Synthesis and biological evaluation of a novel series of furans: Ligands selective for estrogen receptor α

Article abstract of DOI:10.1021/jm010211u

A variety of nonsteroidal systems can function as ligands for the estrogen receptor (ER), in some cases showing selectivity for one of the two ER subtypes, ERα or ERβ. We have prepared a series of heterocycle-based (furans, thiophenes, and pyrroles) ligands for the estrogen receptor and assessed their behavior as ER ligands. An aldehyde enone conjugate addition approach and an enolate alkylation approach were developed to prepare the 1,4-dione systems that were precursors to the trisubstituted and tetrasubstituted systems, respectively. All of the diones were easily converted into the corresponding furans, but formation of the thiophenes and pyrroles from the more highly substituted 1,4-diones was problematical. Of the systems investigated, the tetrasubstituted furans proved to be most interesting. They were ERα bindingand potency-selective agents, with the triphenolic 3-alkyl-2,4,5-tris(4-hydroxyphenyl)furans (15a-d) displaying generally higher subtype binding selectivity than the bisphenolic analogues (15f-i). Binding selectivity for ERα was as high as 50-70-fold, and transcriptional activation studies showed that several members of this series were ERα selective agonists, with the best compound [3-ethyl-2,4,5-tris(4-hydroxyphenyl)furan, 15b] having full transcriptional activity on ERα while being inactive on ERβ. Comparative binding affinity analysis and molecular modeling were used to investigate the preferred binding mode adopted by the furan ligands, which appears to have the C(2) phenol mimicking the important role of the A-ring of estradiol. These ligands should be useful in studying the biological roles of both ERα and ERβ, and they might form the basis for the development of novel estrogen pharmaceuticals.

Full text of DOI:10.1021/jm010211u

3838
J . Med. Chem. 2001, 44, 3838-3848
Syn th esis a n d Biologica l Eva lu a tion of a Novel Ser ies of F u r a n s: Liga n d s
Selective for Estr ogen Recep tor r
Deborah S. Mortensen,† Alice L. Rodriguez,† Kathryn E. Carlson,† J un Sun,‡ Benita S. Katzenellenbogen,‡ and
J ohn A. Katzenellenbogen*,†
Department of Chemistry, and Department of Molecular and Integrative Physiology, University of Illinois,
Urbana, Illinois 61801
Received May 11, 2001
A variety of nonsteroidal systems can function as ligands for the estrogen receptor (ER), in
some cases showing selectivity for one of the two ER subtypes, ERR or ERâ. We have prepared
a series of heterocycle-based (furans, thiophenes, and pyrroles) ligands for the estrogen receptor
and assessed their behavior as ER ligands. An aldehyde enone conjugate addition approach
and an enolate alkylation approach were developed to prepare the 1,4-dione systems that were
precursors to the trisubstituted and tetrasubstituted systems, respectively. All of the diones
were easily converted into the corresponding furans, but formation of the thiophenes and
pyrroles from the more highly substituted 1,4-diones was problematical. Of the systems
investigated, the tetrasubstituted furans proved to be most interesting. They were ERR binding-
and potency-selective agents, with the triphenolic 3-alkyl-2,4,5-tris(4-hydroxyphenyl)furans
(15a -d ) displaying generally higher subtype binding selectivity than the bisphenolic analogues
(15f-i). Binding selectivity for ERR was as high as 50-70-fold, and transcriptional activation
studies showed that several members of this series were ERR selective agonists, with the best
compound [3-ethyl-2,4,5-tris(4-hydroxyphenyl)furan, 15b] having full transcriptional activity
on ERR while being inactive on ERâ. Comparative binding affinity analysis and molecular
modeling were used to investigate the preferred binding mode adopted by the furan ligands,
which appears to have the C(2) phenol mimicking the important role of the A-ring of estradiol.
These ligands should be useful in studying the biological roles of both ERR and ERâ, and they
might form the basis for the development of novel estrogen pharmaceuticals.
The estrogen receptor (ER) is a ligand-regulated
transcription factor whose activity as an inducer or
repressor of gene transcription depends on the nature
of the ligand with which it is bound, as well as the
nature of the coregulator proteins with which it associ-
ates.1 Estrogen action is important in many tissues, and
ER is involved in the development and function of the
reproductive system and plays a role in bone density
maintenance,2-4 regulation of blood-lipid profiles,3-7 and
brain function.8,9 Not surprisingly, ER is a target for
the discovery of new drugs for treating or regulating a
F igu r e 1. Structures and binding affinity data for pyrazoles
1 and 2.
variety of hormone-related conditions.
There are two ER genes, the well-known gene that
produces subtype ERR and the more recently discovered
gene that produces the subtype ERâ.10,11 The tissue
distribution of ERR and ERâ differ, though the biological
significance of this difference is not yet well under-
stood.12 A number of previously known ER ligands,12-15
as well as some recently developed novel ligand sys-
tems,16 have different potencies17 or are able to effect
different response levels through ERR and ERâ.18
Ligands having very high ER subtype selectivity would
be effective probes of the respective biological roles of
ERR and ERâ and might also function as tissue-selective
agents having improved endocrine profiles.
of the residues that define the ligand binding pocket are
the same.19 The two differences are conservative sub-
stitutions, with Met421 in ERR corresponding to an
isoleucine in ERâ and, similarly, Leu384 in ERR being
replaced with a methionine in ERâ. These subtle
changes in the ligand-binding pocket of the two ER
subtypes do not provide a definitive basis for under-
standing the selectivity in either binding or potency that
is seen with some ligands, so further structural analyses
will be required.
In previous investigations, we found that certain
pyrazoles, particularly those displaying a 1,3,5-triaryl-
4-alkyl substitution pattern, were very selective for
ERR, in terms of affinity, potency, and efficacy.20-25
Pyrazole 1 was found to have the highest ERR binding
affinity, but pyrazole 2 had the greatest ERR subtype
selectivity (Figure 1). While certain azoles, such as
oxazoles, thiazoles, isoxazoles, and imidazoles, did not
produce good ER ligands in our hands,20 we wondered
The ER subtypes R and â have only 55% amino acid
identity in the ligand binding domain, yet all but two
* Corresponding author. Address: Department of Chemistry, Uni-
versity of Illinois, 600 S. Mathews Ave., Urbana, IL 61801. Phone:
(217) 333-6310. Fax: (217) 333-7325. E-mail: jkatzene@uiuc.edu.
† Department of Chemistry.
‡ Department of Molecular and Integrative Physiology.
10.1021/jm010211u CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/16/2001

Ligands Selective for Estrogen Receptor R
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3839
Sch em e 1. Preparation of 1,2,4-Trisubstituted Diones 10a -c
Sch em e 2. Preparation of 1,2,3,4-Tetrasubstituted Diones 13a -l
whether other five-membered ring heterocycles bearing
only one heteroatom, namely, furans, thiophenes, and
pyrroles, might provide ER ligands with interesting
biological character. Several furans (3-5) and pyrroles
(6-8) have been described as potential antifertility
agents.26-29 Few, however, showed significant activity,
the most active being the furans (3-5), and binding and
transcriptional activity were reported only in one in-
stance.26
In this report we describe an investigation of single
heteroatom-containing five-membered ring heterocyclic
analogues of the high-affinity pyrazole ligands. We find
that certain furans are ER ligands that display ERR-
selective biological character equivalent to that of the
pyrazoles. We have also developed a structure-activity
relationship within the furan series to optimize ligand
affinity and selectivity and to investigate the binding
mode adopted by this class of ligands.
an R,â-unsaturated ketone 9a -c (Scheme 1). Treatment
of commercially available acetophenones and aromatic
aldehydes with ethanolic KOH afforded chalcones 9a -
c. The R,â-unsaturated ketones 9a -c then underwent
conjugate addition reactions with commercially avail-
able aldehydes, utilizing Stetter’s thiazolium salt cata-
lysts, either 3-ethyl-5-(2-hydroxyethyl)-4-methylthia-
zolium bromide for aromatic aldehydes or 3-benzyl-5-
(2-hydroxyethyl)-4-methylthiazolium chloride for ali-
phatic aldehydes.30 The Stetter reactions produced the
desired 1,2,4-trisubstituted butane-1,4-diones 10a -c in
good yields. However, all of our attempts to use this
approach with an R-substituted enone to produce the
desired tetrasubstituted diones were unsuccessful.
The 1,2,3,4-tetrasubstituted butanediones were in-
stead prepared by the alkylation of an enolate with an
R-bromo ketone (Scheme 2). Treatment of the 1,2-
diarylethanones 11a -d with potassium hexamethyl-
disilyl amide (KHMDS), followed by the addition of an
R-bromo ketone 12a -f, provided the desired 1,3,4-
triaryl-2-alkylbutane-1,4-diones 13a -l. These reactions
afforded the product diones as mixtures of diastereo-
mers, which in some cases were separable by crystal-
lization. Separation was not required, however, because
these centers become trigonal in the final products. The
R-bromo ketones 12a -f were produced in nearly quan-
Resu lts
Syn th esis of Liga n d s. While the literature provides
a plethora of approaches to furans, thiophenes, and
pyrroles, we wished to prepare all three of these
heterocycles from a common precursor, namely, a 1,4-
dione. The trisubstituted 1,4-diketones 10a -c were
prepared by the conjugate addition of an aldehyde to

3840 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Ta ble 1. Synthesis of Furans 14a -n
Mortensen et al.
Ta ble 2. Demethylation of Furans 14a -n to 15a -n
dione furan
R1
R2
R3
R4
% yield
furan deprot. furan
R1
R2
R3
R4
% yield
13a
13b
13c
13d
13e
13f
13g
13h
13i
14a
14b
14c
14d
14e
14f
14g
14h
14i
OCH3 OCH3 CH3
OCH3 OCH3 C2H5
OCH3 OCH3 C3H7
OCH3 OCH3 C4H9
OCH3
OH
OCH3
OCH3
74
70
85
98
85
87
83
95
73
42
75
25
88
92
14a
14b
14c
14d
14e
14f
14g
14h
14i
15a
15b
15c
15d
15e
15f
15g
15h
15i
OH OH CH3
OH
OH
OH
OH
77
93
93
88
85
94
88
93
84
55
70
78
75
77
OH OH C2H5
OH OH C3H7
OH OH C4H9
OCH3 OCH3 p-CH3OC6H4 OCH3
OH OH p-HOC6H4 OH
H
OCH3 C2H5
OCH3 C3H7
OCH3
OCH3
OCH3
H
OCH3
H
H
H
OH C2H5
OH C3H7
OH
OH
OH
H
OH
H
H
H
H
OCH3
OCH3 OCH3 C2H5
H
H
OCH3
OCH3
H
H
C2H5
OH
H
C2H5
OH OH C2H5
13j
14j
H
C2H5
14j
15j
H
H
OH
OH
H
H
C2H5
13k
13l
10a
10b
14k
14l
14m
14n
OCH3 C2H5
14k
14l
14m
14n
15k
15l
15m
15n
OH C2H5
H
C2H5
H
H
H
C2H5
H
H
OCH3
OCH3
OH
OH
OCH3
H
OH
H
titative yields from the parent ketones upon treatment
with bromine and a catalytic amount of aluminum
chloride.
Unfortunately, all attempts to convert the 1,2,3,4-
tetrasubstituted butanediones into thiophenes afforded
solely furan products. Even with the addition of potas-
sium carbonate, Lawesson’s reagent gave only quantita-
tive yields of the furans. An alternate approach using
hydrogen sulfide-hydrogen chloride also failed to give
the desired thiophenes, instead producing furan in low
yield and leaving unreacted starting material. It is
known that these latter reagents can be used to produce
disubstituted thiophenes either from 1,4-diones33,34 or
directly from the corresponding furans.33 However, these
conditions were reported not to work for the conversion
of tetrasubstituted systems, such as 1,4-diaryl-2,3-
dibromobutane-1,4-diones, into the corresponding thio-
phenes.34 While our failure to convert tetrasubstituted
1,4-diones into thiophenes was disappointing, it is not
surprising considering the ample precedent for prefer-
ential furan formation from hindered 1,4-diones.33-36 In
fact, to our knowledge, there has been only one report
of the formation of thiophenes from 1,2,3,4-tetrasubsti-
tuted-1,4-dicarbonyl precursors.37
F u r a n s. The 1,4-diketones were converted into furans
upon treatment with catalytic p-toluenesulfonic acid in
refluxing toluene. Table 1 lists the yields for conversion
of diones 13a -l and 10a -b to furans 14a -n . The
methyl ether protecting groups of furans 14a -n were
removed using boron trifluoride-dimethyl sulfide com-
plex, to afford furans 15a -n as free phenols. The
demethylation reactions and yields for conversion of
14a -n to 15a -n are listed in Table 2.
Th iop h en es. Upon treatment with Lawesson’s re-
agent, diones 10a -b were converted into thiophenes
16a -b.31,32 Furan formation was found to be a competing
process in these reactions, sometimes giving a furan-
to-thiophene ratio as high as a 1:1 when the 1,4-diones
were treated with Lawesson’s reagent alone. Competing
furan formation not only reduced yields, but the furan
byproducts could not be separated from the correspond-
ing thiophenes by flash column chromatography or
recrystallization. Fortunately, the addition of solid
potassium carbonate as an acid scavenger minimized
furan formation, so the thiophenes could be isolated in
pure form. Demethylation of 16a -b with boron trifluo-
ride-dimethyl sulfide complex afforded the phenolic
thiophene analogues 17a -b in good yields (Scheme 3).
P yr r oles. Treatment of diones 10c with p-anisidine
and p-toluenesulfonic acid produced pyrrole 18 in mod-
est yield; however, unreacted dione may be recycled to
increase production of the desired heterocycle. De-
methylation of 18 provided the target pyrrole 19 (Scheme
4). As was the case with the thiophenes, all attempts
Sch em e 3. Preparation of Thiophenes 17a -b

Ligands Selective for Estrogen Receptor R
Sch em e 4. Preparation of Pyrrole 19
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3841
Ta ble 3. Relative Binding Affinity (RBA)a Data for Furans
15a -n , Thiophenes 17a -b, and Pyrrole 19
F igu r e 2. Representative furans and pyrroles explored as
antifertility agents.26-29
ligand
cytosol
ER-R
ER-â
R/â selectivity
15a
15b
15c
15d
15e
15f
15g
15h
15i
15j
15k
15l
15m
15n
17a
17b
19
2.9 ( 0.09
6.5 ( 0.8
3.6 ( 0.6
1.3 ( 0.7
0.84 ( 0.07
4.6 ( 0.5
3.6 ( 0.3
0.7 ( 0
40 ( 6.5
140 ( 38
100 ( 14
21 ( 0.6
8.7 ( 1.1
82 ( 20
0.62 ( 0.02
2.9 ( 0.1
1.8 ( 0.65
3.9 ( 1.1
0.25 ( 0.01
7.1 ( 1.2
15 ( 4.1
65-fold
48-fold
56-fold
5.4-fold
36-fold
12-fold
9.5-fold
5.5-fold
3.3-fold
3.8-fold
2.1-fold
3.4-fold
NA
close structural analogue to a high-affinity pyrazole (cf.,
Figure 1).20
All of the tetrasubstituted furans (15a -l) were as-
sayed for binding to ERR and ERâ, and a number of
structure-affinity trends are of note. All of these furans
had much higher affinity for purified ERR than for ER
from lamb uterine cytosol. We believe that this is due
to reduced nonspecific binding of these relatively lipo-
philic ligands in the assays with the purified receptor,
ERR, which are performed at much lower total protein
concentrations than are the assays using uterine cytosol.
While all of the furans proved to be ERR binding
selective ligands, the data indicates that the presence
of a third phenolic hydroxyl is required to achieve the
highest binding selectivity for ERR versus ERâ (compare
tris-phenols vs the corresponding bisphenols: 15c vs
15g, and 15b vs 15f,h ,i, respectively). As the size of the
3-alkyl substituent is changed in the triphenols, the
highest ERR binding affinity is found with the ethyl and
propyl analogues (15b and 15c with of RBAs 100-140),
but the highest ERR binding selectivity was found with
the methylfuran 15a (RBA ERR/ERâ ) 65). The sym-
metrical tetrakis(4-hydroxyphenyl)furan (15e) has re-
spectable ERR affinity and selectivity, but is not as good
as the best of the 3-alkyl analogues. The homologous
series of bisphenols (15f,h ,i) and monophenols (15j-l)
were prepared to investigate the orientation of these
furans in the ligand binding pocket of ER; their affinities
are lower, and the structure-affinity correlations are
considered further in the Discussion section.
140 ( 13
16.5 ( 1.9
14.8 ( 3.1
3.0 ( 0.6
4.5 ( 1.2
3.4 ( 1.2
0.45 ( 0
0.44 ( 0.09 10.8 ( 2.6
0.02 ( 0.01 0.15 ( 0.01 0.07 ( 0.02
0.10 ( 0.04
0.13
6.8 ( 2.1
NA
2.0 ( 0.4
NA
0.04
NA
NA
NA
0.03
NA
NA
NA
0.04
NA
NA
NA
0.50
NA
NA
NA
a
Determined by a competitive radiometric binding assay with
[3H]estradiol using methods described in the Experimental Section.
Where indicated, values represent the average ((SD or range) of
multiple determinations. In our hands, RBA values are reproduc-
ible with a coefficient of variation of less than 0.3.
to convert 1,2,3,4-tetrasubstituted butane-1,4-diones
into pyrroles resulted in quantitative furan formation.
Biologica l Stu d ies. Rela tive Bin d in g Affin ities.
The relative binding affinity of the heterocyclic ligands
was determined by a competitive radiometric receptor
binding assay first using lamb uterine cytosol as a
source of receptor.38 (The uterus is thought to contain
almost exclusively ERR.)12 The higher affinity ligands
were then studied further with purified human ERR and
ERâ.18,39 The results of these studies, given as relative
binding affinity (RBA) values where the affinity of
estradiol is considered to be 100%, are summarized in
Table 3.
Tr a n scr ip tion a l Activa tion Assa ys. Four high-
affinity tetrasubstituted furan ligands (15a ,b,c,g) were
assessed for transcriptional activation activity through
both ERR and ERâ. These cotransfection assays are
conducted in human endometrial carcinoma (HEC-1)
cells, using expression plasmids for full-length human
ERR or ERâ and an estrogen-responsive luciferase
reporter gene system.16 The results of these initial
screening assays are summarized in Figure 3.
While the members of these series of heterocycles
displayed somewhat lower affinities for uterine cytosol
ER than did those of the pyrazole class,20 they showed
good affinity nevertheless, with the highest RBA of 6.5%
being observed for the 2,4,5-tris(4-hydroxyphenyl)-3-
ethylfuran 15b. As was the case with the pyrazoles,20
the trisubstituted ligands, furans 15m ,n and thiophenes
17a ,b, showed very low affinities and thus were not
assayed further with ERR and ERâ. The one pyrrole we
prepared (19) also had low affinity; this was unexpected,
considering that it is a tetrasubstituted ligand that is a
All four ligands tested were more potent on ERR than
ERâ, and they were fully efficacious on ERR. On ERâ,
however, the three triphenols (15a -c) were inactive, but

3842 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mortensen et al.
F igu r e 4. Dose-response curve for transcriptional activation
by furan 15b though ERR (solid squares) and ERâ (solid
triangles). Antagonist activity through ERâ (circles) was also
assayed. The activation curves for estradiol on those two
receptors is shown for reference (open squares and triangles,
respectively; dotted lines). For details, see the Experimental
Section.
nM for ERR and having essentially no agonist or
antagonist activity on ERâ at concentrations up to 1 µM.
Discu ssion
We have prepared a series of aryl-substituted five-
membered ring heterocycles containing a single het-
eroatom, namely, furans, thiophenes, and pyrroles, and
we have assessed them as ligands for the estrogen
receptor (ER). Certain tetrasubstituted furans proved
to be very selective for ERR, in terms of binding affinity
and potency as agonists, with the triphenolic 3-alkyl-
2,4,5-tris(4-hydroxyphenyl)furans (15a -d ) displaying
generally higher subtype binding selectivity than the
bisphenolic analogues (15f-i) (Table 3). The highest
ERR binding selectivity (65-fold) was obtained with the
2,4,5-tris(4-hydroxyphenyl)-3-methylfuran 15a , but the
best overall combination of high affinity and ERR
selectivity was given by the corresponding 3-ethyl and
3-propyl analogues, furans 15b and 15c (Table 3). The
ethylfuran 15b is sufficiently ERR selective to be used
to activate ERR fully at concentrations (10-100 nM)
where it has no agonist or antagonist activity on ERâ
(Figure 4).
Str u ctu r e-Activity Rela tion sh ip s in th e Tr i-
a r ylfu r a n Ser ies. The heterocycles most closely related
to the furans, thiophenes, and pyrroles studied here are
the pyrazoles that we have investigated extensively
(Figure 1),20,22-25 and certain features that we noted in
the behavior of the pyrazoles are also found with the
new heterocycles. The triaryl-substituted systems have
low binding affinity, as was the case with the pyrazoles
(and other related heterocycles), and high affinity and
ERR selectivity was engendered by having a fourth
group that was aliphatic and moderate in size.20 With
the furans, the most favorable substituents at this
position were ethyl and propyl, whereas with the
pyrazoles it was a propyl substituent. The best furan
(RBA 140 for furan 15b) has considerably higher ERR
binding affinity than the best pyrazole (RBA 75 for
pyrazole 1),24 but in both series, the trend of affinity
increasing with substituent size up to a point, beyond
F igu r e 3. Transcriptional activation data for furans 15a -c
and 15g through ERR and ERâ. Human endometrial cancer
(HEC-1) cells were transfected with expression vectors for ERR
or ERâ and an estrogen responsive reporter gene and were
treated with indicated concentrations of estradiol or ligand for
24 h to assay for agonism (diagonal slashed bars) or antago-
nism (crosshatched bars). Transcriptional activity was normal-
ized to an internal â-gal control plasmid using the luciferase
reporter assay system. Values are expressed as a percent of
the ERR or ERâ response with 10 nM E2, which is set at 100%
(open bars).29 For details, see the Experimental Section.
the bisphenol (15g) showed partial agonist activity.
Thus, while having a third phenol on the C(5) phenyl
group is not essential for high ERR binding affinity
(Table 3), it does appear to be important to ensure that
the furans are highly selective for ERR and not effica-
cious on ERâ.
A full dose-response curve for transcriptional activa-
tion through ERR and ERâ for one of the most selective
of the ligands assayed, 3-ethyl-2,4,5-tris(4-hydroxyphen-
yl)furan (15b), is shown in Figure 4. This furan is highly
selective in its activation of ERR, having an EC50 ) 0.33

Ligands Selective for Estrogen Receptor R
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3843
F igu r e 6. Summary of ERR binding data for furan monophe-
nols and bisphenols for comparative binding analysis. The
binding affinity of the three monophenols is shown in the
shadowed box by each of the three hydroxyl groups, and the
affinity of the three bisphenols is indicated by the italicized
number in parentheses on braces linking two hydroxyl groups.
For a discussion, see the text.
tigate the effect that removing phenolic hydroxyl groups
had on binding affinity. We imagined that deletion of
the hydroxy group from the furan phenol substituent
that was playing the role of the A-ring of estradiol would
have the greatest effect on binding affinity.40 Also, if
two phenols were deleted, we imagined that the
monophenol having the highest affinity would be the
one that retained the hydroxyl on the phenyl substituent
in the furan that mimics the estradiol A-ring. The ERR
binding affinities of the mono- and diphenols are given
in Table 3, but are schematized in Figure 6 for ready
analysis.
In the monophenols (15j-l) (See RBA values shown
in boxes in Figure 6) the C(4) monophenol (15k ) has a
very low affinity, so it is unlikely that the C(4) aryl
group is the mimic of the estradiol A-ring. However, it
is difficult to choose between the C(2) and C(5) phenols
(15j,l), because their affinities are quite comparable. So,
from the monophenols, it appears that either the C(2)
or C(5) phenol rings could be the A-ring mimic. The
highest affinity bisphenol (15f) (see the RBA values
shown in parentheses in italics, Figure 6) has hydroxyl
groups on both the C(2) and C(4) phenyl groups,
consistent with the importance of the C(2) phenol, noted
above. However, the other two bisphenols (15h ,i) again
have very similar affinities, so a definitive distinction
cannot be made between the importance of the other
two phenols, at C(4) and C(5).
On the basis of these data, we make the following
tentative conclusions: (a) the C(4) phenol is unlikely to
be the A-ring mimic (based on the affinity pattern of
the monophenols), (b) the C(2) phenol is most likely to
be the A-ring mimic (based on the higher affinity of the
2,4-bisphenol (15f) than the 2,5- and 4,5-bisphenols
(15h ,i), but (c) the C(5) phenol might also function as
the A-ring mimic. This tentative conclusion is the same
as the one we reached in our analysis of the binding
orientations of the pyrazoles,24 and it is supported by
further work we have done, both in the pyrazole and
the furan series, to develop ERR-selective antagonists
by attaching basic side chains to these heterocyclic
systems.25,43 Thus, we believe that the furan triphenols
bind with the C(2) phenol in the A-ring binding pocket,
but that the C(5) phenol can play this role in certain
analogues (notably furan antagonists).43
F igu r e 5. Three possible mimics for the A-ring of estradiol.
Note, with each A-ring mimic, there are two possible orienta-
tions of the furan core (only one is shown).
which further increase proves detrimental, was noted.
Such trends are well precedented in both steroidal and
nonsteroidal systems,40,41 and they have been inter-
preted to indicate the filling of a preformed pocket of
limited volume within the receptor.40
Where it is possible to make direct comparisons
between furans and thiophenes (15m vs 17a and 15n
vs 17b), it appears that the furans might be higher
affinity ligands. However, this comparison remains
limited by difficulties we encountered in the preparation
of tetrasubstituted thiophenes. We prepared only one
pyrrole (19), and though it was tetrasubstituted, its
affinity was disappointing. This was unexpected, be-
cause it has strong structural analogy to other hetero-
cycles that have high affinity.20 The higher polarity of
the pyrrole compared to the furans and pyrazoles could
account for its lower binding affinity.20
In vestiga tion of th e Or ien ta tion of th e Tr ia r yl-
fu r a n s in th e ER Liga n d Bin d in g P ock et. We have
been interested in the manner in which nonsteroidal ER
ligands having more than one phenol are oriented
within the ligand binding pocket of ER.42 Most of the
furans we studied here that show high binding selectiv-
ity for ERR (15a -c) have three phenols, any one of
which could be serving as the analogue of the phenolic
A-ring of estradiol, a ring that is known to be important
for the high affinity of this natural estrogen (see Figure
5).40 While multiple binding modes could be operative,
we used comparative binding affinity considerations and
molecular modeling to see whether we could determine
which of these three phenols might most likely be
functioning as the A-ring mimic. One should note that
for each of the three potential A-ring mimics, there are
two possible binding modes for the rest of the furan (i.e.,
the ligands could be bound in the orientations shown
in Figure 5 or in alternative binding modes, flipped 180°
around the bond connecting the A-ring mimic phenol
to the furan core, not shown, giving a total of six basic
binding orientations).42
Com p a r a tive Bin d in g An a lysis. In the compara-
tive binding affinity approach, we prepared furan
bisphenols (15f,h ,i) and monophenols (15j-l) to inves-
Molecu la r Mod elin g Stu d ies. We have also used
molecular modeling to investigate the binding orienta-

3844 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mortensen et al.
F igu r e 7. (A) Model of furan 15b in the ERR ligand binding pocket. The surface of the ligand is shown as a continuous green
shape; the surface of the ERR pocket is shown as purple dots. (B) Comparison of the orientation of the furan (gray) with respect
to that of estradiol (purple) in the ERR ligand binding pocket. For details, see the text and Experimental Section.
tion of a furan ligand in ERR. Using the FlexiDock
routine in the modeling program SYBYL, according to
a modification of the protocol we described previously
for the pyrazoles (See Experimental),24 we docked the
ERR-selective furan 15b into the ligand-binding pocket
of ERR, positioning the furan initially in each of six
possible orientations (cf., Figure 5 and earlier discus-
sion). In each case, the FlexiDock routine oriented the
molecule so that the C(2) phenol remained or returned
to the position so as to be the A-ring mimic. With the
C(2) phenol in the A-ring orientation, we obtained the
best fit when the C(3) ethyl group projected downward
in roughly the direction of a 6R- or 7R-substituent in
estradiol. The result of this docking-minimization study
is shown in Figure 7. The purple dots represent the
solvent-accessible surface of the ligand binding pocket,
and the green surface is the molecular volume for the
furan ligand. At the right of the figure is an overlay of
furan 15b (gray) onto estradiol (purple), illustrating
their relative position within the ligand binding pocket.
This final minimized model shows that the hydrogen-
bond contacts found between the phenol and the Glu353
and Arg394 residues in the ERR/estradiol and ERR/
diethylstilbestrol crystal structures44,45 persist with the
C(2) phenol of the furan ligand. However, because of
the overall greater length of the furan, we found that
the phenol appears to be driven more deeply into the
A-ring binding pocket in the furan structure than in the
estradiol structure. The second phenol on C(4) of the
furan makes a hydrogen bond with His524, as is also
found with the distal hydroxyl groups of ligands in the
crystal structures.44,45 The third phenol on C(5) appears
to be making a hydrogen bond with the hydroxy group
of Thr347, which is the only other polar residue in the
ligand binding pocket. As was the case with the pyra-
zoles, it is not clear what interactions are responsible
for the high ERR binding selectivity of this ligand,
although this is obviously a very interesting issue.24
The novel heterocyclic systems we have investigated
here, the 3-alkyl-2,4,5-triarylfurans, in particular, are
ligands for the estrogen receptor that have very high
selectivity for ERR in terms of affinity and potency in
transcription assays. These ligands should be useful in
studying the biological role of both of the ER subtypes,
and they might form the basis for the development of
novel estrogen pharmaceuticals.
Exp er im en ta l Section
Gen er a l Meth od s. All reagents and solvents were pur-
chased from Aldrich or Fischer. Solvents were distilled prior
to use. THF was distilled over sodium/benzophenone. Meth-
ylene chloride was distilled over calcium hydride. Hexane was
distilled over calcium sulfate. Triethylamine was distilled over
calcium hydride. Reactions were all monitored by TLC,
performed on 0.25 mm silica gel glass plates containing F-254
indicator (Merck). Visualization on TLC was achieved by UV
light or phosphomolybdic acid indicator. Column chromatog-
raphy was performed with Woelm 32-63 µm silica gel packing.
Melting points were measured using a Thomas-Hoover capil-
lary melting point apparatus and are uncorrected. NMR
spectra chemical shifts are reported in parts per million
downfield from TMS and referenced to either TMS internal
standard for chloroform-d1 or acetone- d6 solvent peak. NMR
coupling constants are reported in hertz. All compounds used
in structure-activity relationship considerations gave either
satisfactory microanalyses or satisfactory exact mass deter-
minations by high-resolution mass spectrometry and were
shown to be pure by HPLC under two distinct reversed phase
conditions.
Ch em ical Syn th esis. Gen er al P r ocedu r e for th e P r epa-
r a tion of Tr isu bstitu ted Bu ta n e-1,4-d ion es by th e Stetter
Rea ction . Aldehyde, R,â unsaturated ketone, triethlyamine,
and either 3-benzyl-5-(2-hydroxyethyl)-4-methylthiazolium chlo-
ride (with aliphatic aldehydes) or 3-ethyl-5-(2-hydroxyethyl)-
4-methylthiazolium bromide (with aromatic aldehydes) were
combined and heated to reflux in ethanol for 60-96 h. Solvent
was removed under reduced pressure, and the resulting
residue was taken up into CH2Cl2 and extracted with 3 M HCl
(aq), water, saturated NaCl, and dried over sodium sulfate.
Solvent was removed under reduced pressure and crude
product was purified by flash column chromatography and
recrystallization to afford 1,4 diones.
Gen er a l P r oced u r e for th e P r ep a r a tion of Tetr a su b-
stitu ted Bu ta n e-1,4-d ion es. A solution of KHMDS (1.1
equiv, 0.5 M in toluene) was added to a stirring solution of
diarylethanone (1 equiv) in THF (2-5 mL), at -78 °C. The
mixture was stirred for 1 h followed by the dropwise addition
of R-bromo ketone (1.1 equiv). The reaction mixture was
allowed to stir at -78 °C for 1 h and then allowed to gradually
warm to room temperature, with stirring from 5 h to overnight
(monitored for disappearance of starting material) and then
quenched with the addition of H2O. The reaction was further
diluted with EtOAc and washed with H2O and saturated NaCl.
Organic extract was dried over Na2SO4 and filtered and solvent

Ligands Selective for Estrogen Receptor R
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3845
removed under reduced pressure. Purification by flash column
chromatography afforded diones as mixtures of diastereomers
(by NMR).
Gen er a l P r oced u r e for F u r a n s. In toluene, the 1,4-dione
and p-toluenesulfonic acid monohydrate (catalytic amount)
were heated to reflux for 3-12 h. The reaction was cooled and
filtered, the solvent was removed under reduced pressure, and
the crude product was purified by flash column chromatogra-
phy.
Gen er a l Dem eth yla tion P r oced u r e Usin g BF 3‚SMe2. To
a stirring solution of the methyl ether precursor (1 equiv) in
CH2Cl2 (∼8 mL) at room temperature was added BF3‚SMe2
complex (75 equiv). After stirring for 12-18 h, solvent and
excess reagent were evaporated under nitrogen stream in hood.
Residue was taken up in EtOAc and washed with H2O and
saturated NaCl. Organic extract was dried over Na2SO4 and
filtered and solvent removed under reduced pressure. The
resulting residue was purified by silica gel flash column
chromatography.
Gen er a l Dem eth yla tion P r oced u r e Usin g BBr 3. The
methyl ether protected compound was dissolved in CH2Cl2 and
stirred at room temperature. A solution of 1 M boron tribro-
mide in CH2Cl2 was added via syringe, and the reaction was
left to stir overnight or until all starting material had been
consumed. The reaction was poured into a separatory funnel,
extracted with H2O (3 × 10 mL) and saturated NaCl, and dried
over sodium sulfate. Solvent was removed under reduced
pressure and the resulting crude product was purified by flash
column chromatography and recrystallization to afford depro-
tected phenols.
2,3,5-Tr is(4-h yd r oxyp h en yl)-4-m eth ylfu r a n (15a ). Fu-
ran 14a (58.0 mg, 0.14 mmol) was reacted according to the
general demethylation procedure using BF3‚SMe2 to afford
crude 15a . The crude material was purified by flash column
chromatography (1:2 EtOAc:hexanes) and recrystallized from
EtOAc:hexanes to give 15a (40.2 mg, 77% yield): mp 231-
233 °C (dec); 1H NMR (500 MHz, acetone-d6) δ 2.05 (3H, s,
CH3), 6.74 (2H, AA′XX′, J AX ) 8.77, J AA′ ) 2.48, ArH ortho to
OH), 6.94 (2H, AA′XX′, J AX ) 8.46, J AA′ ) 2.42, ArH ortho to
OH), 6.95 (2H, AA′XX′, J AX ) 8.83, J AA′ ) 2.49, ArH ortho to
OH), 7.16 (2H, AA′XX′, J AX ) 8.81, J XX′ ) 2.39, ArH meta to
OH), 7.34 (2H, AA′XX′, J AX ) 8.99, J XX′ ) 2.50, ArH meta to
OH), 7.61 (2H, AA′XX′, J AX ) 8.77, J XX′ ) 2.46, ArH meta to
OH), 8.44 (1H, bs, OH), 8.48 (1H, bs, OH), 8.51 (1H, bs, OH);
13C NMR (125 MHz, acetone-d6) δ 10.5, 116.1(2), 116.4(2),
116.6(2), 117.5, 124.1, 124.6, 124.7, 125.8, 127.5(2), 127.7(2),
132.2(2), 147.4, 147.9, 157.4, 157.5, 157.7; MS (EI, 70 eV) m/z
358.1 (M+); HRMS calcd for C23H18O4 358.120 51, found
358.120 40.
93% yield) as a white powder: 1H NMR (500 MHz, acetone-
d6) δ 0.79 (3H, t, J ) 7.24, CH3CH2CH2), 1.42 (2H, m, CH3CH2-
CH2), 2.48 (2H, m, CH3CH2CH2), 6.73 (2H, AA′XX′, J AX ) 8.89,
J AA′ ) 2.53, ArH ortho to OH), 6.95 (4H, AA′XX′, J AX ) 8.55,
J AA′ ) 2.53, ArH ortho to OH), 7.16 (2H, AA′XX′, J AX ) 8.82,
J XX′ ) 2.41, ArH meta to OH), 7.31 (2H, AA′XX′, J AX ) 8.86,
J XX′ ) 2.47, ArH meta to OH), 7.60 (2H, AA′XX′, J AX ) 9.02,
J XX′ ) 2.42, ArH meta to OH), 8.42 (1H, bs, OH), 8.48 (1H, bs,
OH), 8.51 (1H, bs, OH); 13C NMR (125 MHz, acetone-d6) δ 14.4,
23.8, 26.8, 116.0(2), 116.4(2), 116.6(2), 122.9, 124.1, 124.4,
124.5, 126.1, 127.3(2), 127.7(2), 132.2(2), 147.5, 147.7, 157.4,
157.5, 157.7; MS (EI, 70 eV) m/z 386.2 (M+); HRMS calcd for
C
25H22O4 386.151 81, found 386.152 74.
3-Bu tyl-2,4,5-tr is(4-h yd r oxyp h en yl)fu r a n (15d ). Furan
14d (50.0 mg, 0.11 mmol) was reacted according to the general
demethylation procedure using BF3‚SMe2 to afford crude 15d .
The crude material was purified by flash column chromatog-
raphy (1:1 EtOAc:hexanes) to provide 15d (39.5 mg, 88% yield)
as a solid. A small amount of 15d for biological testing was
further purified by reverse phase HPLC: 1H NMR (500 MHz,
acetone-d6) δ 0.76 (3H, t, J ) 7.41, CH3CH2CH2CH2), 1.22 (2H,
sextet, J ) 7.39, CH3CH2CH2CH2), 1.39 (2H, m, CH3CH2CH2-
CH2), 2.50 (2H, m, CH3CH2CH2CH2), 6.72 (2H, AA′XX′, J AX
8.72, J AA′ ) 2.53, ArH ortho to OH), 6.949 (2H, AA′XX′, J AX
8.83, J AA′ ) 2.49, ArH ortho to OH), 6.952 (2H, AA′XX′, J AX
8.63, J AA′ ) 2.53, ArH ortho to OH), 7.16 (2H, AA′XX′, J AX
8.67, J XX′ ) 2.41, ArH meta to OR), 7.31 (2H, AA′XX′, J AX
9.11, J XX′ ) 2.45, ArH meta to OR), 7.60 (2H, AA′XX′, J AX
)
)
)
)
)
)
9.01, J XX′ ) 2.53, ArH meta to OR), 8.47 (1H, bs, OH), 8.52
(1H, bs, OH), 8.56 (1H, bs, OH); 13C NMR (125 MHz, acetone-
d6) δ 13.9, 23.2, 24.3, 32.7, 116.0(2), 116.4(2), 116.6(2), 123.0,
124.1, 124.4, 124.5, 126.1, 127.3(2), 127.7(2), 132.2(2), 147.5,
147.6, 157.4, 157.5, 157.7; MS (EI, 70 eV) m/z 400.2 (M+);
HRMS calcd for C26H24O4 400.167 46, found 400.167 43.
2,3,4,5-Tetr a k is(4-h yd r oxyp h en yl)fu r a n (15e). Furan
14e (40.0 mg, 0.10 mmol) was reacted according to the general
demethylation procedure using BF3‚SMe2 to afford crude 15e.
The crude material was purified by flash column chromatog-
raphy (1:1 hexane:EtOAc) and recrystallized from EtOAc:
hexanes to give 15e (49.0 mg, 85% yield): mp 247-250 °C
(dec); 1H NMR (500 MHz, acetone-d6) δ 6.76 (4H, AA′XX′, J AX
) 8.72, J AA′ ) 2.45, ArH ortho to OH), 6.77 (4H, AA′XX′, J AX
) 8.65, J AA′ ) 2.45, ArH ortho to OH), 7.37 (4H, AA′XX′, J AX
) 8.62, J XX′ ) 2.32, ArH meta to OH), 8.38 (4H, AA′XX′, J AX
) 8.73, J XX′ ) 2.44, ArH meta to OH), 8.38 (2H, bs, OH), 8.53
(2H, bs, OH); 13C NMR (125 MHz, acetone-d6) δ 116.08(4),
113.12(4), 123.9(2), 124.2(2), 125.5(2), 127.7(4), 132.4(4), 147.7(2),
157.4(2), 157.6(2); MS (EI, 70 eV) m/z 436.2 (M+); HRMS calcd
for C28H20O5 436.131 07, found 436.131 24.
3-Eth yl-2,4-bis(4-h yd r oxyp h en yl)-5-p h en ylfu r a n (15f).
Furan 14f (20.0 mg, 0.05 mmol) was reacted according to the
general demethylation procedure using BF3‚SMe2 to afford
crude 15f. The crude material was purified by flash column
chromatography (1:1 hexane:EtOAc) followed by recrystalli-
zation from CH2Cl2:hexanes to give 15f as a solid (17.4 mg,
94% yield). mp 226-230 °C; 1H NMR (500 MHz, acetone-d6) δ
1.02 (3H, t, J ) 7.49, CH3CH2), 2.53 (2H, q, J ) 7.49, CH3CH2),
6.97 (2H, AA′XX′, J AX ) 8.74, J AA′ ) 2.60, ArH ortho to OH),
6.98 (2H, AA′XX′, J AX ) 8.70, J AA′ ) 2.45, ArH ortho to OH),
7.16 (1H, m, ArH para to furan), 7.19 (2H, AA′XX′, J AX ) 8.75,
J XX′ ) 2.48, ArH meta to OH), 7.24 (2H, m, ArH meta to furan),
7.46 (2H, m, ArH ortho to furan), 7.64 (2H, AA′XX′, J AX ) 8.66,
J XX′ ) 2.53, ArH meta to OH), 8.55 (2H, bs, OH); 13C NMR
(125 MHz, acetone-d6) δ14.9, 17.9, 116.5(2), 116.7(2), 124.2,
124.6, 125.6(2), 125.7, 126.5, 127.5, 127.9(2), 129.2(2), 132.1(2),
132.3, 147.0, 148.4, 157.8, 157.9; MS (EI, 70 eV) m/z 356.1
(M+); HRMS calcd for C24H20O3 356.141 25, found 328.141 29.
3-E t h yl-2,4,5-t r is(4-h yd oxyp h en yl)fu r a n (15b ). Furan
14b (28.0 mg, 0.07 mmol) was reacted according to the general
demethylation procedure using BF3‚SMe2 to afford crude 15b.
The crude material was purified by flash column chromatog-
raphy (1:1 hexane:EtOAc) and recrystallized from EtOAc:
hexanes to give 15b (16.3 mg, 93% yield): mp 219-220 °C;
1H NMR (500 MHz, acetone-d6) δ 1.01 (3H, t, J ) 7.52, CH3-
CH2), 2.51 (2H, q, J ) 7.52, CH3CH2), 6.72 (2H, AA′XX′, J AX
8.99, J AA′ ) 2.52, ArH ortho to OH), 6.95 (2H, AA′XX′, J AX
8.85, J AA′ ) 2.54, ArH ortho to OH), 6.96 (2H, AA′XX′, J AX
8.67, J AA′ ) 2.55, ArH ortho to OH), 7.17 (2H, AA′XX′, J AX
8.89, J XX′ ) 2.41, ArH meta to OH), 7.31 (2H, AA′XX′, J AX
9.01, J XX′ ) 2.47, ArH meta to OH), 7.60 (2H, AA′XX′, J AX
)
)
)
)
)
)
8.95, J XX′ ) 2.49, ArH meta to OH), 8.39 (1H, bs, OH), 8.46
(1H, bs, OH), 8.49 (1H, bs, OH); 13C NMR (125 MHz, acetone-
d6) δ 14.1, 17.1, 115.2(2), 115.6(2), 115.7(2), 123.2, 123.4, 123.5,
123.6, 125.1, 126.5(2), 126.8(2), 131.3(2), 146.5, 146.7, 156.5,
156.7, 156.9; MS (EI, 70 eV) m/z 372.2 (M+); HRMS calcd for
C
24H20O4 372.136 159, found 370.136 761. Anal. Calcd for
2,4-Bis(4-h ydr oxyph en yl)-5-ph en yl-3-pr opylfu r an (15g).
Furan 14g (30.0 mg, 0.075 mmol) was reacted according to
the general demethylation procedure using BF3‚SMe2 to afford
crude 15g. The crude material was purified by flash column
chromatography (2:1 hexane:EtOAc) and recrystallized from
EtOAc:hexanes to give 15g (23.2 mg, 86% yield): 1H NMR (500
MHz, acetone-d6) δ 0.80 (3H, t, J ) 7.39, CH3CH2), 1.44 (2H,
C24H20O4 (·0.5H2O): C, 74.43; H, 6.12. Found: C, 74.82; H, 5.86.
2,3,5-Tr is(4-h yd r oxyp h en yl)-4-p r op ylfu r a n (15c). Fu-
ran 14c (40.0 mg, 0.09 mmol) was reacted according to the
general demethylation procedure using BF3‚SMe2 to afford
crude 15c. The crude material was purified by flash column
chromatography (1:1 hexane:EtOAc) to provide 15c (32.0 mg,

3846 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mortensen et al.
m, CH3CH2CH2), 2.49 (2H, m, CH2-furan), 6.969 (2H, AA′XX′,
J AX ) 8.87, J AA′ ) 2.55, ArH ortho to OH), 6.972 (2H, AA′XX′,
J AX ) 8.55, J AA′ ) 2.41, ArH ortho to OH), 7.16 (1H, m, ArH
para to furan), 7.18 (2H, AA′XX′, J AX ) 8.78, J XX′ ) 2.44, ArH
meta to OH), 7.24 (2H, m, ArH meta to furan), 7.45 (2H, m,
ArH ortho to furan), 7.64 (2H, AA′XX′, J AX ) 8.68, J XX′ ) 2.62,
ArH meta to OH), 8.52 (1H, bs, OH), 8.57 (1H, bs, OH); 13C
NMR (125 MHz, acetone-d6) δ 14.1, 23.6, 26.6, 116.3(2),
116.5(2), 123.0, 124.0, 125.4(2), 125.5, 126.5, 127.3, 127.7(2),
128.9(2), 131.9(2), 132.1, 146.7, 148.5, 157.6, 157.7; MS (EI,
70 eV) m/z 370.2 (M+); HRMS calcd for C25H22O3 370.156 890,
found 370.156 61.
(10H, m, ArH), 7.8 (2H, d, J ) 7.87, ArH meta to OH); 13C
NMR (500 MHz, CDCl3) δ 14.6, 17.4, 115.8, 125.1, 125.2(2),
125.4(2), 125.8, 126.4, 126.8, 126.9, 128.3(2), 128.6(2), 131.1,
131.5(2), 131.6, 147.1, 147.2, 154.9; MS (EI, 70 EV) m/z 340.2
(M+); HRMS calcd for C24H20O2 340.146 33, found 340.146 49.
3-Eth yl-5-(4-h ydr oxyph en yl)-(2,4)-bisph en ylfu r an (15l).
Furan 14l (13.4 mg, 0.04 mmol) was reacted according to the
general demethylation procedure using BF3‚SMe2 to afford
crude 15l. The crude product was purified by flash column
chromatography (95:5 hexane:EtOAc) and recrystallized from
hexane:EtOAc to give furan 15l (10 mg, 78% yield) as a light
1
yellow powder: mp 133-136 °C; H NMR (400 MHz, CDCl3)
δ 1.06 (3H, t, J ) 7.35, CH3), 2.57 (2H, q, J ) 7.48, CH2), 4.65
(1H, s, OH), 6.69 (2H, d, J ) 8.47, ArH ortho to OH), 7.26-
7.46 (10H, m, ArH), 7.74 (2H, d, J ) 7.26, ArH meta to OH);
13C NMR (500 MHz, CDCl3) δ 14.6, 17.4, 115.2(2), 124.7, 124.3,
125.4(2), 125.5, 126.8, 127.0(2), 127.4, 128.5, 128.6(2), 128.8(2),
130.3(2), 131.7, 134.2, 146.6, 154.6; MS (EI, 70 eV) m/z 340.2
(M+); HRMS calcd for C24H20O2 340.146 33, found 340.146 59.
3-Eth yl-2,5-bis(4-h yd r oxyp h en yl)-4-p h en ylfu r a n (15h ).
Furan 14h (30.0 mg, 0.08 mmol) was reacted according to the
general demethylation procedure using BF3‚SMe2 to afford
crude 15h . The crude material was purified by flash column
chromatography (1:1 hexane:EtOAc) to provide 15h (25.8 mg,
93% yield) as a white powder: mp 127-132 °C (dec); 1H NMR
(500 MHz, acetone-d6) δ 1.00 (3H, t, J ) 7.47, CH3CH2), 2.52
(2H, q, J ) 7.50, CH3CH2), 6.72 (2H, AA′XX′, J AX ) 8.86, J AA′
) 2.48, ArH ortho to OH), 6.96 (2H, AA′XX′, J AX ) 8.75, J AA′
) 2.56, ArH ortho to OH), 7.27 (2H, AA′XX′, J AX ) 8.91, J XX′
) 2.47, ArH meta to OH), 7.36 (2H, m, ArH ortho to furan),
7.41 (1H, tt, J ) 7.51, 1.34, ArH para to furan), 7.48 (2H, m,
ArH meta to furan), 7.62 (2H, AA′XX′, J AX ) 8.84, J XX′ ) 2.46,
ArH meta to OH), 8.44 (1H, bs, OH), 8.54 (1H, bs, OH); 13C
NMR (125 MHz, acetone-d6) δ 14.9, 17.9, 116.1(2), 116.5(2),
123.8, 123.9, 124.0, 124.3, 127.5(2), 127.8(2), 128.3, 129.7(2),
131.1(2), 135.4, 147.6, 147.7, 157.5, 157.6; MS (EI, 70 eV) m/z
356.2 (M+); HRMS calcd for C24H20O3 356.141 25, found
356.141 68.
2,5-Bis(4-h yd r oxyp h en yl)-3-p h en ylfu r a n (15m ). Furan
14m (22.0 mg, 0.06 mmol) was reacted according to the general
demethylation procedure using BF3‚SMe2 to afford crude 15m .
The crude material was purified by flash column chromatog-
raphy (1:1 EtOAc:hexane) followed by recrystallization from
CH2Cl2:hexanes to give 15m as a solid (18.7 mg, 92% yield):
1
mp 167-170 °C; H NMR (500 MHz, acetone-d6) δ 6.82 (2H,
AA′XX′, J AX ) 8.77, J AA′ ) 2.32, ArH ortho to OH), 6.83 (1H,
s, furanH), 6.92 (2H, AA′XX′, J AX ) 8.90, J AA′ ) 2.52, ArH ortho
to OH), 7.31 (1H, m, ArH para to furan), 7.39 (2H, m, ArH
meta to furan), 7.43 (2H, AA′XX′, J AX ) 8.817, J XX′ ) 2.45,
ArH meta to OH), 7.46 (2H, m, ArH ortho to furan), 7.67 (2H,
AA′XX′, J AX ) 8.93, J XX′ ) 2.44, ArH meta to OH), 8.56 (2H,
bs, OH); 13C NMR (100 MHz, acetone-d6) δ 107.3, 115.6(2),
115.9(2), 122.78, 122.81, 123.1, 125.4(2), 127.2, 127.8(2),
128.6(2), 128.8(2), 134.9, 147.5, 152.5, 157.36, 158.38; MS
(FAB) m/z 328.2 (M+); HRMS calcd for C22H16O3 328.109 95,
found 328.109 80.
3-Eth yl-4,5-bis(4-h yd r oxyp h en yl)-2-p h en ylfu r a n (15i).
Furan 14i (40.0 mg, 0.10 mmol) was reacted according to the
general demethylation procedure using BF3‚SMe2 to afford
crude 15i. The crude material was purified by flash column
chromatography (1:1 hexane:EtOAc) and recrystallized from
EtOAc:hexanes to give 15i (31.3 mg, 84% yield): mp 204-
206 °C; 1H NMR (500 MHz, acetone-d6) δ 1.04 (3H, t, J ) 7.53,
CH3CH2), 2.57 (2H, q, J ) 7.50, CH3CH2), 6.75 (2H, AA′XX′,
J AX ) 9.14, J AA′ ) 2.46, ArH ortho to OH), 6.97 (2H, AA′XX′,
J AX ) 8.59, J AA′ ) 2.40, ArH ortho to OH), 7.18 (2H, AA′XX′,
J AX ) 8.79, J XX′ ) 2.32, ArH meta to OH), 7.30 (1H, m, ArH
para to furan), 7.34 (2H, AA′XX′, J AX ) 8.86, J XX′ ) 2.47, ArH
meta to OH), 7.47 (2H, m, ArH meta to furan), 7.77 (2H, m,
3,5-Bis(4-h yd r oxyp h en yl)-2-p h en ylfu r a n (15n ). Furan
14n (65.0 mg, 0.18 mmol) was reacted according to the general
demethylation procedure using BF3‚SMe2 to afford crude 15n .
The crude material was purified by flash column chromatog-
raphy (1:1 EtOAc:hexane) followed by recrystallization from
CH2Cl2:hexanes to give 15n as a solid (46.0 mg, 77% yield):
1
mp 160-163 °C; H NMR (400 MHz, acetone-d6) δ 6.79 (1H,
ArH ortho to furan), 8.52 (1H, bs, OH), 8.54 (1H, bs, OH); 13
C
s, furanH), 6.89 (2H, AA′XX′, J AX ) 8.65, J AA′ ) 2.47, ArH ortho
to OH), 6.93 (2H, AA′XX′, J AX ) 8.82, J AA′ ) 2.45, ArH ortho
to OH), 7.23 (1H, m, ArH para to furan), 7.30 (2H, AA′XX′,
J AX ) 8.67, J XX′ ) 2.53, ArH meta to OH), 7.32 (2H, m, ArH
meta to furan), 7.66 (2H, m, ArH ortho to furan), 7.69 (2H,
AA′XX′, J AX ) 8.81, J XX′ ) 2.45, ArH meta to OH), 8.58 (2H,
bs, OH); 13C NMR (100 MHz, acetone-d6) δ108.7, 116.5(2),
116.6(2), 123.4, 125.63, 126.3(2), 126.5(2), 128.0, 129.3(2),
130.7(2), 132.4, 147.0, 153.7, 157.9, 158.36 (2); MS (EI, 70 eV)
m/z 328.2 (M+); HRMS calcd for C22H16O3 328.109 95, found
328.109 55.
NMR (125 MHz, acetone-d6) δ 14.8, 18.0, 116.1(2), 116.6(2),
123.8, 124.4, 125.7, 125.9(2), 126.6, 127.5(2), 127.6, 129.6(2),
132.2(2), 132.7, 146.8, 148.5, 157.7, 157.8; MS (EI, 70 eV) m/z
356.2 (M+); MS (CI, 130 eV) m/z 357.2 (M+ + H); HRMS calcd
for C24H20O3 356.141 25, found 356.140 45.
3-Eth yl-2-(4-h ydr oxyph en yl)-(4,5)-bisph en ylfu r an (15j).
Furan 14j (40 mg, 0.11 mmol) was reacted according to the
general demethylation procedure using BF3‚SMe2 to afford
crude 15j. The crude material was purified by flash column
chromatography (75:25 hexane:EtOAc) and recrystallized from
hexane:EtOAc to give 15j (21 mg, 55% yield) as a white
crystalline solid: mp 159-160 °C; 1H NMR (400 MHz, CDCl3)
δ 1.04 (3H, t, J ) 7.57, CH3), 2.53 (2H, q, J ) 7.51, CH2), 4.76
(1H, s, OH), 6.93 (2H, d, J ) 8.34, ArH ortho to OH), 7.13-
7.47 (10H, m, ArH), 7.65 (2H, d, J ) 8.75, ArH meta to OH);
13C NMR (500 MHz, CDCl3) δ 14.6, 17.3, 115.6(2), 124.1, 124.8,
125.1(2), 126.7, 127.2(2), 127.4, 128.2(2), 128.8(2), 130.2(2),
131.1, 134.2, 146.6, 147.2, 154.6, 164.5; MS (EI, 70 EV) m/ z
340.2 (M+); HRMS calcd for C24H20O2 340.146 33, found
340.146 11.
Gen er a l P r oced u r e for Th iop h en es. The 1,4-dione and
Lawesson’s Reagent were stirred in CH2Cl2 at 40 °C until all
1,4-dione had been consumed, as shown by TLC (2-3 h). The
reaction mixture was then poured into a separatory funnel,
washed with H2O, 10% sodium bicarbonate, and saturated
NaCl, and dried over sodium sulfate. Solvent was removed
under reduced pressure, and the crude product was purified
by flash column chromatography and recrystallization to afford
cyclized thiophenes.
2,5-Bis(4-h yd r oxyp h en yl)-3-p h en ylt h iop h en e (17a ).
Thiophene 16a (22.6 mg, 0.06 mmol) was reacted according
to the general demethylation procedure using BBr3 to afford
crude 17a . The crude material was purified by flash column
chromatography (1:1 hexane:EtOAc) to give 17a as a solid (18.3
mg, 88% yield): mp 125-130 °C; 1H NMR (500 MHz, acetone-
d6) δ 6.78 (2H, AA′XX′, J AX ) 8.73, J AA′ ) 2.47, ArH ortho to
OH), 6.90 (2H, AA′XX′, J AX ) 8.78, J AA′ ) 2.60, ArH ortho to
OH), 7.14 (2H, AA′XX′, J AX ) 8.52, J XX′ ) 2.47, ArH meta to
OH), 7.26 (1H, m, ArH para to thiophene), 7.32 (4H, m, ArH
3-Eth yl-4-(4-h ydr oxyph en yl)-(2,5)-bisph en ylfu r an (15k).
Furan 14k (42 mg, 0.12 mmol) was reacted according to the
general demethylation procedure using BF3‚SMe2 to afford
crude 15k . The crude material was purified by flash column
chromatography (75:25 hexane:EtOAc) and recrystallized from
hexane:EtOAc to give 15k (28 mg, 70% yield) as a white
1
powder: mp 144-147 °C; H NMR (400 MHz, CDCl3) δ 1.11
(3H, t, J ) 7.45, CH3), 2.62 (2H, q, J ) 7.36, CH2), 4.92 (1H,
s OH), 6.97 (2H, d, J ) 8.21, ArH ortho to OH), 7.28-7.49

Ligands Selective for Estrogen Receptor R
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23 3847
meta and ortho to thiophene), 7.34 (1H, s, thiopheneH), 7.56
(2H, AA′XX′, J AX ) 8.79, J XX′ ) 2.52, ArH meta to OH), 8.54
(1H, bs, OH), 8.57 (1H, bs, OH); 13C NMR (125 MHz, acetone-
d6) δ 115.5(2), 115.8(2), 124.9, 125.5, 125.8, 126.7(2), 126.8,
128.3(2), 128.9(2), 130.3(2), 136.6, 136.9, 138.0, 141.9, 157.2,
dextran-coated charcoal as an adsorbant for free ligand.38
Purified ERR and ERâ binding affinities were determined
using a competitive radiometric binding assay using 10 nM
[3H]estradiol as tracer, commercially available ERR and ERâ
preparations (PanVera Inc. Madison, WI), and hydroxylapatite
(HAP) to adsorb bound receptor-ligand complex.39 HAP was
prepared following the recommendations of Williams and
Gorski.46 All incubations were done at 0 °C for 18-24 h.
Unlabeled competitors were prepared in 1:1 DMF:TEA to
ensure solubility. Binding affinities are expressed relative to
estradiol on a percent scale (i.e., for estradiol, RBA ) 100%).
All essays were run in separate, duplicate experiments, which
were reproducible with a coefficient of variation of less than
0.3.
Tr a n scr ip tion a l Activa tion Stu d ies. Transactivation by
ligands on ERR and ERâ was tested in transfected human
endometrial cancer (HEC-1) cells. HEC-1 cells, maintained in
MEM containing 5% CS and 5% FCS, were seeded into 24-
well plates in transfection media (IMEM containing 5% FCS
and transfected at about 50% confluency, using lipofectin-
transferrin. For each well, 1 µg of 4ERE-TATA-LUC, 2.5 µg of
pCMVâGal as internal control, and 50-100 ng of pCMV5-ERR
or pCMV5-ERâ were mixed with 5 µL of lipofectin (GIBCO,
BRL) and 1.6 µL of 1 mg/mL transferrin in 150 µL of HBSS.
The mixture was applied to the cell with 350 µL of serum-free
IMEM media for each well. The cell was incubated at 37 °C in
the 5% CO2 containing incubator for 6 h. Compounds were
prepared as solutions in ethanol and were added to medium
to give a final ethanol concentration of 0.1%. The cell culture
media was replaced by transfection media containing different
concentration of ligands. Cell were incubated for 24 h in the
presence of ligand. The luciferase reporter assay system
(Promega) was used for the luciferase activity assay. The
activity of E2 (10-8 M) on ERR or ERâ was set as 100%, and
the relative activity was adjusted on the basis of the trans-
fection efficiency, which was monitored by the internal control
â-galactosidase activity, as previously described.16,47
157.4; MS (EI, 70 eV) m/z 344.1 (M+); HRMS calcd for C22H16
-
SO2 344.087 10, found 344.086 21.
3,5-Bis(4-h yd r oxyp h en yl)-2-p h en ylt h iop h en e (17b ).
Thiophene 16b (108.0 mg, 0.29 mmol) was reacted according
to the general demethylation procedure using BF3‚SMe2 to
afford crude 17b. The crude material was purified by flash
column chromatography (1:1 hexane:EtOAc) followed by re-
crystallization from CH2Cl2:hexanes to give 17b as a solid (85.1
mg, 85% yield): mp 198-200 °C; 1H NMR (500 MHz, acetone-
d6) δ 6.79 (2H, AA′XX′, J AX ) 8.66, J AA′ ) 2.47, ArH ortho to
OH), 6.90 (2H, AA′XX′, J AX ) 8.80, J AA′ ) 2.60, ArH ortho to
OH), 7.17 (2H, AA′XX′, J AX ) 8.69, J XX′ ) 2.47, ArH meta to
OH), 7.24-7.34 (5H, m, ArH phenyl), 7.32 (1H, s, thiopheneH),
7.57 (2H, AA′XX′, J AX ) 8.72, J XX′ ) 2.52, ArH meta to OH),
8.50 (2H, bs, OH); 13C NMR (125 MHz, acetone-d6) δ 115.3(2),
115.8(2), 125.3, 125.7, 126.8(2), 127.2, 127.9, 128.5(2), 128.8(2),
130.1(2), 134.7, 134.9, 139.1, 142.6, 156.7, 157.5; MS (EI, 70
eV) m/z 344.1 (M+); HRMS calcd for C22H16SO2 344.087 10,
found 344.086 20.
Gen er a l P r oced u r e for N-Su bstitu ted P yr r oles. In
toluene, the 1,4-dione, amine, and p-toluenesulfonic acid
monohydrate were heated to reflux for 24 h, using a Dean-
Stark trap. The reaction was cooled and filtered and solvent
removed in vacuo. Crude product was purified by flash column
chromatography and recrystallization to afford N-substituted
pyrroles.
2-Eth yl-1,3-bis(4-h ydr oxyph en yl)-5-ph en ylpyr r ole (19).
Methyl ether protected pyrrole 18 (38.8 mg, 0.10 mmol) and
boron tribromide (0.6 mL, 0.6 mmol) were reacted as outlined
in the general procedure for demethylation to give 29.0 mg of
brownish solid. Trituration with hexane gave product as off-
white solid (21.8 mg, 60.8% yield): 1H NMR (500 MHz,
acetone-d6) δ 0.90 (3H, t, J ) 7.42, CH3), 2.62 (2H, q, J ) 7.42,
CH2), 6.43 (1, s, pyrroleH), 6.88 (2H, AA′XX′, J AX ) 8.84, J AA′
) 2.63, ArH ortho to OH), 6.90 (2H, AA′XX′, J AX ) 8.80, J AA′
) 2.76, ArH ortho to OH), 7.07 (1H, m, ArH para to pyrrole),
7.11 (2H, AA′XX′, J AX ) 8.70, J XX′ ) 2.73, ArH meta to OH),
7.15 (4H, m, ArH ortho and meta to pyrrole), 7.32 (2H, AA′XX′,
J AX ) 8.63, J XX′ ) 2.51, ArH meta to OH), 8.18 (1H, s, ArOH),
8.67 (1H, s, ArOH); MS (EI, 70 eV) m/z 355.2 (M+); HRMS
calcd for C24H21NO2 355.1572, found 355.1567.
Molecu la r Mod elin g. The protein structure used in the
docking simulations was based on the X-ray crystallographic
structure of the human estrogen receptor ligand binding
domain bound to estradiol (entry 1ere in the Protein Data
Bank). The crystal structure contains three homodimers with
244 residues each. For modeling purposes only one monomer
(chain A) was chosen. The monomer contains several residues
with missing atoms and five residues with alternate conforma-
tions. Missing atoms were added using InsightII, and one
conformation was selected for each of the five residues. Several
residues that are part of two loops were completely missing.
These two loops were modeled on the basis of the crystal
structure of the wild-type estrogen receptor ligand binding
domain complexed to estradiol (entry 1qku) using InsightII.
To eliminate bad contacts and constraints due to crystal
packing and loop reconstruction, the protein was energy
minimized for 1000 steps using the Powell algorithm in the
presence of strong harmonic constraints on the backbone,
followed by an additional 1000 steps without constraints.
Minimization was done with the program SYBYL 6.6 and the
MMFF94 force field. Furan 15b was docked into the minimized
receptor using the FlexiDock routine and the receptor-ligand
complex put through a minimization protocol as previously
described.24
Ack n ow led gm en t. We gratefully acknowledge Do-
rina Kosztin for providing the fully minimized ERR
structure utilized in the FlexiDock processes. We are
grateful for support of this research through grants from
the National Institutes of Health [PHS 5R37 DK15556
(J .A.K.) and 5R01 CA19118 (B.S.K.)] and the U.S. Army
Breast Cancer Research Program [DAMD17-97-1-7076
(J .A.K.)]. NMR spectra were obtained at the Varian
Oxford Instrument Center for Excellence in NMR
Laboratory. Funding for the instrumentation was pro-
vided in part from the W.M. Keck Foundation, the
National Institutes of Health (PHS 1 S10 RR104444-
01), and the National Science Foundation (NSF CHE
96-10502). Mass spectra were obtained on instruments
supported by grants form the National Institute of
General Medical Sciences (GM 27029), the National
Instituted of Health (RR 01575), and the National
Science Foundation (PCM 8121494).
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails for all precursors is available.This material is available
free of charge via the Internet at http/::pubs.acs.org.
Refer en ces
(1) Pollard, J . W. Modifiers of estrogen action. Sci. Med., 38-47.
(2) Cauley, J . A.; Stelley, D. G.; Ensrud, K.; Ettinger, B.; Black, D.;
Cummings, S. R. Estrogen replacement therapy and fractures
in older women. Ann. Intern. Med. 1995, 122, 9-16.
(3) Davidson, N. E. Hormone-replacement therapysBreast versus
heart versus bone. N. Engl. J . Med. 1995, 352, 1638-1639.
(4) J ordan, V. C. Designer estrogens. Sci. Am. 1998, 60-67.
(5) Barrett-Conner, E.; Bush, T. L. Estrogen and coronary heart
disease in women. J AMA 1991, 265, 1861-1867.
(6) Stampfer, M. J .; Colditz, G. A.; Willett, W. C. Postmenopausal
estrogen therapy and cardiovascular DiseasesTen year follow-
up from the nurses’ health study. N. Engl. J . Med. 1991, 325,
756-762.
Biologica l P r oced u r es. Rela tive Bin d in g Affin ities.
Ligand binding affinities (RBAs) using lamb uterine cytosol
as a receptor source were determined by a competitive radio-
metric binding assay using 10 nM [3H]estradiol as tracer and

3848 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 23
Mortensen et al.
(7) Tuck, C. H.; Holleran, S.; Berglund, L. Hormonal regulation of
lipoprotein(a) levels: effects of estrogen replacement therapy on
lipoprotein(a) and actute phase reactants in postmenopausal
women. Arterioscler. Thromb. Vasc. Biol. 1997, 17, 1822-1829.
(8) Sherwin, B. S. Estrogen effects on cognition in menopausal
women. Neurology 1997, 48, S21-S26.
(9) Rice, M. M.; Graves, A. B.; McCurry, S. M.; Larson, E. B.
Estrogen replacement therapy and cognitive function in post-
menopausal women without dementia. Am. J . Med. 1997, 103,
26S-35S.
(10) Kuiper, G. G. J . M.; Enmark, E.; Pelto-Huikko, M.; Nilsson, S.;
Gustafsson, J .-A° . Cloning of a novel receptor expressed in rat
prostate and ovary. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 5925-
5930.
(11) Mosselman, S.; Polman, J .; Dijkema, R. ERâ: Identification and
charaterization of a novel human estrogen receptor. FEBS Lett.
1996, 392, 49-53.
(12) Kuiper, G. G. J . M.; Carlsson, B.; Grandien, K.; Enmark, E.;
Haggbald, J .; et al. Comparison of the ligand binding specificity
and transcipt tissue distribution of estrogen receptors R and â.
Endocrinology 1997, 138, 863-870.
(13) Barkhem, T.; Carlsson, B.; Nilsson, Y.; Enmark, E.; Gustafsson,
J .-A° .; et al. Differential Response of Estrogen Receptor R and
Estrogen Receptor â to Partial Estrogen Agonists/Antagonists.
Mol. Pharmacol. 1998, 54, 105-112.
(14) Makela, S.; Savolainen, H.; Aavik, E.; Myllarnieni, M.; Strauss,
L.; et al. Differentiation between vasculoprotective and utero-
trophic effects of ligands with different binding affinities to
estrogen receptors R and â. Proc. Natl. Acad. Sci. USA 1999,
96, 7077-7082.
(15) Paech, K.; Webb, P.; Kuiper, G. G. J . M.; Nilsson, S.; Gustafsson,
J .-A° .; et al. Differential ligand activation of estrogen receptors
ERR and ERâ at AP1 sites. Science 1997, 277, 1508-1510.
(16) Sun, J .; Meyers, M. J .; Fink, B. E.; Rajendran, R.; Katzenellen-
bogen, J . A.; et al. Novel ligands that function as selective
estrogens or antiestrogens for estrogen receptor-R or estrogen
receptor-â. Endocrinology 1999, 140, 800-804.
(17) Stauffer, S. R.; Katzenellenbogen, J . A. Acyclic Amides as
Estrogen Receptor Ligands: Synthesis, Binding, Activity, and
Receptor Interaction. Bio. Med. Chem. 2000, 8, 1293-1316.
(18) 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.
(19) Pike, A. C. W.; Brzozowski, A. M.; Hubbard, R. E.; Bonn, T.;
Thorsell, A.-G.; et al. Structure of the ligand-binding domian of
oestrogen receptor beta in the presence of a partial agonist and
a full antagonist. EMBO J . 1999, 18, 4608-4618.
(20) Fink, B. E.; Mortensen, D. S.; Stauffer, S. R.; Aron, Z. D.;
Katzenellenbogen, J . A. Novel structural templates for estrogen-
receptor ligands and prospects for combinatorial synthesis of
estrogens. Chem. Biol. 1999, 6, 205-219.
(21) Stauffer, S. R. Development of combinatorial approaches towards
selective estrogen receptor modulators: Investigations of acyclic
amides and tetra-substituted pyazoles. In Department of Chem-
istry; University of Illinios: UrbanasChampaign, 1999; p 208.
(22) Stauffer, S. R.; Huang, Y.; Coletta, C. J .; Tedesco, R.; Katzenel-
lenbogen, J . A. Estrogen pyrazoles: Defining the pyrazole core
structure and the orientation of substituents in the ligand
binding pocket of the estrogen receptor. Bio. Med. Chem. 2001,
9, 141-150.
model for triarylethylenes and triarypropenones. J . Med. Chem.
1989, 32, 1700-1707.
(27) Dikshit, D. K.; Singh, S.; Singh, M. M.; Kamboj, V. P. Synthesis
and biological activity of 2,3- and 3,4-diarylfurans and 2,3,4-
triaryl-2,5-dihydrofurans. Ind. J . Chem. 1990, 29B, 954-960.
(28) Dikshit, D. K.; Munshi, K. L.; Kapil, R. S.; Kamboj, V. P.; Anand,
N. Antifertility agent: Part IXsSynthesis of 2,3,4-triphenyl-
furans & 2,3,4-triphenyl-5-methylfurans. Ind. J . Chem. 1974,
12, 1144-1146.
(29) Iyer, R. N.; Gopalachari, R. S. Studies in Potential Antifertility
Agents: Part VII-Synthesis of Dialkylaminoethyl Derivatives
of 2,4,5-Triphenylpyrroles
& 2,4,5-Triphenylfurans. Ind. J .
Chem. 1973, 11, 1260-1262.
(30) Stetter, H. Catalyzed addition of aldehydes to activated double
bondssa new synthetic approach. Angew. Chem., Int. Ed. Engl.
1976, 15, 639-647.
(31) Thompson, L. A.; Ellman, J . A. Synthesis and Application of
Small Molecule Libraries. Chem. Rev. 1996, 96, 555-600.
(32) Shridhar, D. R.; J ogibhukta, M.; Shanthan Rao, P.; Handa, V.
K. Reaction of 2,4-Bis(4-methoxyphenyl)-1,3,2,4-dithiadiphos-
phetane-2,4-disulphide with substituted butane-1,4-diones: Part
IIsNovel synthesis of substituted thiophenes & furans. Ind. J .
Chem. 1983, 22B, 1187-1190.
(33) Wynberg, H.; Wiersum, U. E. Rearrangements and trans-
alkylation of t-butylthiophenes. J . Org. Chem. 1965, 30, 1058-
1060.
(34) Champaigne, E.; Foye, W. O. The synthesis of 2,5-diaryl-
thiophenes. J . Org. Chem. 1952, 17, 1405-1412.
(35) Wynberg, H.; Metselaar, J . A convenient route to polythiophenes.
Synth. Commun. 1984, 14, 1-9.
(36) Ramasseul, R.; Rassat, A. Preparation d’heterocycles encom-
bres: di-tert-butyl-2,5-furanne, pyrrole et thiophene. Bull. Soc.
Chim. (Fr.) 1965, 3136-3140.
(37) Chakrabarty, M. J . Chem. Soc. 1940, 1385.
(38) Katzenellenbogen, J . A.; J ohnson, H. J .; Myers, H. N. Photo-
affinity labels for estrogen binding proteins of rat uterus.
Biochemistry 1973, 12, 4085-4092.
(39) Carlson, K. E.; Choi, I.; Gee, A.; Katzenellenbogen, B. S.;
Katzenellenboge, J . A. Altered ligand binding properties and
enhanced stability of a constitutively active estrogen receptor:
Evidence that an open pocket conformation is required for ligand
interaction. Biochemistry 1997, 36, 14897-14905.
(40) 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.
(41) Gao, H.; Katzenellenbogen, J . A.; Garg, R.; Hansch, C. Compara-
tive QSAR Analysis of Estrogen Receptor Ligands. Chem. Rev.
1999, 99, 723-744.
(42) Anstead, G. M.; Wilson, S. R.; Katzenellenbogen, J . A. 2-Aryl-
indenes and 2-arylindenones: Molecular structures and con-
siderations in the binding orientation of unsymmetrical non-
steroidal ligands to the estrogen receptor. J Med Chem 1989,
32, 2163-2171.
(43) Mortensen, D. S.; Rodriguez, A. L.; Sun, J .; Katzenellenbogen,
B. S.; Katzenellenbogen, J . A. Furans with basic side chains:
Synthesis and biological evaluation of a novel series of antago-
nists with selectivity for the estrogen receptor alpha. Bio. Med.
Chem. Lett. 2001, to be submitted.
(44) Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard, R. E.;
Bonn, T.; et al. Molecular basis of agonism and antagonism in
the oestrogen receptor. Nature 1997, 389, 753-758.
(45) Shiau, A. K.; Barstad, D.; Loria, P. M.; Cheng, L.; Kushner, P.
J .; et al. The structural basis of estrogen receptor/coactivator
recognition and the antagonism of this interaction by tamoxifen.
Cell 1998, 95, 927-937.
(23) Stauffer, S. R.; Katzenellenbogen, J . A. Solid-phase synthesis
of tetrasubstituted pyrazoles, novel ligands for the estrogen
receptor. J . Comb. Chem. 2000, 2, 318-329.
(24) Stauffer, S. R.; Coletta, C. J .; Tedesco, R.; Sun, J .; Katzenellen-
bogen, B. S.; et al. Pyrazole Ligands: Structure-Affinity/Activity
Relationships Of Estrogen Receptor-R Selective Agonists. J . Med.
Chem. 2000, 43, 4934-4947.
(46) Williams, D.; Gorski, J . Biochemistry 1974, 13.
(47) McInerney, E. M.; Tsai, M. J .; O’Malley, B. W.; Katzenellen-
bogen, B. S. Analysis of estrogen receptor transcriptional
enhancement bya a nuclear hormone receptor coactivator. Proc.
Natl. Acad. Sci. U.S.A. 1996, 93, 10069-10073.
(25) Stauffer, S. R.; Huang, Y. R.; Aron, Z. D.; Coletta, C. J .; Sun, J .;
et al. Triarlpyrazoles with Basic Side Chains: Development of
Pyrazole-Based Estrogen Receptor Antagonists. Bioorg. Med.
Chem. 2001, 9, 151-161.
(26) Durani, N.; J ain, R.; Saeed, A.; Dikshit, D. K.; Durani, S.; et al.
Structure-activity relationship of antiestrogens: a study using
triarylbutenone, benzofuran, and triarylfuran analogues as a
J M010211U