The discovery of tetrahydrofluorenones as a new class of estrogen receptor β-subtype selective ligands

Article abstract of DOI:10.1016/j.bmcl.2006.03.098

Synthesis and derivatization of a series of substituted tetrahydrofluorenone analogs giving potent, ERβ subtype selective ligands are described. Several analogs possessing ERβ binding affinities comparable to 17β-estradiol but with greater than 75-fold selectivity over ERα are reported.

Full text of DOI:10.1016/j.bmcl.2006.03.098

Bioorganic & Medicinal Chemistry Letters 16 (2006) 3489–3494
The discovery of tetrahydrofluorenones as a new class of
estrogen receptor b-subtype selective ligands
R. R. Wilkening,a,* R. W. Ratcliffe,a E. C. Tynebor,a K. J. Wildonger,a A. K. Fried,a
M. L. Hammond,a R. T. Mosley,a P. M. D. Fitzgerald,a N. Sharma,a B. M. McKeever,a
S. Nilsson,c M. Carlquist,c A. Thorsell,c L. Locco,b R. Katz,b K. Frisch,b E. T. Birzin,b
H. A. Wilkinson,b S. Mitra,b S. Cai,b E. C. Hayes,b J. M. Schaefferb and S. P. Rohrerb
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
bDepartment of Molecular Endocrinology, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
cKaro Bio AB, Novum, S-14157 Huddinge, Sweden
Received 6 March 2006; revised 29 March 2006; accepted 30 March 2006
Available online 2 May 2006
Abstract—Synthesis and derivatization of a series of substituted tetrahydrofluorenone analogs giving potent, ERb subtype selective
ligands are described. Several analogs possessing ERb binding affinities comparable to 17b-estradiol but with greater than 75-fold
selectivity over ERa are reported.
Ó 2006 Elsevier Ltd. All rights reserved.
The estrogen receptor ER is comprised of two sub-
types, ERa1,2 and ERb,3,4 which bind 17b-estradiol
with similar affinity and mediate the effects of estrogen
throughout the body. The effects of the non-selective li-
gand 17b-estradiol have been studied extensively dur-
ing hormone replacement therapy (HRT) and have
produced unexpected and controversial results. The
beneficial effects of HRT include the prevention of
osteoporosis, amelioration of hot flashes,5 and reduc-
tion in the risk of colorectal cancer in menopausal
women. The adverse effects, as observed in the womens
health initiative6 (WHI), include a slight elevation in
the incidence of invasive breast cancer, as well as an in-
crease in the risk of coronary heart disease, pulmonary
embolism, and stroke in menopausal women. The clin-
ical usefulness of HRT, and the safety concerns raised
by the WHI, provides a compelling case to delineate
the physiological functions mediated by these two ER
receptors.
while ERa is predominant in the uterus and breast. This
observation suggests that the biological roles of ERa
and ERb receptors might be tissue specific and that an
ER subtype selective ligand might produce a biological
response which is different than the non-selective ligand
17b-estradiol.
These findings prompted an intensive search in both
academia and the pharmaceutical industry for ER sub-
type selective agonists and/or antagonists, to help eluci-
date the pharmacological importance of these receptors.
The recent identification of selective ERa7,9,10 and
ERb11–18 ligands has indeed provided some insight into
the physiological role of the ER receptors. In this paper,
we describe the synthesis and structure–activity relation-
ship (SAR) of a series of tetrahydrofluorenone deriva-
tives leading to the identification of potent ERb
subtype selective analogs.
The tetrahydrofluorenone derivative 419 was identified
as an ERb-selective ligand during a high-throughput
Studies on the tissue distribution of the two estrogen
receptors show that they are widely and to a large extent
differentially expressed in humans.7,8 For instance, ERb
is largely expressed in the lung, prostate, and brain,
OH
O
4
H
9a
H
H
Keywords: Estrogen receptor ligands; Tetrahydrofluorenone.
*
HO
HO
Corresponding author. Tel.: +1 732 594 5729; fax: +1 732 594
9556; e-mail: robert_wilkening@merck.com
17β-Estradiol
4
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.03.098

3490
R. R. Wilkening et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3489–3494
screening of Merck’s sample collection. The ERb
binding (IC50 218 nM) and b-selectivity (12-fold) of lead
compound 4 prompted an investigation into the SAR of
this class of compounds.
ketone and cyclization of the addition product under
strongly acidic conditions provided analogs 8–15 and
20–22 after boron tribromide deprotection of the methyl
ether.
The syntheses of the racemic tetrahydrofluorenone ana-
logs 1–32 began with the preparation of the intermediate
2-substituted indanones 34, as described in Scheme 1.
Reaction of 33 with an alkyl halide and base provided
a low yield of 34 owing to the propensity of 33 to under-
go dialkylation. Consequently, a one-pot reductive
alkylation20 of indanone 33 with an aldehyde21 was
developed. This procedure provided 34 in high yield
without evidence of dialkylated products. In a third
method,19 the indanone 34 was prepared by Friedel–
Crafts acylation of 35 with acid chloride 36, to provide
the para-substituted anisole 37. Reaction of 37 with
formaldehyde and cyclization of the aldol products in
sulfuric acid gave 34.
The synthesis of analog 1, where R1 is hydrogen, was
complicated by the failure of the unsubstituted indanone
33 to alkylate with MVK. Instead, analog 1 was pre-
pared as described in Scheme 3. Reductive alkylation
of the indanone 33 with aldehyde22 39 provided 40.
Acidic hydrolysis of the ketal group gave the ketone
41. Cyclization of 41 under basic conditions was fol-
lowed by an aluminum chloride/ethanethiol deblock of
the methyl ether to give 1.
A variety of R2 substituents were introduced as
described in Scheme 4. Bromination of 38 provided
the bromo enone intermediate 42. Palladium-catalyzed
reactions of 42 provided the 2-furyl (23 and 31), 2-thie-
nyl 24, and phenyl 25 analogs after deblocking of the
methyl ether. Conversion of 42 into the cyano (26 and
32) and trifluoromethyl23 27 analogs was accomplished
using copper(I)-mediated chemistry. The halogenated
products 16–19 and 28–30 were also prepared as de-
scribed below.
Scheme 2 describes the formation of the racemic tetrahy-
drofluorenone platform using the Robinson annulation
reaction. Reaction of 34 with methyl vinyl ketone
(MVK) and cyclization of the addition product with
pyrrolidine/acetic acid gave 38. Deprotection of the
methyl ether with boron tribromide provided analogs
2–7. Reaction of 34 with the appropriate alkyl vinyl
O
O
O
O
O
O
CHO
O
O
39
a or b
a
R1
MeO
MeO
33
40
MeO
MeO
33
34
d,e
b
O
H
O
O
O
c, d
R1
c
MeO
Cl
H
R1
HO
+
1
41
MeO
MeO
O
Scheme 3. Reagents: (a) 39, H2, KOH 10% Pd/C, EtOH; (b) 6 N HCl,
THF; (c) NaOMe, MeOH; (d) AlCl3, EtSH, CH2Cl2.
35
36
37
Scheme 1. Reagents: (a) NaH, R1I, DMF, R1 is Et, i-Pr, Bu, pentyl;
(b) H2, RCHO, KOH or NaOH, 10% Pd/C, EtOH, R1 is Pr, Bu, i-Bu,
Bn, 2-hydroxyethyl; (c) AlCl3, CH2Cl2, R1 is Me, Ph; (d) HCHO,
K2CO3, MeOH; (e) H2SO4.
O
O
H
Br
a
c
O
H
R1
R1
MeO
MeO
O
38
42
a,b
R1
b,c
R1
f,c
or
MeO
MeO
d,e
34
d,e,c
38
R2
R2
O
O
c
O
O
R2
H
R1
16-19, 28-30
R1
23-27, 31, 32
HO
HO
R1
8-15, 20-22
R1
HO
HO
Scheme 4. Reagents: (a) Br2, NaHCO3, CH2Cl2; (b) 2-(Bu3Sn)furan,
PdCl2(PPh3)2, toluene; or 2-(Bu3Sn)thiophene, PdCl2(PPh3)2, toluene;
or C6H5B(OH)2, PdCl2(PPh3)2, Cs2CO3, DMF; (c) BBr3, CH2Cl2; (d)
CuCN, NMP; or FSO2CF2CO2CH3, CuI, DMF; (e) C5H5NÆHCl; (f)
NCS, DMF; or I2, NaHCO3, CH2Cl2.
2-7
Scheme 2. Reagents: (a) MVK, DBU, THF; (b) pyrrolidine, HOAc,
toluene; (c) BBr3, CH2Cl2; (d) R2CH2C(O)CH@CH2, DBU, THF, R2
is Me, Et, Pr, Bu; (e) HOAc, 6 N HCl.

R. R. Wilkening et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3489–3494
3491
The estrogen receptor binding affinities24 of the racemic
tetrahydrofluorenone analogs 1–32 are summarized in
Table 1. Our initial effort to evaluate the SAR of the tet-
rahydrofluorenone series focused on the R1 substitution
of lead compound 4. We found that analogs 1–7, where
R2 is hydrogen, displayed the most favorable ERb bind-
ing when R1 is an ethyl, propyl or butyl group. Smaller
groups, that is, hydrogen and methyl, and larger groups,
that is, pentyl and benzyl, had deleterious effects upon
ERb binding and selectivity. While the ethyl, propyl,
and butyl analogs possessed similar ERb binding, the
ethyl and butyl analogs displayed increased b-selectivity
relative to the propyl lead compound 4.
binding. Branching of the R1 group close to the
bridgehead, analog 11, was not well tolerated, while
substitution further along the alkyl chain was better
tolerated, analog 13. These branched analogs, howev-
er, displayed modest ERb-selectivity when compared
with their linear propyl 10 and butyl 12 counterparts.
Finally, polar groups were observed to greatly reduce
ER binding, as was seen with the 2-hydroxyethyl ana-
log 14.
In a series of analogs where R2 is bromo 16–19, we ob-
served a 50- to 100-fold improvement in ERb binding
relative to the proteo analogs 2–5. Analog 19 was found
to have the most favorable ERb binding affinity
(1.8 nM) and selectivity (76-fold). This analog showed
ERb potency comparable to 17b-estradiol, but with
greatly improved b-selectivity.
Switching the R2 substituent from hydrogen to a
methyl group and examining the R1 group in more
detail produced analogs with greater than 10-fold
improvement in ERb binding. A comparison of deriv-
atives 8–15 showed that the ethyl, propyl, butyl, and
phenyl groups possessed the most favorable ER
It was apparent that variation of the R2 group had a
more profound effect upon improving ER binding than
Table 1. Human estrogen receptor affinity and selectivity
R2
O
R1
HO
Compound
R1
R2
IC50 (nM)
ERa/b
ERa
ERb
17b-Estradiol
1
—
—
H
1.3
1.2
1530
936
374
218
188
945
4220
63
1.1
>7
5
H
Me
>10,000
4245
>10,000
2530
6450
3740
>10,000
1190
1210
455
2
H
3
Et
Pr
H
>27
12
34
4
4
H
5
Bu
Pentyl
H
6
H
7
Benzyl
Me
H
>2
19
43
26
12
39
10
—
32
66
71
18
76
20
20
18
63
45
34
43
85
22
95
45
24
28
8
Me
Me
Me
Me
Me
Me
Me
Me
Br
9
Et
Pr
28
11
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
i-Pr
Bu
1420
630
123
16
i-Bu
2-Hydroxyethyl
195
20
>10,000
15
>10,000
481
883
Phenyl
Me
Et
13.4
4.5
4.4
1.8
13
Br
Br
319
79
Pr
Bu
Br
Et
141
264
Bu
Bu
Pr
Bu
30
273
1.5
15
Bu
Bu
2-Furyl
2-Thienyl
Phenyl
CN
CF3
Cl
196
3.1
1.6
1.2
38
Bu
Bu
70.5
41
Bu
Bu
1650
128
1.5
5.3
1.3
2.5
16
Bu
Bu
114
124
I
Br
Phenyl
Phenyl
Phenyl
113
389
2-Furyl
CN
880
32

3492
R. R. Wilkening et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3489–3494
did variation of the R1 group. Consequently, a more
extensive evaluation of R2 substitution was initiated in
a series where R1 was fixed as either a butyl or phenyl
group. In the series where R1 was butyl, a comparison
of the R2-substituted analogs revealed that the methyl
12, ethyl 20, and butyl 22 analogs possessed similar
ERb binding. The propyl analog 21 possessed the most
potent ERb binding (1.5 nM), while the methyl analog
12 possessed the best ERb-selectivity (39-fold). The cya-
no analog 26 showed a marginal improvement in ERb-
binding potency (5-fold) relative to the proteo analog
5. In contrast, the heteroaryl and phenyl analogs 23–
25, and halogenated analogs 19 and 27–29 showed an
approximately 60- to 160-fold and 35- to 145-fold
improvement in ER binding when compared with the
proteo analog 5.
substituted analogs 12 and 20–22, an increase in chain
length resulted in a decrease in the ERb agonist re-
sponse. This trend was also seen with the heteroaryl ana-
logs 23–25, where an increase in steric size produced a
decrease in agonist response. In the three analogs 22,
24, and 25, where the response decreased to approxi-
mately 50% or less, the ability of these analogs to antag-
onize estradiol’s effects was tested. The butyl 22, thienyl
24, and phenyl 25 analogs were determined to effectively
antagonize the transcriptional activation of estrogen in
HEK 293 cells. The similar EC50 and IC50 of these ana-
logs identified them as mixed agonist/antagonists.
Chiral HPLC resolution26 of the racemic analogs 12 and
19 provided the S-43, -45 and R-44, -46 enantiomers,
respectively. The ER binding values for the resolved
enantiomers are listed in Table 3. It is apparent that
essentially all the binding affinity of the racemates is
derived from the S-enantiomers. The R-enantiomers
are much less active with their binding values compris-
ing 1% or less of the racemates’ binding affinity.
In the 9a-phenyl series, the bromo derivative 30 was
found to be the most potent and b-selective analog.
However, this analog exhibited slightly decreased bind-
ing affinity and b-selectivity compared with the butyl
analog 19. This trend was also seen with the furan deriv-
ative 31 which showed a significant decrease in binding
and selectivity from the related butyl analog 23. The
cyano derivative 32, however, exhibited similar potency
and selectivity to the corresponding butyl analog 26.
X-ray analysis27 of the estrogen b-subtype receptor
complexed with 45 provided evidence for the configura-
tional assignments of the R1 position and suggested an
explanation for the observed ERb-selectivity. The selec-
tivity results from a number of subtle factors that in-
clude the planar nature of the tetrahydrofluorenone
core and the positioning of an R1 substituent orthogonal
to that plane. The first factor is a putative stabilizing
interaction between the planar/aromatic surface of the
tricyclic core and Met336 of hERb which is not possible
with the analogous Leu384 in hERa (see Fig. 1). The Sd
and Ce of Met336 make better van der Waals contacts
with the mostly planar tricyclic platform than is possible
with either Cd of Leu384 in hERa. This feature appears
to be common to other planar ERb-selective molecules
including some phytoestrogens such as genistein28 or
other synthetic molecules including some benzisoxaz-
oles,15 for example. The second selectivity enhancing
interaction is the putative favorable hydrophobic inter-
action between the n-butyl group at R1 which protrudes
orthogonally from the plane of the tricyclic core toward
Ile373 in hERb as depicted in Figure 1. We speculate
that Ile373 in hERb can nicely accommodate the pres-
In addition to the ER binding affinity assay, com-
pounds were evaluated in a cell-based transactivation
assay25 utilizing HEK 293 cells which were stably
cotransfected with human ERb and the alkaline
phosphatase reporter gene. The transcriptional activi-
ty of a series of R2-substituted tetrahydrofluorenone
analogs was determined and compared as a percent
response of 17b-estradiol. These data are summarized
in Table 2. Generally, the potency of these deriva-
tives tracked well with the ER binding values listed
in Table 1 and demonstrated that these analogs be-
haved as functional ERb agonists.
The percent agonist response for the butyl-substituted
analogs varied between 11% and 112%. In the alkyl-
Table 2. Transactivation in ERb cotransfected HEK 293 cells
R2
O
Table 3. Human estrogen receptor affinity and selectivity
Bu
HO
R2
R2
O
O
Compound
R2
EC50
(nM)
% Estradiol
agonism
IC50
(nM)
9a
9a
12
20
21
22
23
24
25
26
27
28
19
29
Me
Et
20
2
112
66
64
11
80
52
41
93
81
78
81
70
—
—
—
40
—
9
Bu
Bu
HO
HO
(S)
9a
(R)
Pr
Bu
4
Compound
R2
IC50 (nM)
ERa ERb
ERa/b
selectivity
30
5
2-Furyl
2-Thienyl
Phenyl
CN
4
12
43
44
Me
Me
Me
R/S
S
630
567
16
19
39
30
>5
5
2
5
—
—
—
—
—
R
>10,000
1810
CF3
Cl
40
11
4
19
45
46
Br
Br
Br
R/S
S
141
129
1.8
1.5
313
76
89
13
Br
I
R
4023
2

R. R. Wilkening et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3489–3494
3493
appears to fill the same space occupied by the aryl group
found in the antagonist sidechain of raloxifene. It could
readily be conjectured then that as the groups replacing
Br get larger that the antagonism should increase as
well.
In summary, starting from lead compound 4, tetrahydro-
fluorenone analogs with low nanomolar affinity for ERb
and greater than 75-fold selectivity over ERa were identi-
fied. Furthermore, compounds of this class were deter-
mined to be functional agonists of ERb in a cell-based
transactivation assay. The new ERb agonists described
herein, particularly 19, should prove to be valuable phar-
macological tools for elucidating the physiological role of
ERb. Further investigation of the tetrahydrofluorenone
class of ERb agonists will be the subject of forthcoming
publications from these laboratories.
Acknowledgments
Figure 1. Superposition of the crystallographic complexes of com-
pound 45 (cyan) with hERb (purple) and 17b-estradiol (yellow) with
hERa (green) (pdb entry: 1ERE) displaying the principal amino acids
involved in ligand recognition. Unless otherwise indicated, residue
numbering is that of hERb.
I thank Drs. Mark Greenlee and Timothy Blizzard for
their many useful suggestions during the preparation
of the manuscript.
References and notes
1. Green, S.; Walter, P.; Kumar, V.; Krust, A.; Bornert, J.
M.; Argos, P.; Chambon, P. Nature 1986, 320, 134.
2. Green, S.; Gilna, P.; Waterfield, M.; Baker, A.; Hort, Y.;
Shine, J. Science 1986, 231, 1150.
3. Kuiper, G. G. J. M.; Gustafuson, J.-A. Proc. Natl. Acad.
Sci. U.S.A. 1996, 93, 5925.
4. Mosselman, S.; Polman, J.; Dijkema, R. FEBS Lett. 1996,
393, 49.
5. Stearns, V.; Ullmer, L.; Lopez, J.; Smith, Y.; Isaacs, C.;
Hayes, D. Lancet 2002, 360, 1851.
6. Women’s Health Initiative. JAMA 2002, 288, 321.
7. Harris, H. A.; Katzenellenbogen, J. A.; Katzenellenbogen,
B. S. Endocrinology 2002, 143, 4172.
8. Enmark, E.; Pelto-Huikko, M.; Grandien, K.; Lagerc-
rantz, S.; Lagercrantz, J.; Fried, G.; Nordenskjold, M.;
Gustafsson, J.-A. J. Clin. Endocrinol. Metab. 1997, 82,
4258.
9. 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. D.; Rohrer, S. A.; Hammond, M. L. J. Med. Chem.
2004, 47, 2171.
10. Kim, S.; Wu, J. Y.; Chen, H. Y.; Birzin, E. T.; Chan, W.;
Yang, Y. T.; Colwell, L.; Li, S.; Dahllund, J.; DiNinno, F.
D.; Rohrer, S. A.; Schaeffer, J. M.; Hammond, M. L.
Bioorg. Med. Chem. Lett. 2004, 14, 2741.
Figure 2. Superposition of the crystallographic complex of compound
45 (cyan) with hERb (purple) and raloxifene (in white, extracted from
pdb entry 1ERR) in the context of hERa (green) (pdb entry: 1ERE)
displaying the principal amino acids involved in ligand recognition.
Unless otherwise indicated, residue numbering is that of hERb.
ence of the n-butyl moiety into space which is not avail-
able in hERa because the sidechain of the analogous
Met421 fills this space. The residue differences in this
region have been used to rationalize some of the
hERb-selectivity exhibited by certain 7-substituted 2-
phenyl-benzofurans16 and 1,3,5-substituted triazines.13
11. Meyers, M. J.; Sun, J.; Carlson, K. E.; Marriner, G. A.;
Katzenellenbogen, B. S.; Katzenellenbogen, J. A. J. Med.
Chem. 2001, 44, 4230.
12. Schopfer, U.; Schoeftter, P.; Bischoff, S. F.; Nozulak, J.;
Feuerbach, D.; Floersheim, P. J. Med. Chem. 2002, 45,
1399.
13. 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. J. Med. Chem. 2002, 45, 5492.
The loss of agonism coupled to a concomitant increase
in the steric bulk of R2 is reasonable in light of the
superposition of raloxifene with compound 45 depicted
in Figure 2. Clearly, the bromine of compound 45

3494
R. R. Wilkening et al. / Bioorg. Med. Chem. Lett. 16 (2006) 3489–3494
14. 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. Bioorg. Med. Chem. Lett. 2004, 14,
2327.
15. Malamas, M. S.; Manas, E. S.; McDevitt, R. E.; Guna-
wan, I.; Xu, Z. B.; Collini, M. D.; Miller, C. P.; Dinh, T.;
Henderson, R. A.; Keith, J. C.; Harris, H. A. J. Med.
Chem. 2004, 47, 5021.
experience, this assay provides IC50 values that are
reproducible to within factor of 2–3. With the
exception of 17b-estradiol and analog 19, the majority
of analogs were tested once in the ER binding assay. The
analogs were tested in duplicate in the transactivation
assay (Table 2).
a
25. Barkhem, T.; Carlsson, B.; Nilsson, Y.; Enmark, E.;
Gustafsson, J.-A.; Nilsson, S. Mol. Pharmacol. 1998, 54,
105.
26. The chiral resolution of the methyl ether precursor of 19
was typical of the tetrahydrofluorenone series. Chiralcel
OJ column (Daicel) 10% ethanol/heptane, retention time
10.4 and 11.7 min.
16. Collini, M. D.; Kaufman, D. H.; Manas, E. S.; Harris, H.
A.; Henderson, R. A.; Xu, Z. B.; Unwalla, R. J.; Miller, C.
P. Bioorg. Med. Chem. Lett. 2004, 14, 4925.
17. Vu, A. T.; Cohn, S. T.; Manas, E. S.; Harris, H. A.;
Mewshaw, R. E. Bioorg. Med. Chem. Lett. 2005, 15, 4520.
18. Chesworth, R.; Wessel, M. D.; Heyden, L.; Mangano, M.;
Zawistoski, M.; Gegnas, L.; Galluzzo, D.; Lefker, B.;
Cameron, K. O.; Tickner, J.; Lu, B.; Castleberry, T. A.;
Petersen, D. N.; Brault, A.; Perry, P.; Ng, O.; Owen, T. A.;
Pan, L.; Ke, H. Z.; Brown, T. A.; Thompson, D. D.;
DaSilva-Jardine, P. Bioorg. Med. Chem. Lett. 2005, 15,
5562.
19. Cragoe, E. J.; Woltersdorf, O. W.; Gould, N. P.;
Pietruszkiewicz, A. M.; Ziegler, C.; Sakurai, Y.; Stokker,
G. E.; Anderson, P. S.; Bourke, R. S.; Kimelberg, H. K.;
Nelson, L. R.; Barron, K. D.; Rose, J. R.; Szarowski, D.;
Popp, A. J.; Waldman, J. B. J. Med. Chem. 1986, 29, 825.
20. International Publication WO 01/82923.
27. A complex of ERb with 45 was prepared by displacing
the ligand binding domain of ERb (residues 260–500)
bound to a weakly absorbing affinity column by elution
with 19. Although the elution was with the racemic
mixture 19, only the active enantiomer 45 bound to the
protein and displaced it from the affinity column. The
complex of 45 with the ligand binding domain of ERb
(residues 260–500) was crystallized by vapor diffusion,
using a precipitant containing 1.8 M NaCl, 6% PEG
6000, 3% isopropanol, and 50 mM Tris buffer, pH 7.2.
Data were measured at beamline 17-ID of the
Advanced Photon Source. The crystal had the symme-
try of space group P6122, with cell dimensions
˚
21. All aldehydes were commercially available with the
exception of glycoaldehyde which was formed in situ from
glyoxaldehyde dimer during the reductive alkylation of 33.
22. Still, I. W.; Yian, S. Tetrahedron Lett. 1987, 28, 2489.
23. Chen, Q.-Y.; Wu, S.-W. J. Chem. Soc. Commun. 1989,
705.
a = b = 64.53, c = 251.16 A. The data were processed
with program X-GEN, which yielded an R-merge of
˚
0.078 for the data from 1 to 2.20 A. The structure was
refined using program SHELXL, with final values for
R-work and R-free of 0.227 and 0.323 for the data
˚
from 10.0 to 2.20 A resolution. Coordinates and struc-
24. The IC50 values were generated in an estrogen receptor
ligand binding assay. This scintillation proximity assay
was conducted in NEN Basic Flash plates using tritiated
estradiol and full-length recombinant human ERa and
ERb proteins with incubation times of 3–23 h. In our
tures factors have been deposited with the Protein Data
Bank (2GIU).
28. Kuiper, G.; Lemmen, J. G.; Carlsson, B.; Corton, J. C.;
Safe, H.; Saag, P. T.; Burg, B.; Gustafsson, J. Endocrinol-
ogy 1998, 139, 4252.