Synthesis and evaluation of estrogen agonism of diaryl 4,5- dihydroisoxazoles, 3-hydroxyketones, 3-methoxyketones, and 1,3-diketones: A compound set forming a 4D molecular library

Article abstract of DOI:10.1021/jm8001795

In this paper, the preparation and systematic evaluation of estrogen receptor α (ERα) and estrogen receptor β (ERβ) activities of some diaryl-1,3-diones and their synthetic intermediates, diaryl-4,5-dihydroisoxazoles, diaryl-3-hydroxyketones, diaryl-3-met

Full text of DOI:10.1021/jm8001795

3562
J. Med. Chem. 2008, 51, 3562–3571
Synthesis and Evaluation of Estrogen Agonism of Diaryl 4,5-Dihydroisoxazoles,
3-Hydroxyketones, 3-Methoxyketones, and 1,3-Diketones: A Compound Set Forming a 4D
Molecular Library
Juha T. Pulkkinen,*,† Paavo Honkakoski,§ Mikael Pera¨kyla¨,‡ Istvan Berczi,| and Reino Laatikainen†
Laboratories of Chemistry, Biochemistry, Department of Biosciences, UniVersity of Kuopio, P.O. Box 1627, FIN-70211 Kuopio, Finland,
Department of Pharmaceutics, UniVersity of Kuopio, P.O. Box 1627, FIN-70211 Kuopio, Finland, Department of Immunology, Faculty of
Medicine, UniVersity of Manitoba, Winnipeg, Manitoba R3E 0W3, Canada
ReceiVed February 20, 2008
In this paper, the preparation and systematic evaluation of estrogen receptor R (ERR) and estrogen receptor
ꢀ (ERꢀ) activities of some diaryl-1,3-diones and their synthetic intermediates, diaryl-4,5-dihydroisoxazoles,
diaryl-3-hydroxyketones, diaryl-3-methoxyketones, and diaryl-2-(dimethyl-λ4-sulfanylidene)-1,3-diones, is
described. The set of 72 compounds constitutes a general schematic structure aryl1-linker1-
spacer-linker2-aryl2, where the linker1-spacer-linker2 length varies between 4 and 8 carbons. The set
of compounds was applied here to map and explore the active sites of subtypes ERR and ERꢀ. The highest
activities were obtained with dihydroisoxazole and hydroxyketone spacers, but even the most flexible diones
with unsubstituted aryl groups showed some agonism. Most compounds were found to be ERR selective or
to activate both receptors, but in some cases we saw also clearly stronger ERꢀ activation.
Introduction
challenging scientific problem. The estrogen receptor binds a
very diverse group of ligands3–41 and thus forms an optimal
target of study the conformational and configurational library
principle, which was the original objective of this work.
Molecular libraries form an essential block of modern drug
development strategies: a lead compound binding to any protein
can be found, in principle, by using a large number of synthetic
molecules. However, combinatorial synthetic methodology still
allows only very simple reaction conditions, which essentially
limits the compound types that can be synthesized. A way to
avoid working with large molecular libraries would be to use
flexible ligands, which cover large areas of conformational
space. If an active ligand is structurally and conformationally
very adaptive, forming its fourth structural dimension, it has
numerous conformations at low populations. The selection of a
suitable conformer, however, has its entropic price tag on the
binding free energy. It has been estimated that the price of fixing
one rotatable bond can be up to 5 kJ/mol,1 which is well
tolerable if the ligand-receptor complementarity is high enough.
Another concern is the complementarity of ligand molecular
motion with the receptor frame: we have recently proposed that
the intramolecular motional properties expand an independent
dimension of the ligand QSARa space2 and, correspondingly,
offer a mechanism for the protein activation. The question of
the motional dimension, together with methodological difficul-
ties, makes the systematic studies of molecular flexibility a
Estrogens have numerous regulating functions in biological
systems, including the reproductive tissues, the cardiovascular
system, the central nervous system, and bone maintenance. In
general, the steroidal regulation forms an important field of drug
research. The estrogen activity is mediated mainly by two
members of the nuclear receptor gene family, the estrogen
receptor isoforms ERR42 and ERꢀ.43 Both forms bind and are
activated by their endogenous ligand 3,17ꢀ-estradiol (E2), and
none of the ER agonists or antagonists currently in use are
specific for either form.44 Because of the serious adverse effects
of ER agonists and antagonists, large variations in ERR and ꢀ
expression in diverse target tissues, and cell- and promoter-
specific functions displayed by the ER subtypes, there are
increasing efforts to explore new chemical scaffolds to develop
subtype-specific ligands and to understand context-specific ER
signaling.45 Indeed, a 500-fold functional selectivity for ERR
has been reported,15 while greater than 200-fold selectivities of
binding affinity for ERꢀ9,17,21 has also been recorded.
Many compounds also exhibit significant estrogen or anti-
estrogen activity, which also has important environmental
biological implications.46 A common structural feature needed
for estrogen and antiestrogen activity seems to be an aromatic
ring or, preferably, a phenol group in the position corresponding
to the A-ring of an estrogen. The most well-known examples
of selective estrogen receptor modulators (SERMs) are tamox-
ifen and raloxifene,3,4 which have rather little in common with
the estrogen structure. Although the structural diversity of the
previously reported synthetic nonsteroidal SERMs is seemingly
remarkable, they can be roughly divided into two main groups.
The first group is formed by polycyclic structures, with some
resemblance to the basic steroidal structure.5–9 The rest of the
compounds have a general structure of ARYL1-CORE
STRUCTURE-ARYL2,10 in which the core may have a third
aromatic ring and an additional substituent, like in tamoxifen.
* To whom correspondence should be addressed. Phone: +358-17-
163247. Fax: +358-17-163259. E-mail: juha.pulkkinen@uku.fi. Address:
Laboratory of Chemistry, Department of Biosciences, University of Kuopio,
P.O. Box 1627, FIN-70211 Kuopio, Finland.
†
Laboratory of Chemistry, Department of Biosciences, University of
Kuopio.
‡
Laboratory of Biochemistry, Department of Biosciences, University
of Kuopio.
§
Department of Pharmaceutics, University of Kuopio.
Department of Immunology, Faculty of Medicine, University of
|
Manitoba.
a Abbreviations: ERR, estrogen receptor R; ERꢀ, estrogen receptor ꢀ;
QSAR, quantitative structure-activity relationship; E2, 3,17ꢀ-estradiol;
SERM, selective estrogen receptor modulator; DMS, dimethyl sulfide; NCS,
N-chlorosuccinimide; HEK293 cells, human embryonic kidney cells; EC50,
half-maximal effective concentration; SEM, standard error of mean; DMSO,
dimethyl sulfoxide; TFAA, trifluoroacetic anhydride; THF, tetrahydrofuran;
DCM, dichloromethane; RT, room temperature.
10.1021/jm8001795 CCC: $40.75
2008 American Chemical Society
Published on Web 06/03/2008

Estrogen Agonism of Diaryl-1,3-diones
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12 3563
Scheme 1a
activation of both ERR and ERꢀ receptors. The fit of the agonists
to the ER binding pocket was explored by molecular modeling.
Results
Syntheses. The general route leading to the compounds is
depicted in Scheme 1 and described previously in detail.48 4,5-
Dihydroisoxazoles C were synthesized by treating a nitrile oxide,
generated in situ with NaOCl from an aldoxime A, with 2-fold
excess of an aryl alkene B in the presence of a catalytic amount
of pyridine as an organic base. The isoxales C were then
converted to hydroxy ketones D by catalytic hydrogenation with
Raney Ni in the presence of H2O and AcOH. The methoxy
ketones E were synthesized by refluxing compounds D in
MeOH overnight in the presence of concentrated hydrochloric
acid. The sulfonium ylides F were prepared from D by using
5-fold excess of the Corey-Kim reagent (S,S-dimethyl-succin-
imidosulfonium chloride, prepared in situ from DMS and NCS)
under argon at -70 °C. Finally, the diketones G were achieved
from F by a reductive desulphurization with zinc in the presence
of acetic acid at room temperature. Unfortunately, we were
unable to synthesize compounds 2, 3, 34, and 35 via this
methodology. However, we were able to make a series of
compounds with two phenyl rings and a varying set of spacers
with different functionalities, flexibilities, and chain lengths.
Compounds 69-76 having aromatic hydroxyl groups were
prepared from corresponding methoxy substituted 4,5-dihy-
droisoxazoles by using an 1.1-fold excess of BBr3 in CH2Cl2
at room temperature overnight (see Scheme 2). All the chiral
compounds were further purified using semipreparative HPLC
techniques with a chiral column and with three exceptions (11,
17, and 19) we were able to separate all the enantiomers of
compounds C (see Table 1). Neither specific rotation nor
absolute configuration of the pure enantiomers was resolved due
to the small amounts of the purified samples. For each separated
compound, the enantiomer having shorter retention time in the
chiral separation is marked in the tables as the enantiomer a
and the other form as b.
Performance and Specificity of the ER Activation
Assay. We performed ER activation assays in HEK293 cells, a
cell line derived from human embryonic kidney. This cell line
has been used to explore estrogenic ligands in ER cotransfection
assays because HEK293 cells do not express endogenous ERs.49
We typically obtained more than 90-fold activation of both of
the ER subtypes by 10 nM E2, indicating a high dynamic range
of this transient transfection assay and its ability to distinguish
even weakly activating compounds. To investigate whether the
observed activation of ER occurs through receptor binding, we
randomly selected a group of test compounds (5, 18b, 64b, and
65). The activation by all compounds was efficiently blocked
by the established ER antagonist ICI-182,780 (Table 5),
indicating that these compounds activated both ERR and ERꢀ
via ligand binding. Further evidence on the ER dependency of
activation is the fact that none of the test compounds, E2, or
ICI-182,780 affected reporter gene activity in the absence of ER
expression vector (data not shown). The estimated EC50 values
for ERR and ERꢀ were close to 2-10 µM in dose-response
assays for the test compounds, while the corresponding values
for E2 were about 0.2 nM or less (data not shown).
a m ) 0-3, n ) 0-2. Reagents and conditions: (i) NaOCl, pyridine,
DCM, 0°C; (ii) H2, Raney-Ni, H2O, AcOH, MeOH-THF, RT; (iii) HCl,
MeOH, refl.; (iv) DMS-NCS, TEA, DCM, -70°C; (v) Zn, AcOH, DCM,
RT.
In the simplest examples, which are biphenyl compounds, the
core is formed by only one carbon-carbon bond.11,12 However,
the most SERMS have a core structure that is nearly planar
and composed of a sp2-hybridized system such as a C-C double
bond,13,14 an amide group,15 or a cyclic core, which makes the
space between the aromatic rings sterically more crowded and
conformationally rigid. The variety of cyclic scaffolds published
so far include five-membered rings like furane,16,17 thiofene,18–20
isoxazole,17,21 cyclopentadiene,22 pyrazole,10 pyrrole,23 cyclo-
pentadienone,24 and pyrrolidinedione,25 as well as six-membered
rings such as oxazine,26 pyridine,27–29 dihydrooxathiine,30,31
isochromane,32 pyrane,33,34 dihydropyrane,35 and dihydrothi-
opyrane.35 The most elaborate cyclic core structures reported
so far are carborane,36,37 bicyclo[2.2.2]octane,36 and
7-oxabicyclo[2.2.1]heptene38 and -heptaediene38 scaffolds. Only
in a few cases5,6,8,26,30–35,38,39 the core structure has sp3-
hybridized atoms that also act as chiral centers. The sp2-
structures are symmetric, and the sp2-sp2 junctions have
intrinsically high rotational barriers due to which many of the
structures can be classified rigid.
The diarylpentanedione structure has six rotatable sp2-sp3
bonds which all exhibit an exceptionally low rotational barrier
so that the structure expands a broad conformational space,
further expanded by the keto-enol tautomerism of the struc-
tures.47 The synthetic pathway,48 offering less flexible, but chiral,
relatives and derivatives of the dione structure, can also be
considered as an expansion of the molecular space. We
synthesized and assayed for ER activation a series of compounds
having a general schematic structure ARYL1-LINKER1-
SPACER-LINKER2-ARYL2, in which the spacers have chiral
sp3-centers and lengths of the CH2-linkers vary in such a way
that the distance between the aromatic rings is 4-8 carbons
(see Scheme 1.). To the best of our knowledge there are only
two previous examples of ER binding compounds in which the
core structure between the aromatic rings is constituted of a
C-C double bond40 or a cyclic triazine spacer41 and flexible
CH2-linkers.
Discussion
The present report describes the preparation and the in vitro
estrogen activity of 72 mainly unsubstituted diaryl-1,3-diones
and their synthetic 4,5-dihydroisoxazole, 1,3-hydroxy ketone,
1,3-methoxy ketone, and dimethylsulfonium ylide intermediates,
some of which in the biological evaluation showed significant
General. The biological evaluation of the compounds with
unsubstituted phenyl groups (see Table 1) show that, within the
group of the diketones G, there are two compounds (5 and 6)
having ERR activity to some extent. The results indicated that

3564 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12
Pulkkinen et al.
Scheme 2
Table 1. In Vitro Estrogen Activity of the Unsubstituted Ligands 1-49
a Receptor activation (a mean of at least three independent transfections, SEM typically <15%) relative to 10 nM E2 corresponding 100, sample concentration
10 µM. b Receptor activation (a mean of at least three independent transfections, SEM typically <15%) relative to 10 nM E2 corresponding 100, sample
concentration 10 µM. c a ) enantiomer with a shorter retention time in the chiral separation. d b ) enantiomer with a longer retention time in the chiral
separation. e Enantiomers not separated, racemic mixture tested.
the optimum length of the spacer for the active ligands would
be 6-7 carbons (m + n ) 3-4). It is also notable that, although
low, the selectivity of compound 1 for ERꢀ is roughly 2-fold.
On the other hand, none of ylides F with the bulky dimethyl
sulfonium group and restricted conformation of the spacer shows
activity. In comparison with other compound groups, the 4,5-
dihydroisoxazoles C show the best overall activities. Many of
them are clearly selective for ERR (13a, 15a, 15b, 18b) and
the difference between the activities of the pure enantiomers is
obvious: the other enantiomer is practically inactive in three
cases (9a, 10b, and 18a). Moreover, both 1 and 11, i.e.,
compounds with the spacer length of four carbons (m + n ) 1)
show ca. 2-fold selectivity for the ERꢀ receptor.
4,5-Dihydroisoxazoles. Ligand docking showed that 4,5-
dihydroisoxazoles C bind in similar conformations to ERR and
ERꢀ in comparison with the other compound groups (G, D,
E). Therefore, the higher overall activities of 4,5-dihydroisox-
azoles C may be partly due to their higher hydrophobicities
and conformational rigidity. Compound 11, which has consider-
able ERꢀ activity, docked to the ligand-binding site of ERꢀ is
shown in Figure 1a. In this conformation the length of 11 is
11.5 Å, close to 10.9 Å of E2, and it makes extensive
interactions with Met336, Leu339, and Leu476 of ERꢀ.
Compounds 15 (Figure 1b) and 19 (Figure 1c), which have the
highest ERR activities among the 4,5-dihydroisoxazoles C, bind
to ERR in a partly folded conformation. The other phenyl group
of 15 and 19 is bound between Leu339 and Phe356, and the
other end of the compounds is fitted in a cavity lined by Leu384,
Met421, His524, and Leu525. This binding mode, which is
better tolerated by ERR than ERꢀ, is common to compounds
having m +n > 2.
Hydroxy Ketones. The activities of hydroxy ketones D are
generally weaker than the activities of the corresponding
isoxazoles. However, both ERR and ERꢀ activities of 23 are
three times higher than the activities of 12 and compound 24 is
the most efficient agonist for both receptors. Interestingly,
enantiomers 24a and b have weak but opposite selectivity for
ERR and ERꢀ. The reason for the high activities of 23 and 24
is that they form a hydrogen bond between their hydroxyl group
and the histidine side chain (His524 of ERR and His475 of ERꢀ)
of the binding pocket (Figure 1d). In some of the docked
conformations 26-29 formed weak hydrogen bonds, i.e., long
hydrogen bonds of unfavorable geometry, with the histidine,
but only 29 shows increased activity compared to the corre-
sponding diketones. Methylation of compound D enhances to
some extent the ERꢀ activity in some cases (32, 38, and 40).
Altogether, the activity results clearly show that the spacer length
of four carbons (m + n ) 1) is a preferable structural feature
for ERꢀ selective compounds, whereas the optimal length of
the spacer for ERR activity is 6-7 carbons (m + n ) 3-4).
The wide range of the active isoxazole compounds suggests
that variable spacer length is tolerated better by the ERR

Estrogen Agonism of Diaryl-1,3-diones
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12 3565
Figure 1. Docked poses of compounds (a) 11 and (d) 24 to ERꢀ and compounds (b) 15, (c) 19, (e) 70 and (f) 72 to ERR.
receptor. This difference reflects the slightly larger volume of
Table 3) and found out that compounds 64 and 65 expressed
the ligand-binding cavity of ERR (450 Å3) than ERꢀ (390 Å3).50
certain ERR activity. This may be due to their spatial similarity
with compounds 38 and 40: the molecules occupy about the
same space in the receptor but the former compounds are
conformationally more restricted. On the other hand, the latter
compounds show also some ERꢀ activity, which is not the case
with the naphthyl ligands.
Aromatic Ring Substitution. To tentatively investigate how
ring substitution affects the ER activity, we made a series of
substituted analogues of compounds 15, 26, and 37 having m
) 2 and n ) 1 (See Table 2). In the present examples, para
substitution of the benzyl group modestly increases the ERꢀ
activity of all the compounds. As to the ERR activity, the
agonistic function of 15 is completely destroyed by para
methoxy substitution, whereas 58 shows higher value in
comparison to the unsubstituted analogue 37. On the other hand,
para fluoro substitution of the benzyl group increases ERR
activation markedly in all three cases, 51 being one of the best
ERR ligands. Replacement of the phenyl ring by 1- or
2-naphthyl group increases the ERꢀ activity of most compounds
to some extent, whereas the change of the ERR activity is
dependent on the functionality of the spacer: 52 and 53 are less
active, 56 and 57 equally active, and 60 and 61 more active
than their phenyl analogues. In addition, we tested some
naphthyl substituted analogues of an inactive compound 36 (see
As the phenol moiety is a common structural feature of
numerous ER ligands, we decided to synthesize and evaluate
some hydroxyl substituted analogues of ligand 15 (see Table
4), which is the best ERR agonist of the unsubstituted 4,5-
dihydroisoxazoles. As seen in Table 4, p-fluoro substitution of
the benzylic ring (51) increased both ERR and ERꢀ activities
of the compound to some extent. Docking simulations showed
that 51 may bind to ER in a similar fashion as 15 (Figure 1b)
and, therefore, the slightly increased activity of 51 is likely due
to enhanced interactions of the p-F-Ph group compared to Ph.
Ligand 69, having a hydroxyl group at the same position,
showed even better activities, ERꢀ activity being 4-fold in
comparison to 51. For both 51 and 69, the activity difference

3566 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12
Pulkkinen et al.
Table 2. In Vitro Estrogen Activity of the Substituted Derivatives 50-61 of Ligands 15, 26, and 37
a Receptor activation (a mean of at least three independent transfections, SEM typically <15%) relative to 10 nM E2 corresponding 100, sample concentration
10 µM. b Receptor activation (a mean of at least three independent transfections, SEM typically <15%) relative to 10 nM E2 corresponding 100, sample
concentration 10 µM. c a ) enantiomer with a shorter retention time in the chiral separation. d b ) enantiomer with a longer retention time in the chiral
separation. e Enantiomers not separated, racemic mixture tested.
Table 3. In Vitro Estrogen Activity of the Naphthyl Derivatives 62-68
of Ligand 36
Table 4. In Vitro Estrogen Activity of the Hydroxyl Substituted
Derivatives 69-76 of ligands 15 and 51
a Receptor activation (a mean of at least three independent transfections,
SEM typically <15%) relative to 10 nM E2 corresponding 100, sample
concentration 10 µM. b Receptor activation (a mean of at least three
independent transfections, SEM typically <15%) relative to 10 nM E2
corresponding 100, sample concentration 10 µM. c a ) enantiomer with a
shorter retention time in the chiral separation. d b ) enantiomer with a
longer retention time in the chiral separation. e Enantiomers not separated,
racemic mixture tested.
between their two enantiomers was significant, being 4-fold for
ERꢀ activity between 69a and 69b. We also prepared com-
pounds 70-72 with hydroxyl group at the o-, m-, or p-position
of the phenethyl group. In comparison to 15a, ERR activity of
71a having an m-hydroxyl was decreased significantly, whereas
ERR activities of both 70a and 72a were markedly increased.
Indeed, compounds 69, 70, and 72 have the highest activities
of the hydroxyl substituted analogues. Ligand docking showed
that the 2-OH of 70 forms a hydrogen bond of favorable
geometry with the histidine (His524, Figure 1e). In contrast,
the 4-OH of 69 and 72 forms hydrogen bonds with Glu353 and
Arg394 of ERR (Figure 1f) mimicking the E2 phenolic A-ring,
which is known to be essential for high affinity ER ligands.17,51,52
a Receptor activation (a mean of at least three independent transfections,
SEM typically <15%) relative to 10 nM E2 corresponding 100, sample
concentration 10 µM. b Receptor activation (a mean of at least three
independent transfections, SEM typically <15%) relative to 10 nM E2
corresponding 100, sample concentration 10 µM. c a ) enantiomer with a
shorter retention time in the chiral separation. d b ) enantiomer with a
longer retention time in the chiral separation.
The activity of 72, which is the highest of the present
compounds, is clearly higher than that of 69 because the p-OH
substituted phenethyl group of 72 fits better to the ligand binding
site (Figure 1f) than the phenethyl group of 69. Docking also
explains why we could not increase the activity by combining
the substituent effects manifested by compounds 73-76.

Estrogen Agonism of Diaryl-1,3-diones
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12 3567
Table 5. Activation of ERR and ERꢀ by Randomly Selected Compounds and Antagonism by 1000 nM ICI-182,780
compound (10 µM)
ER subtype
-fold activation in the absence of ICI-182,780
-fold activation in the presence of ICI-182,780
vehicle
5
65
18b
64b
E2 (1 nM)
vehicle
65
R
R
R
R
R
R
ꢀ
ꢀ
ꢀ
1.0 ( 0.1
8.7 ( 1.1
22.9 ( 3.8
125.5 ( 8.6
26.1 ( 3.4
163.6 ( 15.3
1.0 ( 0.1
1.0 ( 0.1
1.0 ( 0.02
1.6 ( 0.3
1.0 ( 0.2
1.1 ( 0.2
1.2 ( 0.2
0.7 ( 0.2
0.7 ( 0.2
0.9 ( 0.2
4.2 ( 0.4
98.7 ( 6.2
E2 (1 nM)
Ligands 75 and 76 do not fit to the binding site so that hydrogen
bonds of optimal geometry could be formed at the same time
with Glu353, Arg394, and His524, and in the case of 73, the
4-F is close to Glu358, forming a repulsive interaction. The
observation that in the case of 74 the 4-F substituent decreases
activity (72 vs 74), whereas in the case of 51 it increases it (15
vs 51) is likely due to repulsive interactions between 4-F of 74
and Trp383. Namely, the tip of the phenethyl group of 72
(Figure 1f) and 74 is close to the Trp383 (not shown in the
Figure 1). In the case of 51, such repulsive interaction does not
exist because 51 binds to ER in a slightly different orientation
compared to 72 and 74 due to nonexistence of hydrogen bonds
with Glu353 and Arg394.
helix 12 closes the ligand-binding pocket creating a binding
surface for coactivator protein59), there exists increasing amount
of evidence that the simple model of helix 12 acting as a two-
state molecular switch needs to be extended.60 Ligands have
been found to affect differently the conformational dynamics
of ERR61 and modulate ERR and ERꢀ activities by mechanisms
involving subtle differences in receptor-ligand interactions.62,63
The fact that there is not necessarily direct correlation between
the ligand affinity and transcriptional activity highlights the
complexity of receptor activation. Recent studies of peroxisome
proliferator-activated receptor γ have also nicely demonstrated
how partial agonists activate the receptor using a helix 12
independent mechanism by inducing unique changes to the
dynamics of the ligand-binding domain.64,65
Flexibility. A unique feature of these structures, and the
original reason for their synthesis, is that the dibenzyl-1,3-dione
unit contains six low rotational barriers, as demonstrated by an
NMR study of 1,5-diphenylpentane-2,4-dione (4).47 The rota-
tional barrier of CH2 adjacent to aryl is always low and therefore
also the benzyl-4,5-dihydroisoxazole sp3-sp3 junction is rather
freely rotating. Previously, these intramolecular motions of
ligands have been almost completely ignored in drug design
although the importance of intramolecular motions of proteins
and other macromolecules is well approved for their biological
activity.53–55 The present study forms a part of our work on
molecular flexibility and motions2,47,56–58 in which we have
anticipated that also intramolecular motions of ligand can play
a role in molecular regulation and recognition.2 In principle, if
the ligand enhances the motions of protein in the direction of
reaction coordinate, it can act as an agonist, and in the other
case, it can act as antagonist. Also ligand binding affinity may
depend on the motions: if the ligand binding reduces the motions
of the protein, the vibrational entropy of the protein is decreased,
which leads to lower binding affinity. Alternatively, a ligand
may tune the motions of a protein, although at the cost of
binding affinity, so that its motions correspond better the motions
of another protein (such as a coactivator for the ER) and thus
enhance protein-protein recognition. In this case, a poor binding
affinity may mean an agonistic function.
The fact that some activity is observed also with the very
flexible dione structure proposes that neither the motional
complementarity is necessary (although certainly desirable) for
activation: filling the binding cleft is enough. In this way, still
allowing all the motions of the receptor protein, which it has in
the presence of the natural ligands, the flexible ligand cuts away
the motions typical for the empty receptor molecule or even
enhances them to the directions arising from the protein
backbone. One can also say that the binding of the cleft filling
ligand leads to more correlated motions of the receptor, which
can be seen as a condition for activation of the receptor.2
The above considerations are in fair agreement with the
present conception of nuclear receptor activation. Although the
activation can be largely explained by ligand-induced stabiliza-
tion of the active receptor conformation (where the C-terminal
The compounds with a few exceptions were found ERR
selective. One explanation is that the ligand-binding pocket of
ERꢀ is somewhat smaller than that of ERR. However, because
the highest ERꢀ selectivities have been obtained with structures
which are more rigid than the present ones at the site that is
bound to the estrogen binding site, it is tempting to think that
the selectivity reflects also the tolerances of the receptors to
the ligand flexibility.
Conclusions
The compounds having an overall structure of
aryl1-linker1-spacer-linker2-aryl2 studied in this work form
a family of novel estrogen analogues that operate through the
ER receptors. The observation that ERR/ERꢀ active compounds
were found from all the spacer classes indicates that these
compounds excite a wide structural space, forming a 4D
molecular library, supporting our original idea that the com-
pound set can be used to monitor receptor ligand binding
cavities. In this work, the highest activities were obtained with
dihydroisoxazole and hydroxyketone spacers, but even the most
flexible diones without any aryl substituents showed some ER
agonism. The compounds having the shortest spacer lengths (m
+ n ) 1) were found to be ERꢀ selective, whereas almost
without exception the rest of the compounds (m + n ) 2-5)
showed ERR selectivity. This difference in isoform selectivity
can be explained by somewhat smaller ligand-binding cavity
of ERꢀ. The hydroxyl ketone 24 (m + n ) 3), the other
enantiomer of which showed relatively high ERꢀ activity was
an interesting exception to this selectivity rule. The relatively
high concentrations (µM) needed for a significant ER activity
obviously reflect the fact that the binding free energy to the
receptor is not favorable due to entropic reasons and, for most
of the compounds, due to the lack of anchoring phenolic
hydroxyl groups. Rigidifying the dione moiety improved the
overall activity and further improvement was obtained by
introducing an aromatic hydroxyl substituent on the most active
4,5-dihydroisoxazole 15. Computer modeling showed that
increased activities are due to a hydrogen bond formed between

3568 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12
Pulkkinen et al.
H), 7.14-7.11 (m, 3 H), 4.25 (m, 1 H), 3.02 (br s, 1 H, OH), 2.80
(dd, 1 H, J ) 13.6, 7.1), 2.69 (dd, 1 H, J ) 13.6, 7.1), 2.58 (t, 2
H, J ) 7.6), 2.49 (dd, 1 H, J ) 15.7, 2.5), 2.47 (dd, 1 H, J ) 15.7,
6.1), 2.37 (dd, 2 H, J ) 7.9, 6.7), 1.86 (m, 2 H). 13C NMR δ 211.3,
141.4, 137.9 (3 s), 129.4, 128.5, 128.4, 128.4, 126.5, 126.0, 68.7
(7 d), 48.2, 42.9, 42.7, 34.9, 24.8 (5 t). Anal. (C19H22O2 ·0.1H2O)
C, H.
4-Hydroxy-1,6-diphenylhexan-2-one (28). Yield 91%, a yellow
wax. 1H NMR δ 7.35-7.28 (m, 2 H), 7.26-7.21 (m, 3 H),
7.17-7.12 (m, 5 H), 4.00 (m, 1 H), 3.65 (s, 2 H), 3.01 (br s, 1 H,
OH), 2.75 (m, 1 H), 2.65-2.52 (m, 3 H), 1.75 (m, 1 H), 1.63 (m,
1 H). 13C NMR δ 209.3, 141.8, 133.6 (3 s), 129.4, 128.8, 128.4,
128.4, 127.2, 125.8, 66.9 (7 d), 50.7, 48.4, 38.0, 31.7 (4 t). Anal.
(C18H20O2) C, H.
the ligand and the receptor, and that the phenol group of the
most active compound 72 mimics the interactions formed by
the E2 A-ring.
Experimental Section
Materials and Equipment. Aldoximes A were synthesized from
the corresponding aldehydes and hydroxylammonium chloride.66
Benzaldehydes and phenyl acetaldehyde were commercial products
(Aldrich) and distilled prior to use. All the other aldehydes were
prepared from the corresponding alcohols by treatment with DMSO
in the presence of TFAA.67 Allyl arenes B were either commercial
products (Aldrich) or synthesized by Grignard reaction of corre-
sponding aryl magnesium bromides with allyl bromide. All moisture
sensitive reactions were performed under a positive pressure of
argon in oven-dried glassware equipped with rubber septa; solvents
and liquid reagents were dried, distilled, stored under argon, and
transferred using a syringe flushed with argon. Column chroma-
tography was performed with silica gel 60 (SiO2, 70-230 mesh).
Melting points were measured with a Stuart Scientific SMP2
5-Hydroxy-6-(4-fluorophenyl)-1-phenyl-hexan-3-one (55).
1
Yield 66%, a yellow viscous oil. H NMR δ 7.25 (m, 2 H), 7.17
(m, 1 H), 7.15-7.10 (m, 4 H), 6.96 (m, 2 H), 4.23 (m, 1 H), 2.96
(s, 1 H, OH), 2.87 (t, 2 H, J ) 7.5), 2.77-2.70 (m, 3 H), 2.65 (dd,
1 H, J ) 13.8, 5.9), 2.50 (m, 2 H). 13C NMR δ 210.5, 161.7 (d),
140.7 (s), 133.6 (d), 130.8 (dd), 128.5, 128.2, 126.2 (3 d), 115.2
(dd), 68.5 (d), 48.4, 45.0, 42.0, 29.4 (4 t). Anal.
(C18H19FO2 ·0.2H2O) C, H.
1
apparatus and are uncorrected. H and 13C NMR spectra (TMS/
CDCl3) were recorded on a Bruker Avance 500 spectrometer
operating at 500.13 and 125.77 MHz, respectively. All J coupling
constants are given in Hz. Elemental analyses (CHNS) were carried
out with a Thermo Quest CE Instruments EA 1110 CHNS-O
elemental analyzer. Semipreparative HPLC separations of the
enantiomers were performed on a Shimadzu chromatography system
using a Regis Technologies (R,R)-Whelk-O 1 (25 cm × 10 mm
i.d.) chiral column in n-hexane/i-PrOH /AcOH 80/20/0.5 at flow
rate 5 mL/min. Compounds were detected by UV absorption at
254 nm.
Synthesis of 4,5-Dihydroisoxazoles C (Compounds 9-19
and 50-53). To a vigorously stirred solution of oxime A (20
mmol), alkene B (20 mmol) and pyridine (4 mmol) in 100 mL of
DCM at 0 °C was added dropwise, 5% NaOCl solution (60 mL,
40 mmol), keeping the temperature under 5 °C and the mixture
was reacted for 1.5 h. The organic layer was separated and washed
with 2 M HCl, 2 M NaHCO3, and water, and evaporated to give
the crude product, which was purified by column chromatography
using DCM as an eluent.
Synthesis of Methoxyketones E (Compounds 31-33, 36-41
and 58-68). To a stirred solution of hydroxyketone D (1 mmol)
in 25 mL of MeOH was added dropwise conc HCl (0.2 mL) at 0
°C and the mixture was refluxed overnight. Then the solution was
taken up with DCM, washed with saturated NaHCO3 and water,
and evaporated. The crude product was purified by preparative thin
layer chromatography using DCM as an eluent.
2-Methoxy-1,7-diphenyl-heptan-4-one (38). Yield 36%, a red-
dish-brown oil. 1H NMR δ 7.30-7.23 (m, 4 H), 7.21-7.11 (m, 6
H), 3.91 (m, 1 H), 3.29 (s, 3 H), 2.86 (dd, 1 H, J ) 13.7, 5.8), 2.70
(dd, 1 H, J ) 13.7, 6.6), 2.61-2.50 (m, 3 H), 2.42-2.29 (m, 3 H),
1.85 (m, 2 H). 13C NMR δ 209.1, 141.6, 138.0 (3 s), 129.5, 128.4,
128.4, 128.3, 126.3, 125.9, 78.3 (7 d), 57.3 (q), 46.9, 43.0, 39.8,
35.0, 24.9 (5 t). Anal. (C20H24O2) C, H.
5-Methoxy-6-naphthalen-2-yl-1-phenyl-hexan-3-one (61).
Yield 95%, a pale-brown viscous oil. 1H NMR δ 7.80 (d, 1 H, J )
7.3), 7.77 (d, 1 H, J ) 7.1), 7.76 (d, 1 H, J ) 8.3), 7.61 (s, 1 H),
7.44 (m, 2 H), 7.32 (dd, 1 H, J ) 8.3, 1.7), 7.23 (m, 2 H), 7.14 (m,
1 H), 7.11 (m, 2 H), 4.01 (m, 1 H), 3.32 (s, 3 H), 3.03 (dd, 1 H, J
) 13.7, 5.9), 2.88-2.79 (m, 3 H), 2.74-2.59 (m, 3 H), 2.42 (dd,
1 H, J ) 16.1, 4.6). 13C NMR δ 208.4, 141.0, 135.6, 133.5, 132.2
(5 s), 128.4, 128.3, 128.0, 128.0, 127.9, 127.6, 127.5, 126.0, 126.0,
125.5, 78.3 (11 d), 57.5 (q), 47.1, 45.4, 40.1, 29.4 (4 t). Anal.
(C23H24O2) C, H.
5-Benzyl-3-phenethyl-4,5-dihydroisoxazole (15). Yield 51%,
1
a colorless oil. H NMR δ 7.27-7.24 (m, 4 H), 7.21-7.15 (m, 6
H), 4.73 (m, 1 H), 2.95 (dd, 1 H, J ) 13.8, 6.0), 2.84-2.76 (m, 3
H), 2.73 (dd, 1 H, J ) 13.8, 6.9), 2.63-2.52 (m, 3 H). 13C NMR
δ 158.1, 140.5, 137.0 (3 s), 129.4, 128.5, 128.5, 128.2, 126.6, 126.3,
80.5 (7 d), 41.6, 40.9, 32.6, 29.4 (4 t). Anal. (C18H19NO·0.1H2O)
C, H, N.
4-Methoxy-5-naphthalen-2-yl-1-phenylpentan-2-one (65).
1
Yield 28%, a pale-yellow viscous oil. H NMR δ 7.81-7.73 (m,
3,5-Diphenethyl-4,5-dihydroisoxazole (18). Yield 41%, mp )
1
3 H), 7.58 (br s, 1 H), 7.47-7.40 (m, 2 H), 7.32-7.19 (m, 4 H),
7.10 (m, 2 H), 4.00 (m, 1 H), 3.64, 3.62 (2d, 2 H, J ) 15.7), 3.30
(s, 3 H), 3.00 (dd, 1 H, J ) 13.7, 6.0), 2.86 (dd, 1 H, J ) 13.7,
6.3), 2.69 (dd, 1 H, J ) 16.4, 7.5), 2.49 (dd, J ) 16.4, 4.9). 13C
NMR δ 206.8, 135.6, 133.9, 133.5, 132.2 (5 s), 129.5, 128.5, 127.9,
127.9, 127.9, 127.6, 127.5, 127.0, 126.0, 125.4, 78.3 (11 d), 57.5
(q), 50.9, 46.2, 40.0 (3 t). Anal. (C22H22O2) C, H.
56.2-58.0 °C. H NMR δ 7.30-7.25 (m, 4 H), 7.22-7.16 (m, 6
H), 4.49 (m, 1 H), 2.93-2.84 (m, 3 H), 2.74 (m, 1 H), 2.70-2.63
(m, 3 H), 2.47 (dd, 1 H, J ) 16.8, 7.9), 1.95 (m, 1 H), 1.76 (m, 1
H). 13C NMR δ 158.1, 141.3, 140.6 (3 s), 128.5, 128.4, 128.4,
128.3, 126.3, 126.0, 79.2 (7 d), 42.4, 36.9, 32.7, 31.8, 29.6 (5 t).
Anal. (C19H21NO) C, H, N.
5-(4-Fluorobenzyl)-3-phenethyl-4,5-dihydroisoxazole (51).
Yield 30%, mp ) 51.5-52.9 °C. 1H NMR δ 7.28 (m, 2 H),
7.22-7.16 (m, 3 H), 7.14 (m, 2 H), 6.96 (m, 2 H), 4.72 (m, 1 H),
2.89 (dd, 1 H, J ) 14.0, 6.3), 2.86-2.78 (m, 3 H), 2.74 (dd, 1 H,
J ) 14.0, 6.2), 2.65-2.51 (m, 3 H). 13C NMR δ 161.8 (d), 158.1,
140.5 (2 s), 132.8 (d), 130.9 (dd), 128.6, 128.2, 126.3 (3 d), 115.3
(dd), 80.4 (d), 41.6, 40.0, 32.6, 29.4 (4 t). Anal. (C18H18FNO) C,
H, N.
Synthesis of (Dimethyl-λ4-sulfanylidene)-diones F (Com-
pounds 42-49). To a stirred mixture of NCS (20 mmol) in 60 mL
of anhydrous DCM was added dropwise DMS (40 mmol) at -70
°C under argon, and stirring was continued for 1 h. Then a solution
of hydroxyketone D (4.0 mmol) in 6 mL of DCM was added,
maintaining the same temperature, and the solution was stirred for
1 h. TEA (60 mmol) was then added, and after 1 h, the reaction
mixture was treated with saturated NaCl (20 mL). This mixture
was extracted with ether, the organic layer washed with saturated
NaCl, and evaporated. The residue was purified by column
chromatography using 95:5 chloroform-acetone as an eluent.
3-(Dimethyl-λ4-sulfanylidene)-1,7-diphenylheptane-2,4-di-
one (47). Yield 92%, mp ) 92.4-94.8 °C. 1H NMR δ 7.28-7.19
(m, 6 H), 7.19-7.10 (m, 4 H), 3.98 (br s, 2 H), 2.80 (br s, 2 H),
2.64 (s, 3 H), 2.64 (br s, 2 H), 1.92 (tt, 2 H, J ) 7.7, 7.6). 13C
NMR δ 194.0, 190.3, 142.3, 137.1 (4 s), 129.2, 128.4, 128.3, 128.1,
Synthesis of Hydroxyketones D (Compounds 20-30 and
54-57). A solution of isoxazoline C (10 mmol), AcOH (0.1 mol)
and water (1.0 mol) in 100 mL of MeOH-THF (1:1) was stirred
under H2 at RT overnight in the presence of Raney Ni (1.5 g). The
catalyst was removed by filtering through celite, and the filtrate
extracted by DCM. The extracts were washed with saturated
NaHCO3 and water and evaporated to give the pure product.
2-Hydroxy-1,7-diphenyl-heptan-4-one (27). Yield 64%, a
1
colorless oil. H NMR δ 7.31-7.21 (m, 5 H), 7.20-7.14 (m, 2

Estrogen Agonism of Diaryl-1,3-diones
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12 3569
126.2, 125.6 (6 d), 85.9 (s), 48.4, 41.1, 35.5, 27.1 (4 t), 26.6 (q).
Anal. (C21H24O2S·0.1H2O) C, H. S: calcd, 9.37; found, 8.50.
4-(Dimethyl-λ4-sulfanylidene)-1,7-diphenylheptane-3,5-di-
one (48). Yield 60%, mp ) 68.5-71.0 °C. 1H NMR δ 7.26-7.21
(m, 8 H), 7.17-7.13 (m, 2 H), 3.03 (br s, 4 H), 2.93 (m, 4 H), 2.68
(s, 6 H). 13C NMR δ 192.8 (br s), 142.1 (s), 128.7, 128.3, 127.8 (3
d), 87.2 (s), 42.9 (br t), 32.0 (t), 27.0 (q). Anal. (C21H24O2S·0.1H2O)
C, H, S.
5-(4-Fluoro-benzyl)-3-(2-hydroxy-phenethyl)-4,5-dihydro-
isoxazole (73). Yield 99%, mp ) 124.8-126.4 °C. 1H NMR δ
7.12-7.09 (m, 3 H), 7.07 (d, 1 H, J ) 7.8), 6.92 (m, 2 H),
6.71-6.67 (m, 2 H), 6.66 (d, 1 H, J ) 7.7), 4.71 (m, 1 H),
2.87-2.81 (m, 2 H), 2.76-2.68 (m, 3 H), 2.61-2.50 (m, 3 H).
13C NMR δ 161.7 (d), 159.3, 156.4, 141.8 (3 s), 132.6 (d), 130.9
(dd), 129.7, 120.1, 115.4 (dd), 115.2, 113.6, 80.5 (3 d), 41.6, 39.9,
32.3, 28.9 (4 t). Anal. (C18H18FNO2 ·0.5H2O) C, H, N.
Synthesis of diones G (compounds 1 and 4-8). To a stirred
mixture of F (1.5 mmol) and Zn dust (30 mmol) in 25 mL DCM
was added AcOH (30 mmol) at 0 °C, and stirring was continued at
RT overnight. The mixture was filtered, and the filtrate washed
with saturated NaHCO3 and water, eluted with DCM through a ca.
2 cm column of Florisil (100-200 mesh), and evaporated.
1,7-Diphenylheptane-2,4-dione (6). Yield 87%, dark-brown
5-(4-Fluoro-benzyl)-3-(4-hydroxy-phenethyl)-4,5-dihydro-
1
isoxazole (74). Yield 62%, a dark-brown viscous oil. H NMR δ
7.11 (m, 2 H), 6.95-6.92 (m, 4 H), 6.75 (m, 2 H), 4.72 (m, 1 H),
2.90-2.83 (m, 2 H), 2.74-2.66 (m, 3 H), 2.59-2.52 (m, 3 H).
13C NMR δ 161.7 (d), 159.3, 154.8 (2 s), 132.5 (d), 131.5 (s),
130.9 (dd), 129.2, 115.6 (2 d), 115.3 (dd), 80.5 (d), 41.5, 39.8,
31.6, 29.3 (4 t). Anal. (C18H18FNO2 ·H2O) C, N. H: calcd, 6.35;
found, 5.78.
1
viscous oil. H NMR δ 15.42 (br s, 1 H, OH), 7.33-7.29 (m, 2
H), 7.27-7.14 (m, 6 H), 7.13-7.10 (m, 2 H), 5.39 (s, 1 H), 3.56
(s, 2 H), 2.59 (t, 2 H, J ) 7.7), 2.24, (t, 2 H, J ) 7.7), 1.89 (m, 2
H). 13C NMR δ 193.9, 192.4, 141.4, 135.2 (2 s), 129.3, 128.7,
128.4, 128.4, 127.0, 126.0, (6 d), 99.4 (s), 45.2, 37.5, 35.1, 24.7 (4
t). Anal. (C19H20O2 ·0.5H2O) C, H.
5-(4-Hydroxy-benzyl)-3-(2-hydroxy-phenethyl)-4,5-dihydro-
isoxazole (75). Yield 98%, a dark-brown viscous oil. 1H NMR (in
CD3OD) δ 7.04-7.00 (m, 4 H), 6.74 (d, 1 H, J ) 7.6), 6.73-6.69
(m, 3 H), 4.69 (m, 1 H), 2.95 (dd, 1 H, J ) 17.2, 10.1), 2.82-2.78
(m, 3 H), 2.72-2.56 (m, 4 H). 13C NMR (in CD3OD) δ 161.4,
156.9, 156.2 (3 s), 131.3, 131.0 (2 d), 129.1 (s), 128.4 (d), 127.8
(d), 120.5 (s), 116.1, 115.9, 82.3 (3 d), 42.1, 40.8, 28.6, 28.3 (4 t).
Anal. (C18H19NO3 ·H2O) C, H, N.
5-(4-Hydroxy-benzyl)-3-(4-hydroxy-phenethyl)-4,5-dihydro-
isoxazole (76). Yield 99%, a dark-brown viscous oil. 1H NMR (in
CD3OD) δ 7.01 (m, 2 H), 6.99 (m, 2 H), 6.71 (m, 2 H), 6.71 (m,
2 H), 4.69 (m, 1 H), 2.89 (dd, 1 H, J ) 17.2, 10.2), 2.76 (dd, 1 H,
J ) 14.0, 6.1), 2.72-2.62 (m, 4 H), 2.55 (m, 2 H). 13C NMR (in
CD3OD) δ 160.8, 157.0, 156.7, 132.5 (4 s), 131.4, 130.2 (2 d),
129.0 (s), 116.2, 116.1, 82.2 (3 d), 42.1, 40.8, 32.7, 30.4 (4 t). Anal.
(C18H19NO3 ·H2O) C, H, N.
Cell Culture, Transfection, and Reporter Assays. E2 was
bought from Sigma Chemical Co. (St. Louis, MO), and ICI-182,780
was from Tocris (Avonmouth, UK). All other reagents were of
reagent grade from Sigma or Fluka. One day before transfection,
HEK293 cells were seeded in 48-well plates (70 × 103 cells per
well) in phenol-free Dulbecco’s modified Eagle medium supple-
mented with 5% delipidated fetal bovine serum (Sigma) and
antibiotics.68 After a medium change, the cells were transfected
for 4 h with 5 ng ERR or ERꢀ expression vector,69 75 ng reporter
plasmid pERE2TATA-LUC,70 and 20 ng control plasmid pCMVꢀ
by the calcium phosphate method.68 After transfection, the cells
received fresh medium containing either vehicle (0.1% v/v) or test
compound (10 µM). In antagonism studies, 1000 nM ICI-182,780
or vehicle was included in medium. After 24 h, the cells were
washed, lysed, and assayed for luciferase and ꢀ-galactosidase
activities68 with a Victor2 reader (Perkin-Elmer Wallac, Turku,
Finland). After normalization for ꢀ-galactosidase activity, luciferase
activities are expressed relative to that of 10 nM E2 by the formula:
activity ) 100% × [(test compound) - (vehicle)/(E2) - (vehicle)],
where terms in parentheses indicate the corresponding normalized
luciferase activities. Typically, more than 90-fold activation by 10
nM E2 of luciferase with both ER subtypes was seen. The data are
means ( SEM of at least three independent transfections.
Molecular Modeling. To explore the fit of our structures into
the estrogen receptors, all the compounds of Tables 1 and 3 were
docked to the ligand-binding site of the human estrogen receptor
R and ꢀ using the Gold ligand docking program.71 Ligands, which
were built using the Sybyl 7.0 program72 and minimized with the
Tripos force field and default settings, were docked to the ligand-
binding site of ERR and ERꢀ with the GOLD program (version
2.1).71 The X-ray structure of ERR complexed with genistein (PDB
code 1×7R)52 and ERꢀ complexed with ERB-041 (PDB code
1×7B)17 were used for docking calculations. For each ligand,
docking runs were performed with a maximum number of 10 poses
using default parameters. Goldscore was used as the scoring
function. The docked structure having the highest score of the 10
poses was selected to represent the protein-ligand complex.
Because the enantiomers of the compounds were not resolved
experimentally, they were ranked on the basis of the scoring values.
1,8-Diphenyloctane-3,5-dione (8). Yield 81%, a pale-brown
1
viscous oil. H NMR δ 15.46 (br s, 1 H, OH), 7.29-7.24 (m, 4
H), 7.20-7.13 (m, 6 H), 5.42 (s, 1 H), 2.92 (t, 2 H, J ) 7.6), 2.62
(t, 2 H, J ) 7.7), 2.59 (t, 2 H, J ) 7.7), 2.27 (t, 2 H, J ) 7.4), 1.92
(m, 2 H). 13C NMR δ 193.8, 193.3, 141.5, 140.7 (4 s), 128.5, 128.5,
128.4, 128.3, 126.2, 126.0, 99.5 (7 d), 40.1, 37.5, 35.2, 31.5, 27.2
(5 t). Anal. (C20H22O2 ·0.2H2O) C, H.
Synthesis of Phenolic 4,5-Dihydroisoxazoles 69-76. To a
solution of a 4,5-dihydroisoxazole having an aromatic methoxyl
group (2 mmol) in 2 mL of DCM was added 2.2 mL of 1 M BBr3
in DCM (2.2 mmol, for compounds 75 and 76 having two aromatic
methoxyl groups, 4.4 mmol was used), and the solution was stirred
under argon at RT overnight. The organic layer was evaporated to
give the oily product, which was purified by column chromatog-
raphy using DCM as an eluent.
5-(4-Hydroxy-benzyl)-3-phenethyl-4,5-dihydro-isoxazole (69).
Yield 51%, mp ) 124.8-126.4 °C. 1H NMR δ 7.25 (m, 2 H),
7.17 (m, 1 H), 7.13 (m, 2 H), 6.99 (m, 2 H), 6.78 (m, 2 H), 4.71
(m, 1 H), 2.85-2.77 (m, 4 H), 2.68 (dd, 1 H, J ) 14.0, 6.4),
2.60-2.53 (m, 3 H). 13C NMR δ 159.0, 155.1, 140.2 (3 s), 130.5,
128.5, 128.2 (3 d), 128.2 (s), 126.3, 115.6, 80.9 (3 d), 41.5, 39.8,
32.5, 29.3 (4 t). Anal. (C18H19NO2 ·0.75H2O) C, H, N.
5-Benzyl-3-[2-(2-hydroxy-phenyl)-ethyl]-4,5-dihydro-isox-
1
azole (70). Yield 96%, brown viscous oil. H NMR δ 7.28 (m, 2
H), 7.23 (m, 1 H), 7.18 (m, 2 H), 7.11 (m, 1 H), 6.71-6.67 (m, 3
H), 5.57 (br s, 1 H, OH), 4.78 (m, 1 H), 2.96 (dd, 1 H, J ) 13.9,
6.0), 2.85 (dd, 1 H, J ) 17.1, 10.1), 2.78-2.72 (m, 3 H), 2.65-2.55
(m, 3 H). 13C NMR δ 159.4, 156.4, 141.8, 136.7 (4 s), 129.7, 129.3,
128.5, 126.7, 120.1, 115.4, 113.7, 80.7 (8 d), 41.6, 40.7, 32.3, 29.0
(4 t). Anal. (C18H19NO2) H. C: calcd, 76.84; found, 76.00. N: calcd,
4.98; found, 4.31.
5-Benzyl-3-[2-(3-hydroxy-phenyl)-ethyl]-4,5-dihydro-isox-
1
azole (71). Yield 99%, mp ) 120.6-122.8 °C. H NMR δ 7.81
(br s, 1 H, OH), 7.28 (m, 2 H), 7.22 (m, 1 H), 7.18 (m, 2 H), 7.07
(m, 1 H), 6.92 (d, 1 H, J ) 7.9), 6.84 (t, 1 H, J ) 6.9), 4.79 (m,
1 H), 2.96 (dd, 1 H, J ) 13.9, 6.3), 2.91-2.85 (m, 3 H), 2.78 (dd,
1 H, J ) 13.9, 6.5), 2.70-2.62 (m, 3 H). 13C NMR δ 160.2, 154.5,
136.7 (3 s), 130.2, 129.3, 128.5, 127.8 (4 d), 127.2 (s), 126.7, 120.4,
116.9, 81.0 (4 d), 41.8, 40.6, 28.8, 26.6 (4 t). Anal.
(C18H19NO2 ·H2O) N. C: calcd, 72.22; found, 71.68. H: calcd, 7.07;
found, 6.57.
5-Benzyl-3-[2-(4-hydroxy-phenyl)-ethyl]-4,5-dihydro-isox-
1
azole (72). Yield 89%, mp ) 143.0-144.8 °C. H NMR δ 7.29
(m, 2 H), 7.22 (m, 1 H), 7.18 (m, 2H) 6.98 (m, 2 H), 6.79 (m, 2
H), 6.52 (br s, 1 H, OH), 4.78 (m, 1 H), 2.95 (dd, 1 H, J ) 13.9,
6.1), 2.88 (dd, 1 H, J ) 17.1, 10.2), 2.76 (dd, 1 H, J ) 13.9, 6.7),
2.72 (m, 2 H), 2.62 (dd, 1 H, J ) 17.1, 7.5), 2.57 (m, 2 H). 13C
NMR δ 159.3, 155.4, 137.0, 131.4 (4 s), 129.5, 129.3, 128.6, 126.8,
115.5, 80.8 (6 d), 41.7, 40.9, 31.9, 29.7 (4 t). Anal.
(C18H19NO2 ·0.75H2O) C, H, N.

3570 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12
Pulkkinen et al.
(19) Yang, C.; Xu, G.; Li, J.; Wu, X.; Liu, B.; Yan, X.; Wang, M.; Xie, Y.
Benzothiophenes containing a piperazine side chain as selective ligands
for the estrogen receptor alpha and their bioactivities in vivo. Bioorg.
Med. Chem. Lett. 2005, 15, 1505–1507.
(20) Qin, Z.; Kastrati, I.; Chandrasena, R. E. P.; Liu, H.; Yao, P.; Petukhov,
P. A.; Bolton, J. L.; Thatcher, G. R. J. Benzothiophene selective
estrogen receptor modulators with modulated oxidative activity and
receptor affinity. J. Med. Chem. 2007, 50, 2682–2692.
(21) Malamas, M. S.; Manas, E. S.; McDevitt, R. E.; Gunawan, I.; Xu,
Z. B.; Collini, M. D.; Miller, C. P.; Dinh, T.; Henderson, R. A.; Keith,
J. C.,., Jr.; Harris, H. A. Design and Synthesis of Aryl Diphenolic
Azoles as Potent and Selective Estrogen Receptor-ꢀ Ligands. J. Med.
Chem. 2004, 47, 5021–5040.
Acknowledgment. We would like to thank Tarja Ihalainen
and Kaarina Pitka¨nen for expert technical assistance and Tiina
Koivunen for performing the elemental analyses. J.P. (grant no.
75791) and P.H. (grant no. 51610) were supported in part by
the Academy of Finland.
Supporting Information Available: Characterization and el-
emental analysis data; references for the previously reported
compounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
(22) Clegg, N. J.; Paruthiyil, S.; Leitman, D. C.; Scanlan, T. S. Differential
response of estrogen receptor subtypes to 1,3-diarylindene and 2,3-
diarylindene ligands. J. Med. Chem. 2005, 48, 5989–6003.
(23) Dykstra, K. D.; Guo, L.; Birzin, E. T.; Chan, W.; Yang, Y. T.; Hayes,
E. C.; DaSilva, C. A.; Pai, L.-Y.; Mosley, R. T.; Kraker, B.; Fitzgerald,
P. M. D.; DiNinno, F.; Rohrer, S. P.; Schaeffer, J. M.; Hammond,
M. L. Estrogen receptor ligands. 2-Aryl indoles as highly subtype
selective ligands for ERR. Bioorg. Med. Chem. Lett. 2007, 17, 2322–
2328.
(24) McDevitt, R. E.; Malamas, M. S.; Manas, E. S.; Unwalla, R. J.; Xu,
Z. B.; Miller, C. P.; Harris, H. A. Estrogen receptor ligands: design
and synthesis of new 2-arylindene-1-ones. Bioorg. Med. Chem. Lett.
2005, 15, 3137–3142.
(25) Ullrich, J. W.; Unwalla, R. J.; Singhaus, R. R.; Harris, H. A.;
Mewshaw, R. E. Estrogen receptor ꢀ ligands: Design and synthesis
of new 2-phenyl-isoindole-1,3-diones. Bioorg. Med. Chem. Lett. 2007,
17, 118–122.
(26) Yang, W.; Wang, Y.; Ma, Z.; Golla, R.; Stouch, T.; Seethala, R.;
Johnson, S.; Zhou, R.; Gungor, T.; Feyen, J. H. M.; Dickson, J. K.,
Jr. Synthesis and structure-activity relationship of 3-arylbenzoxazines
as selective estrogen receptor ꢀ agonists. Bioorg. Med. Chem. Lett.
2004, 14, 2327–2330.
(27) Hoekstra, W. J.; Patel, H. S.; Liang, X.; Blanc, J.-B. E.; Heyer, D. O.;
Willson, T. M.; Iannone, M. A.; Kadwell, S. H.; Miller, L. A.; Pearce,
K. H.; Simmons, C. A.; Shearin, J. Discovery of Novel Quinoline-
Based Estrogen Receptor Ligands Using Peptide Interaction Profiling.
J. Med. Chem. 2005, 48, 2243–2247.
(28) Vu, A. T.; Cohn, S. T.; Manas, E. S.; Harris, H. A.; Mewshaw, R. E.
ERꢀ ligands. Part 4: Synthesis and structure-activity relationships
of a series of 2-phenylquinoline derivatives. Bioorg. Med. Chem. Lett.
2005, 15, 4520–4525.
(29) Vu, A. T.; Campbell, A. N.; Harris, H. A.; Unwalla, R. J.; Manas,
E. S.; Mewshaw, R. E. ERꢀ ligands. Part 6: 6H-Chromeno[4,3-
b]quinolines as a new series of estrogen receptor ꢀ-selective ligands.
Bioorg. Med. Chem. Lett. 2007, 17, 4053–4056.
(30) Kim, S.; Wu, J. Y.; Birzin, E. T.; Frisch, K.; Chan, W.; Pai, L.-Y.;
Yang, Y. T.; Mosley, R. T.; Fitzgerald, P. M. D.; Sharma, N.;
Dahllund, J.; Thorsell, A.-G.; DiNinno, F.; Rohrer, S. P.; Schaeffer,
J. M.; Hammond, M. L. Estrogen Receptor Ligands. II. Discovery of
Benzoxathiins as Potent, Selective Estrogen Receptor R Modulators.
J. Med. Chem. 2004, 47, 2171–2175.
(31) Blizzard, T. A.; DiNinno, F.; Chen, H. Y.; Kim, S.; Wu, J. Y.; Chan,
W.; Birzin, E. T.; Yang, Y. T.; Pai, L.-Y.; Hayes, E. C.; DaSilva,
C. A.; R, S. P.; Schaeffer, J. M.; Hammond, M. L. Estrogen receptor
ligands. Part. 13: Dihydrobenzoxathiin SERAMs with an optimized
antagonist side chain. Bioorg. Med. Chem. Lett. 2005, 15, 3912–3916.
(32) Wallace, O. B.; Lauwers, K. S.; Dodge, J. A.; May, S. A.; Calvin,
J. R.; Hinklin, R.; Bryant, H. U.; Shetler, P. K.; Adrian, M. D.; Geiser,
A. G.; Sato, M.; Burris, T. P. A selective estrogen receptor modulator
for the treatment of hot flushes. J. Med. Chem. 2006, 49, 843–846.
(33) Norman, B. H.; Dodge, J. A.; Richardson, T. I.; Borromeo, P. S.; Lugar,
C. W.; Jones, S. A.; Chen, K.; Wang, Y.; Durst, G. L.; Barr, R. J.;
Montrose-Rafizadeh, C.; Osborne, H. E.; Amos, R. M.; Guo, S.;
Boodhoo, A.; Krishnan, V. Benzopyrans Are Selective Estrogen
Receptor ꢀ Agonists with Novel Activity in Models of Benign Prostatic
Hyperplasia. J. Med. Chem. 2006, 49, 6155–6157.
(34) Gupta, A.; Dwivedy, A.; Keshri, G.; Sharma, R.; Balapure, A. K.;
Singh, M. M.; Ray, S. Rapid synthesis of 4-benzylidene and 4-[bis-
(4-methoxyphenyl)-methylene-2-substituted phenyl-benzopyrans as
potential selective estrogen receptor modulators (SERMs) using
McMurry coupling reaction. Bioorg. Med. Chem. Lett. 2006, 16, 6006–
6012.
(35) Kanbe, Y.; Kim, M.-H.; Nishimoto, M.; Ohtake, Y.; Kato, N.;
Tsunenari, T.; Taniguchi, K.; Ohizumi, I.; Kaiho, S.; Morikawa, K.;
Jo, J.-C.; Lim, H.-S.; Kim, H.-Y. Discovery of thiochroman and
chroman derivatives as pure antiestrogens and their structure-activity
relationship. Bioorg. Med. Chem. 2006, 14, 4803–4819.
References
(1) Laatikainen, R.; Tuppurainen, K. Torsional entropy as cooperator: a
device for synthetic allostery. Tetrahedron Lett. 1988, 29, 5021–5024.
(2) Tuppurainen, K.; Viisas, M.; Pera¨kyla¨, M.; Laatikainen, R. Ligand
intramolecular motions in ligand-protein interaction: ALPHA, a novel
dynamic descriptor and a QSAR study with extended steroid bench-
mark dataset. J. Comput.-Aided Mol. Des 2004, 18, 175–187.
(3) Grese, T. A.; Dodge, J. A. Selective estrogen receptor modulators
(SERMs). Curr. Pharm. Des. 1998, 4, 71–92.
(4) Lin, X.; Huebner, V. Non-steroidal ligands for steroid hormone
receptors. Curr. Opin. Drug. DiscoVery DeV. 2000, 3, 383–398.
(5) Sun, J.; Meyers, M. J.; Fink, B. E.; Rajendran, R.; Katzenellenbogen,
J. A.; Katzenellenbogen, B. N. Novel ligands that function as selective
estrogens or antiestrogens for estrogen receptor-R or estrogen receptor-
ꢀ. Endocrinology 1999, 140, 800–804.
(6) Meyers, M. J.; Sun, J.; Carlson, K. E.; Katzenellenbogen, B. S.;
Katzenellenbogen, J. A. Estrogen receptor subtype-selective ligands:
asymmetric synthesis and biological evaluation of cis- and trans-5,11-
dialkyl-5,6,11,12-tetrahydrochrysenes. J. Med. Chem. 1999, 42, 2456–
2468.
(7) Miller, C. P.; Collini, M. D.; Harris, H. A. Constrained phytoestrogens
and analogues as ERꢀ selective ligands. Bioorg. Med. Chem. Lett.
2003, 13, 2399–2403.
(8) Wildonger, K. J.; Ratcliffe, R. W.; Mosley, R. T.; Hammond, M. L.;
Birzin, E. T.; Rohrer, S. P. Tetrahydrofluorenones with conforma-
tionally restricted side chains as selective estrogen receptor beta
ligands. Bioorg. Med. Chem. Lett. 2006, 16, 4462–4466.
(9) Blizzard, T. A.; Gude, C.; Morgan, J. D.; Chan, W.; Birzin, E. T.;
Mojena, M.; Tudela, C.; Chen, F.; Knecht, K.; Su, Q.; Kraker, B.;
Mosley, R. T.; Holmes, M. A.; Rohrer, S. P.; Hammond, M. L.
Androstene-3,5-dienes as ER-ꢀ selective SERMs. Bioorg. Med. Chem.
Lett. 2007, 17, 6295–6298.
(10) De Angelis, M.; Stossi, F.; Carlson, K. A.; Katzenellenbogen, B. S.;
Katzenellenbogen, J. A. Indazole Estrogens: Highly Selective Ligands
for the Estrogen Receptor ꢀ. J. Med. Chem. 2005, 48, 1132–1144.
(11) Edsall, R. J. , Jr.; Harris, H. A.; Manas, E. S.; Meshaw, R. E. ERꢀ
Ligands. Part. 1: The discovery of ERꢀ selective ligands which embrace
the 4-hydroxy-biphenyl template. Bioorg. Med. Chem. 2003, 11, 3457–
3474.
(12) Yang, C.; Edsall, R.; Harris, H. A.; Zhang, X.; Manas, E. S.; Mewshaw,
R. E. ERꢀ Ligands. Part 2: Synthesis and structure-activity relation-
ships of a series of 4-hydroxy-biphenyl-carbaldehyde oxime deriva-
tives. Bioorg. Med. Chem. 2004, 12, 2553–2570.
(13) Muthyala, R. S.; Sheng, S.; Carlson, K. E.; Katzenellenbogen, B. S.;
Katzenellenbogen, J. A. Bridged bicyclic cores containing a 1,1-
diarylethylene motif are high-affinity subtype-selective ligands for the
estrogen receptor. J. Med. Chem. 2003, 46, 1589–1602.
(14) Schmidt, J. M.; Tremblay, G. B.; Plastina, M. A.; Ma, F.; Bhal, S.;
Feher, M.; Dunn-Dufault, R.; Redden, P. R. In vitro evaluation of the
anti-estrogenic activity of hydroxyl substituted diphenylnaphthyl alkene
ligands for the estrogen receptor. Bioorg. Med. Chem. 2005, 13, 1819–
1828.
(15) Stauffer, S. R.; Sun, J.; Katzenellenbogen, B. S.; Katzenellenbogen,
J. A. Acyclic amides as estrogen receptor ligands: Synthesis, binding,
activity and receptor interaction. Bioorg. Med. Chem. 2000, 8, 1293–
1416.
(16) Mortensen, D. S.; Rodriguez, A. L.; Sun, J.; Katzenellenbogen, B. S.;
Katzenellenbogen, J. A. Furans with basic side chains: Synthesis and
biological evaluation of antagonists with selectivity for the estrogen
recptor alpha. Bioorg. Med. Chem. Lett. 2001, 11, 2521–2524.
(17) Manas, E. S.; Unwalla, R. J.; Xu, Z. B.; Malamas, M. S.; Miller, C. P.;
Harris, H. A.; Hsiao, C.; Akopian, T.; Hum, W. T.; Malakian, K.;
Wolfrom, S.; Bapat, A.; Bhat, R. A.; Stahl, M. L.; Somers, W. S.;
Alvarez, J. C. Structure-based design of estrogen receptor-ꢀ selective
ligands. J. Am. Chem. Soc. 2004, 126, 15106–15119.
(18) Schopfer, U.; Schoeffer, P.; Bischoff, S. F.; Nozulak, J.; Feuerbach,
D.; Floersheim, P. Toward selective ER ꢀ agonists for central nervous
system disorders: Synthesis and characterization of aryl benz-
thiophenes. J. Med. Chem. 2002, 45, 1399–1401.
(36) Endo, Y.; Yoshimi, T.; Ohta, K.; Suzuki, T.; Ohta, S. Potent Estrogen
Receptor Ligands Based on Bisphenols with a Globular Hydrophobic
Core. J. Med. Chem. 2005, 48, 3941–3944.

Estrogen Agonism of Diaryl-1,3-diones
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 12 3571
(37) Ogawa, T.; Ohta, K.; Yoshimi, T.; Yamazaki, H.; Suzuki, T.; Ohta,
S.; Endo, Y. m-Carborane bisphenol structure as a pharmacophore
for selective estrogen receptor modulators. Bioorg. Med. Chem. Lett.
2006, 16, 3943–3946.
(38) Zhou, H.-B.; Comninos, J. S.; Stossi, F.; Katzenellenbogen, B. S.;
Katzenellenbogen, J. A. Synthesis and evaluation of estrogen receptor
ligands with bridged oxabicyclic cores containing a diarylethylene
motif: estrogen antagonists of unusual structure. J. Med. Chem. 2005,
48, 7261–7274.
(56) Pulkkinen, J. T.; Laatikainen, R.; Ahlgre´n, M. J.; Pera¨kyla¨, M.;
Vepsa¨la¨inen, J. J. Conformational flexibility and role of aromatic-
aromatic interactions in the crystal packing of the coordination
compounds of some novel quadridentate Schiff bases. J. Chem. Soc.
Perkin Trans. 2 2000, 777–784.
(57) Pulkkinen, J.; Zachar, P.; Borek, V. A.; Laatikainen, R.; Kral, V.
Application of novel metallocomplexes for mimicking of cytochrome
P 450. Analysis of product distribution for 1-hexene as a substrate. J.
Mol. Catal. A 2004, 219, 21–27.
(39) Meyers, M. J.; Sun, J.; Carlson, K. E.; Marriner, G. A.; Katzenellen-
bogen, B. S.; Katzenellenbogen, J. A. Estrogen receptor-ꢀ potency-
selective ligands: structure-activity relationship studies of diarylpro-
pionitriles and their acetylene and polar analogues. J. Med. Chem.
2001, 44, 4230–4251.
(40) Smith, H. M.; Knox, A. S.; Zisterer, D. M.; Lloyd, D. G.; Meegan,
M. J. Flexible estrogen receptor modulators: synthesis, biochemistry
and molecular modeling studies for 3-benzyl-4,6-diarylhex-3-ene and
3,4,6-triarylhex-3-ene derivatives. Med. Chem. 2007, 3, 135–155.
(41) Henke, B. R.; Consler, T. G.; Go, N.; Hale, R. L.; Hohman, D. R.;
Jones, S. A.; Lu, A. T.; Moore, L. B.; Moore, J. T.; Orband-Miller,
L. A.; Robinett, R. G.; Shearin, J.; Spearing, P. K.; Stewart, E. L.;
Turnbull, P. S.; Weaver, S. L.; Williams, S. P.; Wisely, G. B.; Lambert,
M. H. A new series of estrogen receptor modulators that display
selectivity for estrogen receptor ꢀ. J. Med. Chem. 2002, 45, 5492–
5505.
(58) Saarela, J. A.; Tuppurainen, K.; Pera¨kyla¨, M.; Santa, H.; Laatikainen,
R. Correlative motions and memory effects in molecular dynamics
simulations of molecules: principal components and rescaled range
analysis suggest that the motions of native BPTI are more correlated
than those of its mutants. Biophys. Chem. 2002, 95, 49–57.
(59) Li, Y.; Lambert, M. H.; Xu, H. E. Activation of nuclear receptors: A
perspective from structural genomics. Structure 2003, 11, 741–746.
(60) Shiau, A. K.; Barstad, D.; Radek, J. T.; Meyers, M. J.; Nettles, K. W.;
Katzenellenbogen, B. S.; Katzenellenbogen, J. A.; Agard, D. A.;
Greene, G. L. Structural characterization of a subtype-selective ligand
reveals a novel mode of estrogen receptor antagonism. Nat. Struct.
Biol. 2002, 9, 359–364.
(61) Hurth, K. M.; Nilges, M. J.; Carlson, K. E.; Tamrazi, A.; Belford,
R. L.; Katzenellenbogen, J. A. Ligand-induced changes in estrogen
receptor conformation as measured by site-directed spin labeling.
Biochemistry 2004, 43, 1891–1907.
(42) Green, S.; Walter, P.; Kumar, V.; Krust, A.; Bornert, J. M.; Argos,
P.; Chambon, P. Human oestrogen receptor cDNA: Sequence, expres-
sion and homology to v-erb-A. Nature 1986, 320, 134–139.
(43) Kuiper, G. G. J. M.; Enmark, E.; Pelto-Huikko, M.; Nilsson, S.;
Gustafsson, J. Å. Cloning of a novel estrogen receptor expressed in
rat prostate and ovary. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 5925–
5930.
(44) Gustafsson, J. Å. What pharmacologists can learn from recent advances
in estrogen signaling. Trends Pharmacol. Sci. 2003, 24, 479–485.
(45) Shang, Y.; Brown, M. Molecular determinants for the tissue specificity
of SERMs. Science 2002, 295, 2380–2381.
(46) Sonnenschein, C.; Soto, A. M. An updated review of environmental
estrogen and androgen mimics and antagonists. J. Steroid Biochem.
Mol. Biol. 1998, 65, 143–150.
(47) Pulkkinen, J. T.; Laatikainen, R.; Vepsa¨la¨inen, J.; and Alhgre´n, M.
Characterization of a conformationally flexible system: NMR spec-
troscopic study of 1,5-diphenyl-2,4-pentanedione. Magn. Reson. Chem.
1999, 37, 119–126.
(48) Pulkkinen, J. T.; Vepsa¨la¨inen, J. J. 3-Unsubstituted 1,5-diaryl-2,4-
pentanediones and-4-methoxypentanones: Synthesis via corresponding
3-hydroxy ketones generated from 2-isoxazolines. J. Org. Chem. 1996,
61, 8604–8609.
(49) Kuiper, G. G. J. M.; Lemmen, J. G.; Carlsson, B.; Corton, J. C.; Safe,
S. H.; van der Saag, P. T.; van der Burg, B.; Gustafsson, J. Å.
Interaction of estrogenic chemicals and phytoestrogens with estrogen
receptor ꢀ. Endocrinology 1998, 139, 4252–4263.
(50) Ruff, M.; Gangloff, M.; Wurtz, J. M.; Moras, D. Estrogen receptor
transcription and transactivation: Structure-function relationship in
DNA- and ligand-binding domains of estrogen receptors. Breast
Cancer Res. 2000, 2, 353–359.
(51) Anstead, G. M.; Carlson, K. E.; Katzenellenbogen, J. A. The estradiol
pharmacophore: ligand structure-estrogen receptor binding affinity
relationships and a model for the receptor binding site. Steroids 1997,
62, 268–303.
(62) Nettles, K. W.; Sun, J.; Radek, J. T.; Sheng, S. B.; Rodriguez, A. L.;
Katzenellenbogen, J. A.; Katzenellenbogen, B. S.; Greene, G. L.
Allosteric control of ligand selectivity between estrogen receptors R
and ꢀ: Implications for other nuclear receptors. Mol. Cell 2004, 13,
317–327.
(63) Margeat, E.; Bourdoncle, A.; Margueron, R.; Poujol, N.; Cavailles,
V.; Royer, C. Ligands differentially modulate the protein interactions
of the human estrogen receptors R and ꢀ. J. Mol. Biol. 2003, 326,
77–92.
(64) Bruning, J. B.; Chalmers, M. J.; Prasad, S.; Busby, S. A.; Karnenecka,
T. M.; He, Y. J.; Nettles, K. W.; Griffin, P. R. Partial agonists activate
PPARγ using a helix 12 independent mechanism. Structure 2007, 15,
1258–1271.
(65) Einstein, M.; Akiyama, T. E.; Castriota, G. A.; Wang, C. F.;
McKeewer, B.; Mosley, R. T.; Becker, J. W.; Moller, D. E.; Meinke,
P. T.; Wood, H. B.; Berger, J. P. The differential interactions of
peroxisome profilerator-activated receptor γ ligands with tyr473 is a
physical basis for their unique biological activities. Mol. Pharmacol.
2008, 73, 62–74.
(66) Ko, S.; Confalone, P. N. Model studies for the total synthesis of the
maytansinoids based on the intramolecular nitrile oxide-olefin [3 +
2] cycloaddition reaction. Tetrahedron 1985, 41, 3511–3518.
(67) Huang, S. L.; Omura, K.; Swern, D. Oxidation of sterically hindered
alcohols to carbonyls with dimethyl sulfoxide-trifluoroacetic acid
anhydride. J. Org. Chem. 1976, 41, 3329–3331.
(68) Ma¨kinen, J.; Frank, C.; Jyrkka¨rinne, J.; Gynther, J.; Carlberg, C.;
Honkakoski, P. Modulation of mouse and human phenobarbital-
responsive enhancer module by nuclear receptors. Mol. Pharmacol.
2002, 62, 366–378.
(69) Kotaja, N.;.; Aittoma¨ki, S.; Silvennoinen, O.; Palvimo, J. J.; Ja¨nne,
O. A. ARIP3 (androgen receptor-interacting protein 3) and other PIAS
(protein inhibitor of activated STAT) proteins differ in their ability to
modulate steroid receptor-dependent transcriptional activation. Mol.
Endocrinol. 2000, 14, 1986–2000.
(70) Honkakoski, P.; Palvimo, J. J.; Penttila¨, L.; Vepsa¨la¨inen, J.; Auriola,
S. Effects of triaryl phosphates on mouse and human nuclear receptors.
Biochem. Pharmacol. 2004, 67, 97–106.
(52) Manas, E. S.; Xu, Z. B.; Unwalla, R. J.; Somers, W. S. Understanding
the selectivity of genistein for human estrogen receptor-ꢀ using X-ray
crystallography and computational methods. Structure 2004, 12, 2197–
2207.
(53) Kitao, A.; Go, N. Investigating protein dynamics in collective
coordinate space. Curr. Opin. Struct. Biol. 1999, 9, 164–169.
(54) Falke, J. J. A moving story. Science 2002, 295, 1480–1481.
(55) Teague, S. J. Implications of protein flexibility for drug discovery.
Nat. ReV. Drug DiscoVery 2003, 2, 527–541.
(71) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible docking.
J. Mol. Biol. 1997, 267, 727–748.
(72) Tripos Inc., St. Louis, Missouri.
JM8001795