Chemical synthesis and evaluation of 17α-alkylated derivatives of estradiol as inhibitors of steroid sulfatase

Article abstract of DOI:10.1016/j.ejmech.2011.06.027

Steroid sulfatase (STS) controls the levels of 3-hydroxysteroids available from circulating steroid sulfates in several normal and malignant tissues. This and the known involvement of active estrogens and androgens in diseases such as breast and prostate cancers thus make STS an interesting therapeutic target. Here we describe the chemical synthesis and characterization of an extended series of 17α-derivatives of estradiol (E2) using different strategies. A variant of the samarium-Barbier reaction with stoichiometric samarium metal and catalytic Kagan reagent formation was used for introducing low reactive benzyl substrates in position 17 of estrone (E1) whereas heterocyclic substrates were metalated and reacted with either the carbonyl or the 17-oxirane of E1. In vitro evaluation of the inhibitory potency of the new compounds against STS identified new inhibitors and allowed a more complete structure-activity relationship study of this family of 17α-derivatives of E2.

Full text of DOI:10.1016/j.ejmech.2011.06.027

European Journal of Medicinal Chemistry 46 (2011) 4227e4237
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Chemical synthesis and evaluation of 17
as inhibitors of steroid sulfatase
a-alkylated derivatives of estradiol
*
Diane Fournier, Donald Poirier
CHUQ-CHUL Research Center (Laboratory of Medicinal Chemistry, Endocrinology and Genomic Unit) and Laval University (Faculty of Medicine), Quebec (Quebec) G1V 4G2, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Steroid sulfatase (STS) controls the levels of 3-hydroxysteroids available from circulating steroid sulfates
in several normal and malignant tissues. This and the known involvement of active estrogens and
androgens in diseases such as breast and prostate cancers thus make STS an interesting therapeutic
Received 25 February 2011
Received in revised form
11 June 2011
Accepted 21 June 2011
Available online 28 June 2011
target. Here we describe the chemical synthesis and characterization of an extended series of 17a-
derivatives of estradiol (E2) using different strategies. A variant of the samarium-Barbier reaction with
stoichiometric samarium metal and catalytic Kagan reagent formation was used for introducing low
reactive benzyl substrates in position 17 of estrone (E1) whereas heterocyclic substrates were metalated
and reacted with either the carbonyl or the 17-oxirane of E1. In vitro evaluation of the inhibitory potency
of the new compounds against STS identified new inhibitors and allowed a more complete structure
Keywords:
Steroid sulfatase inhibitors
Grignard reagent
Kagan reagent
SamariumeBarbier reaction
Estrogen
eactivity relationship study of this family of 17
a
-derivatives of E2.
Ó 2011 Elsevier Masson SAS. All rights reserved.
Cancer
1. Introduction
cleaved in two ways, one acid to yield sulfamates and one nucle-
ophilic to yield phenols, two families of compounds with numerous
interesting biological properties [15e18].
Steroid sulfatase (STS) catalyzes the hydrolysis of steroid-3-O-
sulfates into 3-hydroxysteroids. Since the major part of systemic
circulating precursors for estrogenic and androgenic active steroid
hormones are sulfates such as dehydroepiandrosterone sulfate
(DHEAS) and estrone sulfate (E1S) (Fig. 1), STS plays a major role in
regulating levels of estrogens and androgens. Thus, inhibition of
STS could have important applications mainly against hormone-
dependent breast and prostate cancers, but also against other
hormone-dependent diseases such as acne and alopecia and other
diseases related to less known biological properties of DHEA and
DHEAS such as Alzheimer’s disease and arthritis [1e6].
Among the most efficient STS inhibitors are sulfamates, with the
first, estrone sulfamate (compound 1), reported by Potter and Reed
in 1994 [7]. Arylsulfamates are irreversible type inhibitors of STS
[1e6], but sulfamates have many other biological properties such as
anti-convulsant, antimicrobial, anti-tumor agent, antiviral, antibi-
otic and inhibitor of carbonic anhydrases [8e11]. Such interesting
biological properties have prompted our group to design the sul-
famate linker for solid phase synthesis [12e14]. This linker can be
Although less potent than arylsulfamates, phenolic steroids
such as 17
are also inhibitors of STS [19e21]. Results showed that hydro-
phobic substituents on the 17 -benzyl group of E2 can modulate
a-benzyl derivatives of estradiol (E2), compounds 2aed,
a
the inhibitory activity by interacting with a hypothetic hydro-
phobic pocket in the D-ring area. Thereafter, the published crystal
structure of STS [22,23] showed a hydrophobic tunnel and three
phenylalanines which could have a
p-p interaction with the 17a-
benzyl group. We thus considered two likely ways to increase the
potency of these inhibitors: 1) introducing highly hydrophobic
groups in meta or para position of the benzyl moiety and 2)
introducing
various electronic distributions in 17
maximize a hypothetic interaction between inhibitor and
a
cyclopropyl or
a
heteroaromatic moiety with
a
-position of E2 in order to
pep
enzyme. This resulted in the preparation of two series of
compounds (Figs. 2 and 3), each targeted for one of the two
hypotheses. The first series (phenols 9e13, 17e22) is rather
straightforward and uses t-butyl, benzyloxy, trifluoromethyl and
the lower period, non-electronegative halogens (Br and I). The
choices for the second series (phenols 14e16, 23e25, 27 and 29)
were somewhat complicated by the fact that
rarely exploited in structureeactivity relationship (SAR) studies in
p-p interactions are
* Corresponding author. Tel.: þ1 418 654 2296; fax: þ1 418 654 2761.
E-mail address: donald.poirier@crchul.ulaval.ca (D. Poirier).
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.06.027

4228
D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
2. Results and discussion
2.1. Chemical synthesis
The side-chains needed and not commercially available for the
synthesis of the E2 derivatives were prepared as reported in
Scheme 1. They were next introduced in the 17 -position of E2
a
using three different strategies (Schemes 2 and 3). In the first
strategy (A), the benzyl bromides were submitted to metal-
halogen exchange using either magnesium metal or samarium
metal and subsequently added to the ketone of the t-butyldime-
thylsilyl (TBS) ether of E1 (TBS-E1). In the second strategy (B),
compound 23 was prepared using lithium metalation of 3-picoline
and addition to E1. In the third strategy (C), the 17b-oxirane
generated from E1 was opened using metalated heteroaromatic
substituents. Finally, the preparation of the cyclopropyl
compounds is described in Scheme 4.
2.1.1. Synthesis of halogenated side-chains (Scheme 1)
The side-chains 3c, 4c, 5b, 6b and 8c were synthesized from
available carboxylic acids and aldehydes using standard reduction
procedures followed by bromide substitution of the resulting
alcohol. In the case of the 3-tert-butyl-benzyl bromide (7d),
formation of the triflate 7a followed by palladium-catalyzed carbon
monoxide insertion yielded the methyl ester which was then
submitted to the reduction and bromination procedures.
Fig. 1. Natural substrates (DHEAS and E1S) and products (DHEA and E1) of key
steroidogenic enzyme STS (steroid sulfatase) and two families of inhibitors represented
by sulfamate 1 (EMATE) and phenols 2a-d (estradiol derivatives).
2.1.2. Synthesis of E2 derivatives via strategy A (Scheme 2)
A series of side-chains were introduced in the 17a-position of E2
the context of enzymeeinhibitor interactions. However, it could
be hypothesized that interactions could be maximized if the
electron density is minimal in the aromatic system so that elec-
tronic repulsion is minimized. Aromatic and non-aromatic
systems were thus selected with different sizes and electronic
distribution, some electron-rich and some electron-deficient. The
cyclohexylmethyl group (compound 16) was chosen as a negative
control for aromaticity whereas compounds 2aed represent our
lead compounds. Here we present the chemical synthesis of
eighteen new E2 derivatives and their potency as inhibitors
of STS.
to generate 9e22 from TBS-E1. In all cases, an attempt was first
made at forming the Grignard reagent with powdered magnesium
and the halogenated side-chain. No other methods were attempted
when success was achieved with the above mentioned method.
High excess (8 equiv.) of Grignard reagent (tested using Michler’s
reagent) was used in all cases because the ketone at position 17 of
E1 is known to react poorly with nucleophiles [24]. As many
substrates contained halide substituents on the aromatic ring,
formation of the Grignard reagent using metal-halogen exchange
from a preformed organomagnesium reagent was not thought
Fig. 2. % inhibition of steroid sulfatase (E1S into E1) by series 1 inhibitors.

D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
4229
Fig. 3. % inhibition of steroid sulfatase (E1S into E1) by series 2 inhibitors.
possible. In some cases, yield could however be improved using low
temperature reaction conditions with dry cerium chloride as an
activator of the ketone [24,25]. Of course, as previously observed
steroid was formed, with unreacted TBS-E1 and the product of
carbonyl reduction as the only other detectable materials.
Compounds 9e15 were thus obtained using the Grignard reaction
with yields varying from 21 to 83%, the lower yields being observed
[26e28], only the product of nucleophilic attack on the a-face of the
Scheme 1. Synthesis of halogenated side-chains. Reagents: (a) BnBr, Cs2CO3, CH3CN, reflux; (b) LiAlH4, THF; (c) PPh3, CBr4, CH2Cl2; (d) (CF3SO2)2O, 2,6-lutidine, CH2Cl2; (e) Et3N,
Pd(OAc)2, dppp, CO(g), DMF, MeOH.

4230
D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
Scheme 2. Synthesis of 17a-derivatives of estradiol by three different strategies (AeC). Reagents: (a) Mg, RCH2Br, Et2O, THF; (b) SmI2, RCH2Br, THF; (c) Sm, HgCl2, RCH2Br, THF; (d)
TBAF, THF; (e) 3-picoline, n-BuLi, HMPA, THF; (f) furan or 4-methylthiophene, n-BuLi, THF.
for the three fluorinated compounds. These yields, even in the
presence of an excess of Grignard reagent, are expected for a steri-
cally hindered system such as the ketone of TBS-E1 [24].
Science) than using home-distilled anhydrous THF over sodium-
benzophenone, even if that is the method being used for distil-
lation of the former. That could be due to the presence of a small
amount (25e250 ppm) of 2,6-dimethyl-4-t-butylphenol (butyl-
ated hydroxytoluene, BHT), an anti-oxidant used as a stabilizer
against peroxide formation in the bottled THF. It is known that
anti-oxidants help in the formation of SmI2, but higher concen-
trations (250 ppm) of BHT seemed to slow down the formation of
the Kagan reagent [30]. In the case of compound 16, this method
yielded none of the desired products but only a small amount of
For the cyclohexylmethyl side-chain, the Grignard reaction
using standard conditions or CeCl3 ketone activation failed to
yield the desired product 16, even though the formation of the
Grignard reagent appeared successful from the Michler’s reagent
test. In the case of the CeCl3-assisted Grignard reaction, only an
aldol condensation-dehydration steroid-dimer product was iso-
lated in 35% yield (Scheme 3). To obtain the desired product 16,
a reductive alkylation using samarium iodide (the Kagan reagent)
was thus attempted [29]. This reaction has the advantage of high
regioselectivity of alkyl halides versus aryl halides, increased
reactivity for sterically hindered ketones, and very low basicity,
which allow tolerance for a wider selection of functional groups
[30]. Samarium iodide was generated using purified 1,2-
diiodoethane and powdered samarium metal in anhydrous
degassed THF. It is notable that the reagent was formed more
easily using bottled anhydrous THF (either from Aldrich or EM
E2 (both 17a and 17b-OH). Addition of hexamethylphosphoramide
(HMPA), which is known to increase the reducing power of SmI2
[31,32], and running the reaction in refluxing THF yielded 35% of
the desired product 16. The Kagan reagent used in samarium-
Barbier conditions (preformation of SmI2 followed by addition
of both ketone and halide at the same time) yielded compound 17
in 76% yield, which is excellent in view of the hindered ketone
and low excess (1.5 equiv.) of di-benzyloxybenzyl-halide side-
chain used.
Scheme 3. Formation of unexpected steroid dimer in CeCl3-assisted Grignard reaction.

D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
4231
Scheme 4. Synthesis of cyclopropyl derivatives 27 and 29. Reagents: (a) CF3COOZnEtI, CH2Cl2; (b) TBAF, THF; (c) 4-bromo-1-butene, Grubb’s catalyst, CH2Cl2.
In the case of compounds 18e22, the desired product was not
obtained using the above mentioned conditions. In fact, small
amount of TBS-E2 and both reduced benzyl and dibenzyl coupling
products were isolated, showing that the Kagan reagent was
formed. Addition of HMPA did not yield better results. Both the
samarium-Barbier and the samarium-Grignard (addition of the
halide and then addition of the ketone) conditions were attempted
without success. Another method was then tested following
a publication of Gao et al. [33] on the specific reaction of allyl and
benzyl halides with ketones. This method uses stoichiometric or
excess Sm metal and catalytic HgCl2 in the samarium-Barbier
reaction conditions. An alternative method using catalytic iodine
instead of HgCl2 also worked in a similar way. Compounds 18e22
were obtained using these catalytic samarium-Barbier conditions
in 11e58% yields. The successful synthesis of compounds 18e22
with these reaction conditions could be explained by the lower
concentration of benzyl radical afforded by the catalytic amount of
Sm halide in solution, which would reduce the speed of reaction
between two benzyl radicals, and thus allow reaction of the benzyl
radical with the ketone.
From the results of strategy A, it can be seen that the use of the
Kagan reagent allows the tolerance of a greater range of substitu-
ents, increases reactivity towards hindered ketones while reducing
the excess of halide needed for reaction. For benzyl halides
however, except in very specific cases such as the 3,5-
dibenzyloxybenzyl side-chain used to generate compound 17,
halide reduction and coupling of benzyl radicals happen at a faster
rate than reaction with ketone. In these cases, the preferred reac-
tion method should be the catalytic samarium-Barbier conditions
described by Gao et al. [33].
from reacting the heterocycles (furan and 4-methylthiophen) with
butyllithium. The tertiary alcohols 24 and 25 were thus obtained in
low yields but with the right 17 -OH stereochemistry.
b
2.1.5. Other synthesis methods (Scheme 4)
The cyclopropylmethyl derivative 27 was afforded through
cyclopropanation of 17a-allyl-E2 (26) using in situ generated
CF3COOZnEtI [35]. The completion of the reaction was observed
using NMR since the product showed the same Rf on thin-layer
chromatography as the starting alkene. For the synthesis of 29, the
intermediate alkene 28 was afforded through Grubb’s metathesis of
26 with 4-bromo-1-butene. In a second step, the cyclopropyl 29 was
obtained after cyclopropanation of 28 using the aforementioned
method and the final removal of the TBS protecting group.
2.2. Inhibitory activity of new E2 derivatives on STS
The enzymatic assay was performed using a homogenate of
HEK-293 cells transfected with STS as the source of enzyme. The
transformation of [3H]-E1S into [3H]-E1 was measured using scin-
tillation counting of labeled E1S and E1 in the aqueous and organic
phases, respectively. The best inhibitors from our previous study
[21], compounds 2aed, were used as positive controls for STS
inhibition. Some general tendencies can be deduced when
observing the STS inhibition with compounds of series 1 (Fig. 2) and
the previously synthesized controls. For most substituting groups
that is t-butyl (2b, 10 and 11), benzyloxy (2d, 9 and 17), and tri-
fluoromethyl (12, 13 and 18), potency of the m-disubstituted benzyl
derivatives is lower than both the single substituted meta or para
derivatives. The only exception seems to be substitution by
bromine (2c, 19 and 20), which happens to be the smallest
substituting group of this series. In that case they have roughly the
same inhibitory activity. From these observations we can imagine
that the hydrophobic pocket in that portion of the active site is deep
and narrow. When comparing substitution in meta of the benzyl
group, it can be observed that iodine 21, bromine 2c, benzyloxy 9
2.1.3. Synthesis via strategy B (Scheme 2)
In the case of the pyridine derivative 23, the simple strategy of
employing the 3-picoline anion generated from 3-picoline and
lithium diisopropylamide (LDA) saved us from the necessity to
synthesize the m-pyridylmethyl bromide. The reaction worked only
in presence of HMPA as a disaggregation agent and provided 23 in
14% yield although using an excess of 3-picoline anion.
and trifluoromethyl 12 all have similar potency at 0.1 mM whereas t-
butyl substituted product 11 has lower potency close to that of the
benzyl derivative 2a. The comparison of the different substituting
groups when in para position points to a preference for more highly
hydrophobic groups with t-butyl 2b and iodine 22 superior to
bromine 19 and benzyloxy 2d.
Series 2 compounds (Fig. 3) were synthesized to explore the
influence of different electronic distributions on the aromatic ring
and the necessity of aromaticity in the substituting group. The first
thing that can be observed is that the cyclohexylmethyl 16 has
a similar potency to the benzyl 2a, which seems to indicate that
2.1.4. Synthesis using strategy C (Scheme 2)
For the synthesis of furan and 4-methylthiophene derivatives 24
and 25, the synthesis of the appropriate brominated side-chains
proved to be difficult due to the lack of commercially available
precursors, or simply to their instability. It was then decided to
proceed through the readily available 17
b-oxirane of E1, which is
made in one step from estrone using dimethylsulfonium methylide
[34]. The oxirane was opened using the lithium reagents obtained

4232
D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
only one inhibitor, 17
range as our previous 17
obtained several candidates with higher potency against STS than
the starting 17 -benzyl-E2 (2a), which could be rendered yet more
a-(4-iodobenzyl)-E2 (22), in the same potency
a-(4-t-butylbenzyl)-E2 (2b). We however
a
potent after a sulfamoylation of the phenol. Furthermore, it is
known that the t-butyl group is oxidized in vivo which limits the
application of inhibitor 2b in breast cancer therapy, thus the need
to find an inhibitor which has equal or superior potency and more
stability in vivo.
We did not get a marked increase in inhibitory potency
compared with the previously synthesized compounds. As
observed in our previous work [21], inhibitory potency tends to
increase with hydrophobicity of 17
size and position of benzyl substituting group(s) seems to be
restricted as seen with 17 -(di-meta-t-butyl)-E2 (10), which has
lower inhibitory potency than the starting 17 -benzyl-E2 (2a). The
hypothetical interaction does not seem to be important in our
inhibitor design as seen with our negative control, 17 -cyclo-
hexylmethyl-E2 (16) which showed similar potency to the starting
17 -benzyl-E2 (2a). In the previous studies, it was hypothesized
a substituting groups. However,
a
a
pep
a
a
that there was a hydrophobic pocket neighboring the D cycle of the
enzyme substrate. The x-ray structure of STS confirmed the pres-
ence of this hydrophobic pocket in the form of an access tunnel to
the active site. Moreover, it has been recently reported that inhi-
Fig. 4. Schematic representation of the two hypothetic proximate binding sites of STS.
In red, the active site containing the manually docked substrate E1S. In blue, the
hypothetic access tunnel buried in the bilayer membrane. Figure adapted from [23]
(For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
bition of STS by E1 and 17a-benzyl-E2 has a non-competitive
behavior [37], which is suggestive of an allosteric binding site.
Thus, there may be two proximate binding sites on the enzyme
(Fig. 4), one of which would allow the product of the reaction, E1, to
bind and act as a non-competitive inhibitor to the reaction. Our
inhibitors seem to interact mostly with this allosteric site or
partially with both binding sites. If the second alternative is true, it
may be possible to increase inhibitory potency of the STS inhibitors
by maximizing interactions with both sites.
aromaticity is not necessary for high potency. Similarly, the inhibi-
tory potency of the pentafluorobenzyl derivative 14 is also the same
as that of the benzyl 2a, suggesting that the interaction involved is
not of the
p-p type. When comparing different cycles and sizes,
cyclopropyl 27 was lower than cyclohexyl 16, benzyl 2a, penta-
fluorobenzyl 14 and 5-bromo-2,3-cyclopropane 29, which seems to
4. Experimental
indicate
a hydrophobic-driven affinity as stated in the first
hypothesis. Increased polarity of the aromatic substituent(s) seems
to be detrimental as shown by dibenzylamino 15, which is less
potent than the benzyl 2a. Similarly, introduction of polarity in the
cycle seems detrimental, as seen with low potencies of the m-
pyridine 23 derivative, and when comparing the furan 24 with the
4.1. Chemistry
4.1.1. General
Reagents and the starting steroid (estrone) were purchased
from Aldrich Chemical Co. (Milwaukee, WI, USA) whereas solvents
were obtained from VWR (Ville Mont-Royal, Quebec, Canada).
thiophene 25. In light of the above observations, the hypothetic
p-p
interaction between 17 -benzyl groups and the phenylalanine
a
Thin-layer chromatography (TLC) was performed on 250 mm silica
residues of the active site seems unlikely, since in that case the
cyclohexylmethyl derivative 16 would have shown marked lower
potency. Furthermore, higher potency would be expected for the
pentafluorobenzyl 14, given its known favorable interaction with
benzyl groups [36]. Previous results showed that benzyl substitu-
ents were generally better than flexible alkyl chains, simple phenyl
and phenethyl substituents [21]. This is not unexpected since flex-
ible ligands have high entropic energy punishment upon binding
due to loss of degrees of freedom, when compared to rigid ligands. It
thus seems that inhibitor/enzyme interaction is best with a rela-
gel 60 F254 plates (E. Merck, Darmstadt, Germany), and compounds
were visualized with a solution of ammonium molybdate/sulfuric
acid/water with heating. 1,2-Diiodoethane was purified by dis-
solving in CH2Cl2, washing with as saturated solution of sodium
sulfite, drying over magnesium sulfate, filtration and evaporation.
Purification of final compounds was performed by flash-column
chromatography using 230e400 mesh ASTM silica gel 60 (Sili-
cycle, Quebec, Canada). Infrared spectra (IR) were obtained with
a PerkineElmer 1600 spectrophotometer and data expressed in
cmꢀ1. 1H and 13C NMR spectra were recorded either with a Bruker
(AC/F) 300 spectrometer (when mentioned), or with a Bruker
AVANCE 400 spectrometer (Billerica, MA, USA). The chemical shifts
tively rigid 17a-moiety separated from the steroid E2 (a rigid group
itself) by one rotatable bond and that, as aforementioned, the
hydrophobic cavity is long and narrow, which make 4-t-butylbenzyl
2b and 4-iodobenzyl 22 the best interacting inhibitors in our study.
(d
) were expressed in ppm and referenced to chloroform (7.26 and
77.0 ppm for 1H and 13C, respectively) or acetone (29.0 ppm for 13
C
NMR). Low-resolution mass spectra (LRMS) were recorded with an
LCQ Finnigan apparatus (San Jose, CA, USA) equipped with an
atmospheric pressure chemical ionization (APCI) source.
3. Conclusion
From this work, we have seen that the synthesis of 17a-benzyl
derivatives of E2 can be achieved using the catalytic samarium-
Barbier method, which tolerates a greater range of functional
groups while increasing the reactivity toward sterically hindered
ketones compared to the classic Grignard reaction. We obtained
4.1.2. Synthesis of non-commercially available substrates (building
blocks) for alkylation
4.1.2.1. 3-Benzyloxybenzyl bromide (3c) [38]. 3-Hydroxybenzalde-
hyde (5.00 g, 0.041 mol) was dissolved in anhydrous acetonitrile

D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
4233
(130 mL) under argon atmosphere. Cesium carbonate (20.01 g,
0.061 mol) was added and the suspension stirred for 5 min. Benzyl
bromide (11.69 mL, 0.102 mol) was then added and the solution
heated at reflux for 16 h. The solution was concentrated on rotary
evaporator, water was added and the mixture was extracted with
EtOAc. The organic phase was washed twice with water, once with
brine, dried over MgSO4, filtered and concentrated. Water was
added and the mixture was extracted with CH2Cl2. The crude
product was purified by flash chromatography on silica gel with
hexanes/EtOAc (80/20) to yield 3a as a white solid (8.59 g). 1H NMR
on silica gel with hexanes to yield 4.62 g (86%) of bromide 5b. 1H
NMR
d (CDCl3) 1.33 (s, 18H, di-t-butyl), 4.52 (s, 2H, PhCH2Br), 7.23
(d, 2H, J ¼ 1.8 Hz, 2-CH and 6-CH), 7.37 (t, 1H, J ¼ 1.8 Hz, 4-CH); 13C
NMR (75 MHz)
(2ꢂ), 136.8, 151.3 (2ꢂ).
d
(CDCl3) 31.4 (6ꢂ), 34.8 (2ꢂ), 34.9, 122.7, 123.3
4.1.2.4. 3,5-Di-bromobenzyl bromide (6b) [41]. 3,5-Di-bromo-
benzoic acid (2.53 g, 9.00 mmol) was submitted to the same reac-
tion sequence as described for the synthesis of 5b to yield 2.5 g of
bromide 6b. 1H NMR
d (CDCl3) 4.37 (s, 3H, PhCH2Br), 7.48 (s, 2H,
d
(CDCl3) 5.13 (s, 2H, PhCH2O), 5.33 (s, 2H, COOCH2Ph), 7.35e7.50
2-CH and 6-CH), 7.60 (s, 1H, 4-CH); 13C NMR (75 MHz)
d (CDCl3)
(m, 9H, 2-CH, 4-CH, 5-CH, 6-CH and PhCH2O), 10.00 (s, 1H, PhCHO).
The aldehyde 3a was dissolved in anhydrous THF (200 mL) under
argon and cooled to 0 ꢁC. Lithium aluminum hydride (1.55 g,
0.041 mol) was added in small portions and the solution stirred at
room temperature for 2 h. The reaction was then quenched using
water (0.8 mL), a 10% wt aqueous NaOH solution (1.15 mL) and
water again (1.9 mL) and left to settle. The suspension was then
filtered and concentrated. Water was added and the mixture
extracted with EtOAc, the organic phase dried over MgSO4, filtered
30.7, 123.0 (2ꢂ), 130.8 (2ꢂ), 134.0, 141.3.
4.1.2.5. 3-tert-Butylbenzyl bromide (7d) [42]. 3-t-Butylphenol
(1.00 g, 6.66 mmol) was dissolved in anhydrous CH2Cl2 (130 mL)
and cooled to 0 ꢁC in an ice bath. 2,6-Lutidine (1.7 mL, 14.64 mmol)
and triflic anhydride (2.2 mL, 13.31 mmol) were added successively.
The mixture was stirred overnight at room temperature, quenched
with water and extracted with CH2Cl2. The organic phase was
washed successively with aqueous saturated NaHCO3 solution and
brine, dried with MgSO4, filtered and evaporated to yield 1.109 g
and concentrated to yield 7.07 g crude alcohol 3b. 1H NMR
d (CDCl3)
4.68 (s, 2H, PhCH2OH), 5.08 (s, 2H, PhCH2O), 6.90e7.46 (m, 9H,
2-CH, 4-CH, 5-CH, 6-CH and PhCH2O). Crude alcohol 3b (7.06 g)
was dissolved in anhydrous CH2Cl2 (330 mL) and the solution
cooled to 0 ꢁC. Triphenylphosphine (17.28 g, 0.066 mol) and
carbon tetrabromide (21.85 g, 0.066 mol) were then added and
the solution stirred at room temperature for 2 h. The reaction
mixture was quenched with water and extracted with CH2Cl2.
The organic phase was dried over MgSO4, filtered and
concentrated. The product was purified by flash chromatography
on silica gel with hexanes/EtOAc (9/1) to yield 5.85 g (64%) of
(59%) of 7a. 1H NMR
d (CDCl3) 1.33 (s, 9H, t-butyl), 7.09 (ddd, 1H,
J1 ¼ 1.4 Hz, J2 ¼ 2.2 Hz, J3 ¼ 7.8 Hz 4-CH), 7.25 (t, 1H, J1 ¼ 2.0 Hz,
2-CH), 7.40 (m, 2H, 5-CH and 6-CH). The triflic ester 7a (1.11 g,
3.93 mmol) was dissolved in anhydrous DMF (150 mL) and anhy-
drous MeOH (50 mL). Et3N (1.6 mL, 11.76 mmol), Pd(OAc)2 (265 mg,
1.18 mmol) and 1,3-bis(diphenylphosphino) propane (486 mg,
1.18 mmol) were added. Gaseous CO was bubbled for 1 h through
the mixture while it was heated to 90 ꢁC. The reaction mixture was
then stirred overnight at room temperature under a CO atmo-
sphere, after which it was poured into brine and extracted with
Et2O. The organic phase was washed with water, brine, dried over
MgSO4, filtered and evaporated. The crude product was purified by
flash chromatography on silica gel using hexanes/EtOAc (95/5) to
bromide 3c. 1H NMR
PhCH2O), 6.90e7.46 (m, 9H, 2-CH, 4-CH, 5-CH, 6-CH and
PhCH2O); 13C NMR (75 MHz)
(acetone-d6) 32.5, 69.4, 114.3,
d (CDCl3) 4.47 (s, 2H, PhCH2Br), 5.07 (s, 2H,
d
114.8, 121.0, 127.0 (2ꢂ), 127.2, 127.8, 128.0, 129.3, 136.1, 138.6, 158.3.
yield 554 mg (73%) of 7b. 1H NMR
d (CDCl3) 1.35 (s, 9H, t-butyl), 3.92
(s, 3H, COOCH3), 7.37 (t, 1H, J ¼ 7.8 Hz, 5-CH), 7.60 (dq, 1H,
J1 ¼ 1.2 Hz, J2 ¼ 7.9 Hz, 4-CH), 7.86 (dt, 1H, J1 ¼ 1.3 Hz, J2 ¼ 7.7 Hz,
6-CH), 8.08 (t, 1H, J ¼ 1.8 Hz, 2-CH). The ester 7b (554 mg,
2.88 mmol) was dissolved in anhydrous THF (30 mL) and cooled to
0 ꢁC. Lithium aluminum hydride (219 mg, 5.76 mmol) was added in
small portions and the mixture stirred 2 h under argon at room
temperature. Water was then added, concentrated HCl added to
acidify the solution and the mixture was extracted with Et2O, dried
over MgSO4, filtered and concentrated to achieve quantitatively 7c.
4.1.2.2. 3,5-Dibenzyloxybenzyl bromide (4c) [39]. 3,5-Dihydroxy-
benzoic acid (3.70 g, 0.024 mol) was submitted to the same reac-
tion sequence as described for the synthesis of 3c, except that
cesium carbonate and benzyl bromide quantities were tripled in
the first step to get the tribenzyl derivative 4a. The bromide 4c
(1.58 g) was obtained as final product. 1H NMR
d (CDCl3) 4.42 (s, 2H,
CH2Br), 5.03 (s, 4H, 2ꢂ PhCH2O), 6.56 (t, 1H, J ¼ 2.2 Hz, 4-CH), 6.65
(d, 2H, J ¼ 2.2 Hz, 2-CH and 6-CH), 7.41 (m, 10H, 2 ꢂ PhCH2O); 13
C
NMR (75 MHz)
d
(acetone-d6) 33.5, 69.7 (2ꢂ), 101.8, 108.3 (2ꢂ),
1H NMR
d (CDCl3) 1.34 (s, 9H, t-butyl), 4.70 (s, 2H, PhCH2OH), 7.19 (d,
127.6 (4ꢂ), 127.8 (2ꢂ), 128.4 (4ꢂ), 137.2 (2ꢂ), 140.2, 160.1 (2ꢂ).
1H, J1 ¼ 6.9 Hz, 6-CH), 7.33 (m, 2H, 4-CH and 5-CH), 7.40 (s, 1H,
2-CH). The alcohol 7c (700 mg, 4.26 mmol) was dissolved in
anhydrous CH2Cl2 (100 mL) and cooled at 0 ꢁC. Triphenylphosphine
(2.235 g, 8.52 mmol) and carbon tetrabromide (2.827 g, 8.52 mmol)
were added. The mixture was stirred at room temperature for 1 h.
The reaction mixture was quenched with water and extracted with
CH2Cl2. The organic phase was dried over MgSO4, filtered and
concentrated. The product was purified by flash chromatography
on silica gel with hexanes to yield 903 mg (93%) of bromide 7d. 1H
4.1.2.3. 3,5-Di-tert-butylbenzyl bromide (5b) [40]. 3,5-Di-t-butyl-
benzoic acid (5.00 g, 0.021 mol) was dissolved in anhydrous THF
(250 mL) under argon atmosphere and the solution cooled to 0 ꢁC.
Lithium aluminum hydride (1.62 g, 0.043 mol) was added in small
portions and the solution stirred at room temperature overnight.
The reaction was quenched with water, Et2O (100 mL) was added
and the mixture acidified with concentrated HCl solution until the
solid residue was dissolved. The medium was extracted with Et2O
and the organic phase dried over MgSO4, filtered and concentrated
NMR
d (CDCl3) 1.33 (s, 9H, t-butyl), 4.51 (s, 2H, PhCH2Br), 7.22 (d,
J ¼ 7.6 Hz, 1H, 6-CH), 7.30 (t, 1H, J ¼ 7.4 Hz, 5-CH), 7.33 (d, 1H,
to yield 4.31g of 5a. 1H NMR
d (CDCl3) 1.34 (s, 18H, di-t-butyl), 4.70
J ¼ 7.9 Hz, 4-CH), 7.40 (s, 1H, 2-CH).
(s, 2H, PhCH2OH), 7.23 (d, 2H, J ¼ 1.8 Hz, 2-CH and 6-CH), 7.38 (t,1H,
J ¼ 1.8 Hz, 4-CH). Crude alcohol 5a (4.25 g) was dissolved in
anhydrous CH2Cl2 (500 mL) and cooled at 0 ꢁC. Triphenylphosphine
(10.23 g, 0.039 mol) and carbon tetrabromide (12.93 g, 0.039 mol)
were added and the mixture was stirred at room temperature for
1 h. The reaction mixture was quenched with water and extracted
with CH2Cl2. The organic phase was dried over MgSO4, filtered and
concentrated. The product was purified by flash chromatography
4.1.2.6. 3-Dibenzylaminobenzyl bromide (8c). 3-Aminobenzoic acid
(1.00 g, 7.29 mmol) was submitted to the same reaction sequence as
described for the synthesis of 3c. The crude product was purified by
flash chromatography on silica gel with hexanes/EtOAc (90/10) to
yield 917 mg (89%) of bromide 8c. 1H NMR
d (CDCl3) 4.38 (s, 2H,
PhCH2Br), 4.65 (s, 4H, (PhCH2)2N), 7.14 (t, 1H, J ¼ 8.2 Hz, 5-CH), 7.34
(m, 13H, 2-CH, 4-CH, 6-CH and (PhCH2)2N); 13C NMR (75 MHz)

4234
D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
d
(acetone-d6) 34.5, 54.2 (2ꢂ), 112.7, 113.3, 117.4, 126.8 (4ꢂ), 128.6
33.73, 34.74 (6ꢂ), 39.61, 42.72, 43.84, 46.65, 49.46, 62.41 (2ꢂ),
(5ꢂ), 129.4 (2ꢂ), 138.9 (3ꢂ), 149.3.
82.78, 112.65, 115.23, 120.30, 125.28 (2ꢂ), 126.52, 132.70, 136.88,
138.28, 150.59 (2ꢂ), 153.34; LRMS for [M þ NH4]þ 492.4 m/z.
4.1.3. Procedure for the Grignard reaction (synthesis of compounds
9e15)
4.1.3.3. (17
(11). White powder; IR
0.98 (s, 3H, 18-CH3), 1.34 (s, 9H, t-butyl), 2.69 and 2.94 (2d, 2H,
b
)-17-(3-tert-Butylbenzyl)-estra-1(10),2,4-triene-3,17-diol
Powdered magnesium (705 mg, 28.99 mmol) was flame acti-
vated under argon in a dry tri-necked flask and left to cool down to
room temperature. Dry Et2O (5.6 mL) was added to the activated
Mg powder and a small portion (0.1 mL) of bromide solution
(2.26 g of 3c in 4.0 mL of Et2O) was added and the reaction started
with the heat from the hand or with a few drops of MeI (gas
evolution and cloudy solution with heat). The rest of the bromide
was then added slowly taking care not to boil off the solvent. The
mixture was stirred at room temperature for 2 h. A small amount
of the solution was used for a test with Michler’s reagent [43]. A
blue-green coloration indicated that the Grignard’s reagent was
formed. The Grignard reagent solution was added slowly at room
temperature to a solution of 3-t-butyldimethylsilyl-O-estrone
(TBS-E1) [44] (400 mg, 0.96 mmol) in anhydrous THF (24 mL) and
the mixture was stirred overnight at room temperature under
argon atmosphere. The mixture was poured in a saturated aqueous
NH4Cl solution, extracted with EtOAc, dried over MgSO4, filtered
and concentrated. Since the Rf of starting TBS-E1 and the final
product were very similar, the remaining starting product was
reduced to TBS-E2 by dissolving the crude product in anhydrous
MeOH (5 mL) and adding excess (4e5 eq.) of NaBH4. After 1 h at
room temperature, water was added, and the mixture was
extracted with EtOAc, dried over MgSO4, filtered and concentrated.
Purification by flash chromatography on silica gel with hexanes/
EtOAc (9/1) yielded 364 mg (65%) of 3-TBS-9 and 3-TBS-E2, which
was not recovered. Only the alkylated compound (3-TBS-9) was
submitted to the deprotection procedure. The TBS ether of 9
(364 mg, 0.677 mmol) was dissolved in anhydrous THF (7 mL)
under argon and cooled to 0 ꢁC. Tetrabutylammonium fluoride
(TBAF) in THF (0.81 mL, 0.81 mmol) was added dropwise and the
solution stirred at room temperature for 35 min. The reaction was
quenched with water, extracted with EtOAc, dried over MgSO4,
filtered and concentrated. The residue was purified with flash
chromatography on silica gel using hexanes/EtOAc (85/15) to yield
118 mg (83%) of 9. The same procedure was used for the synthesis
of 3-TBS-10 (65%), 3-TBS-11 (44%), 3-TBS-12 (21%), 3-TBS-13 (37%),
3-TBS-14 (27%) and 3-TBS-15 (65%), which after hydrolysis of the
TBS group afforded 10e15 (83e87%).
y
, (film on NaCl) 3342 (OH); 1H NMR
d
(CDCl3)
J ¼ 13.4 Hz, 17 -CH2), 2.85 (m, 2H, 6eCH2), 6.58 (d, 1H, J ¼ 2.2 Hz,
a
4eCH), 6.64 (dd, 1H, J1 ¼ 2.6 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.11 (d, 2H,
J ¼ 6.8 Hz, 20-CH), 7.19 (d,1H, J ¼ 8.4 Hz,1eCH), 7.28 (m, 3H, 40-CH, 50-
CH and 60-CH); 13C NMR (75 MHz)
d (CDCl3) 14.54, 23.37, 26.34, 27.48,
29.67, 31.40 (3ꢂ), 31.45, 33.70, 34.57, 39.61, 42.55, 43.84, 46.73, 49.47,
83.00, 112.65, 115.23, 123.26, 126.52, 127.65, 128.14 (2ꢂ), 132.62,
137.65, 138.25, 151.03, 153.37; LRMS for [MeH2OþH]þ 401.1 m/z.
4.1.3.4. (17
diol (12). White powder; IR
b
)-17-(3-Trifluoromethylbenzyl)-estra-1(10),2,4-triene-3,17-
y
, (film on NaCl) 3330 (OH); 1H NMR
d
(CDCl3) 0.97 (s, 3H, 18-CH3), 2.72 and 3.01 (2d, 2H, J ¼ 13.3 Hz,
17
a
-CH2), 2.85 (m, 2H, 6eCH2), 6.58 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.65
(dd,1H, J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.19 (d, 1H, J ¼ 8.0 Hz,1eCH),
7.43 (m, 1H, 60-CH), 7.51 (d, 2H, J ¼ 8.0 Hz, 40-CH and 50-CH), 7.59 (s,
1H, 20-CH); 13C NMR (75 MHz)
d (CDCl3) 14.39, 23.26, 26.32, 27.45,
29.62, 31.33, 33.94, 39.68, 42.13, 43.84, 46.97, 49.48, 83.31, 112.65,
115.25, 123.06, 125.64, 126.50, 127.69, 128.28, 130.22 (q, J ¼ 31.9 Hz),
132.53, 134.46, 138.25, 139.53, 153.38; LRMS for [M þ NH4]þ
448.2 m/z.
4.1.3.5. (17
triene-3,17-diol (13). White powder; IR
1H NMR
(CDCl3) 0.97 (s, 3H, 18-CH3), 2.76 and 3.08 (2d, 2H,
J¼ 13.7 Hz, 17 -CH2), 2.85(m, 2H, 6eCH2), 4.60(s,1H, OH), 6.58(d,1H,
b
)-17-[3,5-Bis(trifluoromethyl)benzyl]-estra-1(10),2,4-
y, (film on NaCl) 3389 (OH);
d
a
J ¼ 2.8 Hz, 4eCH), 6.65 (dd,1H, J1 ¼ 2.8 Hz, J2 ¼ 8.6 Hz, 2eCH), 7.19 (d,
1H, J ¼ 8.6 Hz,1eCH), 7.76 (s,1H 40-CH), 7.82 (s, 2H, 20-CH and 60-CH);
13C NMR (75 MHz)
d (CDCl3) 14.28, 23.24, 26.27, 27.42, 29.60, 31.32,
34.25, 39.70, 41.91, 43.83, 47.06, 49.45, 83.39, 112.71, 115.27, 120.18,
123.52 (q, J¼ 272.7 Hz) (2ꢂ),126.50,130.88(q,J¼ 33.0 Hz) (2ꢂ),131.18
(2ꢂ), 132.41, 138.23, 141.30, 153.38; LRMS for [M þ NH4]þ 516.3 m/z.
4.1.3.6. (17
3,17-diol (14). White powder; IR
NMR (CDCl3) 0.96 (s, 3H, 18-CH3), 2.85 (m, 2H, 6eCH2), 2.94 (s, 2H,
b
)-17-(2,3,4,5,6-Pentafluorobenzyl)-estra-1(10),2,4-triene-
y
, (film on NaCl) 3307 (OH); 1H
d
penta-FPhCH2), 4.56 (s, 1H, OH-phenol), 6.58 (d, 1H, J ¼ 2.8 Hz,
4eCH), 6.65 (dd, 1H, J1 ¼ 2.8 Hz, J2 ¼ 8.6 Hz, 2eCH), 7.18 (d, 1H,
J ¼ 8.6 Hz, 1eCH); 13C NMR (75 MHz)
d (CDCl3) 14.37, 22.85, 26.28,
4.1.3.1. (17
diol (9). White powder, IR
d
b
)-17-(3-Benzyloxybenzyl)-estra-1(10),2,4-triene-3,17-
, (film on NaCl) 3330 (OH); 1H NMR
27.38, 29.58, 30.15, 31.33, 33.72, 39.73, 43.77, 47.34, 49.76, 84.02,
112.16,112.67,115.67,126.45,132.33,138.24,153.45, CF signals are not
visible due to multiple F couplings; LRMS [MeH2OþH]þ 435.1 m/z.
y
(CDCl3) 0.97 (s, 3H,18eCH3), 2.65 and 2.91(2d, 2H, J ¼ 13.3 Hz,
17
a-CH2), 2.85 (m, 2H, 6eCH2), 4.57 (s, 1H, OH), 5.08 (s, 2H,
PhOCH2Ph), 6.58 (d, 1H, J ¼ 2.6 Hz, 4-CH), 6.64 (dd, 1H, J1 ¼ 2.8 Hz,
J2 ¼ 8.4 Hz, 2-CH), 6.89 (dd, 2H, J1 ¼ 2.5 Hz, J2 ¼ 8.1 Hz, 40-CH and 60-
CH), 6.95 (d, 1H, J ¼ 1.9 Hz, 20-CH), 7.18 (d, 1H, J ¼ 8.4 Hz, 1-CH), 7.23
(d, 1H, J ¼ 7.9 Hz, 50-CH), 7.39 (m, 4H, OCH2Ph); 13C NMR (75 MHz)
4.1.3.7. (17
3,17-diol (15). White powder; IR
NMR (CDCl3) 0.91 (s, 3H, 18eCH3), 2.53 and 2.79 (2d, 2H,
J ¼ 13.1 Hz, 17 -CH2), 2.84 (m, 2H, 6eCH2), 4.66 (br s, 4H,
b
)-17-[3-(Dibenzylamino)benzyl]-estra-1(10),2,4-triene-
y
, (film on NaCl) 3325 (OH); 1H
d
a
d
(CDCl3) 14.47, 23.30, 26.33, 27.47, 29.64, 31.57, 33.77, 39.63, 42.51,
N(CH2Ph)2), 4.78 (s, 1H, OH), 6.57 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.63
(dd, 1H, J1 ¼ 2.6 Hz, J2 ¼ 8.3 Hz, 2eCH), 7.12 (d, 1H, J ¼ 7.6 Hz,
40eCH), 7.16 (d, 1H, J ¼ 8.3 Hz, 1eCH), 7.29 (m, 13H 20-CH, 50-CH,
43.84, 46.84, 49.49, 69.91, 83.13,112.65, 112.75, 115.23, 117.64,123.68,
126.51, 127.50 (2ꢂ), 127.90, 128.54 (2ꢂ), 129.08, 132.70, 137.06,
138.28, 139.97, 153.32, 158.62; LRMS for [MeH2OþH]þ 451.2 m/z.
60-CH and N(CH2Ph)2); 13C NMR (75 MHz)
d (CDCl3) 14.45, 23.20,
26.31, 27.41, 29.65, 31.36, 33.59, 39.56, 42.81, 43.77, 46.67, 49.30,
54.31(2ꢂ), 82.89, 110.77, 112.60, 115.18, 119.41, 126.50, 126.71,
126.88 (2ꢂ), 128.12 (4ꢂ), 128.61 (4ꢂ), 129.03, 132.78, 138.29,
138.61 (2ꢂ), 139.17, 149.07, 153.26; LRMS [M þ H]þ 558.3 m/z.
4.1.3.2. (17
3,17-diol (10). White powder; IR
NMR (CDCl3) 0.99 (s, 3H,18-CH3),1.34 (s, 18H, di-t-butyl), 2.68 and
2.92 (2d, 2H, J ¼ 13.1 Hz, 17 -CH2), 2.85 (m, 2H, 6eCH2), w4.6
b
)-17-[3,5-Bis(tert-butyl)benzyl]-estra-1(10),2,4-triene-
y
, (film on NaCl) 3318 (OH); 1H
d
a
(broad s, 1H, OH), 6.58 (d, 1H, J ¼ 2.7 Hz, 4eCH), 6.64 (dd, 1H,
J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.11 (d, 2H, J ¼ 1.8 Hz, 20eCH and
60eCH), 7.19 (d, 1H, J ¼ 8.4 Hz, 1eCH), 7.33 (t, 1H, J ¼ 1.7 Hz, 40eCH);
4.1.4. Procedure for classic samariumeBarbier reaction (synthesis
of compounds 16 and 17)
In a dry flask, 40 mesh samarium powder (68 mg, 0.45 mmol)
and ICH2CH2I (85 mg, 0.33 mmol) were weighed under a nitrogen
13C NMR (75 MHz)
d (CDCl3) 14.58, 23.42, 26.36, 27.48, 29.69, 31.50,

D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
4235
atmosphere. The flask was purged with argon and anhydrous
degassed THF (5 mL) was added with vigorous stirring. After 2 h of
stirring, a dark blue SmI2 solution was obtained. Anhydrous
degassed hexamethylphosphoramide (HMPA) (0.21 mL, 1.20 mmol)
was then added, turning the blue solution to deep violet. A solution
of TBS-E1 [44] (50 mg, 0.12 mmol) and cyclohexylmethyl bromide
27.45, 29.60, 31.31, 33.86, 39.64, 42.19, 43.81, 46.99, 49.47, 83.41,
112.68, 115.25, 124.80 (2ꢂ), 125.74, 126.48, 128.54 (q, J ¼ 32.2 Hz),
131.33 (2ꢂ), 132.43, 138.21, 142.79, 153.40; LRMS for [MeH]ꢀ
429.5 m/z.
4.1.5.2. (17
White powder; IR
b
)-17-(4-Bromobenzyl)-estra-1(10),2,4-triene-3,17-diol (19).
, (film on NaCl) 3342 (OH); 1H NMR
(CDCl3)
-CH2),
(33
mL, 0.24 mmol) in anhydrous degassed THF (5 mL) was then
y
d
added with a cannula to the SmI2/HMPA solution. The mixture was
refluxed overnight, quenched with saturated aqueous NH4Cl solu-
tion, extracted with EtOAc, dried over MgSO4, filtered and evapo-
rated. The crude product was purified by flash chromatography on
silica gel with hexanes/EtOAc (93/7) to yield 22 mg (35%) of TBS-16.
This TBS ether was treated with a solution of TBAF (1.0 M) in THF as
reported above to yield 14 mg (88%) of 16.
0.96 (s, 3H, 18eCH3), 2.62 and 2.89 (2d, 2H, J ¼ 13.4 Hz, 17
a
2.84 (m, 2H, 6eCH2), 6.58 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.64 (dd, 1H,
J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.18 (d, 1H, J ¼ 7.5 Hz, 1eCH), 7.19 (d,
2H, J ¼ 8.3 Hz, 20eCH and 60eCH), 7.44 (d, 2H, J ¼ 8.3 Hz, 30eCH and
50eCH); 13C NMR (75 MHz)
d (CDCl3) 14.82, 23.68, 26.75, 27.89,
30.04, 31.79, 34.33, 40.11, 42.25, 44.28, 47.34, 49.98, 83.56, 113.10,
115.66, 120.73, 126.90, 131.49 (2ꢂ), 133.07, 133.15 (2ꢂ), 137.92,
138.67, 153.75; LRMS for [MeH]ꢀ 441.3 and 439.3 m/z.
4.1.4.1. (17
(16). White powder; IR
(CDCl3) 0.88 (s, 3H, 18-CH3), 2.82 (m, 2H, 6-CH2), 6.56 (d, 1H,
b
)-17-(Cyclohexylmethyl)-estra-1(10),2,4-triene-3,17-diol
y
, (film on NaCl) 3306 (OH); 1H NMR
4.1.5.3. (17
diol (20). White powder; IR
b
)-17-(3,5-Dibromobenzyl)-estra-1(10),2,4-triene-3,17-
, (film on NaCl) 3378 (OH); 1H NMR
d
y
J ¼ 2.6 Hz, 4-CH), 6.63 (dd, 1H, J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2-CH), 7.15 (d,
d
(CDCl3) 0.95 (s, 3H, 18eCH3), 2.57 and 2.89 (2d, 2H, J ¼ 13.4 Hz,
1H, J ¼ 8.4 Hz, 1-CH); 13C NMR (75 MHz)
d
(CDCl3) 14.17, 23.33,
17
a
-CH2), 2.84 (m, 2H, 6eCH2), 6.58 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.64
26.31 (2ꢂ), 26.57, 26.64, 27.39, 29.63, 31.24, 33.95, 34.32, 35.69,
36.15, 39.64, 43.74, 43.92, 46.90, 49.35, 84.35, 112.56, 115.17, 126.49,
132.77, 138.30, 153.24; LRMS for [M þ H]þ 369.1 m/z.
(dd,1H, J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.17 (d, 1H, J ¼ 8.4 Hz,1eCH),
7.44 (d, 2H, J ¼ 1.7 Hz, 20eCH and 60eCH), 7.55 (t, 1H, J ¼ 1.7 Hz,
40eCH); 13C NMR (75 MHz)
d (CHCl3) 14.33, 23.23, 26.25, 27.42,
29.60, 31.24, 34.05, 39.64, 41.79, 43.80, 47.02, 49.44, 83.37, 112.68,
115.23, 122.32 (2ꢂ), 126.49, 131.81, 132.44, 132.74 (2ꢂ), 138.22,
142.79, 153.33; LRMS for [MeH]ꢀ 519.1 m/z.
4.1.4.2. (17b)-17-[3,5-Bis(benzyloxy)benzyl]-estra-1(10),2,4-triene-3,17-
diol (17). TBS-E1 (200 mg, 0.48 mmol) and 4c (370 mg, 0.96 mmol)
were submitted to the same reaction sequence as described for the
synthesis of 16, except that there was no addition of HMPA and the
reaction was run at room temperature to yield 262 mg (76%) of 3-
TBS-17 which was treated with TBAF (1.0 M) as reported above
4.1.5.4. (17
White powder; IR
b
)-17-(3-Iodobenzyl)-estra-1(10),2,4-triene-3,17-diol (21).
, (film on NaCl) 3388 (OH); 1H NMR
(CDCl3)
-CH2),
y
d
0.96 (s, 3H, 18eCH3), 2.60 and 2.88 (2d, 2H, J ¼ 12.6 Hz, 17
a
for giving 17 in quantitative yield. White powder; IR
y
, (film on
2.86 (m, 2H, 6eCH2), 6.58 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.64 (dd, 1H,
J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.05 (t, 1H, J ¼ 7.8 Hz, 50eCH), 7.18
(d, 1H, J ¼ 8.4 Hz, 1eCH), 7.29 (d, 1H, J ¼ 7.8 Hz, 60eCH), 7.59 (d,
1H, J ¼ 7.9 Hz, 40eCH), 7.70 (s, 1H, 20eCH); 13C NMR (75 MHz)
NaCl) 3342 (OH); 1H NMR
d
(CDCl3) 0.96 (s, 3H, 18-CH3), 2.61 and
2.88 (2d, 2H, J ¼ 13.2 Hz, 17
a-CH2), 2.84 (m, 2H, 6-CH2), 5.04 (s, 4H,
2 ꢂ PhCH2O), 6.57 (m, 3H, 4-CH, 20-CH and 60-CH), 6.64 (dd, 1H,
J1 ¼ 2.5 Hz, J2 ¼ 8.4 Hz, 2-CH), 7.17 (d, 1H, J ¼ 8.4 Hz, 1-CH), 7.37
d
(CDCl3) 14.42, 23.28, 26.30, 27.46, 29.63, 31.33, 33.89, 39.65,
(m, 11H, 40-CH and 2 ꢂ PhCH2O); 13C NMR (75 MHz)
d
(CDCl3) 14.48,
41.95, 43.83, 46.94, 49.48, 83.22, 94.18, 112.67, 115.24, 126.50,
129.70, 130.31, 132.58, 135.31, 138.26, 139.88, 141.03, 153.35;
LRMS for [M-H2OþH]þ 471.0 m/z; [MeH]ꢀ 487.3 m/z.
23.28, 26.30, 27.45, 29.63, 31.34, 33.76, 39.59, 42.80, 43.81, 46.83,
49.44, 70.00 (2ꢂ), 83.19, 100.18, 110.19 (2ꢂ), 112.65, 115.23, 126.49,
127.55 (4ꢂ), 127.95 (2ꢂ), 128.54 (4ꢂ), 132.61, 136.88, 138.24, 140.67
(2ꢂ), 153.36, 159.65 (2ꢂ); LRMS for [M þ H]þ 575.3 m/z.
4.1.5.5. (17
White powder; IR
b
)-17-(4-Iodobenzyl)-estra-1(10),2,4-triene-3,17-diol (22).
, (film on NaCl) 3346 (OH); 1H NMR
(CDCl3) 0.96
-CH2), 2.85
y
d
4.1.5. Procedure for HgCl2-catalyzed samarium-Barbier reaction
(synthesis of compounds 18e21)
(s, 3H, 18eCH3), 2.61 and 2.88 (2d, 2H, J ¼ 13.4 Hz, 17
a
(m, 2H, 6eCH2), 6.58 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.64 (dd, 1H,
J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.07 (d, 2H, J ¼ 8.2 Hz, 20eCH and
60eCH), 7.18 (d, 1H, J ¼ 8.4 Hz, 1eCH), 7.64 (d, 2H, J ¼ 8.2 Hz, 30eCH,
In a dry flask, 40 mesh samarium powder (108 mg, 0.72 mmol),
TBS-E1 [44] (200 mg, 0.48 mmol) and 4-trifluoromethylbenzyl
bromide (172 mg, 0.72 mmol) were weighed under a nitrogen
atmosphere. The flask was purged with argon, anhydrous degassed
THF (1.5 mL) was added and the solution cooled to 0 ꢁC. HgCl2
(29 mg, 0.107 mmol) was dissolved in anhydrous degassed THF
(0.2 mL) and added to the cool mixture. The mixture was stirred
under argon at 0 ꢁC for 2 h, at room temperature for 2 h, then
filtered on celite and evaporated to give the crude TBS-18, which
was submitted to the deprotection procedure (TBS hydrolysis with
TBAF) as reported above to yield 18 (14%, two steps). The same
procedure was used for the synthesis of 3-TBS-19 (57%), 3-TBS-20,
3-TBS-21 (49%) and TBS-22 (58%) which after hydrolysis of the TBS
group afforded 20 (11%, two steps), 19, 21 and 22 (91e94%).
50eCH); 13C NMR (75 MHz)
d (CDCl3) 14.44, 23.27, 26.32, 27.48, 29.64,
31.35, 33.84, 39.66, 41.91, 43.85, 46.92, 49.51, 83.24, 91.76, 112.70,
115.26, 126.51, 132.55, 133.14 (2ꢂ), 137.08 (2ꢂ), 138.14, 138.25,
153.41; LRMS for [MeH2OþH]þ 471.1 m/z; [M-H]ꢀ 487.8 m/z.
4.1.6. Synthesis of pyridin-3-ylmethyl derivative 23
An LDA solution was prepared by dissolving diisopropylamine
(662 mL, 4.72 mmol) in anhydrous THF (6 mL) cooling the solution
to 0 ꢁC, adding a solution of n-BuLi (2.1 mL, 4.72 mmol) in hexanes
and stirring the mixture at 0 ꢁC for 30 min. Anhydrous HMPA
(821
mL, 4.72 mmol) was added and the mixture stirred an addi-
tional 15 min at 0 ꢁC. 3-Picoline (454
mL, 4.72 mmol) was dissolved
in anhydrous THF (6 mL) and added dropwise to the LDA/HMPA
solution and the resulting solution stirred for 30 min at 0 ꢁC.
Estrone (319 mg,1.18 mmol) was dissolved in anhydrous THF (3 mL)
and added dropwise to the 3-picoline anion solution. The mixture
was stirred 1 h at room temperature, then quenched with a satu-
rated aqueous NH4Cl solution and extracted with EtOAc. The
organic phase was washed with water, dried over MgSO4, filtered
and evaporated. The crude product was purified by flash
4.1.5.1. (17
diol (18). White powder; IR
b
)-17-(4-trifluoromethylbenzyl)-estra-1(10),2,4-triene-3,17-
y
(film on NaCl) 3330 (OH); 1H NMR
d
(CDCl3) 0.97 (s, 3H, 18eCH3), 2.73 and 3.00 (2d, 2H, J ¼ 13.3 Hz,
17
a
-CH2), 2.84 (m, 2H, 6eCH2), 6.59 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.65
(dd,1H, J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.18 (d, 1H, J ¼ 8.5 Hz,1eCH),
7.44 (d, 2H, J ¼ 8.0 Hz, 20eCH and 60eCH), 7.57 (d, 2H, J ¼ 8.0 Hz,
30eCH and 50eCH); 13C NMR (75 MHz)
d (CDCl3) 14.39, 23.22, 26.28,

4236
D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
chromatography on silica gel with hexanes/EtOAc (60/40 to 70/30)
which was treated with TBAF (1.0M) as reported above to yield
to yield 58 mg (14%) of 23.
94 mg (93%) of 27.
4.1.6.1. (17
(23). White powder; IR
b
)-17-(Pyridin-3-ylmethyl)-estra-1(10),2,4-triene-3,17-diol
4.1.8.1. (17
b
)-17-(Cyclopropylmethyl)-estra-1(10),2,4-triene-3,17-diol
y
, (film on NaCl) 3440 (OH); 1H NMR
(27). White powder; IR
y
, (film on NaCl) 3320 (OH); 1H NMR
d
(DMSO-d6) 0.84 (s, 3H,18eCH3), 2.55 and 2.85 (2d, 2H, J ¼ 13.6 Hz,
d
(CDCl3) 0.13 and 0.55 (2m, 4H, 20eCH2 and 30eCH2), 0.88 (t, 1H,
17
a
-CH2), 2.70 (m, 2H, 6eCH2), 6.48 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.55
J ¼ 6.6 Hz, 20eCH), 0.92 (s, 3H, 18eCH3), 1.48 (d, 2H, J ¼ 6.0 Hz,
(dd,1H, J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.02 (d,1H, J ¼ 8.4 Hz,1eCH),
7.17 (dd, 1H, J1 ¼ 4.9 Hz, J2 ¼ 7.7 Hz, 50eCH), 7.66 (d, 1H, J ¼ 7.8 Hz,
60eCH), 8.37 (dd, J1 ¼ 4.9 Hz, J2 ¼ 1.6 Hz, 40eCH), 8.47 (d, 1H,
17
a
-CH2), 2.81 (m, 2H, 6eCH2), 6.56 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.63
(dd,1H, J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.14 (d,1H, J ¼ 8.4 Hz,1eCH);
13C NMR (75 MHz)
d
(CDCl3) 4.28, 4.66, 5.81, 14.15, 23.32, 26.28,
J ¼ 1.7 Hz, 20eCH); 13C NMR (75 MHz)
d
(DMSO-d6) 14.72, 22.90,
27.42, 29.64, 31.60, 34.73, 39.50, 41.68, 43.79, 46.34, 49.38,
84.11, 112.58, 115.18, 126.48, 132.72, 138.29, 153.28; LRMS for
[MeH2OþH]þ 309.1 m/z and [M þ H]þ 327.1 m/z.
26.18, 27.20, 29.20, 31.04, 32.06, 39.66, 43.34, 46.82 (2ꢂ), 48.82,
82.00, 112.72, 114.93, 122.63, 126.05, 130.46, 134.95, 137.20, 138.42,
146.75, 151.75, 154.92; LRMS for [M þ H]þ 364.3 m/z.
4.1.8.2. (17b)-17-{[2-(2-Bromoethyl)cyclopropyl]methyl}estra-1(10),
4.1.7. Procedure for oxirane opening (synthesis of compounds 24
and 25)
2,4-triene-3,17-diol (29). Allyl derivative 26 [45] (200 mg,
0.44 mmol) was dissolved in anhydrous CH2Cl2 (8 mL). 4-
Bromobutene (0.44 mL, 4.37 mmol) and 2nd generation Grubb’s
A 2-furyllithium solution was prepared by cooling a solution of
n-BuLi (1.76 mL) in hexanes solution to ꢀ20 ꢁC, adding a solution of
catalyst
([1,3-bis(2,4,6-trimethylphenyl)-2-imidazolidinylidene]
furan (143
mL, 1.96 mmol) in anhydrous Et2O (4 mL) and refluxing
dichloro(phenylmethylene)(tricyclohexylphosphine) ruthenium)
(37 mg, 0.04 mmol) were added, the mixture refluxed overnight
and then concentrated on a rotary evaporator. The crude product
was purified using a silica gel 25 þ M column on a Biotage chro-
matographer eluted and hexanes/EtOAc (97.5/2.5 to 85/15) as
eluent to yield 166 mg (71%) of 28. This latter was submitted to
cyclopropanation and TBS hydrolysis as described above for
the solution for 4 h. Oxirane-E1 [34] (60 mg, 0.196 mmol) was
dissolved in anhydrous THF (4.4 mL) and added to the furyllithium
solution at ꢀ78 ꢁC. The mixture was stirred overnight, letting it
reach room temperature. It was then quenched with water,
extracted with EtOAc, dried over MgSO4, filtered and evaporated.
The crude product was purified by flash chromatography on silica
gel with hexanes/EtOAc (80/20) to yield 25 mg (36%) of 24.
compound 27 to yield 29. White powder; IR
(OH); 1H NMR
(CDCl3) 0.39 (t, 4H, J ¼ 6.5 Hz, CH2 of cyclopropyl),
0.88 (m, 2H, 10-CH and 20-CH), 0.91 (s, 3H, 18eCH3), 1.47 and 1.50
-CH2), 2.81 (m, 2H, 6eCH2), 3.48 (m, 2H,
y, (film on NaCl) 3305
d
4.1.7.1. (17
White powder; IR
b
)-17-(2-Furylmethyl)-estra-1(10),2,4-triene-3,17-diol (24).
, (film on NaCl) 3301 (OH); 1H NMR
(CDCl3)
-CH2),
y
d
(2d, J ¼ 8.8 Hz, 17
a
0.96 (s, 3H, 18eCH3), 2.84 and 2.94 (2d, 2H, J ¼ 14.7 Hz, 17
a
CH2CH2Br), 6.56 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.63 (dd, 1H, J1 ¼ 2.7 Hz,
2.83 (m, 2H, 6eCH2), 6.18 (d, 1H, J ¼ 3.0 Hz, 50eCH), 6.35 (dd, 1H,
J1 ¼ 2.0 Hz, J2 ¼ 3.0 Hz, 30eCH), 6.56 (d, 1H, J ¼ 2.6 Hz, 4eCH), 6.63
(dd, 1H, J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 7.17 (d, 1H, J ¼ 8.4 Hz,
J2 ¼ 8.4 Hz, 2eCH), 7.14 (d, 1H, J ¼ 8.4 Hz, 1eCH); 13C NMR (75 MHz)
d
(CDCl3) 11.23, 13.54, 14.15, 17.82, 23.38, 26.27, 27.41, 29.64, 31.57,
33.25, 34.64, 37.50, 39.51, 40.98, 43.77, 46.41, 49.41, 83.98, 112.60,
115.19, 126.48, 132.70, 138.30, 153.29; LRMS for [MeH2OþH]þ 415.1
and 417.1 m/z and [M þ H]þ 433.1 and 435.1 m/z.
1eCH), 7.38 (d, 1H, J ¼ 1.2 Hz, 30eCH); 13C NMR (75 MHz)
d (CDCl3)
14.29, 23.41, 26.24, 27.42, 29.62, 31.52, 34.72, 35.94, 39.48,
43.78, 46.52, 49.56, 82.79, 108.26, 110.38, 112.62, 115.20, 126.49,
132.59, 138.25, 141.70, 153.24, 153.30; LRMS for [MeH2OþH]þ
335.1 m/z.
4.2. Enzymatic assay
As described in the previous study [21], the steroid sulfatase
(STS) assay was performed using a human embryonic kidney
(HEK)-293 cell transiently transfected with a sulfatase expression
vector (pCMV-sulfa) as the source of enzyme. The cells were
prepared by performing 5 freezing (ꢀ80 ꢁC) and thawing cycles
and homogenization using a Dounce homogenizer. The reaction
4.1.7.2. (17b)-17-(4-Methyl-2-thienyl)-estra-1(10),2,4-triene-3,17-diol
(25). This compound was obtained as described above for 24, but
using 4-methylthiophene instead of furan. White powder (22%
yield); IR y d (CDCl3) 0.97 (s, 3H,
, (film on NaCl) 3340 (OH); 1H NMR
18eCH3), 2.24 (d, 3H, J ¼ 0.7 Hz, 60eCH), 2.83 (m, 2H, 6eCH2), 2.91
and 3.05 (2d, 2H, J ¼ 14.4 Hz, 17 -CH2), 6.58 (d, 1H, J ¼ 2.6 Hz,
a
was carried out using 100
mM of estrone sulfate (E1S) (0.5% of
4eCH), 6.64 (dd, 1H, J1 ¼ 2.7 Hz, J2 ¼ 8.4 Hz, 2eCH), 6.75 (d, 2H,
which were [3H]-E1S) in 1.0 mL of tris-acetate buffer (pH 7.4)
containing 5 mM ethylenediaminetetraacetic acid (EDTA), 10%
glycerol and an ethanolic solution of test compound. Only ethanol
was used for the control. After 2 h of incubation at 37 ꢁC, the
reaction was stopped by adding 1.0 mL of xylene. The tubes were
vortexed and centrifugated at 3500 g for 20 min to separate
J ¼ 14.6 Hz, 30eCH and 50eCH), 7.17 (d, 1H, J ¼ 8.4 Hz, 1eCH); 13C
NMR (75 MHz)
d (CDCl3) 14.51, 15.72, 23.47, 26.30, 27.45, 29.64,
31.50, 34.45, 37.75, 39.61, 43.81, 46.70, 49.81, 82.76, 112.65, 115.23,
119.86, 126.52, 129.91, 132.64, 137.16, 138.28, 140.01, 153.33; LRMS
for [MeH2OþH]þ 365.1 m/z.
organic and aqueous phases. A 400 mL aliquot of each phase was
4.1.8. Synthesis of cyclopropyl derivatives 27 and 29
used for radioactivity measurement using a Wallac 1400 scintil-
lation counter (Ramsey, MN, USA). The results were expressed as
the percent of estrone (E1) produced (100% for control without
inhibitor) and the percent of inhibition then determined for three
An Et2Zn solution in toluene (0.67 mL, 0.74 mmol) was added
to anhydrous CH2Cl2 (0.75 mL) and cooled to 0 ꢁC. A trifluoro-
acetic acid solution (0.37 mL, 0.74 mmol) in CH2Cl2 was added
dropwise and the solution was stirred 20 min at 0 ꢁC. A CH2I2
solution in CH2Cl2 (0.37 mL, 0.74 mmol) was then added and the
solution stirred a further 20 min at 0 ꢁC. Allyl derivative 26 [32]
(170 mg, 0.372 mmol) was dissolved in anhydrous CH2Cl2
(0.4 mL) and added to the CF3CO2ZnCH2I solution. The mixture
was stirred 30 min at room temperature, quenched with a satu-
rated aqueous NH4Cl solution, extracted with EtOAc, dried over
MgSO4, filtered and evaporated to yield 141 mg (80%) of 3-TBS-26
inhibitor concentrations (0.01, 0.1 and 1 mM).
Acknowledgments
This work was supported by an operating grant from the
Canadian Institutes of Health Research (CIHR). Careful reading of
the manuscript by Micheline Harvey is also greatly appreciated.

D. Fournier, D. Poirier / European Journal of Medicinal Chemistry 46 (2011) 4227e4237
4237
[21] R.P. Boivin, V. Luu-The, R. Lachance, F. Labrie, D. Poirier, J. Med. Chem. 43
(2000) 4465e4478.
References
[22] D. Gosh, Cell. Mol. Life Sci. 64 (2007) 2013e2022.
[23] F.G. Hernandez-Guzman, T. Higashiyama, W. Pangborn, Y. Osawa, D. Gosh,
J. Biol. Chem. 278 (2003) 22989e22997.
[24] S.M. Singh, L. Xun, F. Labrie, Tetrahedron Lett. 35 (1994) 1157e1160.
[25] V. Dimitrov, K. Kostrova, M. Genov, Tetrahedron Lett. 37 (1996) 6787e6790.
[26] M. Salman, B.R. Reddy, S. Ray, P.L. Stotter, G.C. Chamness, J. Steroid Biochem.
24 (1988) 539e548.
[1] M.J. Reed, A. Purohit, L.W. Woo, S.P. Newman, B.V. Potter, Endroc. Rev. 26
(2005) 171e202.
[2] A. Horvath, A. Billich, Expert Opin. Ther. Patents 15 (2005) 1541e1553.
[3] P. Nussbaumer, A. Billich, Expert Opin. Ther. Patents 13 (2003) 605e625.
[4] K. Aidoo-Gyamfi, T. Cartledge, K. Shah, S. Ahmed, Anti-Cancer Agents Med.
Chem. 9 (2009) 599e612.
[5] D.Poirier,L.C.Ciobanu,R.Maltais, ExpertOpin. Ther. Patents9(1999)1083e1099.
[6] M.J. Reed, A. Purohit, L.W.L. Woo, B.V.L. Potter, Endocr.-Rel. Cancer 3 (1996) 9e23.
[7] N.M. Howarth, A. Purohit, M.J. Reed, B.V.L. Potter, J. Med. Chem. 37 (1994)
219e221.
[8] J.-Y. Winum, A. Scozzafava, J.-L. Montero, C.T. Supuran, Expert Opin. Ther.
Patents 14 (2004) 1273e1308.
[27] M. Salman, B.R. Reddy, P. Delgado, P.L. Stotter, L.C. Fulcher, G.C. Chamness,
Steroids 56 (1991) 375e387.
[28] M. Salman, B.R. Reddy, S. Ray, P.L. Stotter, J. Steroid Biochem. 24 (1986)
539e548.
[29] A. Krief, A.M. Laval, Chem. Rev. 99 (1999) 745e777.
[30] H.B. Kagan, Tetrahedron 59 (2003) 10351e10372.
[31] E. Prasad, R.A. Flowers, J. Am. Chem. Soc. 124 (2002) 6895e6899.
[32] A. Dahlen, G. Hilmersson, Eur. J. Inorg. Chem. 17 (2004) 3393e3403.
[33] X. Gao, X. Wang, R.F. Cai, J.D. Wei, S.H. Wu, Acta Chim. Sinica 51 (1993)
1139e1144.
[34] C.E. Cook, R.C. Corley, M.E. Wall, Steroids 33 (1968) 2789e2793.
[35] J.C. Lorenz, J. Long, Z. Yang, S. Xue, Y. Xie, Y. Shi, . J. Org. Chem. 69 (2004)
327e334.
[36] M.H. Bracey, B.F. Cravatt, R.C. Stevens, FEBS Lett. 567 (2004) 159e165.
[37] V. Ahmed, Y. Liu, C. Silvestro, S.D. Taylor, Bioorg. Med. Chem. 14 (2006)
8564e8573.
[38] D.M. Bender, R.M. Williams, J. Org. Chem. 62 (1997) 6690e6691.
[39] T.K. Venkatachalam, H. Huang, G. Yu, F.M. Uckun, Synth. Commun. 34 (2004)
1489e1497.
[40] A. Gourdon, Eur. J. Org. Chem. (1998) 2797e2801.
[41] E. Díez-Barra, J.V. García-Martinez, S. Merino, R. del Rey, J. Rodríguez-López,
P. Sánchez-Verdú, J. Tejeda, J. Org. Chem. 66 (2001) 5664e5670.
[42] J. Adams, M. Belley, J. Org. Chem. 51 (1986) 3878e3881.
[43] H. Gilman, F. Schulze, Bull. Soc. Chim. 41 (1927) 1479e1481.
[44] J.D. Pelletier, F. Labrie, D. Poirier, Steroids 59 (1994) 536e547.
[45] M.R. Tremblay, D. Poirier, J. Comb. Chem. 2 (2000) 41e65.
[9] S.R. Hanson, M.D. Best, C.-H. Wong, Angew. Chem. Int. Ed. Engl. 43 (2004)
5736e5763.
[10] W.J. Spillane, C.A. Ryder, R.M. Walsh, P.J. Curran, D.G. Concagh, S.N. Wall, Food
Chem. 56 (1996) 255e261.
[11] B.E. Maryanoff, M.J. Costanzo, S.O. Nortey, M.N. Greco, R.P. Shank,
J.J. Schupsky, M.P. Ortegon, J.L. Vaught, J. Med. Chem. 41 (1998) 1315e1343.
[12] L.C. Ciobanu, R. Maltais, D. Poirier, Org. Lett. 2 (2000) 445e448.
[13] D. Poirier, M. Bérubé, D. Fournier, Innovation & Perspectives in Solid Phase
Synthesis & Combinatorial Libraries (Peptides, Proteins and Nucleic Acids
Small Molecules Organic Chemical Diversity). Mayflower Worldwide Limited,
Kingswinford, England, 2004, 281e285.
[14] D. Fournier, P.A. Breuil, D. Poirier, Adv. Exp. Med. Biol. 611 (2009) 219e220.
[15] D. Poirier, L.C. Ciobanu, M. Bérubé, Bioorg. Med. Chem. Lett. 12 (2002)
2833e2838.
[16] L.C. Ciobanu, D. Poirier, J. Comb. Chem. 5 (2003) 429e440.
[17] L.C. Ciobanu, D. Poirier, ChemMedChem 1 (2006) 1249e1259.
[18] M. Bérubé, F. Delagoutte, D. Poirier, Molecules 15 (2010) 1590e1631.
[19] L.C. Ciobanu, R.P. Boivin, V. Luu-The, F. Labrie, D. Poirier, J. Med. Chem. 42
(1999) 2280e2286.
[20] L.C. Ciobanu, V. Luu-The, D. Poirier, J. Steroid Biochem. Mol. Biol. 80 (2002)
339e353.