Steroidal affinity labels of the estrogen receptor α. 4. Electrophilic 11β-aryl derivatives of estradiol

Article abstract of DOI:10.1021/jm990179s

Ten electrophilic estradiol 11β-aryl derivatives were synthesized, with three different types of 11β-substituent: (i)p?O(CH₂)₂X (compounds: 6, X = OSO₂CH₃; 7, X = I; 13, X = NHCOCH₂-Cl; 15, X = N(CH

Full text of DOI:10.1021/jm990179s

J . Med. Chem. 2000, 43, 613-628
613
Ster oid a l Affin ity La bels of th e Estr ogen Recep tor r. 4. Electr op h ilic 11â-Ar yl
Der iva tives of Estr a d iol
Sigrid Aliau,† Georges Delettre,§ He´le`ne Mattras,† Driss El Garrouj,†,‡ Franc¸ois Nique,§ Georges Teutsch,§ and
J ean-Louis Borgna*,†
INSERM Unite´ 439, 70 rue de Navacelles, 34090 Montpellier, France, and Hœchst Marion Roussel, 102 route de Noisy,
93235 Romainville Cedex, France
Received April 15, 1999
Ten electrophilic estradiol 11â-aryl derivatives were synthesized, with three different types of
11â-substituent: (i) pØO(CH2)2X (compounds: 6, X ) OSO2CH3; 7, X ) I; 13, X ) NHCOCH2-
Cl; 15, X ) N(CH3)COCH2Br; and 16, X ) N(CH3)COCH2Cl); (ii) pØO(CH2)5X (compounds:
17, X ) I; 20, X ) NHCOCH2Br; and 22, X ) N(CH3)COCH2Br); and (iii) pØCtCCH2X
(compounds: 27, X ) NHCOCH2Cl; and 29, X ) N(CH3)COCH2Cl). The range of their apparent
affinity constants for binding the lamb uterine estrogen receptor R (ERR) was 3-40% that of
estradiol. Six electrophiles, chloroacetamides 13, 16, 27, and 29, iodide 17, and bromoacetamide
20 (whose arm linking the electrophilic carbon to the 11â-phenyl group includes at least six
bonds), were able to irreversibly inhibit the binding of [3H]estradiol to ER (25-60% decrease
in binding sites), with the following compound effectiveness order: 17 < 13 < 16 ∼ 20 ∼ 27 ∼
29. Mesylate 6, iodide 7 (whose linking arm includes only three bonds), and bromoacetamides
15 and 22 (which differ from 16 by the Cl to Br change and from 20 by the NH to NCH3 change,
respectively) were much less effective (<10% decrease in binding sites, if any). The fact that
the inactivation of estradiol-binding sites by the six electrophiles was totally prevented by
estradiol indicated that they were ER affinity labeling agents. When ER was modified by methyl
methanethiosulfonate, an SH-specific reagent, the different compounds led to very contrasting
results in ER affinity labeling. With modified ER, iodide 17 and chloroacetamides 27 and 29
were practically inactive, chloroacetamides 13 and 16 and bromoacetamide 20 were still active
but less effective than on the native ER, whereas tertiary bromoacetamides 15 and 22, found
to be practically inactive on native ER, became the most effective electrophiles (∼45% and
∼65% binding sites inactivated, respectively). The results indicate that in the steroid-filled
hormone-binding pocket: (i) nucleophilic residues are localized on the â-side but relatively
remote from the steroid nucleus (distance from C-11 > “seven bonds”); (ii) relatively discrete
changes in the electrophilic functionality, such as Cl to Br or NH to NCH3 of haloacetamido
compounds, can markedly modify the positioning of the electrophilic center which could no
longer react with the nucleophilic residues; and (iii) cysteine residues (probably homologues of
human ERR cysteine 381 and/or cysteine 530) are, at least partly, the covalent attachment
sites of the electrophiles. Moreover, modification of cysteine residues by methyl methaneth-
iosulfonate changes the structure of the hormone-binding pocket, whose labeling by the various
electrophiles is profoundly altered.
In tr od u ction
Recently, the crystal structures of the hormone-
binding domain of the human ERR bound to various
estrogens or antiestrogens were established. Tanen-
baum et al.8 determined the structure of covalently
dimerized domains bound to estradiol, whereas the
structure of S-carboxymethylated domain bound: (i) to
estradiol or raloxifene (a benzothiophene partial anti-
estrogen) or (ii) to diethylstilbestrol (a diphenylethylene
estrogen) or 4-hydroxytamoxifen (a triphenylethylene
partial antiestrogen) was determined by Brzozowski et
al.9 and Shiau et al.,10 respectively. The overall archi-
tecture of the ERR ligand-binding domain was similar
to those seen in the crystal structure of other nuclear
receptor hormone-binding domains.11-16 The ERR hor-
mone-binding domain appears to comprise 12 R-helices
and 2 â-strands. This domain is folded into a three-layer
structure with three internal helices sandwiched be-
tween the most external helices. The ligand-binding
pocket is predominantly formed by hydrophobic residues
The biological effects of estrogens in target cells are
mediated by nuclear receptors. Until 1996, it was
generally accepted that only one type of estrogen recep-
tor (ER), cloned in 19851,2 and now termed ERR, existed.
Recently, cDNA clones encoding another subtype of ER,
termed ERâ, were obtained from various tissues from
rat,3 mouse,4 and human.5 Interestingly, the ERâ
mRNA was found in organs such as prostate6 and
testis,7 which are not considered to be major estrogen
target organs, whereas it was much lower than the ERR
mRNA in the uterus,6 which is the major estrogen target
organ in mammals.
* Corresponding author: J ean-Louis Borgna. Tel: 33 467 04 37 14.
Fax: 33 467 04 37 15. E-mail: borgna@u439montp.inserm.fr.
† INSERM Unite´ 439.
§ Hœchst Marion Roussel.
‡ Present adress: Universite´ Sidi Mohamed Ben Abdalah, Faculte´
des Sciences et Techniques, Fe´s-Sa¨ıs, Maroc.
10.1021/jm990179s CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/27/2000

614 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Aliau et al.
located in six distinct structural elements, whereas the
polar amino acids E353 and R394 are directly involved
in hydrogen bonds with the phenolic hydroxyl of the
estradiol A-ring (or diethylstilbestrol, raloxifene, or
4-hydroxytamoxifen counterpart) and H524 is involved
in a single hydrogen bond with the 17â-hydroxyl of the
estradiol D-ring (or diethylstilbestrol or raloxifene
counterpart). Agonists and antagonists bind at the same
site within the core of the ligand-binding domain but
demonstrate different binding modes.9,10
electrophiles the remoteness of potential ER covalent
attachment sites; and (iii) various terminal electrophilic
functionalities in order to favor ER affinity labeling by
the compounds. Ten electrophilic 11â-arylestradiol de-
rivatives were prepared in which various electrophilic
functionalities, e.g. iodide, (methylsulfonyl)oxy, haloac-
etamido, or N-methylhaloacetamido, are linked to the
steroid by an 11â-p-ethoxyphenyl, -p-pentoxyphenyl, or
-p-propynylphenyl group. Note that since several of
these compounds are related to steroidal pure anties-
trogens,23,24 and since there is presently no model for
the structure of ER bound to pure antiestrogens, affinity
labeling of ER by the compounds could give useful
information to define the binding mode of such com-
pounds. In this paper, we report: (i) the synthesis of
these 10 electrophilic 11â-estradiol derivatives; (ii) their
affinity for the lamb ERR; and (iii) their ability to
become irreversibly bound to lamb and human ERRs,
either native or modified by MMTS.
Affinity labeling of receptors enables direct identifica-
tion of nucleophilic amino acid residues of the hormone-
binding site which are in contact or in close proximity
to electrophilic ligands. This approach, much more
versatile than the crystallization approach, can be used
to demonstrate, from the whole ER in its soluble state,
the different positions of various hormone agonists and
antagonists in the ER hormone-binding pocket. Using
both wild-type and various cysteine to alanine mutants
of the human ERR, three out of the four cysteines (C381,
C417, C447, C530) of the hormone-binding domain were
found to be covalent attachment sites of electrophilic
nonsteroidal and steroidal ligands of ERR. In wild-type
and C530A mutant ERRs, C530 and C381 were respec-
tively the covalent attachment sites of an aziridine
derivative of tamoxifen,17,18 whereas C417 and C530
were alkylated by 17R-[(haloacetamido)alkyl]estra-
diols,19-21 compounds which displayed more agonistic
characters than the former. The results are in agree-
ment with the crystal structure established for the ERR
ligand-binding domain, since all three cysteines are
located within or in very close proximity to structural
elements involved in delineation of the hormone-binding
pocket.9
Resu lts
Syn th esis of Electr op h ilic 11â-Estr a d iol Der iva -
tives. All of the electrophilic estradiol derivatives
described here were prepared according to a common
strategy starting from the appropriate hydroxy- or
iodine-substituted 11â-(alkoxyphenyl)- or 11â-(alky-
nylphenyl)estrane derivative.
Scheme 1 exemplifies the synthesis of estradiol 11â-
(p-ethoxyphenyl) derivatives. The already reported 11â-
(4-hydroxyphenyl)estradienedione (1)23 was used as
starting material. The hydroxyethyl chain was intro-
duced in 1 through alkylation of the phenol function by
bromoethanol protected as its tert-butyldimethylsilyl
(TBDMS) ether. Acidic hydrolysis of the TBDMS group
afforded the primary alcohol 2 (61%), which was esteri-
fied by methanesulfonyl chloride to give mesylate 3
(98%). The A-ring in 3 was aromatized with acetyl
bromide and acetic anhydride, through formation of the
2-en-3-ol acetate, and alkaline hydrolysis of the ester24,25
gave estratrienolone 4 (74%). This reaction also gave
rise to the formation of the nonconjugated estra-5(10),9-
(11)-dien-3-one 5 (16%) resulting from abstraction of the
11R proton instead of a proton on C-2, in R position to
the 3-keto function. Borohydride reduction of the 17-
keto function of mesylate 4 provided mesylate 6 (88%),
which was converted to iodide 7 (72%) by the action of
sodium iodide. The latter was used as a template to
prepare bromoacetamido (12 and 15) and chloroaceta-
mido (13 and 16) derivatives, respectively. Protection
of the two hydroxyls of 7 as tetrahydropyranyl (THP)
ethers gave 8 (mixture of stereoisomers). Substitution
of the iodine atom in 8 by the trifluoroacetamido group
provided secondary trifluoroacetamide 9 (65%) along
with a small amount of vinyl ether 10 (12%), resulting
from dehydroiodination of 8. Alkaline hydrolysis of 9
yielded primary amine 11 (97%), which was acylated
by bromoacetyl bromide. Deprotection of the 3- and 17â-
hydroxyls by hydrochloric acid in a mixture of methanol
and THF led to a partial halogen exchange, yielding a
mixture (70:30) of the expected secondary bromoaceta-
mide 12 together with secondary chloroacetamide 13.
The presence of 12 and 13 in the obtained product was
observed by NMR spectroscopy: two singlets corre-
sponding to CH2-Br and CH2-Cl appeared at 3.86 and
With the aim of gaining further insight into the
positioning of ligands in the hormone-binding pocket,
we undertook to synthesize and then evaluate electro-
philic 11â-estradiol derivatives as ERR affinity labeling
agents. Recently, we reported on the synthesis and
properties of six new 11â-estradiol derivatives in which
an 11â-ethyl, 11â-butyl, or 11â-decyl chain was substi-
tuted with various electrophilic functionalities: bro-
mide, (methylsulfonyl)oxy, (p-tolylsulfonyl)oxy, or bro-
moacetamido.22 In these six electrophilic compounds,
only 11â-[(tosyloxy)decyl]estradiol displayed significant
ER affinity labeling activity. The covalent attachment
site of the 11â-decyl derivative was not identified. It did
not seem to occur at cysteine residues since the ER
alkylation process was not prevented by the thiol-
specific reagent methyl methanethiosulfonate (MMTS).
Moreover, due to the length and mobility of the n-decyl
chain, the relative location of the electrophilic carbon
from the steroid nucleus was not clear. This reduced
the interest of the compound for analysis of the ER
binding pocket.
To overcome this drawback, we decided to synthesize
and then evaluate a series of electrophilic 11â-aryl
derivatives of estradiol in which the 11â-substituent
included: (i) a phenyl ring, to confer rigidity from C-11
(contrary to an 11â-aliphatic arm) to the axial substitu-
ent and to restrict localization of the electrophilic carbon
to the â-side of the steroid nucleus; (ii) on the phenyl
ring, various linear para chains with differing lengths
and mobilities in order to determine from ER-bound

Steroidal Affinity Labels of ERR
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 615
Sch em e 1. Synthesis of Estradiol 11â-Ethoxyphenyl Derivatives

616 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Aliau et al.
Sch em e 2. Synthesis of Estradiol 11â-Pentoxyphenyl Derivatives
4.06 ppm, respectively. As the two compounds were
unseparable by chromatography, bromoacetamide 12
was converted to chloroacetamide 13 by the action of
lithium chloride. The N-methyl homologues 15 and 16
of 12 and 13, respectively, were prepared analogously
by substitution of the iodine atom of 8 by the N-
methyltrifluoroacetamido group. Due to the higher
basicity/nucleophilicity ratio of the secondary vs the
primary amide anion, a higher amount of the vinyl
compound 10 was produced in this case (35% vs 12%).
The yield of tertiary trifluoroacetamide 14 (49%) was
lower than that obtained for 9 in the unsubstituted
series (65%). Alkaline hydrolysis of the amide bond of
14 gave the corresponding N-methyl secondary amine
which was acylated with bromoacetyl bromide. Hydro-
bromic acid was used in order to avoid any halogen
exchange during deprotection of THP ethers. Moreover,
brine washing of the organic extracts during the workup
was omitted. The tertiary N-methylbromoacetamide 15
was thus prepared in 85% yield. As above, the corre-
sponding N-methylchloroacetamide 16 (81%) was ob-
tained by substitution of the bromine atom of 15 with
chlorine.
1 was previously described.24 Iodide 17 was used as
starting material for the preparation of bromoaceta-
mides 20 and 22. Protection of the two hydroxyls of 17
as THP ethers gave 18 (mixture of stereoisomers).
Substitution of the iodine atom in 18 by the trifluoro-
acetamido group yielded the secondary trifluoroaceta-
mide 19 (40%). In this case, no dehydroiodinated
product was noticed, but untransformed starting mate-
rial (20%) was recovered. Alkaline hydrolysis of the
amide bond of 19 yielded the corresponding primary
amine, which was acylated with bromoacetyl bromide.
Hydrobromic acid was used to deprotect the 3- and 17â-
hydroxyls, producing bromoacetamide 20 (57%). Since
substitution of the iodine atom of 18 by the N-methyl-
trifluoroacetamido group, to give the tertiary trifluoro-
acetamide 21, was even less efficient than in the
previous p-ethoxyphenyl series, 21 was nicely prepared
(98%) by N-methylation of the secondary trifluoroac-
etamide 19. The tertiary N-methylbromoacetamide 22
was prepared (54%) by the same method as that used
to obtain 20.
Scheme 3 illustrates the synthesis of electrophilic
estradiol 11â-(p-propynylphenyl) derivatives. The pro-
pargyl alcohol 24 was prepared in 87% yield by hy-
droxyalkylation (butyllithium, paraformaldehyde) of
the previously described, bis-protected 11â-(4-ethy-
The synthesis of electrophilic estradiol 11â-(p-pen-
tyloxyphenyl) derivatives is shown in Scheme 2. Prepa-
ration of 11â-[4-(5-iodopentyl)phenyl]estradiol (17) from

Steroidal Affinity Labels of ERR
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 617
Sch em e 3. Synthesis of Estradiol 11â-Propynylphenyl Derivatives
nylphenyl)estradiol (23);26 it was further converted to
mesylate 25 (69%). The secondary trifluoroacetamide 26
was obtained (53%) from 25 by substitution of the
methanesulfonoxy group with the trifluoroacetamido
group. N-Methylation of 26 afforded the tertiary tri-
fluoroacetamide 28 (84%). In that p-propynylphenyl
series, chloroacetamides 27 and 29 were prepared
directly through alkaline hydrolysis of 26 and 28,
respectively, followed by acylation with chloroacetyl
chloride and deprotection of the 3- and 17â-hydroxyls
(81% and 78% yields, respectively).
For all secondary haloacetamides (13, 20, and 27), in
DMSO, the NMR signal of the CH2Cl/Br group appeared
as a singlet, whereas two singlets (∼1:1 ratio) were
obtained for each of the CH2Cl/Br and NCH3 groups of
tertiary haloacetamides (15, 16, 22, and 29). These
results suggested that, at least in DMSO at 20 °C,
secondary haloacetamides displayed a single conforma-
tion (presumably the Z conformation) whereas, due to
hindered rotation about the N-CO bond, tertiary ha-
loacetamides displayed both the E and Z conformations.
ER Bin d in g Affin ity. The apparent relative affini-
ties of electrophiles were determined by competitive
binding between increasing concentrations of com-
pounds and a constant [3H]estradiol concentration,
using lamb uterine cytosol as an ERR source. The fact
that electrophiles could irreversibly bind ER and cytosol
components (which would tend to overestimate and
underestimate, respectively, the affinity of the com-
pound) could markedly alter the compound’s affinity
value. Therefore, to minimize potential irreversible
binding of compounds to ER and to cytosolic compo-
nents, binding experiments were performed at pH 7.0,
since the reactivity of nucleophilic amino acid residues
generally increases with pH. Table 1 gives the apparent
affinity constants of the compounds for the lamb uterine
ER, relative to that of estradiol (100%), which were
calculated according to Korenman.27 There was no clear
relationship between the compound affinity, which
according to compound ranged from 3% to 40%, and the
size of the 11â-chain since the above extreme values
were obtained for chloroacetamide 13 and mesylate 6
whose 11â-substituent lengths are similar, whereas
compounds with either a shorter chain such as iodide 7
or a longer chain such as iodide 17 and bromoaceta-
mides 20 and 22 showed intermediate values. In the
haloacetamide series, methylation of the nitrogen in the
amide function or the Cl to Br substitution did not

618 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Aliau et al.
Ta ble 1. Structural and Binding Characteristics of Electrophilic 11â-Estradiol Derivatives to ER
electrophile-bearing chaina
compd
11â-substituent
class
length
degree of freedom
apparent RACb
covalent bindingc
estradiol
100
6
7
ØO(CH2)2OSO2CH3
ØO(CH2)2I
1
2
3
6
2
4
39.9 ( 8.4
10.6 ( 2.0
3.30 ( 0.66
4.22 ( 0.69
6.96 ( 0.52
8.23 ( 1.54
17.1 ( 5.9
9.70 ( 2.49
14.5 ( 2.8
19.4 ( 0.2
-
-
+
-
+
+
+
-
+
+
13
15
16
17
20
22
27
29
ØO(CH2)2NHCOCH2Cl
ØO(CH2)2N(CH3)COCH2Br
ØO(CH2)2N(CH3)COCH2Cl
ØO(CH2)5I
ØO(CH2)5NHCOCH2Br
ØO(CH2)5N(CH3)COCH2Br
ØCtCCH2NHCOCH2Cl
ØCtCCH2N(CH3)COCH2Cl
3
4
6
9
5
7
5
6
2
a
The 10 electrophiles were classified according to the type of the electrophile-bearing chain. Five different classes were defined, which
are characterized: (i) by the remoteness of the electrophilic carbon from the 11â-phenyl ring (the length of the linking arm was evaluated
according to the number of bonds); and (ii) by the degree of freedom of the chain. b Apparent RACs (relative affinity constants) of compounds
for the cytosolic lamb uterine ER were determined by competitive (20 h, 20 °C) radioreceptor assay using [3H]estradiol as tracer, as
described in Materials and Methods. Data are means of duplicate determinations ( SD, in two to five separate experiments. c Covalent
binding of compounds to ER was established from their ability to irreversibly inhibit binding of [3H]estradiol (cf. Results, Irreversible
Binding to ER). The symbol (-) means that the decrease in estradiol-binding sites induced by the compound, if any, was <10%.
markedly change the affinity of the 11â-derivatives.
Therefore in each of the three haloacetamide series, the
apparent affinities of the compounds were similar
(compare 13/15/16, 20/22, and 27/29). They were how-
ever lower in the ethoxy series than in the pentoxy or
propynyl series. The apparent affinities of the 10
electrophiles were roughly in the same range as previ-
ously found for electrophilic estradiol 11â-alkyl deriva-
tives.22 Compared to the affinities of chemically inert
11â-estradiol derivatives, the compound affinities ap-
peared slightly lower than those of pure antiestrogens
bearing a bulky amidoalkyl28 or amidoalkoxyphenyl23
11â-substituent and much lower than those of estrogens
derived from estradiol by introducing various short
aliphatic chains at the 11â-position.25,29
Ir r ever sible Bin d in g to ER. Affinity labeling re-
ceptors is usually a concentration- and time-dependent
process which, depending on the type of electrophile,
could be influenced by temperature and pH.19,20 Specific
ER alkylating activities of the 10 electrophiles were
evaluated from their ability to irreversibly inhibit
specific estradiol binding in cytosol by means of a
procedure which was validated previously.20 This pro-
cedure includes three steps: (i) incubation of cytosol
with the electrophile to allow the ER affinity labeling;
(ii) removal of unbound electrophile by charcoal absorp-
tion; and (iii) measurement of the residual concentration
of specific [3H]estradiol-binding sites in cytosol, under
exchange conditions to allow displacement of the non-
covalently ER-bound electrophile. With high-affinity
electrophiles, such as those evaluated in this study, it
could be difficult to evaluate their ER affinity labeling
ability since incomplete [3H]estradiol labeling of ERR
noncovalently occupied by the electrophile would result
in a false positive ER affinity labeling assessment. Two
criteria, which distinguish between covalent and non-
covalent binding of the electrophile to ER, can be used
to avoid such misinterpretations. The effect of electro-
phile on the estradiol-binding site concentration in
cytosol can be determined either: (a) according to the
time of cytosol exposure to the electrophile;20 or (b) using
increasing concentrations of [3H]estradiol at the ex-
change step.19
requires conventional binding and then involves a time-
dependent chemical reaction. Therefore, if according to
increasing times of cytosol exposure to the compounds
the decrease in [3H]estradiol-binding sites is much more
prolonged for the electrophile than for control com-
pounds (such as a chemically inert analogue of the
electrophile or unlabeled estradiol), then the electrophile
likely alkylated ER.
(b) Contrary to the classical binding of a chemically
inert ligand to ER, the covalent binding of a reactive
ligand cannot be reversed by an increase in the amount
of [3H]estradiol used for displacement of noncovalently
bound ligand. Therefore, an irreversible decrease in
specific [3H]estradiol binding in cytosol preexposed to
an electrophilic ligand (decrease evaluated from the
specific binding in cytosol samples, incubated without
ligand or with control compounds) probably signals an
ER affinity labeling process. This approach could,
however, be limited by the increase of nonspecific [3H]-
estradiol binding (proportional to the applied concentra-
tion of [3H]estradiol) to cytosol components, which
restricts the accuracy of the measurements.
Another criterion could be used to determine the
specificity of the alkylation process. Occupation of the
hormone-binding pocket by a ligand such as estradiol
must totally inhibit specific covalent attachment of the
electrophile to ER.19
To determine the alkylation level of ER, the cytosol
was thus incubated for increasing times with various
electrophile concentrations at either 0 or 25 °C and at
either pH 7.0 or 8.5. Cytosol was concomitantly incu-
bated without steroid or with 100 nM unlabeled estra-
diol to obtain reference values for the specific estradiol-
binding site concentration. Six out of the 10 electrophiles
elicited marked decreases in the estradiol-binding site
concentration. Figure 1 shows the results of a repre-
sentative experiment in which cytosol was incubated for
16 h at 0 °C, pH 8.5, without steroid, with 100 nM
estradiol or with 200 nM 11â-estradiol derivatives.
Mesylate 6 and iodide 7, whose structures include the
ethoxyphenyl group (Table 1), did not have any signifi-
cant effect on the binding site concentration. Among the
three haloacetamido homologues of 6 and 7, the two
chloroacetamides 13 and 16 induced a marked effect
(40-50% decrease in the binding site concentration
(a) Affinity labeling of receptors is usually a slower
process than classical noncovalent binding since it first

Steroidal Affinity Labels of ERR
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 619
F igu r e 1. Inactivation of specific estradiol-binding sites in
lamb uterine cytosol by 11â-estradiol derivatives. Uterine
cytosol (4 mg protein/mL, pH 8.5) was incubated for 16 h at 0
°C, without steroid (0), with 100 nM estradiol (E2), or with
200 nM of the synthesized electrophilic 11â-estradiol deriva-
tives. Samples were then charcoal treated. Charcoal was
pelleted by centrifugation, and supernatant aliquots were
incubated under exchange conditions with 20 nM [3H]estradiol
in the absence and presence of 5 µM unlabeled estradiol. Total
binding and nonspecific binding of [3H]estradiol were deter-
mined as described in Materials and Methods. Total binding
(unfilled bars) and nonspecific binding (filled bars), which did
not significantly vary according to the compound incubated
with cytosol, are represented. Values are means of duplicate
determinations. The top of the unfilled bars also indicates the
percentage decrease in the concentration of specific estradiol-
binding sites (right scale), using the concentration of binding
sites measured in cytosol exposed to estradiol as the reference
binding site concentration.
F igu r e 2. Time-course of inactivation of specific estradiol-
binding sites. Uterine cytosol was incubated at 0 °C, pH 8.5,
without steroid, with 100 nM estradiol, or with 200 nM
chloroacetamide 16 or 29, iodide 17, or bromoacetamide 20.
After various periods of time, aliquots were removed and
immediately charcoal treated. After sample centrifugation, the
total binding and nonspecific binding of [3H]estradiol (20 nM)
in supernatants occurring under exchange conditions were
determined. The total binding of [3H]estradiol in supernatants
corresponding to cytosol incubated without steroid (0) or with
estradiol (b, E2), chloroacetamide 16 (2) or 29 ([), iodide 17
(1), or bromoacetamide 20 (9) and the nonspecific binding of
[3H]estradiol (X, NS), which did not significantly vary accord-
ing to the compound incubated with cytosol, are represented
as functions of the incubation time. Values are means of
duplicate determinations.
relative to that measured in cytosol exposed to estra-
diol), whereas the tertiary bromoacetamide 15, which
is directly derived from 16 by the Cl to Br substitution,
had no effect. The iodopentyloxyphenyl derivative 17
induced a 30% effect. Of the two bromoacetamido
homologues of 17, the secondary amide 20 was effective
(∼55%), whereas the tertiary amide 22 was not. Finally,
the secondary and tertiary chloroacetamides 27 and 29,
whose structures include the propynylphenyl group,
induced a 50-60% effect. The effect of compounds
depended on the compound concentration (not shown);
however, it did not markedly increase for compound
concentrations > 100 nM (cf Figure 3). Depending on
the ER/protein ratio in cytosol (0.9-2.2 pmol ER/mg
protein in the various experiments performed), the
percentage decreases in binding site concentration
varied markedly, i.e., the higher the relative ER titer,
the higher the decrease. However, the same qualitative
results were obtained in the various experiments per-
formed, i.e., compounds 6, 7, 15, and 22 did not induce
a significant decrease (which would be g10%) in the
binding site concentration, whereas for the other com-
pounds the order for increasing effectiveness was al-
ways: 17 < 13 < 16 ∼ 20 ∼ 27 ∼ 29. With compound
concentration e 300 nM, regardless of the conditions
used (temperature, pH, incubation time), the maximum
decrease in the binding site concentration never ex-
ceeded 70%. Results almost identical to those shown in
Figure 1 were obtained when incubation of cytosol with
11â-estradiol derivatives was performed at 25 °C instead
of at 0 °C (not shown). The major difference between
the two series of results concerned the binding site
concentration measured from control cytosol incubated
without compound. Due to the unprotected hormone-
binding sites, at 25 °C there was a slight time-dependent
decrease in the binding site concentration, which was
not observed at 0 °C. When exposure of cytosol to the
11â-estradiol derivatives was performed at pH 7.0
instead of at pH 8.5, decreases in binding sites, although
qualitatively similar to those shown in Figure 1, were
less marked. Finally, similar qualitative results were
obtained with compounds 13, 16, 17, 20, 27, and 29

620 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Aliau et al.
slight decrease which was stabilized after only 2 h (effect
which probably reflected completion of the formation of
the ER:estradiol complex). Electrophiles 16, 17, and 29
induced much more gradual and pronounced decreases
than that induced by estradiol, since at least 8 h was
required for binding stabilization. A similar result was
obtained with electrophiles 13 and 27 (not shown).
Bromoacetamide 20 induced a marked decrease, which
did not show any time dependence.
Kinetic studies would not be suitable in some in-
stances to reveal an affinity labeling process19 (cf.
Discussion). Therefore, to potentially distinguish be-
tween conventional binding and covalent attachment of
20 to the ER hormone-binding site, experiments were
performed in which the hormone-binding site concen-
tration was determined using increasing amounts of
[3H]estradiol. Figure 3 shows the binding site concen-
tration in cytosol preincubated with 30 nM (Figure 3A),
100 nM (Figure 3B), or 300 nM (Figure 3C) bromoac-
etamide 20 or its N-methyl homologue 22 or 100 nM
unlabeled estradiol, as a function of the amount of [3H]-
estradiol (20, 40, or 80 nM) used for exchange. Under
equilibrium conditions, the deficit in binding sites
measured with 20 nM [3H]estradiol in cytosol preincu-
bated with unlabeled estradiol or with 22 appeared to
result from conventional binding of residual steroid,
since this deficit was progressively (at the expected
ratios) resorbed when 40 and 80 nM [3H]estradiol,
instead 20 nM, were used. This was not observed with
cytosol preincubated with 20. There were substantial
differences between the [3H]estradiol binding measured
with 40 and 80 nM [3H]estradiol and those calculated
assuming that the deficit measured with 20 nM [3H]-
estradiol would result only from conventional occupancy
of ER by bromoacetamide 20. Following incubation of
cytosol with compound 20, the deficit in binding sites
measured with 80 nM [3H]estradiol differed only slightly
from that measured with 20 nM. Figure 3D shows the
variation in this deficit measured with 80 nM [3H]-
estradiol, according to the concentration of 20; relative
to the deficit in binding sites obtained for 300 nM, the
decrease was >50% and >80% for 30 and 100 nM 20,
respectively.
F igu r e 3. Irreversible inactivation of specific estradiol-
binding sites by bromoacetamide 20. Uterine cytosol was
incubated for 16 h at 0 °C, pH 8.5, without steroid, with 100
nM estradiol, or with various concentrations (30, 100, or 300
nM) of bromoacetamide 20 or 22. After charcoal treatment and
then centrifugation, supernatant aliquots were incubated for
16 h at 20 °C with 20, 40, or 80 nM [3H]estradiol in the absence
and presence of 10 µM unlabeled estradiol. Total binding and
nonspecific binding of [3H]estradiol in the aliquots were
determined. (A-C) Specific [3H]estradiol-binding site concen-
trations determined from cytosol incubated with 30 nM (A),
100 nM (B), or 300 nM (C) estradiol (b) or bromoacetamide
20 (2) or 22 (1) are expressed as percentages of the specific
[3H]estradiol-binding site concentration measured from cytosol
incubated without steroid. Percentage inhibitions of [3H]-
estradiol binding elicited by the various concentrations of the
three steroids at 20 nM [3H]estradiol concentration were used
to determine theoretical binding levels which would be ob-
tained at 40 and 80 nM [3H]estradiol if decreases in specific
binding measured at 20 nM [3H]estradiol were due exclusively
to classical competitive inhibition of the compounds; cytosol
incubated with estradiol (O) or bromoacetamide 20 (4) or 22
(3). (D) Deficit in estradiol-binding site concentration induced
by bromoacetamide 20 and determined using 80 nM [3H]-
estradiol is plotted against the concentration of 20. Values are
means of duplicate determinations; experimental variation was
under 10%.
using wild-type human ERR30 expressed in COS cells;
the effects of the compounds (60-85% decrease in the
binding site concentration according to the compounds)
were more pronounced than those obtained with the
lamb uterine ER, possibly due to the higher ER/protein
ratio in COS extracts than in uterine cytosols. Further
studies of ER affinity labeling by 11â-aryl derivatives
of estradiol were then performed with the lamb uterine
cytosol at 0 °C, pH 8.5, using compounds at 100-300
nM.
According to the kinetics or the exchange criterion,
the six electrophiles appeared to bind covalently to ER.
To check whether this covalent binding occurred at or
very close to the hormone-binding site, the effects of the
compounds were then determined using the estradiol-
filled ER. In this case, none of the six compounds was
able to significantly decrease the hormone-binding site
concentration (Figure 4). We therefore concluded that
electrophiles 13, 16, 17, 20, 27, and 29 were indeed
affinity labels of ERR.
Since the ER covalent attachment sites of reactive
ligands identified so far are exclusively cysteines, we
then examined the possible involvement of cysteine
residues in the alkylation process of ERR by 11â-
estradiol derivatives. The effects of the compounds were
determined using cytosol treated by MMTS, a thiol-
specific reagent which transforms SH into S-S-CH3.
Eight out of the 10 electrophiles were able to signifi-
cantly decrease the binding site concentration in MMTS-
treated cytosol (Figure 5). Of the four electrophiles found
that had no significant effect on the native ER, mesylate
Kinetic studies were performed to determine whether
the observed decreases in [3H]estradiol-binding site
concentration following cytosol exposure to certain of
the electrophiles were due to affinity labeling of ER or
to classical binding of electrophiles. Figure 2 shows the
time-course variation of the estradiol-binding site con-
centration in cytosol incubated without steroid, with
unlabeled estradiol or with electrophile 16, 17, 20, or
29. For control cytosol incubated without steroid, ac-
cording to the incubation time there was no variation
in the binding site concentration. Estradiol elicited a

Steroidal Affinity Labels of ERR
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 621
F igu r e 5. Inactivation of specific estradiol-binding sites in
MMTS-treated cytosol by 11â-estradiol derivatives. Effect of
binding site occupancy by estradiol. Uterine cytosol was
preincubated for 4 h at 0 °C, pH 8.5, with 2 mM MMTS and
then for an additional 4 h with or without 100 nM estradiol.
Aliquots of the two samples were incubated for 16 h at 0 °C
without steroid (0) or with 200 nM of each of the 10 synthesized
electrophilic estradiol derivatives (6-29). The total binding
and nonspecific binding of [3H]estradiol (30 nM) occurring
under exchange conditions were determined. The specific [3H]-
estradiol-binding site concentration in samples corresponding
to cytosol preincubated without (unfilled bars) or with (filled
bars) estradiol is represented. Values are means of duplicate
determinations.
F igu r e 4. Inability of electrophilic 11â-estradiol derivatives
to inactive estradiol-filled hormone-binding site of ER. Uterine
cytosol was preincubated for 4 h at 0 °C, pH 8.5, with or
without 100 nM estradiol. Aliquots of the two samples were
then incubated for 16 h at 0 °C without steroid (0) or with 200
nM chloroacetamide 13, 16, 27, or 29, iodide 17, or bromoac-
etamide 20. After charcoal removal of steroid, the total binding
and nonspecific binding of [3H]estradiol (30 nM) occurring
under exchange conditions in the various samples were
determined. The specific [3H]estradiol-binding site concentra-
tion in samples corresponding to cytosol preincubated without
(unfilled bars) or with (filled bars) estradiol is represented.
Values are means of duplicate determinations.
binding site concentrations measured between unfilled
and estradiol-filled ER samples exposed to 17, 27, or
29. In this case, the deficit in binding site concentration
resulting from exposure of cytosol to the compounds
would essentially result from conventional binding and
not from ER affinity labeling. This was confirmed using
increasing amounts of [3H]estradiol for the exchange
step. The deficit in binding sites resulting from cytosol
exposure to 17, 27, or 29 progressively disappeared as
the [3H]estradiol concentration increased (not shown).
We concluded that cysteine blockade by MMTS had very
contrasted effects on the ability of electrophiles to
covalently interact with the hormone-binding site of
ERR: (i) iodide 17 and chloroacetamides 27 and 29,
which were initially effective, became practically inac-
tive; (ii) chloroacetamides 13 and 16 and bromoaceta-
mide 20 were still active but with lower effectivenesses;
and (iii) tertiary bromoacetamides 15 and 22, first
inactive, became the most effective compounds.
6 and iodide 7 were still inactive, whereas tertiary
bromoacetamides 15 and 22 became very effective
(∼45% and ∼65% effectiveness, respectively). The six
other electrophiles induced decreases in the binding site
concentration which were less pronounced than those
obtained with the native ER (compare Figure 5 with
Figure 1). For chloro- and bromoacetamides 13, 16, and
20, the decreases (25-35%) appeared to reflect a specific
effect, since they were totally prevented by estradiol
occupancy of the hormone-binding site, whereas hormone-
binding site occupancy by estradiol only partially pre-
vented the decrease in the binding site concentration
induced by iodide 17 and chloroacetamides 27 and 29
(30-50%). This latter result suggested that after char-
coal treatment of cytosol, the non-ER-bound residual
concentration of these particular electrophiles or deriva-
tives was sufficient to reduce the specific binding of [3H]-
estradiol measured in cytosol exposed to each of the
other electrophiles. In the absence of estradiol, due to
complexing by ER and then protection from charcoal
absorption, the concentration of residual electrophile
would be higher than in the presence of estradiol. This
difference between residual concentrations of electro-
phile could explain, via classical competitive binding
with [3H]estradiol, the relatively minor difference in the
Discu ssion
The 10 high-affinity steroidal electrophiles synthe-
sized and evaluated as affinity labeling agents of ERR
in this study are closely related; i.e., they are derived
from 11â-phenylestradiol by introducing on the 11â-

622 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Aliau et al.
phenyl group, in the para position, a linear substituent
that includes a terminal electrophilic functionality. Five
different classes of compounds can be defined according
to the 11â-substituent type (Table 1): (1) ØO(CH2)2X
(6 and 7), (2) ØO(CH2)2N(H/CH3)COCH2X (13, 15, and
16), (3) ØO(CH2)5X (17), (4) ØO(CH2)5N(H/CH3)COCH2X
(20 and 22), and (5) ØCtCCH2N(H/CH3)COCH2X (27
and 29), where X ) Cl, Br, I, or OSO2CH3. However,
three parameters, i.e., the type of electrophilic function-
ality and the length and flexibility of the arm linking
the electrophilic carbon to the 11â-phenyl group, should
largely account for the reactivity of these molecules in
the ERR hormone-binding pocket. There are four dif-
ferent types of electrophilic functionality in these mol-
ecules, with the following chemical reactivity order:
methanesulfonate < iodide < chloroacetamide < bro-
moacetamide. According to the compound, the electro-
philic center is three (6 and 7), six (13, 15-17, 27, and
29), or nine (20 and 22) bonds away from the 11â-phenyl
group, whereas the degree of freedom of the linking arm
is two (6, 7, 27, and 29), four (13, 15, and 16), five (17),
or seven (20 and 22).
philic center and C-11 was quite similar to the corre-
sponding distance in 6 and 7, although including a
strong electrophilic carbon, was unable to alkylate the
ERR hormone-binding site.22 Conversely, compound 17,
whose electrophilic center is identical but more remote
from C-11 than that of 7, was an effective ER affinity
labeling agent.
(ii) A six- or nine-bond linking arm allows the elec-
trophilic center to be localized in the neighborhood of a
reactive nucleophilic amino acid residue in the hormone-
binding pocket. However, with this type of linking arm,
relatively discrete changes in the arm (such as NH to
NCH3 change) or in the electrophilic functionality (such
as CH2Cl to CH2Br change) modify the positioning of
the electrophilic center, which can no longer be attacked
by the nucleophilic amino acid residue of the hormone-
binding pocket.
Similar variations in ER affinity labeling efficiency
were previously reported for a series of closely related
electrophilic hexestrol derivatives;31 i.e., the activity of
the compounds markedly varied according to the length
and type of the chain linking an aziridine function to
the hexestrol nucleus. The inability of class 1 compounds
to alkylate ER suggests that the covalent attachment
sites of electrophilic 11â-estradiol derivatives are rela-
tively remote (at least “seven bonds” away) from the
steroid nucleus when the derivatives are bound to ER.
This assumption agrees with previous results obtained
with other electrophilic 11â-estradiol derivatives, since
compounds whose linear arm was at most seven bonds
long were ineffective for alkylating ERR, whereas a 10-
bond linking arm compound was very effective.22 More-
over, with the exception of class 4 compounds (in which
nine bonds separate the electrophilic center from the
11â-phenyl ring), for all other 11â-derivatives the direct
branching of the phenyl ring (with an axial configura-
tion) on the steroid nucleus restricts the location of the
electrophilic center to the â-side of the steroid. There-
fore, when these 11â-derivatives are in the hormone-
binding pocket, relative to the steroid nucleus, the target
amino acid residues are localized on the â-side.
ERR was modified by MMTS with the aim of deter-
mining the potential involvement of cysteine residues
of the hormone-binding pocket in the ER affinity label-
ing process. This approach was successfully used to
highlight that the covalent attachment of tamoxifen
aziridine and of electrophilic 17R-estradiol derivatives
to ERR occurred at cysteine residues, since: contrary
to native or wild-type ERR, MMTS-modified ER19,20 and
the double (C381A/C530A) and quadruple (C381A/
C417A/C447A/C530A) cysteine to alanine ERR mu-
tants18,21 were no longer alkylated by the electrophiles.
The effects of 11â-estradiol derivatives on MMTS-
modified ER were striking: (i) class 1 compounds were
still ineffective; (ii) compounds 15 and 22, found to be
ineffective on native ER, became the most effective
compounds; and (iii) the activity of the other classes 2-5
compounds decreased, i.e., classes 2 and 4 compounds
were still active but at a slight lesser extent, whereas
classes 3 and 5 compounds displayed very little, if any,
activity. These results can be interpreted as follows: (i)
for iodide 17 and chloroacetamides 27 and 29, cysteinyl
residues in native ER would be quasi-exclusive covalent
attachment sites without any site in MMTS-modified
On the basis of results obtained in kinetic and binding
displacement experiments with lamb uterine cytosol, 6
out of the 10 electrophiles appeared to be effective
affinity labeling agents of native ERR, since: (i) ERR is
the major ER form expressed in the uterus; and (ii) very
similar results were obtained using extracts from cells
transfected with a human ERR expression vector. The
two class 1 compounds 6 and 7, which include the
shorter linking arm (comprising three bonds), did not
alkylate the ER hormone-binding site, whereas with the
unique exception of the two tertiary bromoacetamides
15 and 22, all members of classes 2-5 (including one
iodide and five haloacetamides, whose linking arm
comprises six or nine bonds) were able to effectively
alkylate the ERR hormone-binding site. The inefficacy
of 15 and 22, which only differ from active compounds
16 and 20 by a CH2Cl to CH2Br and a NH to NCH3
change in the haloacetamido function, respectively, was
astonishing. The effect of the NH to NCH3 change found
with class 4 bromoacetamides was not observed with
classes 2 and 5 chloroacetamides, since 16 and 29 were
found to be at least as effective as 13 and 27, respec-
tively, for alkylating ER. These results indicate that,
depending on the compound class, the positioning of the
compound within the hormone-binding pocket would be
strongly influenced by discrete compound modifications.
For classes 2 and 5 chloride compounds, the NH to
NCH3 change would have no marked effect on the
compound positioning; whereas the Cl to Br change for
the class 2 tertiary chloroacetamide 16 and the NH to
NCH3 change for the class 4 bromoacetamide 20 would
modify the compound positioning in the hormone-
binding pocket. The results suggest that starting from
the 11â-phenylestradiol structure:
(i) A three-bond linking arm is insufficient for the
electrophilic center to be localized near a reactive
nucleophilic amino acid residue in the hormone-binding
pocket. The absence of ERR affinity labeling by class 1
compounds 6 and 7 probably resulted from the short
linking arm in these compounds and not from their type
of electrophilic functionality, since 11â-(bromoacetami-
dobutyl)estradiol, whose distance between the electro-

Steroidal Affinity Labels of ERR
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 623
ER; (ii) for bromoacetamides 15 and 22, there would be
no covalent attachment site in native ER, whereas in
MMTS-modified ER non-cysteinyl residues would pro-
vide such sites; and (iii) for chloroacetamides 13 and
16 and bromoacetamide 20, both cysteinyl residues and
non-cysteinyl residues in native ER would be the
covalent attachment sites, or cysteinyl residues in native
ER and non-cysteinyl residues in MMTS-modified ER
would be the covalent attachment sites. Since the lamb
ERR hormone-binding domain has >95% identity with
its human counterpart32 with only one nonconservative
change (R503 in lamb ER is changed to Q502 in human
ERR) and perfect conservation of the four cysteines
(C382, C418, C448, and C531 in lamb ERR), both the
crystal structure and the results of the human ERR
affinity labeling studies with tamoxifen aziridine could
likely be extended to the lamb ER. If we assume that
the binding mode of electrophilic 11â-estradiol deriva-
tives to ERRs would be similar to those of raloxifene9
and 4-hydroxytamoxifen10 with the 11â-substituent
acting as the aminoethoxyphenyl group of the two
nonsteroidal antiestrogens, then the cysteinyl residues
involved in the covalent attachment sites of 11â-estra-
diol derivatives would probably be homologues of human
ERR C381 and/or C530. This was deduced from the fact
that these two residues: (i) were located in structural
elements of the hormone-binding domain (helix 5/helix
6 and helix 11/helix 12 loop, respectively) which contact
the aminoethoxy chain of nonsteroidal antiestrogens;9
and (ii) constituted the exclusive covalent attachment
sites of tamoxifen aziridine.17,18 The lamb homologues
of human ERR C447 and C417 did not seem to be
involved since C447, located at the core of the hormone-
binding domain, was not accessible to iodoacetic acid,33
a very small thiol reagent, and since C417 was found
to be a covalent attachment site only for reactive
estrogenic 17R-estradiol derivatives.21 The assumption
that the homologues of human ERR C381 and/or C530
would probably be the covalent attachment sites of the
electrophilic 11â-estradiol derivatives to lamb ERR is
in accordance with the fact that human ERR C530 was
recently found to be the major covalent attachment site
of two aziridine ligands directly obtained from iodides
7 and 17 and closely related to haloacetamides 13, 15,
and 16 and to bromoacetamides 20 and 22, respec-
tively.34 The presence of many potential non-cysteinyl
attachment sites for the electrophiles, e.g., 12 conserved
histidyl residues and 9 conserved lysyl residues in both
lamb and human ERR hormone-binding domains, makes
it difficult to determine which of these residues would
constitute the non-cysteinyl attachment sites; some of
these nucleophilic residues located close to C381 and
C530 may be the potential sites.
from those that would be obtained using unmodified
hormone-binding domain.
In conclusion, we were successful in preparing a series
of electrophilic 11â-arylestradiol derivatives which are
ERR affinity labeling agents. The covalent attachment
sites of these electrophiles have not yet been identified.
However, both cysteinyl (probably homologues of human
ERR C381 and/or C530) and non-cysteinyl residues
located on the â-side and remote from C-11 of the steroid
(distance > “seven bonds”) appear to be the electrophile
covalent attachment sites. Studies involving the expres-
sion of recombinant ERR hormone-binding domain are
under way to identify, by mass spectrometry analysis,
the covalent attachment sites of the 17R- and 11â-
estradiol derivatives we developed. The expected results
could be used, in conjunction with the three-dimensional
model of the human ERR hormone-binding domain,
established from crystallized ER fragments, to model
the interaction of 17R- and 11â-derivatives with ERR.
Such studies should enhance the knowledge of ERR/
ligand interactions, which result in full or defective
activation of ERR, thus accounting for the estrogen or
antiestrogen activity of ER ligands.
Ma ter ia ls a n d Meth od s
Ch em ica l Syn th esis a n d Ch a r a cter iza tion of 11â-
Su bstitu ted Estr a d iol Der iva tives. All solvents and re-
agents were used as received from commercial sources. Reac-
tion progress was monitored using Merck silica gel 60 F254 TLC
plates (thickness 0.5 mm). Column chromatography was
carried out on Merck silica gel 60H (particle size < 60 µm).
Melting points (mp) were determined on a Kofler apparatus
and were uncorrected. Infrared (IR) spectra of compounds were
recorded on either a Nicolet 5SX or a Perkin-Elmer 580
spectrophotometer as chloroform solutions or Nujol suspen-
sions. Ultraviolet (UV) spectra were recorded on a Perkin-
Elmer lambda 9 or on a Varian Cary 2200 spectrophotometer.
Proton nuclear magnetic resonance (NMR) spectra were
recorded at 300 MHz on a Bruker AC300 spectrometer. Unless
otherwise mentioned, compounds were dissolved in deuterio-
chloroform (with 1-2 drops of deuteriopyridine in a few cases).
Chemical shifts (δ) were expressed downfield from tetrameth-
ylsilane (used as internal standard). Mass spectra were
obtained on the following mass spectrometers: Finnigan 4500
(EI, 70 eV); Autospec E (LSIMS) or ZAB-HFQ (FAB) (Micro-
mass Ldt). Combustion analyses were carried out by the
Analytical Department of Hœchst Marion Roussel (Romain-
ville) and are within (0.4%, unless otherwise noted. The usual
workup involved dilution of the reaction mixture in an aqueous
solution (generally 2 M sodium bicarbonate, unless otherwise
stated) followed by extraction with an organic solvent (ethyl
acetate, unless otherwise mentioned), washing with small
portions of brine, and drying over magnesium sulfate. After
filtration, the solvent was evaporated to dryness under reduced
pressure.
11â-[4-(2-Hyd r oxyeth oxy)p h en yl]estr a -4,9-d ien e-3,17-
d ion e (2). Under an argon atmosphere, a 50% oily suspension
of sodium hydride (2.78 g, 57.93 mmol) was added to a
suspension of phenolic compound 123 (15 g, 41.38 mmol) in
DMF (200 mL). After 25 min stirring, a brown solution was
obtained. (2-Bromoethoxy)-tert-butyldimethylsilane (19.8 g,
82.76 mmol) in DMF (110 mL) was added to the latter and
the mixture was stirred for 22 h at room temperature; 6 N
hydrochloric acid (50 mL) was added and the resulting solution
was stirred for an additional hour. The usual workup and then
chromatography (EtOAc-cyclohexane, 8:2) afforded 10.3 g of
pure primary alcohol 2 (61%) as an amorphous solid, which
was crystallized (CH2Cl2-isopropyl ether) to yield 9.51 g (57%)
of whitish crystals: mp 201 °C; IR 3605, 1735, 1658, 1609,
1582, 1509 cm-1; NMR 0.56 (s, 3H, CH3-18), 2.04 (t, 1H, OH),
As revealed by the striking changes in the ER
alkylating properties of bromoacetamides 15 and 22,
methylthiolation of cysteine residues (presumably all
except the C447 homologue) changed the accessibility
of nucleophilic amino acids in the hormone-binding
pocket. Presently we do not know whether the modifica-
tion of cysteines induced only local or remote changes
in the ER structure. Nevertheless, this effect suggests
that models of the ERR hormone-binding domain bound
to estrogen or antiestrogen, which were established from
S-carboxymethylated crystallized domain, may differ

624 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Aliau et al.
3.96 (m, 2H, CH2-OH), 4.05 (d, 2H, CH2-OΦ), 4.39 (d, 1H,
H-11), 5.80 (s, 1H, H-4), 6.84-7.10 (2d, 4H, C6H4) ppm.
nm, (EtOH/NaOH 0.1 N) 228 (ꢀ 16900), 280 (ꢀ 3100), 287 (ꢀ
3100), 301 (ꢀ 3000) nm. Anal. (C26H31O3 I) C, H, I.
3,17â-Bis[tetr a h yd r o(2H)-2-p yr a n yloxy]-11â-[4-(2-iod o-
eth oxy)p h en yl]estr a -1,3,5(10)-tr ien e (8). p-Toluenesulfonic
acid (510 mg, 2.96 mmol) was added to iodide 7 (4.39 g, 8.47
mmol) in THF (86 mL) and distilled dihydropyran (26 mL) and
the mixture was stirred for 1.5 h; then triethylamine (8.5 mL)
was added. The usual workup and then chromatography
(cyclohexane-EtOAc-NEt3, 90:10:0.1) gave 5.01 g of pure
compound 8 (86%) as a white amorphous solid: IR 1609, 1580,
1512, 1500 cm-1; NMR 0.35 (s, 3H, CH3-18), 3.35 (t, 2H, CH2I),
3.40-3.70 and 3.90 (m, 6H, H-11, H-17 and 2 CH2-OCO), 4.14
(t, 2H, CH2-OΦ), 4.58 (t) - 4.68 (t) - 4.82 (m) - 4.85 (m)
(1H, OCHO-17), 5.32 (t, 1H, OCHO-3), 6.62 (m, 1H, H-2), 6.79
(m, 1H, H-4), 6.84 (m, 1H, H-1), 6.62-6.98 (2d, 4H, C6H4) ppm.
2-[4-(3,17-Dioxoest r a -4,9-d ien -11â-yl)p h en oxy]et h yl
Met h a n esu lfon a t e (3). Methanesulfonyl chloride (2.8 mL,
36.17 mmol) was added to an ice-cooled solution of 2 (9.49 g,
23.34 mmol) in dichloromethane (200 mL) containing triethy-
lamine (5 mL, 36.1 mmol) and 4-(dimethylamino)pyridine (200
mg, 1.6 mmol). The mixture was then stirred for 2 h at 0 °C.
The usual workup and then recrystallization of the crude
product (CH2Cl2-isopropyl ether) gave 11.12 g of pure mesy-
late 3 (98%) as whitish crystals: mp 164 °C; IR 1736, 1659,
1609, 1584, 1509, 1360, 1176 cm-1; NMR 0.56 (s, 3H, CH3-
18), 3.09 (s, 3H, CH3-SO2), 4.22 and 4.56 (2m, 4H, CH2-OΦ
and CH2-OSO2), 4.39 (d, 1H, H-11), 5.80 (s, 1H, H-4), 6.82-
7.11 (2d, 4H, C6H4) ppm.
N-[2-[4-[3,17â-Bis[tetr a h yd r o(2H)-2-p yr a n yloxy]estr a -
1,3,5(10)-t r ie n -11â-yl]p h e n oxy]e t h yl]t r iflu or oa c e t a -
m id e (9). Under an argon atmosphere, a 50% oily suspension
of sodium hydride (230 mg, 4.8 mmol) was added to trifluo-
roacetamide (494 mg, 4.36 mmol) in THF (12 mL). After 20
min stirring, iodide 8 (1.5 g, 2.18 mmol) in DMF (12 mL) was
added and the mixture was stirred for 1.5 h at 65 °C. The usual
workup (NH4Cl) and then chromatography (cyclohexane-
EtOAc-NEt3, 75:25:01) gave 148 mg of the vinyl ether 10
(12%) derived from dehydroiodination of 8 and 956 mg of pure
trifluoroacetamide 9 (65%) as a white amorphous solid. Com-
2-[4-(3-Hyd r oxy-17-oxoestr a -1,3,5(10)-tr ien -11â-yl)p h e-
n oxy]eth yl Meth a n esu lfon a te (4). A mixture of acetyl
bromide (5.3 mL, 71 mmol) and acetic anhydride (10.6 mL,
112 mmol) was added to an ice-cooled solution of compound 3
(10.63 g, 21.94 mmol) in dichloromethane (100 mL). The
reaction mixture was stirred for 1.5 h, while allowing it to
reach room temperature, and then the solvents were removed
under reduced pressure. The remaining oil was solubilized in
THF (100 mL); the solution was cooled to 0 °C and methanol
(100 mL) was slowly added followed by 2 N sodium hydroxide
(74.5 mL). The mixture was stirred for 45 min at room
temperature, and then 6 N hydrochloric acid (31 mL) was
added. After concentration under reduced pressure, the usual
workup and then chromatography (CH2Cl2-EtOAc, 9:1) af-
forded first 1.75 g of 5 (16%) and then 7.87 g of pure mesylate
4 (74%) as amorphous solids. Compound 4: IR 3598, 1733,
1611, 1584, 1512, 1359, 1175 cm-1; NMR 0.44 (s, 3H, CH3-
18), 3.04 (s, 3H, CH3-SO2), 4.00 (t, 1H, H-11), 4.13 and 4.51
(2m, 4H, CH2-OΦ and CH2-OSO2), 6.49 (dd, 1H, H-2), 6.67
(d, 1H, H-4), 6.79 (d, 1H, H-1), 6.63-7.02 (2d, 4H, C6H4) ppm.
Compound 5: IR 1733, 1713, 1607, 1572, 1508, 1360, 1175
cm-1; NMR 1.03 (s, 3H, CH3-18), 2.83 (m, 2H, H-4), 3.11 (s,
3H, CH3-SO2), 4.24 and 4.59 (2m, 4H, CH2-OΦ and CH2-
OSO2), 6.82-7.11 (2d, 4H, C6H4) ppm.
pound 9: IR (Nujol) 1721, 1613, 1574, 1561, 1512, 1500 cm-1
;
NMR 0.36-0.37 (2s, 3H, CH3-18), 3.40-3.95 (m, 6H, H-11,
H-17 and 2 CH2-OCO), 3.72 (q, (t after D2O addition), 2H,
CH2-N), 4.00 (t, 2H, CH2-OΦ), 4.58-4.68 (2t, 1H, OCHO-
17), 5.31 (t, 1H, OCHO-3), 6.61 (m, 1H, H-2), 6.79 (m, 1H, H-4),
6.83 (m, 1H, H-1), 6.61-7.00 (2d, 4H, C6H4) ppm. Compound
10: IR 1645, 1622, 1605, 1580, 1574, 1507, 1499 cm-1; NMR
0.35 (s, 3H, CH3-18), 3.45-3.95 (m, 6H, H-11, H-17 and 2
CH2-OCO), 4.33 and 4.58-4.68 (d, 1H and m, 2H, OCHO-17
and vinyl CH2), 4.96-5.31 (2t, 1H, OCHO-3), 6.56 (m, 2H, H-2
and CH-OΦ), 6.79 (m, 2H, H-1 and H-4), 6.72-7.02 (2d, 4H,
C6H4) ppm; UV (EtOH) 233 (ꢀ 21200), 279 (ꢀ 3000), 285 (ꢀ 2600)
nm.
3,17â-Bis[tetr a h yd r o(2H)-2-p yr a n yloxy]-11â-[4-(2-a m i-
n oeth oxy)p h en yl]estr a -1,3,5(10)-tr ien e (11). Trifluoroac-
etamide 9 (1.06 g, 1.58 mmol) in methanol (5 mL) and 1 N
sodium hydroxide (3.1 mL) were stirred for 3 h at room
temperature. The usual workup (NH4Cl) yielded 886 mg of
primary amine 11 (97%) as a white amorphous solid: IR 3385,
1610, 1581, 1512, 1498 cm-1; NMR 0.36 (s, 3H, CH3-18), 3.42-
3.90 (m, 6H, H-11, H-17 and 2 CH2-OCO), 3.90 (t, 2H, CH2-
OΦ), 4.59-4.68 (2t, 1H, OCHO-17), 5.31 (t, 1H, OCHO-3), 6.60
(m, 1H, H-2), 6.78 (m, 1H, H-4), 6.84 (m, 1H, H-1), 6.63-6.97
(2d, 4H, C6H4) ppm.
2-Ch lor o-N-[2-[4-(3,17â-d ih yd r oxyestr a -1,3,5(10)-tr ien -
11â-yl)p h en oxy]eth yl]a ceta m id e (13). Bromoacetyl bro-
mide (64 µL, 0.74 mmol) in dichloromethane (4.5 mL) was
added over a 5-min period to primary amine 11 (406 mg, 0.705
mmol) in dichloromethane (4.5 mL) containing pyridine (62
µL, 0.785 mmol). After 1 h stirring at 0 °C, the usual workup
gave 488 mg of bromoacetamide, whose 3- and 17â-hydroxyls
were immediately deprotected: the crude product was solu-
bilized in methanol (3.2 mL) and THF (3.2 mL); 2 N hydro-
chloric acid (1 mL) was added and the mixture was stirred for
1.5 h at room temperature. The usual workup and then
chromatography (CH2Cl2-MeOH-NH4OH, 95:5:0.25) gave 220
mg of an amorphous solid containing a mixture (70:30) of
bromoacetamide 12 and chloroacetamide 13: NMR (DMSO-
d6) 3.86 (s, CH2Br) and 4.06 (s, CH2Cl); MS (LSIMS) 528(MH+),
527(M+) and 484(MH+), 483(M+), respectively. Conversion of
bromoacetamide 12 to chloroacetamide 13 was performed as
follows: a mixture of 12 and 13 (200 mg) in DMF (4.8 mL)
was stirred for 20 h at room temperature with lithium chloride
(160 mg, 3.78 mmol). Evaporation of the solvent under reduced
pressure and then chromatography afforded 110 mg of pure
chloroacetamide 13 (35%) as a white amorphous solid: IR
(Nujol) ≈3550, 1658, 1608, 1578, 1538, 1510 cm-1; NMR
(DMSO-d6) 0.18 (s, 3H, CH3-18), 3.43 (m, 3H, H-17 and CH2N),
2-[4-(3,17â-Dih yd r oxyestr a -1,3,5(10)-tr ien -11â-yl)p h e-
n oxy]eth yl Meth a n esu lfon a te (6). Sodium borohydride (638
mg, 16.87 mmol) was added to an ice-cooled solution of
compound 4 (8.18 g, 16.88 mmol) in methanol (81 mL) and
THF (81 mL). After 1 h stirring at 0 °C, acetone (6 mL) was
added, followed 5 min later by 0.5 N hydrochloric acid (until
to reach pH 2). The usual workup afforded 7.7 g of crude
product, which was purified by crystallization: the crude
product was solubilized in a hot mixture of methanol and
dichloromethane (1:1, 200 mL); isopropyl ether (300 mL) was
added and the resulting solution was concentrated under
reduced pressure, then cooled in an ice bath. Pure mesylate 6
(7.25 g, 88%) was obtained as white crystals: mp 211 °C; IR
(Nujol) 3374, 1620, 1609, 1580, 1510, 1497, 1348, 1175 cm-1
;
NMR 0.34 (s, 3H, CH3-18), 3.04 (s, 3H, CH3-SO2), 3.70 (t, 1H,
H-17), 3.93 (t, 1H, H-11), 4.12 and 4.50 (2m, 4H, CH2-OΦ and
CH2-OSO2), 6.48 (dd, 1H, H-2), 6.66 (d, 1H, H-4), 6.78 (d, 1H,
H-1), 6.62-7.01 (2d, 4H, C6H4) ppm; UV (EtOH) 228 (ꢀ 17400),
281 (ꢀ 3500), 286 (ꢀ 3650) nm, (EtOH/NaOH 0.1 N) 228 (ꢀ
16900), 280 (ꢀ 3200), 287 (ꢀ 3300), 302 (ꢀ 3100) nm. Anal.
(C27H34O6S) C, H, S.
11â-[4-(2-Iod oet h oxy)p h en yl]est r a -1,3,5(10)-t r ien e-3,-
17â-d iol (7). Sodium iodide (4.94 g, 32.95 mmol) and mesylate
6 (6.41 g, 13.17 mmol) were dissolved in 2-butanone (214 mL).
The mixture was stirred under reflux for 17 h, then the solvent
was evaporated under vacuum, and the residue was dissolved
in a mixture of ethyl acetate and DMF (9:1, 1.5 L). The usual
workup (Na2S2O3) and then crystallization (EtOH) of the crude
product afforded 4.94 g of pure iodide 7 (72%) as white
crystals: mp 242 °C; IR (Nujol) 3343, 1611, 1579, 1507 cm-1
;
NMR 0.34 (s, 3H, CH3-18), 3.34 (t, 2H, CH2I), 3.72 (t, 1H,
H-17), 3.94 (t, 1H, H-11), 4.13 (t, 2H, CH2-OΦ), 6.49 (dd, 1H,
H-2), 6.66 (d, 1H, H-4), 6.79 (d, 1H, H-1), 6.62-7.00 (2d, 4H,
C6H4) ppm; UV (EtOH) 230 (ꢀ 16400), 281 (ꢀ 3500), 286 (ꢀ 3500)

Steroidal Affinity Labels of ERR
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 625
3.89 (m, 3H, H-11 and CH2-OΦ), 4.06 (s, 2H, CH2Cl), 4.40 (d,
1H, OH-17), 6.29 (dd, 1H, H-2), 6.45 (m, 1H, H-4), 6.66 (m,
1H, H-1), 6.60-6.95 (2d, 4H, C6H4), 8.41 (t, 1H, NH), 8.89 (s,
1H, OH-3) ppm. Anal. (C28H34NO4Cl) H, N, Cl; C: calcd 69.48,
found 68.6.
enesulfonic acid (338 mg, 1.78 mmol) was added to iodide 1724
(3.55 g, 6.33 mmol) in THF (34 mL) and distilled dihydropyran
(17 mL) and the mixture was stirred for 1.5 h at room
temperature; triethylamine (6.4 mL) was then added. The
usual workup and then chromatography (cyclohexane-EtOAc-
NEt3, 90:10:0.1) gave 5.1 g of crude iodide 18 (quantitative
yield) as a colorless oil: IR 1610, 1581, 1512, 1498 cm-1; NMR
0.37 (s, 3H, CH3-18), 3.19 (t, 2H, CH2I), 3.40-3.70 and 3.80-
4.05 (m, 5H, H-11 and 2 CH2-OCO), 3.67 (m, 1H, H-17), 3.85
(t, 2H, CH2-OΦ), 4.59-4.68 (2m, 1H, OCHO-17), 5.31 (m, 1H,
OCHO-3), 6.61 (m, 1H, H-2), 6.77 (m, 1H, H-4), 6.85 (m, 1H,
H-1), 6.61-6.96 (2d, 4H, C6H4) ppm.
N -[5-[4-[3,17â-Bis[t e t r a h yd r o(2H )-2-p yr a n yloxy]e s-
t r a -1,3,5(10)-t r ie n -11â-y l]p h e n o x y ]p e n t y l]t r iflu o r o -
a ceta m id e (19). Under an argon atmosphere, a 50% oily
suspension of sodium hydride (37 mg, 0.77 mmol) was added
to trifluoroacetamide (118 mg, 1.05 mmol) in THF (2 mL). After
20 min stirring, iodide 18 (535 mg, 0.734 mmol) in DMF (2
mL) was added and the mixture was stirred for 5.5 h at room
temperature. The usual workup (NH4Cl) and then chroma-
tography (hexanes-EtOAc, 7:3) gave 107 mg of unreacted
iodide 18 (20%) along with 210 mg of pure trifluoroacetamide
19 (40%) as a white amorphous solid: IR 3340, 1727, 1610,
1576, 1512, 1498, 1180 cm-1; NMR 0.37 (s, 3H, CH3-18), 3.70-
4.00 (m, 10H, H-11, H-17, 2 CH2-OCO, CH2-N and CH2-
OΦ), 4.59-4.67 (m, 1H, OCHO-17), 5.31 (m, 1H, OCHO-3),
6.31 (s, 1H, NH), 6.61 (m, 1H, H-2), 6.70-6.90 (m, 2H, H-1
and H-4), 6.61-6.97 (2d, 4H, C6H4) ppm.
2-Br om o-N-[5-[4-(3,17â-d ih yd r oxyestr a -1,3,5(10)-tr ien -
11â-yl)p h en oxy]p en tyl]a ceta m id e (20). Trifluoroacetamide
19 (550 mg, 0.77 mmol) solubilized in THF (5 mL) and
methanol (3 mL) was stirred for 2.5 h at room temperature
with 1 N sodium hydroxide (1.54 mL). The usual workup (NH4-
Cl) yielded 480 mg of primary amine (quantitative yield) as a
white amorphous solid. Bromoacetyl bromide (45 µL, 0.52
mmol) in dichloromethane (4 mL) was added over a 5-min
period to primary amine (306 mg, 0.495 mmol) in dichlo-
romethane (4 mL) containing pyridine (45 µL, 0.551 mmol).
After 1 h stirring at 0 °C, the usual workup (no brine washing)
gave 328 mg of bromoacetamide whose 3- and 17â-hydroxyls
were immediately deprotected: the crude product was solu-
bilized in methanol (2.5 mL) and THF (2.5 mL); 2 N hydro-
bromic acid (0.66 mL) was added and the mixture was stirred
for 1.5 h at room temperature. The usual workup (no brine
washing) and then chromatography (EtOAc-hexanes-MeOH-
NH4OH, 80:20:2.5:0.25) gave 160 mg of pure bromoacetamide
20 (57%) as an amorphous solid: IR (Nujol) ≈3550, 1658, 1612,
1580, 1543, 1511, 1505 cm-1; NMR (DMSO-d6) 0.18 (s, 3H,
CH3-18), 3.07 (q, 2H, CH2N), 3.46 (m, 1H, H-17), 3.81 (s, 2H,
CH2Br), 3.83 (m, 3H, H-11 and CH2-OΦ), 4.40 (d, 1H, OH-
17), 6.29 (dd, 1H, H-2), 6.45 (d, 1H, H-4), 6.70 (d, 1H, H-1),
6.62-6.94 (2d, 4H, C6H4), 8.24 (t, 1H, NH), 8.89 (s, 1H, OH-3)
ppm; UV (EtOH) 230 (ꢀ 17400), 282 (ꢀ 3700), 288 (ꢀ 3600) nm,
(EtOH/NaOH 0.1 N) 230 (ꢀ 17000), 281 (ꢀ 3100), 288 (ꢀ 3200),
301 (ꢀ 2900) nm. Anal. (C31H40NO4Br) H, N; C: calcd 65.26,
found 64.8; Br: calcd 14.00, found 13.0.
N-[2-[4-[3,17â-Bis[tetr a h yd r o(2H)-2-p yr a n yloxy]estr a -
1,3,5(10)-tr ien -11â-yl]p h en oxy]eth yl]-N-m eth yltr iflu or o-
a ceta m id e (14). Under an argon atmosphere, a 50% oily
suspension of sodium hydride (38 mg, 0.8 mmol) was added
to N-methyltrifluoroacetamide (139 mg, 1.09 mmol) in THF
(2 mL). After 20 min stirring, iodide 8 (0.5 g, 0.728 mmol) in
DMF (2.5 mL) was added and the mixture was stirred for 2 h
at 80 °C. The usual workup (NH4Cl) and then chromatography
(cyclohexane-EtOAc-NEt3, 80:20:0.1) gave 141 mg of vinyl
ether 10 (35%) derived from dehydroiodination of compound
8 and 245 mg of pure amide 14 (49%) as a white amorphous
solid. Compound 14: IR 1692, 1610, 1582, 1576, 1512, 1498
cm-1; NMR 0.35-0.36 (2s, 3H, CH3-18), 3.13-3.14-3.25-3.26
(4s, 3H, CH3-N), 3.43-4.15 (m, 10H, H-11, H-17, 2 CH2-OCO,
CH2-N and CH2-OΦ), 4.58-4.68 (2t, 1H, OCHO-17), 5.31 (t,
1H, OCHO-3), 6.60 (m, 1H, H-2), 6.78 (m, 1H, H-4), 6.84 (m,
1H, H-1), 6.60-6.98 (2d, 4H, C6H4) ppm.
2-Br om o-N-[2-[4-(3,17â-d ih yd r oxyestr a -1,3,5(10)-tr ien -
11â-yl)p h en oxy]eth yl]-N-m eth yla ceta m id e (15). Trifluo-
roacetamide 14 (652 mg, 0.951 mmol) in THF (7.5 mL) and 1
N sodium hydroxide (1.22 mL) were stirred for 1.5 h at room
temperature. The usual workup (NH4Cl) yielded 561 mg of
secondary amine (quantitative yield) as an oil, which was
immediately bromoacetylated: the crude product was solubi-
lized in dichloromethane (6 mL) containing pyridine (85 µL,
1.06 mmol) and bromoacetyl bromide (87 µL, 1.0 mmol) in
dichloromethane (6 mL) was added over a 5-min period. After
1 h stirring at 0 °C, the usual workup gave 636 mg of
bromoacetamide (94%), whose 3- and 17â-hydroxyls were
immediately deprotected: the crude product was solubilized
in methanol (4 mL) and THF (4 mL); 2 N hydrobromic acid (2
mL) was added and the mixture was stirred for 1.5 h at room
temperature. The usual workup (no brine washing) and then
chromatography (CH2Cl2-MeOH-NH4OH, 95:5:0.25) gave
438 mg of pure bromoacetamide 15 (85%) as an amorphous
solid. Crystallization (CH2Cl2-isopropyl ether) gave 352 mg
of compound 15 (68%) as white crystals: mp 158 °C; IR
(Nujol) ≈3550, 1612, 1584, 1512, 1502 cm-1; NMR (DMSO-
d6) 0.18 (s, 3H, CH3-18), 2.88-3.08 (2s, 3H, CH3N), 3.47 (m,
3H, H-17), 3.59-3.67 (2t, 2H, CH2N), 3.86 (t, 1H, H-11),
3.95-4.04 (2t, 2H, CH2-OΦ), 4.11-4.16 (2s, 2H, CH2Br), 4.41
(d, 1H, OH-17), 6.29 (dd, 1H, H-2), 6.45 (d, 1H, H-4), 6.67 (m,
1H, H-1), 6.67-6.95 (2d, 4H, C6H4), 8.89 (s, 1H, OH-3) ppm;
UV (EtOH) 230 (ꢀ 19300), 281 (ꢀ 3500), 286 (ꢀ 3500) nm,
(EtOH/NaOH 0.1 N) 230 (ꢀ 18500), 280 (ꢀ 3000), 287 (ꢀ 3100),
302 (ꢀ 2900) nm; MS (EI) 541(M+). Anal. (C29H36NO4Br) C, H,
N, Br.
2-Ch lor o-N-[2-[4-(3,17â-d ih yd r oxyestr a -1,3,5(10)-tr ien -
11â-yl)p h en oxy]eth yl]-N-m eth yla ceta m id e (16). A solu-
tion of bromoacetamide 15 (165 mg, 0.304 mmol) and lithium
chloride (129 mg, 3.05 mmol) in DMF (4 mL) was stirred for
20 h at room temperature and then the solvent was removed
under reduced pressure. The usual workup and then chroma-
tography (CH2Cl2-MeOH-NH4OH, 95:5:0.25) afforded 135 mg
of pure chloroacetamide 16 (89%) as a white amorphous solid.
Crystallization (CH2Cl2-isopropyl ether) gave 123 mg of
compound 16 (81%) as white crystals: mp 229 °C; IR (Nujol)
≈3550, 1640, 1612, 1582, 1510 cm-1; NMR (DMSO-d6) 0.18
(s, 3H, CH3-18), 2.88-3.05 (2s, 3H, CH3N), 3.48 (m, 3H, H-17),
3.60-3.67 (2t, 2H, CH2N), 3.95-4.02 (2t, 2H, CH2-OΦ), 3.87
(m, 1H, H-11), 4.36-4.40 (2s, 2H, CH2Cl), 4.40 (d, 1H, OH-
17), 6.29 (dd, 1H, H-2), 6.45 (d, 1H, H-4), 6.66 (d, 1H, H-1),
6.66-6.96 (2d, 4H, C6H4), 8.88 (s, 1H, OH-3) ppm; UV
(EtOH) 228 (ꢀ 18000), 280 (ꢀ 3400), 286 (ꢀ 3500) nm, (EtOH/
NaOH 0.1 N) 229 (ꢀ 18000), 280 (ꢀ 3000), 287 (ꢀ 3100), 301 (ꢀ
2900) nm. Anal. (C29H36NO4Cl) H, N, Cl; C: calcd 69.93, found
69.4.
N-[5-[4-[3,17â-Bis[tetr a h yd r o(2H)-2-p yr a n yloxy]estr a -
1,3,5(10)-tr ien -11â-yl]p h en oxy]p en tyl]-N-m eth yltr iflu o-
r oa ceta m id e (21). Under an argon atmosphere, a 50% oily
suspension of sodium hydride (108 mg, 2.25 mmol) was added
to trifluoroacetamide 19 (1.42 g, 1.99 mmol) in DMF (34 mL).
After 20 min stirring at room temperature, methyl iodide (243
µL, 3.9 mmol) in THF (2 mL) was added and the mixture was
stirred for 1 h at room temperature. The usual workup (NH4-
Cl) and then chromatography (cyclohexane-EtOAc-NEt3, 75:
25:0.1) gave 1.423 g of pure trifluoroacetamide 21 (98%) as a
white amorphous solid: IR 1691, 1610, 1580, 1512, 1498 cm-1
;
NMR 0.36-0.37 (CH3-18), 3.00-3.10 (2s, 3H, CH3-N), 3.30-
3.95 (m, 10H, H-11, H-17, 2 CH2-OCO, CH2-N and CH2-
OΦ), 4.58-4.67 (2t, 1H, OCHO-17), 5.31 (t, 1H, OCHO-3), 6.60
(m, 1H, H-2), 6.78 (m, 1H, H-4), 6.85 (m, 1H, H-1), 6.60-6.96
(2d, 4H, C6H4) ppm.
3,17â-Bis[t et r a h yd r o(2H )-2-p yr a n yloxy]-11â-[4-[(5-io-
d op en tyl)oxy]p h en yl]estr a -1,3,5(10)-tr ien e (18). p-Tolu-
2-Br om o-N-[5-[4-(3,17â-d ih yd r oxyestr a -1,3,5(10)-tr ien -
11â-yl)p h en oxy]p en tyl]-N-m eth yla ceta m id e (22). Trifluo-

626 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Aliau et al.
roacetamide 21 (464 mg, 0.637 mmol) solubilized in THF (3
mL) and methanol (3 mL) was stirred for 2.5 h at room
temperature with 1 N sodium hydroxide (0.9 mL). The usual
workup (NH4Cl) yielded 431 mg of secondary amine (quantita-
tive yield) as a white amorphous solid. Bromoacetyl bromide
(58 µL, 0.67 mmol) in dichloromethane (5 mL) was added over
a 5-min period to the crude secondary amine (431 mg) in
dichloromethane (5 mL) containing pyridine (57 µL, 0.7 mmol)
and the mixture was stirred for 1 h at 0 °C. The usual workup
(no brine washing) gave 405 mg of bromoacetamide whose 3-
and 17â-hydroxyls were immediately deprotected: the crude
product was solubilized in methanol (3 mL) and THF (3 mL);
2 N hydrobromic acid (0.8 mL) was added and the mixture
was stirred for 1.5 h at room temperature. The usual workup
(no brine washing) and then chromatography (EtOAc-hex-
anes-MeOH-NH4OH, 80:20:2.5:0.25) gave 201 mg of pure
bromoacetamide 22 (54%) as an amorphous solid: IR (Nujol)
≈3550, 1635, 1611, 1580, 1510, 1502 cm-1; NMR (DMSO-d6)
0.19 (s, 3H, CH3-18), 2.81-2.99 (2s, 3H, CH3-N), 3.20-3.50
(m, 3H, CH2N and H-17), 3.84 (m, 3H, H-11 and CH2-OΦ),
4.07-4.10 (2s, 2H, CH2Br), 4.34-4.39 (2d, 1H, OH-17), 6.29
(dd, 1H, H-2), 6.45 (d, 1H, H-4), 6.69 (d, 1H, H-1), 6.62-6.94
(2d, 4H, C6H4), 8.89 (s, 1H, OH-3) ppm; UV (EtOH) 230 (ꢀ
19900), 282 (ꢀ 3800), 288 (ꢀ 3700) nm, (EtOH/NaOH 0.1 N)
230 (ꢀ 19500), 281 (ꢀ 3200), 288 (ꢀ 3300), 302 (ꢀ 2900) nm; MS
(FAB) 583(M+), 503(M+ - Br). Anal. (C32 H42 NO4 Br) C, H,
N; Br: calcd 13.67, found 13.1.
2-Ch lor o-N-[3-[4-(3,17â-d ih yd r oxyestr a -1,3,5(10)-tr ien -
11â-yl)p h en yl]-2-p r op yn -1-yl]a cet a m id e (27). Secondary
trifluoroacetamide 26 (504 mg, 0.757 mmol) in THF (5 mL)
and methanol (5 mL) was stirred for 4.5 h at room temperature
with 1 N sodium hydroxide (2.25 mL). The usual workup (NH4-
Cl) yielded 422 mg of secondary amine (98%) as a yellow
amorphous solid. Chloroacetyl chloride (37 µL, 0.457 mmol)
in dichloromethane (4 mL) was added over a 5-min period to
an ice-cooled solution of secondary amine (248 mg, 0.435 mmol)
in dichloromethane (4 mL) containing pyridine (10 µL, 0.484
mmol). After 1 h stirring at 0 °C, the usual workup gave 276
mg of chloroacetamide (98%) whose 3- and 17â-hydroxyls were
immediately deprotected: the crude product was solubilized
in methanol (2.5 mL) and THF (2.5 mL); 2 N hydrochloric acid
(1 mL) was added and the mixture was stirred for 1.5 h at
room temperature. The usual workup and then chromatogra-
phy (CH2Cl2-MeOH-NH4OH, 95:5:0.5; then EtOAc-hex-
anes-MeOH-NH4OH, 80:20:2.5:0.25) gave 172 mg of pure
chloroacetamide 27 (81% from 26) as an amorphous solid.
Crystallization (CH2Cl2-isopropyl ether) afforded 156 mg of
27 (74%) as white crystals: mp 168 °C; IR (Nujol) ≈3550, 1663,
1611, 1583, 1535, 1506 cm-1; NMR (DMSO-d6) 0.15 (s, 3H,
CH3-18), 3.49 (m, 1H, H-17), 3.94 (t, 1H, H-11), 4.09 (s, 2H,
CH2Cl), 4.11 (d, 2H, CH2N), 4.43 (d, 1H, OH-17), 6.28 (dd, 1H,
H-2), 6.46 (d, 1H, H-4), 6.67 (d, 1H, H-1), 7.05-7.13 (2d, 4H,
C6H4), 8.71 (t, 1H, NH), 8.91 (s, 1H, OH-3) ppm; UV (EtOH)
250 (ꢀ 23500), 261 (ꢀ 21800), 278 (ꢀ 2800), 284 (ꢀ 2700), 290 (ꢀ
2400) nm, (EtOH/NaOH 0.1 N) 250 (ꢀ 29700), 260 (ꢀ 25000),
302 (ꢀ 2700) nm; MS (El) 477(MH+), 442(M+ - Cl). Anal.
(C29H32NO3Cl) H, N, Cl; C: calcd 72.87, found 72.0.
3-[4-[3,17â-Bis[tetr ah ydr o(2H)-2-pyr an yloxy]estr a-1,3,5-
(10)-tr ien -11â-yl]p h en yl]-2-p r op yn -1-ol (24). Butyllithium
in hexanes (11.8 mL, 1.6 M) was slowly added to a stirred
solution of ethynylphenyl steroid 2326 (7.2 g, 13.32 mmol) in
THF (60 mL) at -30 °C. Five min after the addition was
complete, paraformaldehyde (662 mg, 22 mmol) was added and
the mixture was allowed to reach room temperature, while
stirring for 2 h. The usual workup (NH4Cl) and then chroma-
tography (cyclohexane-EtOAc, 8:2) gave 6.64 g of propargyl
alcohol 24 (87%) as a white solid. Crystallization (Et2O)
afforded 5.06 g of 24 (67%) as white crystals: mp 234 °C; IR
3608, 1607, 1574, 1555, 1506, 1498 cm-1; NMR 0.31-0.33 (2s,
3H, CH3-18), 3.40-3.60 and 3.80-4.00 (2m, 4H, 2 CH2-OCO),
3.68 (dt, 1H, H-17), 4.43 (s, 2H, CH2-OH), 4.58-4.67 (2m, 1H,
OCHO-17), 5.32 (m, 1H, OCHO-3), 6.60 (dd, 1H, H-2), 6.79
(m, 2H, H-1 and H-4), 7.04-7.16 (2d, 4H, C6H4) ppm.
3-[4-[3,17â-Bis[tetr ah ydr o(2H)-2-pyr an yloxy]estr a-1,3,5-
(10)-tr ien -11â-yl]ph en yl]-2-pr opyn -1-yl Meth an esu lfon ate
(25). Triethylamine (0.19 mL, 1.39 mmol) and methanesulfonyl
chloride (0.11 mL, 1.39 mmol) were added to an ice-cooled
solution of propargyl alcohol 24 (508 mg, 0.89 mmol) in
dichloromethane (8 mL) containing 4-(dimethylamino)pyridine
(7 mg, 0.06 mmol) and the mixture was then stirred for 1 h at
0 °C. The usual workup and then chromatography (cyclohex-
ane-EtOAc-NEt3, 70:30:0.1) gave 400 mg of pure mesylate
25 (69%) as a white amorphous solid: IR 2220, 1606, 1575,
1507, 1498, 1366, 1175 cm-1; NMR 0.33 (s, 3H, CH3-18), 3.14
(s, 3H, CH3-SO2), 3.55 and 3.93 (2m, 5H, H-11 and 2CH2-
OCO), 3.69 (t, 1H, H-17), 4.57 and 4.66 (2m, 1H, OCHO-17),
5.31 (m, 1H, OCHO-3), 6.60 (dd, 1H, H-2), 6.80 (m, 2H, H-1
and H-4), 7.08-7.18 (2d, 4H, C6H4) ppm.
N-[3-[4-[3,17â-Bis[tetr a h yd r o(2H)-2-p yr a n yloxy]estr a -
1,3,5(10)-tr ien -11â-yl]p h en yl]-2-p r op yn -1-yl]tr iflu or oa c-
eta m id e (26). Under an argon atmosphere, a 50% oily
suspension of sodium hydride (364 mg, 7.59 mmol) was added
to trifluoroacetamide (780 mg, 6.9 mmol) in THF (18 mL). After
20 min stirring, mesylate 25 (2.27 g, 3.5 mmol) in DMF (18.5
mL) was added and the mixture was stirred for 1 h at 60 °C.
The usual workup (NH4Cl) and then chromatography (cyclo-
hexane-EtOAc-NEt3, 70:30:0.1) gave 1.24 g of pure secondary
trifluoroacetamide 26 (53%) as a whitish amorphous solid: IR
3440, 1730, 1606, 1575, 1539, 1534, 1507, 1498 cm-1; NMR
0.34 (s, 3H, CH3-18), 3.45-3.60 and 3.85-4.00 (2m, 5H, H-11,
and 2 CH2-OCO), 3.68 (t, 1H, H-17), 4.33 (d, 2H, CH2-N),
4.58-4.67 (2m, 1H, OCHO-17), 5.31 (m, 1H, OCHO-3), 6.59
(dd, 1H, H-2), 6.80 (m, 2H, H-1 and H-4), 7.06-7.16 (2d, 4H,
C6H4) ppm.
N-[3-[4-[3,17â-Bis[tetr a h yd r o(2H)-2-p yr a n yloxy]estr a -
1,3,5(10)-tr ien -11â-yl]p h en yl]-2-p r op yn -1-yl]-N-m eth yltr i-
flu or oa ceta m id e (28). Under an argon atmosphere, a 50%
oily suspension of sodium hydride (52 mg, 1.09 mmol) was
added to secondary trifluoroacetamide 26 (660 mg, 0.991
mmol) in DMF (15 mL). After 10 min stirring, methyl iodide
(123 µL, 1.98 mmol) in THF (1.2 mL) was added and the
mixture was stirred for 1 h at room temperature. The usual
workup (NH4Cl) and then chromatography (cyclohexane-
EtOAc-NEt3, 80:20:0.1) gave 569 mg of pure tertiary trifluo-
roacetamide 28 (84%) as a white amorphous solid: IR 2235,
2215, 1696, 1617, 1607, 1575, 1555, 1507, 1498 cm-1; NMR
0.34 (s, 3H, CH3-18), 3.14-3.22 (2s, 3H, CH3N), 3.45-3.65 and
3.85-4.00 (2m, 5H, H-11, and 2CH2-OCO), 3.68 (m, 1H,
H-17), 4.36-4.45 (2s, 2H, CH2N), 4.58-4.66 (2m, 1H, OCHO-
17), 5.31 (m, 1H, OCHO-3), 6.59 (dd, 1H, H-2), 6.80 (m, 2H,
H-1 and H-4), 7.06-7.16 (2d, 4H, C6H4) ppm.
2-Ch lor o-N-[3-[4-(3,17â-d ih yd r oxyestr a -1,3,5(10)-tr ien -
11â-yl)p h en yl]-2-p r op yn -1-yl]-N-m et h yla cet a m id e (29).
Tertiary trifluoroacetamide 28 (626 mg, 0.921 mmol) in THF
(5.5 mL) and methanol (5.5 mL) was stirred for 1 h at room
temperature with 1 N sodium hydroxide (1.85 mL). The usual
workup (NH4Cl) yielded 525 mg of secondary amine (98%) as
a yellow amorphous solid. Chloroacetyl chloride (44 µL, 0.57
mmol) in dichloromethane (4 mL) was added over a 5-min
period to an ice-cooled solution of secondary amine (317 mg,
0.543 mmol) in dichloromethane (4 mL) containing pyridine
(84 µL, 0.604 mmol) and the mixture was stirred for 1 h at 0
°C. The usual workup gave 347 mg of chloroacetamide (97%)
whose 3- and 17â-hydroxyls were immediately deprotected:
the crude product was solubilized in methanol (2.5 mL) and
THF (2.5 mL); 2 N hydrochloric acid (1.2 mL) was added and
the mixture was stirred for 2 h at room temperature. The usual
workup and then chromatography (CH2Cl2-MeOH-NH4OH,
95:5:0.25) gave 213 mg of pure chloroacetamide 29 (78% from
28) as an amorphous solid. Crystallization (CH2Cl2-isopropyl
ether) afforded 203 mg of chloroacetamide 29 (74%) as white
crystals: mp 240 °C; IR (Nujol) ≈3580, ≈3520, 1645, 1620,
1580, 1501 cm-1; NMR (DMSO-d6) 0.16 (s, 3H, CH3-18), 2.92-
3.06 (2s, 3H, CH3N), 3.49 (m, 1H, H-17), 3.93 (t, 1H, H-11),
4.36-4.43 (2s, 5H, CH2Cl, CH2N and OH-17), 6.28 (dd, 1H,
H-2), 6.46 (d, 1H, H-4), 6.66 (d, 1H, H-1), 7.06-7.14 (2d, 4H,
C6H4), 8.71 (t, 1H, NH), 8.91 (s, 1H, OH-3) ppm; UV (EtOH)
251 (ꢀ 26400), 261 (ꢀ 23900), 277 (ꢀ 3200), 284 (ꢀ 3100), 289 (ꢀ

Steroidal Affinity Labels of ERR
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4 627
(8) Tanenbaum, D. M.; Wang, Y.; Williams, S. P.; Sigler, P.
Crystallographic comparison of the estrogen and progesterone
receptor’s ligand binding domains. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 5998-6003.
(9) Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard, R. E.;
Bonn, T.; Engstro¨m, O.; O¨ hman, L.; Greene, G. L.; Gustafsson,
J .-A.; Carlquist, M. Molecular basis of agonism and antagonism
in the estrogen receptor. Nature 1997, 389, 753-758.
(10) Shiau, A. K.; Barstad, D.; Loria, P. M.; Cheng, L.; Kushner, P.
J .; Agard, D. A.; Greene, G. L. The structural basis of estrogen
receptor/coactivator recognition and the antagonism of this
interaction by tamoxifen. Cell 1998, 95, 927-937.
(11) Bourguet, W.; Ruff, M.; Chambon, P.; Gronemeyer, H.; Moras,
D. Crystal structure of the ligand-binding domain of the human
nuclear receptor RXRR. Nature 1995, 375, 377-382.
(12) Renaud, J .-P.; Rochel, N.; Ruff, M.; Vivat, V.; Chambon, P.;
Gronemeyer, H.; Moras, D. Crystal structure of the RAR-γ ligand
binding domain bound to all-trans retinoic acid. Nature 1995,
378, 681-689.
2700) nm, (EtOH/NaOH 0.1 N) 250 (ꢀ 32900), 260 (ꢀ 27400),
302 (ꢀ 3000) nm. Anal. (C30H34NO3Cl) C, H, N, Cl.
Biologica l Ma ter ia ls. [6,7-3H]Estradiol (specific activity
2.00 PBq/mol, radiochemical purity > 98%) was purchased
from Amersham International (Amersham, England). Estra-
diol and 11â-derivatives used for binding studies were solu-
bilized in absolute ethanol. Solutions were stored at -20 °C
in the dark. The purity of solubilized compounds was checked
before use by TLC.
Cytosolic ER. Cytosol was prepared in Tris-HCl buffer
(T20), pH 7 (competitive binding experiments) or 8.5 (irrevers-
ible binding experiments), from immature lamb uteri as
described previously.19 The protein concentration, determined
according to Layne,35 was adjusted to 2 mg/mL (competitive
binding experiments) or 4 mg/mL (irreversible binding experi-
ments).
(13) Wagner, R. L.; Apriletti, J . W.; McGrath, M. E.; West, B. L.;
Baxter, J . D.; Fletterick, R. J . A structural role for hormone in
the thyroid hormone receptor. Nature 1995, 378, 690-697.
(14) Uppenberg, J .; Svensson, C.; J aki, M.; Bertilsson, G.; J enderberg,
L.; Berkenstam, A. Crystal structure of the ligand binding
domain of the human nuclear receptor PPARγ. J . Biol. Chem.
1998, 273, 31108-31112.
(15) Williams, S. P.; Sigler, P. B. Atomic structure of the progesterone
complexed with its receptor. Nature 1998, 393, 392-396.
(16) Nolte, R. T.; Wisely, G. B.; Westin, S.; Cobb, J . E.; Lambert, M.
H.; Kurokawa, R.; Rosenfeld, M. G.; Willson, T. M.; Glass, C.
K.; Milburn, M. V. Ligand binding and co-activator assembly of
the peroxisome proliferator-activated receptor-γ. Nature 1998,
395, 137-143.
(17) Harlow, K. W.; Smith, D. N.; Katzenellenbogen, J . A.; Greene,
G. L.; Katzenellenbogen, B. S. Identification of cysteine 530 as
the covalent attachment site of an affinity-labeling estrogen
(ketononestrol aziridine) and antiestrogen (tamoxifen aziridine)
in the human estrogen receptor. J . Biol. Chem. 1989, 264,
17476-17485.
(18) Reese, J . C.; Wooge, C. H.; Katzenellenbogen, B. S. Identification
of two cysteines closely positioned in the ligand binding pocket
of the human estrogen receptor: roles in ligand binding and
transcriptional activation. Mol. Endocrinol. 1992, 6, 2160-2166.
(19) El Garrouj, D.; Aumelas, A.; Borgna, J .-L. Steroidal affinity
labels of the estrogen receptor. 1. 17R-(bromoacetoxy)alkyl/
alkynylestradiols. J . Med. Chem. 1993, 36, 2973-2983.
(20) El Garrouj, D.; Aliau, S.; Aumelas, A.; Borgna, J .-L. Steroidal
affinity labels of the estrogen receptor. 2. 17R-[(haloacetamido)-
alkyl]estradiols. J . Med. Chem. 1995, 38, 2339-2348.
(21) Aliau, S.; El Garrouj, D.; Yasri, A.; Katzenellenbogen, B. S.;
Borgna, J .-L. 17R-[(Haloacetamido)alkyl]estradiols alkylate the
human estrogen receptor at cysteine residues 417 and 530.
Biochemistry 1997, 36, 5861-5867.
Com p etitive Bin d in g Assa y: Ap p a r en t Rela tive Af-
fin ity Con sta n ts. Competition assays between 5 nM [3H]-
estradiol and increasing concentrations of nonradioactive
estradiol or 11â-derivatives for binding the lamb uterine
cytosol (2 mg protein/mL, pH 7) were performed for 20 h at 20
°C, as described previously.19 Apparent relative affinity con-
stants (RACs) of competitors were calculated according to
Korenman27 using concentrations of unbound and specifically
bound [3H]estradiol at 50% specific binding inhibition, as well
as the concentration of unlabeled estradiol and 11â-derivatives
which inhibited 50% of the specific binding of [3H]estradiol.19
Sta n d a r d Ir r ever sible Bin d in g Assa y. Cytosol (4 mg
protein/mL, pH 8.5) was first incubated for 16 h at 0 °C with
estradiol, with 11â-derivatives, or without any steroid. Samples
were then treated with an equal volume of charcoal suspension
(1% charcoal, 0.1% dextran T70 in T20, pH 7) for 30 min at 0
°C, and charcoal was removed by centrifugation. The total and
nonspecific [3H]estradiol binding formed under exchange
conditions (16 h, 20 °C) in supernatant aliquots were deter-
mined by charcoal assay, as described previously19 using,
except where otherwise stated, 20 nM [3H]estradiol in the
absence and presence of 5 µM nonradioactive estradiol.
Ra d ioa ctivity Deter m in a tion s. The radioactivity of the
various cytosol samples (200 µL) was counted in 4 mL of
Emulsifier Safe (Packard).
Ack n ow led gm en t. The authors thank M. J oanni-
dou, B. Khe´mis, C. Lang, F. Maquin, and their teams
(Structural Analysis Laboratory of Hœchst Marion
Roussel at Romainville) for acquisition and interpreta-
tion of the spectral data and N. Moine for performing
elemental analyses.
(22) Lobaccaro, C.; Pons, J .-F.; Duchesne, M.-J .; Auzou, G.; Pons, M.;
Nique, F.; Teutsch, G.; Borgna, J .-L. Steroidal affinity labels of
the estrogen receptor. 3. Estradiol 11â-n-alkyl derivatives bear-
ing a terminal electrophilic group: antiestrogenic and cytotoxic
properties. J . Med. Chem. 1997, 40, 2217-2227.
(23) Nique, F.; Van de Velde, P.; Bre´maud, J .; Hardy, M.; Philibert,
D.; Teutsch, G. 11â-Amidoalkoxyphenyl estradiols, a new series
of pure antiestrogens. J . Steroid Biochem. Mol. Biol. 1994, 50,
21-29.
(24) Van de Velde, P.; Nique, F.; Bouchoux, F.; Bre´maud, J .; Hameau,
M.-C.; Lucas D.; Moratille C.; Viet, S.; Philibert, D.; Teutsch, G.
RU 58668, a new pure antiestrogen inducing a regression of
human mammary carcinoma implanted in nude mice. J . Steroid
Biochem. Mol. Biol. 1994, 48, 187-196.
(25) Be´langer, A.; Philibert, D.; Teutsch, G. Regio and stereospecific
synthesis of 11â-substituted 19-nor-steroids. Steroids 1981, 37,
361-382.
(26) Claussner, A.; Nedelec, L.; Philibert, D.; Van de Velde, P. New
19-norsteroids. European Patent EP 0 384 842 (Priority 24/02/
89, FR 8 902 384); Chem. Abstr. 1990, 115, 256464.
(27) Korenman, S. G. Relation between estrogen receptor inhibitory
activity and binding to cytosol of rabbit and human uterus.
Endocrinology 1970, 87, 1119-1123.
(28) Claussner, A.; Nedelec, L.; Nique, F.; Philibert, D.; Teutsch, G.;
Van de Velde, P. 11â-Amidoalkyl estradiols, a new series of pure
antiestrogens. J . Steroid Biochem. Mol. Biol. 1992, 41, 609-
614.
(29) 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.
(30) Reese, J . C.; Katzenellenbogen, B. S. Mutagenesis of cysteines
in the hormone binding domain of the human estrogen receptor.
J . Biol. Chem. 1991, 266, 10880-10887.
Refer en ces
(1) Green, S.; Walter, P.; Kumar, V.; Krust, A.; Bornet, J .-M.; Argos,
P.; Chambon, P. Human oestrogen receptor cDNA: sequence,
expression and homology to v-erb-A. Nature 1986, 320, 134-
139.
(2) Greene, G. L.; Gilna, P.; Waterfield, M.; Baker, A.; Hort, Y.;
Shine, J . Sequence and expression of human estrogen receptor
complementary DNA. Science 1986, 231, 1150-1154.
(3) Kuiper, G. G. J . M.; Enmark, E.; Pelto-Huikko, M.; Nilsson, S.;
Gustafsson, J .-A. Cloning of a novel estrogen receptor expressed
in rat prostate and ovary. Proc. Natl. Acad. Sci. U.S.A. 1996,
93, 5925-5930.
(4) Tremblay, G. B.; Tremblay, A.; Copeland, N. G.; Gilbert, D. J .;
J enkins, N. A.; Labrie, F.; Gigue`re, V. Cloning, chromosomal
localization and functional analysis of the murine estrogen
receptor â. Mol. Endocrinol. 1998, 11, 353-365.
(5) Mosselman, S.; Polman, J .; Dijkema, R. ERâ: identification and
characterization of a novel human estrogen receptor. FEBS Lett.
1996, 392, 49-53.
(6) Kuiper, G. G. J . M.; Carlson, B.; Grandien, K.; Enmark, E.;
Ha¨ggblad, J .; Nilsson, S.; Gustafsson, J .-A. Comparison of the
ligand binding specificity and transcript tissue distribution of
estrogen receptors R and â. Endocrinology 1997, 138, 863-870.
(7) Enmark, E.; Pelto-Huikko, M.; Grandien, K.; Lagercrantz, S.;
Lagercrantz, J .; Fried, G.; Nordenskjo¨ld, M.; Gustafsson, J .-A.
Human estrogen receptor â gene structure, chromosomal local-
ization and expression pattern. J . Clin. Endocrinol. Metab. 1997,
82, 4258-4265.

628 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 4
Aliau et al.
(31) Zablocki, J . A.; Katzenellenbogen, J . A.; Carlson, K. E.; Norman,
M. J .; Katzenellenbogen, B. S. Estrogenic affinity labels: syn-
thesis, irreversible receptor binding, and bioactivity of aziridine
substituted hexestrols derivatives. J . Med. Chem. 1987, 30, 829-
838.
(32) Madigou, T.; Tiffoche, C.; Lazennec, G.; Pelletier, J .; Thieulant,
M.-L. The sheep estrogen receptor: cloning and regulation of
expression in the hypothalamo-pituitary axis. Mol. Cell. Endo-
crinol. 1996, 121, 153-163.
(33) Hegy, G.; Shackleton, C.; Carlquist, M.; Bonn, T.; Engstro¨m, O.;
Sjoholm, P.; Witkowska, H. Carboxymethylation of the human
estrogen receptor ligand-binding domain-estradiol complex:
HPLC/ESMS peptide mapping shows that cysteine 447 does not
react with iodoacetic acid. Steroids 1996, 61, 367-373.
(34) Aliau, S.; Mattras, H.; Richard, E.; Borgna, J .-L. Cysteine 530
of the human estrogen receptor R is the main covalent attach-
ment site of 11â-(aziridinylalkoxyphenyl)estradiols. Biochemistry
1999, 38, 14752-14762.
(35) Layne, E. Spectrophotometric and turbidimetric methods for
measuring proteins. Methods Enzymol. 1957, 3, 444-454.
J M990179S