled to the corresponding aryl iodide or bromide
using the Stille coupling conditions.36 As the results in
Table 1 indicate, yields of the Z-phenylvinyl estradiol
3-acetates 5a,c-h were low using this method (2-35%).
Under these conditions some starting material 2 was
recovered and the coupled products accounted for only
a small percentage of the reaction mixture. The major
product was the protiodestannylated vinyl estradiol
3-acetate 4. The coupled intermediates 5 were hydro-
lyzed with sodium hydroxide-methanol to give the
target 17R,Z-(4-substituted-phenyl)vinyl estradiols 6a,c-
Figure 2. Para substitution on phenyl ring.
expressing a more appropriate transcriptional response.
The results from this work described the synthetic
feasibility of Pd(0)-catalyzed reactions. The molecular
modeling methods used for the docking studies sug-
gested that the steroidal portion of the compounds
bound in a manner similar to estradiol in the published
crystal structures. However, the 17R-substituent oc-
cupied a space bounded by phenylalanine-525 and three
methionine residues, a region not normally accessed by
other steroidal and nonsteroidal ligands. This docking
process generated calculated binding energies that gave
a strong correlation with the observed RBA values. In
this study we report the extension of the synthetic
methods to the preparation of 17R,Z-(4-substituted-
phenyl)vinyl estradiols, their evaluation as estrogenic
agents, and the use of molecular modeling to describe
their use as probes for the ERR-HBD. As the results
indicate, these compounds are potent estrogenic agents
with structure-activity relationships that differ from
1
h, which were characterized by H and 13C NMR and
elemental analysis. The Z-stereochemistry was estab-
lished by 1H NMR, where the coupling constant for the
vinylic protons was J ) 12-13 Hz. No isomerization of
products was noted during the reaction or purification
steps.
The low yields and difficulties encountered in separa-
tion led us to examine alternative synthetic methods.
We previously reported the essentially quantitative
conversion of the 17R,E- and Z-stannylvinyl estradiols
to the corresponding 17R,E- and Z-iodovinyl estra-
diols.25,29 We anticipated that Z-iodovinyl estradiol 7
would then participate in the reverse coupling reaction
with either the arylstannane (reverse Stille coupling,
method B) or arylboronic acid (Suzuki coupling, method
C). Although iododestannylation proceeded quantita-
tively, the product 7 proved to be photosensitive, requir-
ing protection against ambient exposure to light. The
iodovinyl estradiol 7 was purified by flash chromatog-
raphy and used immediately with the arylstannane or
arylboronic acid. One reverse Stille coupling with the
4-tributylstannylacetophenone was performed to dem-
onstrate the feasibility of this method. Although aryl-
boronic acids are more readily available and more stable
than the stannanes, only one Suzuki coupling was
performed with 4-fluorophenylboronic acid. Yields of the
Z-phenylvinyl estradiol 3-acetates 5b,i were 34% and
52%, respectively, although the vinyl estradiol 4 was
still a significant component of the reaction mixture. The
intermediates were hydrolyzed to remove the 3-acetyl
group, and the products 6b,i were characterized by
NMR spectrometry and elemental analysis.
those
of
17R,E-analogues. This difference can be explained by
an alternative binding mode as determined by docking
studies.
Results
Synthesis of Estrogenic Ligands. The target com-
pounds selected in this study were the Z-isomers of the
17R,E-(4-substituted-phenyl)vinyl estradiols from the
previous series.36 We had observed in earlier studies
that synthesis of the Z-phenylvinyl estradiols was more
complex that of the E-isomers. The requisite Z-tri-n-
butylstannylvinyl estradiol was more difficult to obtain
in high yields and, being the thermodynamically less
favored product, was also much more prone to pro-
teolytic decomposition. Therefore, different synthetic
strategies were used to prepare the target compounds
(Scheme 1).
The initial approach, method A, was a variation of
that used for the E-isomers. Hydrostannation of ethynyl
estradiol 3-acetate 1 with tri-n-butyltin hydride using
triethylborane as the radical initiator gave reasonable
yields of the Z-vinylstannane 2 when tin/alkyne ratios
were low (<2), when the reaction was run at ambient
Relative Binding Affinities. Our evaluative strat-
egy involved an initial screening of the new compounds
and standard ligands for their relative binding affinities
(RBAs) at 4 and 25 °C using the ERR-HBD isolated from
the transfected BL21 cells.39 Competitive radiometric
binding assays using this protein have provided repro-
ducible RBA values for ERR-HBD. Ligands that dem-
onstrate significant affinity (RBA > 10) for the ER-HBD

4302 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Hanson et al.
Scheme 1. Synthesis of Z-17R-(4-Substituted-phenyl)vinyl Estradiols
Table 1. Yields of 17R-Z-(4-Substituted-phenyl)vinyl Estradiol
3-Acetates and 3-OH Products
Table 2. Relative Binding Affinity for
17R-Z-(4-Substitituted-phenyl)vinyl Estradiols
X
yield of -3-acetate, %
yield of -3-OH, %
RBA, %
H
F
10
44
10
2
32
25
35
5
89
31
55
90
44
94
68
51
at 4 °C
at 4 °C
at 25 °C
at 25 °C
compd
X
for Z
for E
for Z
for E
OH
6a
6b
6c
6d
6e
6f
6g
6h
6i
H
F
57
43
12
20
10
5
20
22
22
16
24
21
9
10
5
36
66
44
25
12
9
9
22
25
27
18
8
CN
CH3
OCH3
NH2
COCH3
OH
CN
CH3
CF3
9
OCH3
NH2
COCH3
39
26
29
32
would subsequently be evaluated for efficacy in the in
vivo rat uterotrophic growth assay.40 The products were
compared to both estradiol and the unsubstituted
phenylvinyl estradiol 6a using this assay in order to
evaluate the effects of substituents on both the basic
steroidal interactions and on the phenylvinyl group. The
results of the competitive binding assays, shown in
Table 2, indicated that all of the compounds retained
significant affinity for the estrogen receptor. In most
cases, the differences between the 4 °C RBA values
(kinetic) and the 25 °C RBA values (equilibrium) were
not dramatic.41 RBA values for the corresponding
E-isomers36 are included for comparison purposes.
As previously noted, the introduction of the
Z-phenylvinyl group at the 17R-position of estradiol
53
60
results in a modest reduction in the relative binding
affinity at 4 and 25 °C (RBA ) 57%, 66%) while the
E-isomer produces
a
more significant decrease
(RBA ) 16%, 9%). Introduction of substituents at the
para position of the phenyl ring leads to a further
reduction in the RBA values. Compounds 6d-f have
only 1/8 to 1/6 the binding affinity (RBA ) 9-12%) of the
unsubstituted compound, while compounds 6c,h,i are
approximately 30-40% as effective as 6a as estrogen
receptor ligands. Only two compounds in this series, 6b
and 6g, had affinities (RBA ) 39-57) greater than half
of the parent compound, and none were more potent.

Diols for Hormone Binding Domain
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4303
Table 3. Uterotrophic Growth Assay for Selected Z- and
E-Isomers
binding interactions observed with the Z-isomers to
those predicted for the corresponding E-phenylvinyl
isomers. We used a molecular modeling approach simi-
lar to that described in our earlier study36 to evaluate
the interaction of the new probes with the ERR-HBD
and explain the observed biological effects. On the basis
of the known pharmacology of the Z-isomers, we selected
the estradiol-ERR-HBD complex (PDB code 1G50)42 for
which the coordinates were readily available. The
estradiol-ERR-HBD monomer was optimized using
molecular mechanics methods and then was used for
the binding studies. Our ligands were superimposed on
the estradiol structure (steroidal components), the estra-
diol was deleted, and the new complex was subjected
to a docking procedure previously described. Best dock-
ing poses, selected from the docking study, and ad-
ditional poses, where the steroidal skeleton was super-
imposed on estradiol and the phenylvinyl moiety had
different starting conformations, were optimized using
simulated annealing procedures. Alternative modes in
which the molecules were rotated around the 3-17 axis
were also examined.
The binding mode in which the steroid maintained
the “standard” orientation and the dihedral angle
between it and the vinyl group was about -80° had the
lowest binding energy (Figure 3). In this ligand-ERR-
HBD conformation, the steroidal portion of the ligand
occupies essentially the same position as estradiol, only
translated by about 1.4 Å toward Glu-353 and Arg-394
at the end of the binding pocket. This shift increases
the distance between the 17-OH and His-524, possibly
reducing the hydrogen-bonding effect at that site. The
phenyl ring is located in a hydrophobic pocket on the
R-face of the steroidal C,D-region, bounded by two
phenylalanine residues (-404 and -425), two leucines
(-391 and -402), isoleucine-424, and methionine-421.
This pocket is formed by the conjunction of the â sheet,
helices 3, 5, 6, 7 and the 6-7 loop. The leucines in
particular provide a cap to that portion of the binding
pocket juxtaposed to the para position of the phenyl
group. Therefore, we examined the influence of the
4-substituent on the ligand-receptor binding. Although
the 4-substituent had little effect on the ultimate
conformation of the ligand-HBD complex, interactions
between the 4-substituent and the leucine side chains
were observed. Even though all of the substituents
evaluated were sterically undemanding, still some
movement of the receptor side chains was required to
generate optimal binding.
ED50,a nmol
R
for Z
for E
CH3
CF3
F
4.2
1.1
2.3
5.5
10.6
72
a ED50
: dose equal to 50% response of 10 nmol estradiol.
ED50(estradiol) ) 0.06 nmol.
This is in contrast to the E-(4-substituted-phenyl)vinyl
estradiol series in which almost every compound had
an RBA value greater than that of the unsubstituted
parent compound. In that series the best derivatives
were 3-6 times more effective than the parent com-
pound. Therefore, while there is a general similarity
between the two series of compounds, i.e., the individual
RBA values rarely differed by more than a factor of 2,
there exist significant differences in ERR-HBD binding
that require reconciliation.
Uterotrophic Growth Assays. The second compo-
nent of our biological evaluation process involved as-
saying the compounds possessing significant RBA val-
ues for estrogenic efficacy. Initially we had selected RBA
> 10 as the criterion for in vivo evaluation; however,
because all of the compounds had appropriate binding
values, we elected to completely evaluate only three
compounds from each series for this study. The im-
mature rat uterotrophic growth assay is well-established
for demonstrating estrogenic and antiestrogenic re-
sponses mediated though ERR.40 Although reporter
assays have been developed more recently to define
estrogenic activity, they were not available at the time
to assay this series of compounds. Three compounds in
each series were evaluated over a 4 log dose range
(0.01-100 nmol) for their uterotrophic potency. All of
the compounds in the series were full agonists and at
the highest tested doses yielded uterotrophic growth
that was comparable to 10 nmol of estradiol. Therefore,
the complex that the compounds formed with the intact
receptor was competent to elicit a full agonist response.
The results for the uterotrophic assay are shown in
Table 3. Although the compounds are significantly less
potent than estradiol in this assay, the results indicate
two trends. First, the Z-isomers are generally more
potent than the E-isomers. Compounds 6b and 6f are
approximately 30 and 10 times more potent than the
corresponding E-isomer. The Z- and E-4-methylphenyl-
vinyl estradiols are essentially equipotent in this assay
(4.2 vs 5.5 nmol). The second trend is that there exists
essentially no obvious correlation between the RBA
values and the ED50 values obtained in this study. The
Z-4-methyl and 4-fluoro compounds had higher RBA
values than the Z-4-trifluoromethyl ligand but ex-
pressed roughly 1/2 to 1/4 the in vivo potency. The same
absence of a correlation between the in vitro binding
affinity and in vivo uterotrophic growth potency was
also observed for the E-4-fluoro and E-4-trifluoromethyl
compounds.
Initial efforts to obtain a direct correlation between
the calculated binding energy and the observed RBA
values were not as successful as for the E-isomers
(r2 ) 0.83); nevertheless, a general trend existed indi-
cating that lower protein flexibility at this site led to
lower binding affinity (Figure 4A). However, when the
protein energy was factored in, the correlation between
the RBA values and the calculated binding energy was
more significant (r2 ) 0.96) (Figure 4B).
Because the correlations for this series were different
from those observed for the E-isomers, we compared the
two predicted binding modes. For this study we selected
the 4-fluorophenylvinyl estradiol isomers in their most
probable binding poses. The protein backbones were
superimposed, and residues that exhibited essentially
Molecular Modeling. The objectives of the molecu-
lar modeling component of the study were threefold. The
first aim was to determine the preferred ligand-HBD
conformation compared to estradiol. Then, on the basis
of that binding mode, the effect of the 4-substituent
could be evaluated. Third, we would compare the

4304 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Hanson et al.
Figure 3. Docking of Z-4-fluorophenylvinyl estradiol (green) and estradiol (purple) in ERR-LBD. An approximate 1 Å translation
of the steroid ring yields movement of Arg-394 and Glu-353 but permits positioning of substituted phenylvinyl group in a
hydrophobic pocket bounded by Phe-, Leu-428, Leu-402 residues.
no deviation were removed (Figure 5). The Z-4-fluoro-
phenylvinyl group occupies a hydrophobic pocket on the
R-face of the ligand binding pocket, roughly in the same
region as that observed for the E-4-fluorophenylvinyl
estradiol. However, the identity of the amino acids lining
the binding pocket for the 4-fluorophenylvinyl group is
different for each isomer. The Z-stereochemistry, coupled
with the torsional rotation around the 17R-vinyl bond,
forces the phenyl ring into a binding pocket bounded
by Phe-404 and -425, Leu-402 and -391, Ile-424, and
Met-421, as opposed to the Phe-425, Met-341, -342, and
-421, Leu-346 and -410, and Val-418 grouping previ-
ously identified for the E-isomers. Although the steroi-
dal portions of the two series occupy similar domains,
having a number of residues in common, the environ-
ment for the terminal phenyl rings is significantly
different. For example, the degree of interaction with
Met-421 is different for the two isomers and the pocket
termini are different.
propriate coupling conditions that were developed for
the more stable E-vinylstannanes. The use of the
standard Pd(0) catalyst [tetrakis(triphenylphosphine)-
palladium(0)] to provide the Z-coupled materials
(2-35% isolated yields) required reaction conditions,
usually elevated temperatures, that gave the protiodest-
annylated material 4 and other unidentified compounds
as the major products. Recent studies of Pd(0)-catalyzed
coupling reactions suggest that more sterically hindered,
electron-rich phosphines may be even better Pd(0)
ligands.43,44 These catalysts are compatible with aryl
chlorides, opening this methodology to a wider variety
of substrates. We also were able to demonstrate that
the target compounds were accessible by alternative
means. Stille coupling in the reverse fashion (method
B), with vinyl iodide and arylstannane, was also suc-
cessful as was the Suzuki coupling (method C) with the
vinyl iodide and arylboronic acid. Substituted arylstan-
nanes and boronic acids are commercially available and
therefore also provide access to many substituted phe-
nylvinyl estradiols. However, should more highly sub-
stituted phenylvinyl derivatives become necessary, meth-
ods B and C would be limited by the availability or
synthesis of the requisite arylstannanes and boronic
acids. Efforts to improve coupling yields remain to be
done, but the concept is valid. As in the case of the
E-isomers, saponification of the 3-acetate intermediates
was uneventful and gave the final products in yields of
31-94%. The Z-stereochemistry of the olefin was es-
Discussion
Chemistry. The choice of Z-(4-substituted-phenyl)-
vinyl estradiols was based on the biological activity
expressed by the corresponding Z-halovinyl and
Z-phenyl(seleno or thio)vinyl estradiols. However, the
chemistry of C-C coupling reactions with Z-vinylstan-
nanes, especially functionalized vinylstannanes is less
well-described than for E-vinylstannanes. Our studies
were hampered by the fact that hydrostannation under
radical conditions gave the Z-isomer as the minor
product unless the time, temperature, and stoichiometry
of the reaction were carefully controlled. Under kineti-
cally dominating conditions we were able to isolate the
Z-vinylstannane as the major product although in
modest yields. The second problem involved the ap-
1
tablished by H NMR, which gave a coupling constant
of J ) 12.5-13.5 Hz.
Biological Evaluation. The results of the biological
assays indicated that the Z-(4-substituted-phenyl)vinyl
estradiol derivatives that we prepared were potent
estrogenic ligands both in vitro and in vivo. The range

Diols for Hormone Binding Domain
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4305
however, the Z-isomer is 10 times more potent in vivo
(Z ED50 ) 1.1 nmol vs E ED50 ) 10.6 nmol). On the other
hand, of the methylphenyl isomers, the E-isomer has a
higher RBA value but the compounds are essentially
equipotent in vivo. Even within the isomeric series, the
order of ED50 values does not match the order of RBA
values. As has been suggested by others,45 results from
ER-HBD studies do not necessarily reflect the intact ER.
The presence of the additional functional group at the
para position and the stereochemistry of the vinyl group
may influence the orientation of the peptide backbone
or the surface features. Such minor alterations in
secondary or tertiary structure may affect the ability
of the ER-HBD to elicit appropriate conformational
adaptations of the other domains of the intact ER
responsible for dimerization and DNA recognition.
Therefore, evaluation of the data provide important
information not only for structure-activity relationships
in this series and its relationship to the corresponding
E-series but also for understanding the effect of ligand
binding on the subsequent biological response.
Structure-Activity Relationships in the Z-(4-
Substituted-phenyl)vinyl Estradiols. We prepared
and evaluated the E-(4-substituted-phenyl)vinyl estra-
diols as part of our program to develop chemical probes
to explore the binding pocket of the estrogen receptor-
R. Because the compounds demonstrated significant
ERR-HBD affinity, we believed that interactions be-
tween the ligand and the residues within the binding
pocket could be evaluated by molecular modeling. Other
investigators have applied a variety of computational
methods to evaluate ligand binding to the estrogen
receptor subtypes.46-52 We had synthesized several
Z-phenylvinyl estradiols and examined their preferred
solution and gas-phase conformations using a combina-
tion of NMR and computational methods.53 Those
results suggested that the Z-isomers should exhibit
different binding modes compared to the E-isomers.
The first relevant finding from this study was that
all of the 4-substituted derivatives that were evaluated
retained significant affinity for the ER-HBD. Although
previous studies indicated that Z-phenyl-(X)-vinyl estra-
diols generally displayed reasonable affinity for the
receptor, the effect of para substitution had not been
examined. The data clearly indicate that the presence
of a 4-substituent did not disrupt receptor binding.
However, in contrast to the E-isomers, where substitu-
tion generally increased receptor affinity compared to
the unsubstituted parent compound, in the Z-series all
of the substituted derivatives had RBA values less than
the parent compound. Because individual electronic,
hydrophobic, or steric parameters of the substituents
did not correlate directly with the observed order of
binding affinity, our interpretation of the results re-
quired evaluation of the presumed binding modes as
determined by molecular modeling and its comparison
with our previous modeling study.
Figure 4. (A) Calculated binding energy vs RBA at 25 °C
without p-CH3. (B) Sum of (0.2)(protein) plus binding energies
vs RBA at 25 °C without p-CH3.
of RBAs at 25 °C was from 9 to 66, and these values
were, except for 6d (X ) CN), equal to or greater than
the values obtained at 4 °C. Although there was a
variation in the order of potencies, the binding behavior
was generally similar to that expressed by the corre-
sponding E-isomers (Table 2) Where differences existed,
e.g., X ) F, CN, CH3, and COCH3, the magnitude was
usually less than a factor of 2. The data suggested that
the compounds may interact with the receptor in similar
binding modes.
On the basis of the in vitro data, three pairs of
compounds from the Z- and E-series were selected for
evaluation in the immature female rat uterotrophic
growth assay. In this case, there was no obvious direct
relationship between the RBAs and the in vivo response.
Comparing the corresponding E- and Z-phenylvinyl
isomers is illustrative. Both 4-fluorophenyl isomers are
potent in vitro ER-HBD binding agents (Z RBA ) 44
vs E RBA ) 22), but the Z-isomer is more potent in vivo
Docking studies were performed with the E- and
Z-(4-fluorophenyl)vinyl estradiol isomers as a basis for
interpreting the observed biological results. Both iso-
mers adopted low-energy complexes with the ERR-HBD
structure, with the steroidal rings essentially overlap-
ping the estradiol skeleton (Figure 3). The most notable
effect for the steroidal component was an approximately
by a factor of 30-fold (Z ED50 ) 2.3 nmol vs E ED50
)
72 nmol). The two trifluoromethyl isomers have es-
sentially the same in vitro RBA values (RBA ) 9 vs 8);

4306 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Hanson et al.
Figure 5. Docking of E- (purple) and Z--4-fluorophenylvinyl estradiols (green) in ERR-HBD. The position of steroid skeleton
within the binding pocket is similar for both isomers; however, the orientation of the 4-substituted phenylvinyl groups is different.
The E-isomer is bounded by Met-421, -341, -342 and Phe-425, whereas the Z-isomer is bounded by Phe-404 and -425 and Leu-402
and -428.
1.4 Å translational movement toward the Arg-394, Glu-
353 residues that hydrogen-bond to the A-ring phenolic
hydroxyl. Such movements have been observed in the
crystal structures of other ER-HBD complexes and
apparently are well-tolerated.48 Other amino acids
intimately associated with estradiol binding and the
agonist response, His-524, Phe-404, and Met-342, -343,
421, undergo only minor perturbations, i.e., <1 Å. The
movement of the ligand along that axis allows the
Z-phenylvinyl ring to adopt a conformation that inserts
the aromatic ring into a previously unidentified hydro-
phobic pocket similar but not identical to that occupied
by the E-phenylvinyl moiety (Figure 5). The pocket that
surrounds the terminal phenyl group comprises Phe-
404 and -425, Leu-391 and -402, Ile-424, and Met-421.
Unlike the binding pocket for the E-isomer, the presence
of an amino acid residue (Leu-402) immediately juxta-
posed to the para position imposes significant steric
constraints. As the receptor binding values (RBAs) with
the ERR-HBD suggest, introduction of the 4-substituent
causes a reduction in binding affinity. That binding is
not affected more indicates that the phenylvinyl group
or the binding pocket or both can adapt to achieve a
lower energy complex. The influence of protein energy
on ligand binding is reflected in its contribution to the
correlation between calculated binding energies and
observed RBA values (Figure 4). It is notable that for
the E-isomers where no receptor residue is juxtaposed
to the para substituent and steric factors are not
significantly involved, there was essentially no contri-
bution of the protein energy to the correlation equation.
The adaptive response by the ligand and receptor
clearly occurs within the agonist conformation because
all of the compounds, E- and Z-isomers, generate full
estrogenic responses in vivo. The interaction between
the ligand and the intact receptor present within the
normal cell is not fully represented by the model
developed with the ER-HBD. This is apparent from the
lack of a clear correlation between the in vitro RBA
values and the in vivo ED50 values. The presence of the
other receptor domains in the full-length receptor and
the coregulatory (activator) proteins provides influences
that modify the biological response in ways not readily
accessible to computational analysis. Slight perturba-
tions in the receptor peptide backbone may enhance or
suppress the interactions between the DBD and the
ERE of the DNA. Similarly, perturbations at the surface
of the HBD can enhance, suppress, or modify the
selectivity of the coactivator protein binding required
for the transcriptional response. Therefore, it remains
an important consideration of binding studies not to
attach excessive significance to binding affinities and
computational modeling in predicting the in vivo bio-
logical activity of NR ligands.
The docking study does provide the basis for ad-
ditional considerations of these compounds as potential
ER-subtype selective ligands. Although the homology
between the ERR-HBD and ERâ-HBD is less than
60%,54 within the binding site itself there are only two
mutations. One of these, Leu-384 f Met- occurs on the
â-face of the steroid and would not be significantly
involved with interactions with the phenylvinyl group
of our compounds. The second, however, is Met-421f
Ile-, and this specific amino acid is one of three meth-
ionine residues that we predicted would line the hydro-
phobic pocket enfolding the E-phenylvinyl moiety. It is
also associated with the Z-phenylvinyl binding mode,
although not in the same manner. If our proposed
binding model is correct, the E-phenylvinyl estradiol
derivatives should display greater ER-subtype selectiv-
ity than the corresponding Z-isomers.
In summary, using three Pd(0)-catalyzed coupling
reactions, we have prepared and characterized a series
of unique probes for the ERR-HBD. Competitive binding

Diols for Hormone Binding Domain
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4307
(10 mL) solution was added, and the mixture was stirred for
8 h. The mixture was partitioned between ethyl acetate and
water (100 mL:100 mL), and the Bu3SnF was removed by
filtration. The organic layer was separated, and the aqueous
layer was extracted twice with ethyl acetate. The organic
layers were combined, washed with NH4Cl/H2O (100 mL),
water (2 × 100 mL), and brine (2 × 100 mL), dried over
magnesium sulfate, and concentrated. The residue was puri-
fied by chromatography on silica gel using hexanes/ethyl
acetate (5:1) as the eluent. Evaporation of the solvent gave
the purified 5. In most cases the purified material was used
directly for the hydrolysis step.
3-Acetoxy-(17R,20Z)-21-(phenyl)-19-norpregna-1,3,5-
(10),20-tetraene-17â-ol (5a). A mixture of bromobenzene
(0.17 g, 1.1 mmol), tetrakis(triphenylphosphine)palladium(0)
(0.024 g, 0.02 mmol), one crystal of 2,6-di-tert-butyl-4-methyl-
phenol, and 4 (0.6 g, 0.9 mmol) was reacted using method A.
The residue was chromatographed on silica gel (16 g), using
hexane/ethyl acetate (5:1) to afford recovered 4 (0.1 g) and 5a
(0.06 g, 18%, based on recovered staring material): Rf ) 0.35
(hexane/ethyl acetate 5:1).
assays indicated that the new substituted compounds
possessed significant receptor binding affinity compared
to the unsubstituted parent compound and to estradiol.
All of the new compounds expressed full estrogenic
activity in vivo, but potency levels did not correlate with
the observed in vitro RBA values. Molecular modeling
of the ligand-ERR-HBD complexes provided a rationale
for the observed binding results but not the in vivo
biological activity. The model suggests that the E- and
Z-substituted phenylvinyl estradiols access different
hydrophobic pockets within the ERR-HBD and that
these pockets are the result of an adaptive response by
the receptor to the presence of the ligand. Modeling of
the interactions between putative ligands and the
individual receptor subtype binding pockets provides the
basis for designing ligands with enhanced receptor
affinity and receptor subtype selectivity. Studies di-
rected toward these goals are in progress.
3-Acetoxy-(17R,20Z)-21-[(4-hydroxy)phenyl]-19-norpre-
gna-1,3,5(10),20-tetraene-17â-ol (5c). A mixture of 4-iodo-
phenol (0.29 g, 1.4 mmol), tetrakis(triphenylphosphine)-
palladium(0) (0.024 g, 0.02 mmol), and 4 (0.41 g, 0.65 mmol)
in anhydrous toluene (25 mL) was reacted using method A.
The residue was chromatographed on silica gel (6 g) using
hexane/ethyl acetate (5:1) to afford recovered 4 (0.1 g) and 5c
as a mixture (0.04 g, 14%; based on recovered staring material,
19%): Rf ) 0.12 (hexane/ethyl acetate 5:1).
3-Acetoxy-(17R,20Z)-21-[(4-cyano)phenyl]-19-norpregna-
1,3,5(10),20-tetraene-17â-ol (5d). A mixture of 4-bromo-
benzonitrile (0.13 g, 0.7 mmol), tetrakis(triphenylphosphine)-
palladium(0) (0.016 g, 0.014 mmol), and 2 (0.31 g, 0.5 mmol)
in toluene (5 mL) was reacted under method A conditions for
72 h. The residue was chromatographed on silica gel (50 g)
and eluted with hexane/ethyl acetate (5:1) to yield 5d
(0.005 g, 2%).
Experimental Section
General Methods. All reagents and solvents were pur-
chased from Aldrich or Fisher Scientific. THF and toluene were
distilled from sodium/benzophenone. Reactions were monitored
by TLC, performed on 0.2 mm silica gel plastic backed sheets
containing F-254 indicator. Visualization on TLC was achieved
using UV light, iodine vapor, and/or phosphomolybdic acid
reagent. Column chromatography was performed with 32-
63 µm silica gel packing. Melting points were determined using
an Electrotherm capillary melting point apparatus and are
uncorrected. NMR spectra chemical shifts are reported in parts
per million downfield from TMS and referenced to TMS
internal standard for deuterochloroform (DCCl3) or deutero-
acetone (CD3COCD3) solvent peak. Coupling constants are
reported in hertz. All compounds gave satisfactory elemental
analyses, (0.4% (Atlantic Microchemical Laboratories, Inc.,
Norcross, GA), unless otherwise stated.
3-Acetoxy-(17R,20Z)-21-[(4-methyl)phenyl]-19-norpre-
gna-1,3,5(10),20-tetraene-17â-ol (5e). A mixture of p-iodo-
toluene (0.32 g, 1.5 mmol), tetrakis(triphenylphosphine)-
palladium(0) (0.005 g, 0.004 mmol), 2 (0.69 g, 1.1 mmol), and
1 crystal of 2,6-di-tert-butyl-4-methylphenol were heated with
refluxing anhydrous toluene (15 mL) for 7 h using method A.
The residue was chromatographed on silica gel using hexane/
ethyl acetate (5:1) to afford 5e (0.14 g, 32%): Rf ) 0.25 (hexane/
Synthetic Methods. 3-Acetoxy-(17R,20Z)-21-(tri-n-bu-
tylstannyl)-19-norpregna-1,3,5(10),20-tetraene-17â-ol (2).
To a solution of 1 (6.76 g, 20 mmol) in tetrahydrofuran
(20 mL) was added tri-n-butyltin hydride (8.5 mL, 31 mmol)
and 1 M triethylborane (1 mL, 8.8 mmol). The reaction mixture
was stirred under nitrogen at room temperature for 10 h. The
THF was removed under reduced pressure, and the resulting
oil was separated via silica gel column chromatography using
hexane/ethyl acetate (5:1) as both the packing and eluting
solvent. The product 2 was isolated as an oil that solidified
upon standing (4.6 g, 36%; 63% based on recovered starting
material): Rf ) 0.68 (hexane/ethyl acetate 5:1); mp 80-82 °C;
1H NMR in CDCl3 δ 0.8-2.8 (m, b, 42H, steroid and tri-
1
ethyl acetate 5:1); H NMR in CDCl3 δ 0.90 (s, 3H, 18-CH3),
1.2-2.8 (m, 18H, steroid nucleus), 5.89 (d, 1H, J20-21
)
12.80 Hz, 20-H), 6.49 (d, 1H, J21-20 ) 12.72 Hz, 21-H), 6.56
(d, 1H, J4-2 ) 2.6 Hz, 4-H), 6.62 (dd, 1H, J2-1 ) 8.49 Hz, J2-4
) 2.73 Hz, 2-H), 7.14 (m, 3H, 1-H, 23-H, 27-H), 7.34 (d, 2H,
J24-23 and J26-27 ) 7.89 Hz, 24-H and 26-H).
3-Acetoxy-(17R,20Z)-21-[(4-methoxy)phenyl]-19-nor-
pregna-1,3,5(10),20-tetraene-17â-ol (5g). A mixture of
p-iodoanisole (0.93 g, 4 mmol), tetrakis(triphenylphosphine)-
palladium(0) (0.016 g, 0.014 mmol), 4 (1.26 g, 2 mmol), and
one crystal of 2,6-di-tert-butyl-4-methylphenol were heated in
refluxing anhydrous toluene (25 mL) for 12 h. The residue was
chromatographed on silica gel using with hexane/ethyl acetate
(4:1) to afford 5g (0.22 g, 25%): Rf ) 0.35 (hexane/ethyl acetate
4:1); 1H NMR in acetone-d6 δ 0.90 (s, 3H, 18-CH3), 1.2-2.8
(m, 15H, steroid nucleus), 3.79 (s, 3H, OCH3), 5.87 (d, 1H,
J20-21 ) 13.08 Hz, 20-H), 6.39 (d, 1H, J21-20 ) 13.08 Hz,
21-H), 6.82 (m, 4H, 2-H, 4-H, 24-H, and 26-H), 7.30 (d, 1H,
J1-2 ) 8.79).
3-Acetoxy-(17R,20Z)-21-[(4-amino)phenyl]-19-norpregna-
1,3,5(10),20-tetraene-17â-ol (5h). A mixture of p-iodoaniline
(0.22 g, 5.7 mmol), tetrakis(triphenylphosphine)palladium(0)
(0.028 g, 0.02 mmol), 4 (1.7 g, 2.8 mmol), and one crystal of
2,6-di-tert-butyl-4-methylphenol were heated in refluxing an-
hydrous toluene (25 mL) for17 h. The residue was chromato-
graphed on silica gel using hexane/ethyl acetate (4:1) to afford
5h (0.06 g pure and 0.5 g mixture, >35%): Rf ) 0.32 (hexane/
ethyl acetate 5:1); 1H NMR in acetone-d6 δ 0.97 (s, 3H,
butyltin), 2.89 (s, 3H, -OCOCH3), 5.87 (d, 1H, J20-21
)
13.2 Hz, 20-H), 6.78 (d, 1H, J21-20 ) 12.0 Hz, 21-H), 6.79
(s, 1H, 4-H), 6.84 (d, 1H, J2-1 ) 8.4 Hz, 2-H), 7.28 (d, 1H +
CDCl3, J1-2 ) 9 Hz, 1-H); 13C NMR in acetone-d6 δ 13.00
(-(CH2CH2CH2CH3)3), 14.07 (-(CH2CH2CH2CH3)3), 14.63
(C-18), 20.93 (-OCOCH3), 24.07 (C-15), 27.10 (C-11), 28.02
(C-7), 28.16 (-(CH2CH2CH2CH3)3), 29.01 (-(CH2CH2CH2-
CH3)3), 33.77 (C-12), 39.58 (C-16), 40.29 (C-8), 44.83(C-9), 47.69
(C-13), 50.47 (C-14), 85.43 (C-17), 119.59 (C-2), 122.37 (C-4),
124.99 (C-21), 126.98 (C-1), 138.55 (C-10), 138.72 (C-5), 149.72
(C-20), 151.45 (C-3), 169.69 (-OCOCH3). Further elution gave
3-acetoxy-(17R,20E)-21-(tri-n-butylstannyl)-19-norpregna-1,3,5-
(10),20-tetraene-17â-ol (3) (1.2 g, 10%; based on recovered
starting material, 26%).
Method A. Stille Coupling. General Procedure. A
mixture of the 4-substituted aryl bromide or iodide (1.0 mmol),
tetraksi(triphenylphosphine)palladium(0) (0.02 mmol), one
crystal of 2,6-di-tert-butyl-4-methylphenol, and 2 (0.9 mmol)
was stirred in anhydrous toluene (10 mL). The reaction
mixture was heated at reflux under a nitrogen atmosphere
and monitored for the disappearance of 2. The reaction mixture
was cooled to ambient temperature, a 10% KF-CH3OH

4308 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Hanson et al.
18-CH3), 1.2-2.9 (m, 15H, steroid nucleus), 5.98 (d, 1H, J20-21
) 13.2 Hz, 20-H), 6.46 (d, 1H, J20-21 ) 12.9 Hz, 21-H), 6.78 (d,
Method C. Suzuki Coupling. 3-Acetoxy-(17R,20Z)-21-
[(4-fluoro)phenyl]-19-norpregna-1,3,5(10),20-tetraene-
17â-ol (5b). To a solution of 7 (0.46 g, 1 mmol) in THF
(15 mL) was added tris(dibenzylideneacetone)dipalladium
(0.15 g, 0.17 mmol), sodium bicarbonate (0.43 g, 4 mmols,
4 equiv, in 10 mL of water), and 4-fluorobenzene boronic acid
(0.29 g, 2.1 mmol). The reaction mixture was protected from
light and stirred at room temperature for 14 h. The mixture
was extracted with ethyl acetate (2 × 50 mL), washed with
water (2 × 100 mL) and brine (2 × 100 mL), dried over
magnesium sulfate, filtered, and concentrated to yield a brown
gummy residue. The residue was chromatographed on a silica
gel column (30 g) using 98:2 chloroform/methanol as the
eluting solvent to give 5b as a crude product (0.19 g, 44%):
Rf ) 0.45 (chloroform/methanol 98:2); 1H NMR in CDCl3 δ 0.97
(s, 3H 18-CH3), 1.2-2.9 (m, b, 15H, steroid nucleus), 5.90 (d,
1H, J20-21 ) 13.08 Hz, 20-H), 6.45 (d, 1H, J21-20 ) 12.7 Hz,
21-H), 6.78 (d, 1H, J4-2 ) 2.0 Hz, 4-H), 6.83 (dd, 1H, 2-H), 6.9
(∼t, 2H, 24-H and 26-H), 7.04 (d, contaminant), 7.26 (d, 1H
and CDCl3 peak, 1-H), 7.4 (m, 2H, 25-H and 27-H), 7.6-7.7
(m, contaminant). Further elution yielded the E-isomer 5o
(0.8 g, 17.5%).
Hydrolysis of 3-Acetoxy Intermediates. General Pro-
cedure. Compound 5 (0.3 mmol) was stirred in methanol
(10 mL) with 10 N sodium hydroxide (0.1 mL) for 20 min. The
solution was reacidified with a few drops of acetic acid. The
methanol was removed under vacuum, and the residue was
partitioned between ethyl acetate and water (100 mL:100 mL).
The aqueous layer was extracted with ethyl acetate. Organic
layers were combined, washed with water (100 mL), dried over
magnesium sulfate, and concentrated to yield 6. Recrystal-
lization from acetone/hexane afforded colorless crystals
(50-90%).
(17R,20Z)-21-(Phenyl)-19-norpregna-1,3,5(10),20-tet-
raene-3,17â-diol (6a). Compound 5a (0.06 g, 0.15 mmol) was
deprotected to yield an off-white powder (0.057 g, 0.15 mmol,
100%). Recrystallization in ethyl acetate afforded a white
powder (0.05 g, 0.13 mmol, 89%): Rf ) 0.03 (hexane/ethyl
acetate 5:1); mp 125-127 °C; 1H NMR in CDCl3 δ 0.90 (s, 3H,
18-CH3), 1.2-2.9 (m, 15H, steroid nucleus), 4.58 (s, 1H,
17-OH), 5.93 (d, 1H, J20-21 ) 12.78 Hz, 20-H), 6.53 (d, 1H,
J21-20 ) 12.81 Hz, 21-H), 6.57 (s, 1H, J4-2 ) 2.19 Hz), 6.62
(dd, 1H, J2-1 ) 8.58 Hz and J2-4 ) 2.7 Hz, 2-H), 7.17 (d, 1H,
J1-2 ) 8.79 Hz, 1-H), 7.25 (1H and CDCl3, m, 24-H), 7.27 (t,
2H, J ) 7.71 and 7.05 Hz, 23-H and 25-H), 7.44 (d, J )
7.35 Hz, 22-H and 26-H); 13C NMR in CDCl3 δ 14.07 (C-18)
23.30 (C-15), 27.46 (C-11), 29.67 (C-7), 26.52 (C-6), 32.58
(C-12), 38.62 (C-16), 39.66 (C-8), 43.83 (C-9), 48.04 (C-13),
49.72 (C-14), 85.09 (C-17), 112.71 (C-2), 115.27 (C-4), 126.55
(C-25), 127.13 (C-1), 127.91 (C-24 and C-26), 128.10 (C-23 and
C-27), 129.09 (C-21), 136.04 (C-22), 132.78 (C-10), 137.44
(C-20), 138.33 (C-5), 153.65 (C-25), 153.34 (C-3).
(17R,20Z)-21-[(4-Fluoro)phenyl]-19-norpregna-1,3,5(10),-
20-tetraene-3,17â-diol (6b). The crude acetylated product 5b
(0.19 g, mmol) was deprotected to yield 6b (0.053 g, 31%):
Rf ) 0.15 (hexane/ethyl acetate 5:1); mp 105-108 °C; 1H NMR
in CDCl3 δ 0.91 (s, 3H, 18-CH3), 1.2-2.9 (m, 15H, steroid
nucleus), 4.6 (s, b, 17-OH), 5.90 (d, 1H, J20-21 ) 12.84 Hz,
20-H), 6.45 (d, 1H, J21-20 ) 12.72 Hz, 20-H), 6.56 (d, 1H, J4-2
) 2.67 Hz, 4-H), 6.63 (dd, 1H, J2-1 ) 8.46 Hz, J2-4 ) 2.67 Hz,
1H, J4-2 ) 2.1 Hz, 4-H), 6.84 (dd, 1H, J2-4 ) 2.4 Hz, J2-1
8.7 Hz, 2-H), 7.21 (d, 1H, J1-2 ) 7.2 Hz, 1-H), 7.30 (m, 4H,
24-H, 26-H and CDCl3 peak), 7.60 (d, 4H, J23-24 and J27-26
)
)
7.2 Hz, 23-H and 27-H), 7.62 (d, 2H, J23-24 and J27-26 ) 8.91
Hz, 23-H and 27-H); 13C NMR in acetone-d6 δ 14.43 (C-18),
20.80 (-OCOCH3), 23.65 (C-15), 27.02 (C-11), 27.80 (C-7),
28.89 (C-6), 32.50 (C-12), 38.19 (C-16), 40.20 (C-8), 44.57
(C-9), 48.60 (C-13), 49.82 (C-14), 55.25 (-OCH3), 83.68 (C-17),
113.53 (C-24 and C-26), 119.46 (C-2), 122.25 (C-4), 126.88
(C-1), 129.71 (C-21), 130.42 (C-22), 132.35 (C-23 and C-27),
134.92 (C-10), 138.34 (C-20), 138.61 (C-5), 149.53 (C-3), 159.36
(C-25), 169.58 (-OCOCH3).
3-Acetoxy-(17R,20Z)-21-(iodo)-19-norpregna-1,3,5(10),-
20-tetraene-17â-ol (7). To a solution of 2 (0.77 g, 1.22 mmol)
in chloroform/methylene chloride (1:1, 10 mL) was added a
slurry of N-iodosuccinimide (0.36 g, 1.6 mmol) in the same
solvent solution. The reaction mixture was stirred, under
aluminum foil, at -78 °C for 2 h. The reaction mixture was
washed with saturated sodium bicarbonate/water (20 mL).
Aqueous and organic layers were separated. The aqueous layer
was extracted with chloroform (20 mL × 2). Organic layers
were combined washed with water (20 mL × 2) and brine
(20 mL × 2), dried over magnesium sulfate, and concentrated.
The yellow oil (1.26 g) was purifed on a silica gel column
(40 g) and covered with aluminum foil, using hexane/ethyl
acetate (5: 1) as the eluting solvent to give 7 as a colorless
crystals (0.34 g, 83%): Rf ) 0.25 (hexane/ethyl acetate 5:1);
1H NMR in CDCl3-d6 δ 0.96 (s, 3H, 18-CH3), 1.2-2.9 (m, 15H,
steroid nucleus), 6.38 (d, 1H, J21-20 ) 8.73 Hz, 21-H), 6.79 (d,
1H, J4-2 ) 2.46 Hz, 4-H), 6.85 (dd, 1H, J2-4 ) 2.55 Hz, 2-H),
6.85 (d, 1H, J20-21 ) 8.52 Hz, 20-H), 7.27 (d, 1H and CDCl3
peak, J1-2 ) 8.31 Hz, 1-H); 13C NMR in CDCl3 δ 14.13 (C-18),
23.24 (-OCOCH3), 21.11 (C-15), 26.09 (C-11), 27.20 (C-7),
29.50 (C-6), 32.02 (C-12), 37.72 (C-16), 39.10 (C-8), 43.83
(C-9), 48.41 (C-13), 49.67 (C-14), 72.91 (C-21), 84.88 (C-17),
118.61 (C-2), 121.50 (C-4), 126.39 (C-1), 137.77 (C-10), 138.17
(C-5), 143.40 (C-20), 148.41 (C-3), 169.85 (-OCOCH3).
4-Tributylstannylacetophenone (8).
A solution of
4-bromoacetophenone (1.08 g, 5.4 mmol), bis(tributyltin) (5 mL,
10 mmol), tetrakis(triphenylphosphine)palladium(0) (0.08 g,
0.07 mmol), and toluene (50 mL) was stirred under nitrogen
for 24 h at reflux. The black solution was concentrated and
chromatographed on a 45 g silica column packed with hexane
and eluted with 5:1 hexane/ethyl acetate to yield an oil 12
1
(1.22 g, 55%): Rf ) 0.82 (hexane/ethyl acetate 5:1); H NMR
in CDCl3 δ 0.8-1.7 (m, 27H, SnBu3), 2.58 (s, 3H, COCH3), 1.59
(d, 2H, J2-3 ) 8 Hz, 2-H and 6-H), 7.88 (d, 2H, J3-2 ) 8 Hz,
3-H and 5-H).
Method B. Reversed Stille Coupling. 3-Acetoxy-(17R,-
20Z)-21-[(4-acetyl)phenyl]-19-norpregna-1,3,5(10),20-tet-
raene-17â-ol (5i). A mixture of 12 (0.40 g, 1 mmol), tetrakis-
(triphenylphosphine)palladium(0) (0.009 g, 0.008 mmol), and
8 (0.24 g, 0.51 mmol) was stirred under nitrogen in anhydrous
toluene (20 mL) at reflux for 15 min. The reaction vessel was
protected from light by wrapping it in aluminum foil. A 10%
KF/H2O (20 mL) solution was added, and the mixture was
stirred for 1 h. The solution was partitioned between ethyl
acetate and water (100 mL:100 mL), and the precipitate
(SnBu3F) was filtered off. The aqueous layer was extracted
with ethyl acetate (2 × 100 mL). Organic layers were combined
and washed with water (100 mL) and brine (2 × 100 mL), dried
over magnesium sulfate, and concentrated. The residue was
chromatographed on a silica gel column (30 g), using hexane/
ethyl acetate (5:1) to afford 5i (0.012 g, 5%): Rf ) 0.05 (hexane/
2-H), 7.00 (ddd∼t, 2H, 24-H and 26-H), 7.16 (d, 1H, J1-2
)
8.31 Hz, 1-H), 7.26 (s, CDCl3), 7.45 (dd∼t, 2H, 23-H and
27-H); 13C NMR in CDCl3 δ 14.04, (C-18), 24.25 (C-15), 27.45
(C-11), 27.41 (C-7), 29.65 (C-6), 32.35 (C-12), 38.45 (C-16),
39.59 (C-8), 43.76 (C-9), 48.02 (C-13), 49.55 (C-14), 84.88
(C-17), 112.68 (C-2), 114.81 (d, JCCF ) 21 Hz, C-24 and C-26),
115.25 (C-4), 126.54 (C-22), 127.26 (C-1), 131.03 (C-21), 131.08
(d, JCCCF ) 7.7 Hz, C-23 and C-27), 132.61 (C-10), 135.86
(C-20), 138.29 (C-5), 153.34 (C-3), 161.88 (d, JC-F ) 246 Hz,
C-25); 19F NMR in acetone-d6 δ -16.42;
1
ethyl acetate 5:1); H NMR in CDCl3 δ 0.91 (s, 3H, 18-CH3),
1.2-2.8 (m, 21H, steroid nucleus), 6.01 (d, 1H, J20-21
)
12.93 Hz, 20-H), 6.52 (d, 1H, J21-20 ) 13.1 Hz, 21-H), 6.79 (d,
1H, J4-2 ) 2.4 Hz, 4-H), 6.86 (dd, 1H, J2-1 ) 8.3 Hz, J2-4
)
(17R,20Z)-21-[(4-Hydroxy)phenyl]-19-norpregna-1,3,5-
(10),20-tetraene-3,17â-diol (6c). Compound 5c (0.04 g,
0.09 mmol) was deacetylated using the hydrolytic procedure
to yield a yellow powder (0.023 g, 63%). Recrystallization in
2.46 Hz, 2-H), 7.29 (d, 1H, J1-2 ) 8.79 Hz, 1-H), 7.54 (d, 2H,
J23-24 and J27-26 ) 8.40 Hz, 23-H, 27-H), 7.90 (d, 2H, J24-23
and J26-27 ) 8.31 Hz, 24-H and 26- Hz,).

Diols for Hormone Binding Domain
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4309
acetone/hexane afforded an off-white powder (0.020 g, 55 %):
Rf ) 0.03 (hexane/ethyl acetate 4:1); mp 133-136 °C; 1H NMR
in acetone-d6 δ 0.96 (s, 3H, 18-CH3), 1.2-2.9 (m, 15H, steroid
nucleus), 5.83 (d, 1H, J20-21 ) 12.93 Hz, 20-H), 6.35 (d, 1H,
J21-20 ) 12.93 Hz, 21-H), 6.55 (s, 1H, J4-2 ) 2.34 Hz), 6.60
(dd, 1H, J2-1 )8.4 Hz and J2-4 ) 2.43 Hz, 2-H), 6.77 (d, 2H,
J23-24 ) 8.64 Hz, 23-H and 27-H), 7.11 (d, 1H, J1-2 ) 8.31 Hz,
1-H), 7.53 (d, 2H, J24-23 ) 8.61 Hz, 24-H and 26-H), 8.01 (s,
1H, 3-OH), 8.43 (s, 1H, 25-OH); 13C NMR in acetone-d6 δ 14.27
(C-18) 23.52 (C-15), 28.11 (C-11), 28.16 (C-7), ∼29 under
acetone peak (C-6), 32.39 (C-12), 38.00 (C-16), 40.58 (C-8),
44.57 (C-9), 48.62 (C-13), 50.14 (C-14), 83.36 (C-17), 113.31
(C-2), 114.95 (C-24 and C-26), 115.66 (C-4), 126.80 (C-23 and
C-27), 127.82 (C-1), 129.20 (C-21), 131.32 (C-22), 132.32
(C-10), 134.22 (C-20), 138.16 (C-5), 153.65 (C-25), 154.00
(C-3).
acetone/chloroform to yield a light-orange powder (0.31 g,
68%): Rf ) 0.47 (1:1 hexane/ethyl acetate); mp 140-142 °C;
1H NMR in CDCl3 δ 0.91 (s, 3H, 18-CH3), 1.2-2.9 (m, 15H,
steroid nucleus), 3.8 (s, b, 2H, -NH2), 4.6 (s, b, 17-OH), 5.79
(d, 1H, J20-21 ) 12.90 Hz, 20-H), 6.40 (d, 1H, J21-20 ) 12.60
Hz, 20-H), 6.56 (d, 1H, J4-2 ) 2.7 Hz, 4-H), 6.64 (m, 4H, 2-H,
4-H, 24-H and 26-H), 7.16 (d, 1H, J1-2 ) 8.4 Hz, 1-H), 7.26 (s,
CDCl3), 7.32 (d, 2H, J ) 8.4 Hz, 23-H and 27-H); 13C NMR in
acetone-d6 δ 15.04 (C-18), 24.17 (C-15), 27.79 (C-11), 28.66
(C-7), under acetone peak (C-6), 32.94 (C-12), 38.51 (C-16),
41.25 (C-8), 45.03 (C-9), 49.12(C-13), 50.12 (C-14), 83.99
(C-17), 113.90 (C-2), 114.58 (C-24 and C-26), 116.25 (C-4),
127.41 (C-1), 131.14 (C-21), 132.33 (C-10), 132.66 (C-23 and
C-27), 133.03 (C-20), 138.76 (C-5), 150.02 (C-25), 156.29
(C-3).
(17R,20Z)-21-[(4-Acetyl)phenyl]-19-norpregna-1,3,5(10),-
20-tetraene-3,17â-diol (6i). Compound 5i (0.012 g,
0.026 mmol) was hydrolyzed using the general procedure to
yield 6i (0.0056 g, 51%): Rf ) 0.12 (hexane/ethyl acetate 5:1);
1H NMR in CDCl3 δ 0.91 (s, 3H, 18-CH3), 1.2-2.8 (m, 18H,
steroid nucleus), 6.01 (d, 1H, J20-21 ) 12.99 Hz, 20-H), 6.51
(d, 1H, J21-20 ) 12.81 Hz, 21-H), 6.57 (d, 1H, J4-2 ) 2.7 Hz,
(17R,20Z)-21-[(4-Cyano)phenyl]-19-norpregna-1,3,5(10),-
20-tetraene-3,17â-diol (5d). Compound 5d (0.005g) was
hydrolyzed to yield 6d (0.004 g, 90%): Rf ) 0.22 (hexane/ethyl
1
acetate 3:1); mp 99-105 °C; H NMR in acetone-d6 δ 0.90 (s,
3H, 18-CH3), 1.2-2.8 (m, 15H, steroid nucleus), 6.15 (d, 1H,
J20-21 ) 13.20 Hz, 20-H), 6.48 (d, 1H, J21-20 ) 13.20 Hz,
21-H), 6.54 (d, 1H, J4-2 ) 2.79, 4-H), 6.60 (dd, 1H, J2-4 ) 2.8
Hz, J2-1 ) 8.32 Hz, 2-H), 7.11 (d, 1H, J1-2 ) 8.31 Hz, 1-H), (d,
2H, J23-24 and J27-26 ) 8.58 Hz, 23-H and 27-H) (d, 2H, J24-23
4-H), 6.64 (dd, 1H, J2-1 ) 8.3 Hz, 2-H), 7.16 (d, 1H, J1-2
)
8.31 Hz, 1-H), 7.55 (d, 2H, J23-24 and J27-26 ) 8.46 Hz, 23-H,
27-H), 7.90 (d, 2H, J24-23 and J26-27 ) 8.37 Hz, 24-H and 26-
H); 13C NMR in CDCl3 δ 14.04 (C-18), 23.31 (C-15), 26.43
(-COCH3), 26.62 (C-11), 27.43 (C-7), 29.63 (C-6), 32.45 (C-12),
38.69 (C-16), 39.59 (C-8), 43.75 (C-9), 48.10 (C-13), 49.68
(C-14), 85.07 (C-17), 112.71 (C-2), 115.26 (C-4), 126.55 (C-21),
127.31 (C-1), 127.91 (C-23 and C-27), 129.52 (C-24 and C-26),
132.53 (C-10), 135.50 (C-25), 137.77 (C-20), 138.27 (C-5), 142.69
(C-22), 153.37 (C-3), 197.71 (-COCH3).
and J26-27 ) 8.55 Hz, 24-H and 26-H), 7.96 (s, 1H, 3-OH); 13
C
NMR in acetone-d6 δ 14.42 (C-18), 23.81 (C-15), 27.23 (C-11),
28.16 (C-7), ∼29 under acetone peak (C-6), 32.93 (C-12), 38.91
(C-16), 40.67 (C-8), 44.46 (C-9), 48.87 (C-13), 50.17 (C-14),
84.51 (C-17), 110.23 (C-25), 113.46 (C-2), 115.80 (C-4), 119.52
(-CN), 126.91 (C-24 and C-26), 127.80 (C-1), 131.58 (C-21),
131.71 (C-23 and 27), 131.85 (C-10), 138.27 (C-5), 140.41
(C-20), 143.61 (C-22), 155.82 (C-3).
Molecular Modeling and Dynamics. We initially evalu-
ated the conformations of our ligands 6a-i using the Builder
module from Insight II. Potentials for each atom were assigned
automatically or manually, when necessary. Low-energy con-
formations were generated using the molecular mechanics
method (Discover program, 100 steps, 0.001 final convergence)
and compared to solution conformations determined by NMR.39
The ERR-HBD used in our study was obtained from the
Protein Data Bank (PDB code 1G50, wild-type ERR-HBD
cocrystallized with estradiol). From the available monomers
(A, B), monomer A from the A/C homodimer was selected for
the docking and molecular dynamics studies. All water mol-
ecules present in the crystal structure were deleted. The
monomer contains all the amino acid residues between ASN
304 and HIS 550. All manipulations were performed using the
Builder module in Insight II. The complex of the ERR-HBD
monomer and estradiol bound within the binding cavity was
minimized using the molecular mechanics method with re-
straints applied to the backbone atoms of the protein (Dis-
cover_3 module, CVFF force field, dielectric constant 2.0,
conjugate gradient minimization 10 000 steps or until 0.001
final convergence). All ligands used in this study were con-
structed using the Builder module from Insight II. Potentials
for each atom were assigned automatically or manually when
necessary. Each ligand was optimized using the molecular
mechanics method as done with the receptor. Partial charges
for each atom were calculated using the Mopac program from
the Ampac/Mopac module in the Insight II package. In
addition, ligands were further optimized using the semi-
emperical method (calculation method. PM3; calculation type,
optimization; optimizer type, native).
(17R,20Z)-21-[(4-Methyl)phenyl]-19-norpregna-1,3,5-
(10),20-tetraene-3,17â-diol (6e). Compound 5e (0.14 g,
0.32 mmol) was hydrolyzed to yield 6e (0.060 g, 48%). Recrys-
tallization in chloroform afforded a white powder (0.055 g,
44%): Rf ) 0.38 (hexane/ethyl acetate 4:1); mp 95-96 °C;
1H NMR in CDCl3 δ 0.90 (s, 3H, 18-CH3), 1.2-2.8 (m, 18H,
steroid nucleus), 5.89 (d, 1H, J20-21 ) 12.81 Hz, 20-H), 6.50
(d, 1H, J21-20 ) 12.75 Hz, 21-H), 6.56 (d, 1H, J4-2 ) 2.46 Hz,
4-H), 6.62 (dd, 1H, J2-1 ) 8.40 Hz, J2-4 ) 2.82 Hz, 2-H), 7.15
(m, 3H, 1-H, 23-H, 27-H), 7.34 (d, 2H, J24-23 and J26-27
)
8.19 Hz, 24-H and 26-H); 13C NMR in CDCl3-d6 δ 14.07
(C-18), 21.18 (C-28), 23.26 (C-15), 26.48 (C-11), 27.44 (C-7),
29.68 (C-6), 32.49 (C-12), 38.49 (C-16), 39.61 (C-8), 43.79
(C-9), 47.99 (C-13), 49.58 (C-14), 85.08 (C-17), 112.70 (C-2),
115.26 (C-4), 126.55 (C-24 and C-26), 127.91 (C-1), 128.87
(C-21), 129.01 (C-23 and C-27),132.66 (C-10), 134.31 (C-22),
135.46 (C-25), 136.95 (C-20), 138.30 (C-5), 153.38 (C-3).
(17R,20Z)-21-[(4-Methoxy)phenyl]-19-norpregna-1,3,5-
(10),20-tetraene-3,17â-diol (6g). Compound 5g (0.14 g,
0.314 mmol) was hydrolyzed using the general procedure to
yield 6g (0.125 g, 98%), and recrystallization in acetone/hexane
afforded white crystalline needles (0.120 g, 94%): Rf ) 0.21
(hexane/ethyl acetate 4:1); mp 164-165 °C; 1H NMR in
acetone-d6 δ 0.97 (s, 3H 18-CH3), 1.2-2.9 (m, b, 15H, steroid
nucleus), 3.79 (s, 3H, OCH3), 5.87 (d, 1H, J20-21 ) 12.96 Hz,
20-H), 6.38 (d, 1H, J21-20 ) 12.96 Hz, 21-H), 6.53 (d, 1H, J4-2
) 2.55 Hz, 4-H), 6.62 (dd, 1H, J2-4 ) 2.43 Hz, J2-1 ) 8.43 Hz,
2-H), 6.86 (d, 2H, J24-23 and J26-27 ) 8.91 Hz, 24-H and 26-H),
7.11 (d, 1H, J1-2 ) 8.26 Hz, 1-H), 7.62 (d, 2H, J23-24 and J27-26
) 8.64 Hz, 23-H and 27-H), 7.95 (s, 1H, 3-OH); 13C NMR in
acetone-d6 δ 14.50 (C-18), 23.71 (C-15), 27.27 (C-11), 26.16
(C-7), ∼29 under acetone peak (C-6), 32.60 (C-12), 38.26
(C-16), 40.73 (C-8), 44.52 (C-9), 48.70 (C-13), 49.87 (C-14),
55.29 (-OCH3), 83.78 (C-17), 113.46 (C-2), 113.58 (C-24 and
C-26), 115.81 (C-4), 126.95 (C-1), 129.67 (C-21), 130.52 (C-22),
131.89 (C-10), 132.37 (C-23 and C-27), 135.06 (C-20), 138.32
(C-5), 155.80 (C-3), 159.41 (C-25).
The Affinity program within the Docking module in In-
sightII was used to perform the docking studies of the ligands
with the ERR-HBD.55 This module includes elements from
Monte Carlo, simulated annealing, and minimization for
automatically docking and finding the best structures of the
ligand complexed to the receptor based on the energy of the
ligand-receptor complex. The ligand was superimposed on the
estradiol molecule (A-ring over A-ring), and the estradiol was
then deleted. The complex was subjected to energy minimiza-
tion to obtain a starting structure in which bad steric contacts
are removed and internal energies are relieved. During the
(17R,20Z)-21-[(4-Amino)phenyl]-19-norpregna-1,3,5(10),-
20-tetraene-3,17â-diol (6h). The crude acetylated product 5h
(∼0.5 g) was deprotected to yield a dark-orange solid 6h
(0.34 g, 75.3%) followed by three recrystalizations from

4310 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13
Hanson et al.
docking procedure both the ligand and the protein residues
within the ligand binding cavity (amino acids within 15 Å of
the ligand as well as all amino acids in helix-12, loops 11-12,
1-3, 6-7) were allowed to flex while the backbone atoms and
the rest of the protein were restrained in their original
positions. In addition, the phenylvinyl side chain of the ligand
was rotated by a maximum of 180° increments in order to more
fully explore the potential binding modes of the conformational
choices of the ligand. After each docking procedure, structures
within 10 kcal/mol of the lowest energy structure and with an
rms distance of greater than 0.125 Å were selected and used
in simulated annealing studies. At the beginning of each run,
the ligand-receptor complex was minimized over 5000 steps
or until 0.001 final convergence. Then each structure was
heated from 300 to 500 K over 5000 fs and allowed to
equilibrate for and additional 5000 fs. Each structure was
allowed to cool to 300 K in 20 stages with 10 K decrements
for each stage and 100 fs long equilibration periods for each
stage. The structure at the end of the final stage was recorded
in an archive file and further minimized with 200 steps. Each
of the dynamics and simulated annealing cycles was repeated
10 times. During these calculations additional restraints were
applied to amino acids facing the outer surface of the protein.
All calculations involving docking and refinement of generated
structures were performed with a dielectric constant of 2.0.
Results of the docking studies were analyzed using a
combination of modules: Analysis, Discover_3, Docking, and
Viewer. Each structure generated during the docking, simu-
lated annealing, and dynamics runs was analyzed in terms of
binding energy, ligand energy, and protein energy. Values of
the binding energy ∆Ebinding were calculated as the difference
between the potential energy of the complex (Ecomplex) and the
potential energy of the ligand (Eligand) and receptor (Ereceptor).52,56
Binding energy calculations were performed using the Energy
Analysis macro in the Discover_3 module.
Receptor Binding Studies. In Vitro Competitive Bind-
ing Assay. The compounds were screened for their affinity
for the ERR-HBD isolated from BL 21 cells that overexpressed
the 33 kDa PER-23d ERG vector. The cells were induced with
0.6 mM isopropyl-â-thiogalactopyranoside for 3 h at room
temperature, pelleted by centrifugation, frozen, and stored at
-75 °C. The cells were thawed and lysed by sonication (4 ×
20 s) in four volumes of lysis buffer (50 mM Tris, 50 mM NaCl,
1 mM EDTA, 1 mM dithiothreitol, 1 M urea, pH 7.4) several
times. Clarified fractions, obtained at 30000g for 30 min, were
pooled, assayed for receptor binding, and diluted to 50 nM in
ER, and 100 µL aliquots were frozen and stored at -75 °C
until ready for use. Then 80 µL of the ERR-HBD-containing
extract was incubated with 10 µL of 10 nM 6,7-[H-3]estradiol
(specific activity of 51 Ci/mmol) and 10 µL of buffer, unlabeled
estradiol, or test ligand in 100 µL total volume. The final
concentrations were 1 nM 6,7-[H-3]estradiol, 2 nM unlabeled
estradiol (using 200 nM estradiol to define specific binding),
and 0.5-5000 nM of the test ligand. In all cases, 10 µL of each
incubation solution was removed for assay of the actual initial
concentration of [H-3]estradiol and the remainder was incu-
bated at 4 or 25 °C for 18 h. After incubation, 100 µL of
dextran-coated charcoal suspension (fines removed) was added
to adsorb the unbound [H-3]estradiol, incubated for 10 min,
and centrifuged, and 100 µL samples were taken from the
supernatant fraction for assay of radioactivity. The results
were calculated and plotted as % specific binding as a function
of log of competitor concentration using the best fit equation
for the binding inhibition to define the 50% inhibition level.
The relative binding affinity (RBA) was calculated as 100 times
[E]/[C], where [E] was the concentration of unlabeled estradiol
needed to reduce the specific binding of [H-3]estradiol by 50%
and [C] was the concentration of test ligand needed to reduce
the specific binding by 50%.
of Health (Grants PHS 1RO1 CA81049 and CA 89469
to R.N.H.), the U.S. Army Breast Cancer Research
Program (Grants DAMD 17-99-1-9333 and 17-00-1-
00384 to R.N.H.), and the Boothroyd Foundation (to
E.R.D.). Molecular modeling was performed on instru-
ments supported in part by a grant from the National
Science Foundation (Grant CHE-9974642).
Supporting Information Available: Results from el-
emental analysis of the target compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Evans, R. M. The steroid and thyroid hormone receptor super-
family. Science 1988, 240, 889-895.
(2) Tsai, M.-J.; O’Malley, B. W. Molecular mechanisms of action of
steroid/thyroid receptor superfamily members. Annu. Rev. Bio-
chem. 1994, 63, 451-486.
(3) Nilsson, S.; Gustafsson, J.-A. Biological role of estrogen and
estrogen receptors. Crit. Rev. Biochem. Mol. Biol. 2002, 37, 1-28.
(4) Hart, L. L.; Davie, J. R. The estrogen receptor: more than the
average transcription factor. Biochem. Cell Biol. 2002, 80, 335-
341.
(5) Pike, A. C. W.; Brzozowski, A. M.; Hubbard, R. E. A structural
biologist’s view of the oestrogen receptor. J. Steroid Biochem.
Mol. Biol. 2000, 74, 261-268.
(6) Katzenellenbogen, B. S.; Choi, I.; Delage-Mourroux, R.; Ediger,
T. R.; Martini, P. G. V.; Montano, M.; Sun, J.; Weis, K.;
Katzenellenbogen, J. A. Molecular mechanisms of estrogen
action: selectve ligands and receptor pharmacology. J. Steroid
Biochem. Mol. Biol. 2000, 74, 279-285.
(7) McDonnell, D. P.; Dana, S. L.; Hoener, P. A.; Lieberman, B. A.;
Imhof, M. O.; Stein, R. B. Cellular mechanisms which distin-
guish between hormone- and antihormone-activated estrogen
receptor. Ann. N.Y. Acad. Sci. 2001, 949, 121-137.
(8) Dutertre, M.; Smith, C. L. Molecular mechanisms of selective
estrogen receptor modulator (SERM) action. J. Pharmacol. Exp.
Ther. 2000, 295, 431-437.
(9) Krishnan, V.; Heath, H.; Bryant, H. U. Mechanism of action of
estrogens and selective estrogen receptor modulators. Vitam.
Horm. 2001, 60, 123-147.
(10) Katzenellenbogen, B. S.; Sun, J.; Harrington, W. R.; Kraichely,
D. M.; Ganessunker, D.; Katzenellenbogen, J. A. Structure-
function relationships in estrogen receptors and the character-
ization of novel selective estrogen receptor modulators with
unique pharmacological profiles. Ann. N.Y. Acad. Sci. 2001, 949,
6-15.
(11) Leclercq, G. Molecular forms of the estrogen receptor in breast
cancer. J. Steroid Biochem. Mol. Biol. 2002, 80, 259-272.
(12) Miner, J. N.; Tyree, C. M. Drug discovery and the intracellular
receptor family. Vitam. Horm. 2001, 62, 253-280.
(13) Karamouzis, M. V.; Gorgoulis, V. G.; Papavassiliou, A. G.
Transcription factors and neoplasia: vistas in novel drug design.
Clin. Cancer Res. 2002, 8, 949-961.
(14) Johnston, S. R. D.; Riddle, S.; Haynes, B. P.; Hern, R. A.; Smith,
I. E.; Jarman, M.; Dowsett, M. The novel anti-estrogen iodoxifene
inhibits the growth of human cancer xenografts and reduces the
frequency of acquired anti-estrogen resistance. Br. J. Cancer
1997, 75, 804-809.
(15) Jones, C. D.; Jevnikar, M. G.; Pike, A. J.; Peters, M. K.; Black,
L. J.; Falcone, J. F.; Clemens, J. A. Anti-estrogens. 2. Structure-
activity studies in a series of 3-aroyl-2-arylbenzo[b]thiophene
derivatives leading to [6-hydroxy-2-(4-hydroxyphenyl)benzo[b]-
thien-2-yl][4-[2-(1-piperidinyl)ethoxy]phenyl]methanone hydro-
chloride LY 156758, a remarkably effective estrogen antagonist
with only minimal intrinsic activity. J. Med. Chem. 1984, 27,
1057-1066.
(16) Palkowitz, A. D.; Glasebrook, A. L.; Thrasher, K. J.; Hauser, K.
L.; Short, L. L.; Phillips, D. L.; Muehl, B. S.; Sato, M.; Shetler,
P. K.; Cullinan, G. L.; Pell, T. R.; Bryant, H. U. Discovery and
synthesis of [6-hydroxy-3-[4-[2-(1-piperidinyl)-ethoxy]phenoxy]-
2-(4-hydroxyphenyl)]benzo[b]thiophene: a novel, highly potent,
selective estrogen receptor modulator. J. Med. Chem. 1997, 40,
1407-1416.
(17) LaBrie, F.; LaBrie, C.; Belanger, A.; Simard, J.; Gauthier, S.;
Luu-The, V.; Merand, Y.; Giguere, V.; Candas, B.; Luo, S.;
Martel, C.; Singh, S. M.; Fournier, M.; Coquet, A.; Richard, V.;
Charbonneau, R.; Charpanet, G.; Tremblay, A.; Cusan, L.;
Villeux, R. EM-652 (SCH 57068), a third generation SERM
acting as a pure anti-estrogen in the mammary gland and
endometrium. J. Steroid Biochem. Mol. Biol. 1999, 69, 51-84.
(18) Greenberger, L. M.; Annablr, T.; Collins, K. I.; Komm, B. S.;
Lyttle, C. R.; Miller, C. P.; Satyaswaroop, P. G.; Zhang, Y.; Frost,
P. A new anti-estrogen, 2-(4-hydroxy-phenyl)-3-methyl-1-[4-(2-
Acknowledgment. We gratefully acknowledge Dr.
Roger Krautz for his assistance with the 500 MHz NMR
spectrometry. We are grateful for the support of this
research through grants from the National Institutes

Diols for Hormone Binding Domain
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 13 4311
piperidin-2-yl-ethoxy-)benzyl]-1H-indol-5-ol hydrochloride (ERA-
923), inhibits the growth of tamoxifen-sensitive and -resistant
tumors and is devoid of uterotrophic effects in mice and rats.
Clin. Cancer Res. 2001, 7, 3166-3177.
substituted-phenyl)-19-norpregna-1,3,5(10),20-tetraene-3,17â-di-
ols as probes for the estrogen receptor-R (ERR) hormone binding
domain. J. Med. Chem. 2003, 46, 2865-2876.
(37) Lebl, T.; Holecek, J.; Dymal, M.; Steinborn, D. Synthesis and
characterization and reactivity of 2-functionalized vinylstan-
nanes. J. Organomet. Chem. 2001, 625, 86-94.
(38) Lipshutz, B. H.; Keil, R.; Barton, J. C. Preparation of
Z-vinylstannanes via hydrozirconation of stannylacetylenes.
Tetrahedron Lett. 1992, 33, 5861-5864.
(19) Meegan, M. J.; Hughes, R. B.; Lloyd, D. G.; Williams, D. C.;
Zisterer, D. M. Ethyl side-chain modifications in novel flexible
anti-estrogenssdesign, synthesis and biological efficacy in assay
against the MCF-7 breast tumor cell line. Anti-Cancer Drug Des.
2001, 16, 57-69.
(39) Sielstad, D. A.; Carlson, K. E.; Katzenellenbogen, J. A.; Kushner,
P. J.; Greene, G. L. Molecular characterization by mass spec-
trometry of the human estrogen receptor ligand-binding domain
expressed in Escherichia coli. Mol. Endocrinol. 1993, 9, 647-
658.
(40) DeSombre, E. R.; Mease, R. C.; Sanghavi, J.; Singh, T.; Seevers,
R. H.; Hughes, A. Estrogen receptor binding affinity and
uterotrophic activity of triphenylhaloethylenes. J. Steroid Bio-
chem. 1988, 29, 583-590.
(41) Raynaud, J. P.; Ojasoo, T.; Bouton, M. M.; Philibert, D. Receptor
binding as a tool in the development of new bioactive steroids.
In Drug Design; Ariens, E. J., Ed.; Academic Press: New York,
1979; Vol. VIII, pp 169-214.
(42) Eiler, S.; Gangloff, M.; Duclaud, S.; Moras, D.; Ruff, M. Over-
expression, purificiation and crystal structure of native ER alpha
LBD. Protein Expression Purif. 2001, 20, 165-
(43) Littke, A. F.; Fu, G. C. The first general method for Stille cross-
couplings of aryl chlorides. Angew. Chem., Int. Ed. 1999, 38,
2411-2416.
(44) Littke, A. F.; Fu, G. C. Palladium-catalyzed coupling reactions
of aryl chlorides. Angew. Chem., Int. Ed. 2002, 41, 4176-4211.
(45) Bapat, A. R.; Frail, D. E. Full-length estrogen receptor R and
its ligand-binding domain adopt different conformations upon
ligand binding. J. Steroid Biochem. Mol. Biol. 2003, 86, 143-
149.
(20) Weatherman, R. V.; Carroll, D. C.; Scanlan, T. S. Activity of a
tamoxifen-raloxifene hybrid ligand for estrogen receptors at an
AP-1 site. Bioorg. Med. Chem. Lett. 2001, 11, 3129-3131.
(21) Rosati, R. L.; Da Silva, J.; Cameron, K. O.; Thompson, D. D.;
Ke, H. Z.; Toler, S. M.; Brown, T. A.; Pan, L. C.; Ebbinghaus, C.
F.; Reinhold, A. R.; Elliott, N. C.; Newhouse, B. N.; Tjoa, C. M.;
Sweetnam, P. M.; Cole, M. J.; Arriola, M. W.; Gauthier, J. W.;
Crawford, D. T.; Nickerson, D. F.; Pirie, C. M.; Qi, H.; Simmons,
H. A.; Tkalcevic, G. T. Discovery and preclinical pharmacology
of a novel, potent, nonsteroidal estrogen receptor agonist/
antagonist, CP 336156, a diaryltetrahydronaphthalene. J. Med.
Chem. 1998, 41, 2928-2931.
(22) Bury, P. S.; Christiansen, L. B.; Jacobsen, P.; Jorgensen, A. S.;
Kanstrup, A.; Naerum, L.; Bain, S.; Fledelius, C.; Gissel, B.;
Hansen, B. S.; Korsgaard, N.; Thorpe, S. M.; Wasserman, K.
Synthesis and pharmacological evaluation of novel cis-3,4-diaryl-
hydroxychromanes as high affinity partial agonists for the
estrogen receptor. Bioorg. Med. Chem. 2002, 10, 125-145.
(23) Levenson, A. S.; MacGregor-Schafer, J. I.; Bentrm, D. J.; Pease,
K. M.; Jordan, V. C. Control of the estrogen-like action of
tamoxifen-estrogen receptor complex by the surface amino acid
at position 351. J. Steroid Biochem. Mol. Biol. 2001, 76, 61-70.
(24) Jordan, V. C.; MacGregor-Schafer, J.; Levenson, A. S.; Liu, H.;
Pease, K. M.; Simons, L. A.; Zapf, J. W. Molecular classification
of estrogens. Cancer Res. 2001, 61, 6619-6623.
(46) Wurtz, J.-M.; Egner, U.; Heinrich, N.; Moras, D.; Mueller-
Fahrnow, A. Three-dimensional models of estrogen receptor
ligand binding domain complexes, based on related crystal
structures and mutational and structure-activity relationships.
J. Med. Chem. 1998, 41, 1803-1814.
(47) Sippl, W. Receptor-based 3-D QSAR analysis of estrogen receptor
ligandssmerging the accuracy of receptor-based alignments with
the computational efficiency of ligand-based methods. J. Com-
put.-Aided Mol. Des. 2000, 14, 559-572.
(48) Egner, U.; Heinrich, N.; Ruff, M.; Gangloff, M.; Mueller-Fahrnow,
A.; Wurtz, J.-M. Different ligands-different receptor conforma-
tions: modeling of the hERR-LBD in complexes with agonists
and antagonists. Med. Res. Rev. 2001, 21, 523-539.
(49) Wolohan, P.; Reichert, D. E. CoMFA and docking study of novel
estrogen receptor subtype selective ligands. J. Comput.-Aided
Mol. Des. 2003, 17, 313-328.
(25) Napolitano, E.: Fiaschi, R.; Hanson, R. N. Structure-activity
relationships of estrogenic ligands. Synthesis and evaluation of
(17R,20E/Z)-21-halogenated-19-norpregna-1,3,5(10),20-tetraene-
3,17â-diols. J. Med. Chem. 1991, 34, 2754-2759.
(26) Napolitano, E.; Fiaschi, R.; Herman, L. W.; Hanson, R. N.
Synthesis and estrogen receptor binding of (17R,20E)- and (17R,-
20Z)-21-phenylthio- and phenylseleno-19-norpregna-1,3,5(10),-
20-tetraene-3,17â-diols. Steroids 1996, 61, 384-387.
(27) Hanson, R. N.; Herman, L. W.; Fiaschi, R.; Napolitano, E.
Stereochemical probes for the estrogen receptor: Synthesis and
receptor binding of (17R,20E/Z)-21-phenyl-19-norpregna-1,3,5-
(10),20-tetraene-3,17â-diols. Steroids 1996, 61, 718-722.
(28) Hanson, R. N.; Napolitano, E.; Fiaschi, R. Novel high affinity
steroidal estrogenic ligands: Synthesis and receptor binding of
11â-vinyl-17R-E/Z-phenylselenovinyl estradiols. Steroids 1998,
63, 479-483.
(50) Schipira, M.; Abagyan, R.; Totrov, M. Nuclear hormone receptor
targeted virtual screening. J. Med. Chem. 2003, 46, 3045-3059.
(51) van Lipzig, M. M. H.; ter Laak, A. M.; Jongejan, A.; Vermeulen,
N. P. E.; Wamelink, M.; Geerke, D.; Meerman, J. H. N.
Prediction of ligand binding affinity and orientation of xeno-
estrogens to the estrogen receptor by molecular dynamics
simulations and the linear interaction energy method. J. Med.
Chem., in press.
(52) DeLisle, R. K.; Yu, S.-J.; Nair A. C.; Welsh, W. J. Homology
modeling of the estrogen receptor subtype â (ERâ) and calcula-
tion of ligand binding affinities. J. Mol. Graphics Modell. 2001,
20, 155-167.
(29) Hanson, R. N.; Napolitano, E.; Fiaschi, R. Synthesis and
evaluation of 11â-substituted-21-chloro/iodo-(17R,20E/Z)-19-nor-
pregna-1,3,5(10),20-tetraene-3,17â-diols: High affinity ligands
for the estrogen receptor. J. Med. Chem. 1998, 41, 4686-4692.
(30) McKenna, N. J.; O’Malley, B. W. From ligand to response:
generating diversity in nuclear coregulator function. J. Steroid
Biochem. Mol. Biol. 2000, 74, 351-356.
(31) Connor, C. E.; Norris, J. D.; Broadwater, G.; Willson, T. M.;
Gottardis, M. M.; Dewhirst, M. W.; McDonnell, D. P. Circum-
venting tamoxifen resistance in breast cancers using antiestro-
gens that induce unique conformational changes in the estrogen
receptor. Cancer Res. 2001, 61, 2917-2922.
(53) Sebag, A. B.; Friel, C. J.; Hanson, R. N.; Forsyth, D. A.
Conformational studies on (17R,20Z)-21-(X-phenyl)-19-norpre-
gna-1,3,5(10),20-tetraene-3,17â-diols using 1D and 2D NMR
spectroscopy and GIAO calculations of 13C shieldings. J. Org.
Chem. 2000, 65, 7902-7912.
(54) Katzenellenbogen, B. S.; Sun, J.; Harrington, W. R.; Kraichely,
D. M.; Ganessunker, D.; Katzenellenbogen, J. A. Structure-
function relationships in the estrogen receptors and the char-
acterization of novel selective estrogen receptor modulators with
unique pharmacological profiles. Ann. N. Y. Acad. Sci., submit-
ted.
(55) Insight II, 97.2 and Discover 2.9.7; MSI Inc.: San Diego, CA.
(56) Nair, A. C.; Miertus, S.; Tossi, A.; Romeo, D. A computational
study of the resistance of HIV-1 aspartic protease to inhibitors
ABT-538 and VX-478 and design of new analogues. Biochem.
Biophys. Res. Commun. 1998, 242, 545-551.
(32) Sathya, G.; Jansen, M. S.; Nagel, S. C.; Cook, C. E.; McDonnell,
D. P. Identification and characterization of novel estrogen
receptor-â-sparing antiprogestins. Endocrinology 2002, 143,
3071-3082.
(33) Liu, H.; Lee, E.-S.; De Los Reyes, A.; Zapf, J. W.; Jordan, V. C.
Silencing and reactivation of the selective estrogen receptor
modulator-estrogen receptor R complex. Cancer Res. 2001, 61,
3632-3639.
(34) Levenson, A. S.; Svoboda, K. M.; Pease, K. M.; Kaiser, S. A.;
Chen, B.; Simons, L. A.; Jovanovic, B. D.; Dyck, P. A.; Jordan,
V. C. Gene expression profiles with activation of the estrogen
receptor R-selective estrogen receptor modulator complex in
breast cancer cells expressing wild-type estrogen receptor.
Cancer Res. 2002, 62, 4419-4426.
(35) Hanson, R. N.; Lee, C. Y.; Friel, C. J.; DeSombre, E. R.; Hughes,
A. Evaluation of 17R-(trifluoromethylphenyl)vinyl estradiols as
novel estrogen receptor ligands. Steroids 2003, 68, 143-148.
(36) Hanson, R. N.; Lee, C. Y.; Friel, C. J.; Dilis, R.; DeSombre, E.
R.; Hughes, A. Synthesis and evaluation of (17R,20E)-21-(4-
JM040157S