Solid phase synthesis of 17α-E/Z-(X-phenyl)-vinyl estradiols using the stille coupling reaction

Article abstract of DOI:10.1016/S0040-4020(00)00004-1

As a continuation of our program to develop probes for the hormone binding domain (HBD) of the estrogen receptor (ER), we designed a series of novel 17α-E/Z-(X-phenyl)-vinyl estradiols. Based upon our experience with solution chemistry we applied solid phase synthesis using carboxylated resins to synthesize the new compounds. The Stille coupling reaction permitted the introduction of a variety of functional groups and positional isomers on the terminal phenyl group. Subsequent cleavage from the resin generated a series of novel estradiol derivatives. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00004-1

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 1623–1629
Solid Phase Synthesis of 17a-E/Z-(X-Phenyl)-Vinyl Estradiols
Using the Stille Coupling Reaction
†
*
Choon Young Lee and Robert N. Hanson
´
Department of Pharmaceutical Sciences, Bouve College of Health Sciences, Northeastern University, Boston, MA 02115 USA
Received 23 July 1999; accepted 22 December 1999
Abstract—As a continuation of our program to develop probes for the hormone binding domain (HBD) of the estrogen receptor (ER), we
designed a series of novel 17a-E/Z-(X-phenyl)-vinyl estradiols. Based upon our experience with solution chemistry we applied solid phase
synthesis using carboxylated resins to synthesize the new compounds. The Stille coupling reaction permitted the introduction of a variety of
functional groups and positional isomers on the terminal phenyl group. Subsequent cleavage from the resin generated a series of novel
estradiol derivatives. ᭧ 2000 Elsevier Science Ltd. All rights reserved.
Introduction
As a part of our ongoing program to design and develop new
therapeutic agents for the treatment of breast cancer, we
have focused on new steroidal derivatives that interact at
the hormone binding domain (HBD) of the estrogen recep-
tor (ER). While many of our initial studies confirmed the
established estrogen receptor structure activity relation-
ships, derivatives with the E- and Z-X-vinyl group at
the 17a-position particularly demonstrated unusual
properties.1 Further explorations with phenylvinyl (I) and
phenylselenovinyl (II) estradiol suggested that receptor
affinities comparable to estradiol itself could be maintained
The synthesis of our target compounds to date had relied on
traditional solution phase chemistry. In order to prepare new
derivatives containing a variety of functional groups or
existing as positional isomers, we considered approaches
in spite of the apparent steric bulk of the 17a substituent.2
Recent publications of the crystal structure of the liganded
HBD of the ER3 suggested that the 17a groups project into a
region that may accommodate significant steric tolerance.
that could generate a large number of compounds more
We have elected to develop new estradiol derivatives that
easily. The logical choice was solid phase synthesis. We
could exploit that tolerance.
envisioned that we could append our steroid to the inert
polymer support, divide it into discrete aliquots, perform
the requisite synthetic transformation, remove its individual
products from the support and then characterize them.
While a significant body of literature existed for solid
phase synthesis (SPS) with steroids4–9 and for Stille
coupling,10–12 there were no prior reports on the specific
application that we wished to carry out. For example, Poirer
et al., has described solid phase transformations of both
androstanes and 16a-substituted estradiols,4 however,
neither employed transformations comparable to those we
would require. Similarly, several groups have reported the
use of the Stille reaction to couple aromatic and alkyl
groups10–12 but with fewer structural constraints than those
Keywords: solid phase synthesis; estrogen receptor probes; carboxylation;
hydrostannylation; Stille reaction.
imposed by the estrogen scaffold. Therefore, this work
involved developing new methods to achieve our objectives.
* Corresponding author. E-mail: r.hanson@nunet.neu.edu
†
E-mail: cholee@lynx.dac.neu.edu
0040–4020/00/$ - see front matter ᭧ 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00004-1

1624
C. Y. Lee, R. N. Hanson / Tetrahedron 56 (2000) 1623–1629
Scheme 1. Reagent: (a) Jones reagent (H2Cr2O4, H2SO4, acetone); (b) n-BuLi, TMEDA, cyclohexane, 50ЊC; (c) Dry ice, THF; (d) 17a-Ethynyl estradiol, DCC,
DMAP, CH2Cl2; (e) HSnBu3, Et3B, THF, 50–60ЊC; (f) 17a-Ethynyl estradiol, HSnBu3, Et3B, THF, 50–60ЊC; (g) DCC, DMAP, CH2Cl2; (h) R-Aryl-X, Pd
(PPh3)4, BHT, toluene, N2, reflux; (i) 5 N-NaOH in CH3OH–Dioxane (1:3); (j) 5%-CH3COOH; (k) 10%-NaHCO3.
In this report we demonstrate our approach to developing
the solid phase synthesis of the 17a-substituted phenylvinyl
estradiols. This involved coupling the steroid intermediates
to the resin, identifying appropriate reaction conditions and
cleaving the final products from the resin. The result is a
reliable method for generating a novel series of functional-
ized estradiols which can be evaluated for their biological
properties.
selection of a linker that could be both formed and cleaved
under mild conditions. This was based on our observations
that 17a-substituted estradiols were unstable under strongly
acidic conditions such as those frequently used to release
products from the resins. Therefore our resin of choice was
carboxylated polystyrene which could be esterified under
neutral conditions and ultimately cleaved with mild base.
Our first example (compound 8a) was prepared using the
carboxylated resin obtained either by oxidation of a Wang
resin using Jones reagent13 or by carboxylation of a poly-
styrene resin via lithiation with n-butyl lithium.14 The
reactions for both methods were easily monitored by the
appearance of the 1700 cmϪ1 absorption in the FT-IR
spectrum. The loading capacity of our carboxylated resins
was determined by coupling 17a-ethynyl estradiol onto the
resins using DCC in the presence of catalytic amount of
DMAP and measuring its subsequently cleaved estradiol
derivatives from the aliquot of the resins. The loading15 of
oxidized Wang resin was 0.4–0.6 mmol gϪ1 and that of
carboxylated polystyrene was 1.5–1.9 mmol gϪ1. Once we
confirmed the utility of coupling through the ester linkage
using carboxy polystyrene resin we employed the commer-
cially available carboxy polystyrene for the remainder of
our studies. The loading yield of the reaction using
the resins with already known loading capacity
(2.47 mmol gϪ1) was 82%. The yield was determined by
‘cleave and characterize’ methods.
The approach that we selected incorporated several features.
First, we chose the carboxylated resins because the estrogen
could be selectively coupled through its phenolic linkage to
the polymer and the ultimate cleavage of the ester bond at
the end of the synthetic sequence would pose few problems.
Use of an ether linkage would require either acidic or reduc-
tive cleavage, which would not be compatible with the
functional groups present in the intermediates or final
products. Similarly, amides, carbamates and photolabile
links could also present potential problems at various steps
of the process. Esterification at the 3-position, however,
would not interfere with either the hydrostannylation or the
palladium (0) catalyzed coupling reactions that would occur
at the 17a-position. The integrity of the tertiary alcohols, E/Z-
styryl groups, or functionality on the terminal phenyl group
would be compromised if conditions other than a mild base
were used to remove the product from the resin.
Synthesis of the analogs (Scheme 1) commenced by
coupling the 3-phenolic group of 17a-ethynyl estradiol to
the carboxy polystyrene resin. An antimony (III) chloride
Results and Discussion
One of the key elements of the synthetic scheme was the

C. Y. Lee, R. N. Hanson / Tetrahedron 56 (2000) 1623–1629
Table 1. Yields (%) of Stille coupling reaction using solid phase synthesis
1625
tributyltin hydride to alkyne. At 60ЊC the reaction generated
greater than 20:1 (E/Z) ratio bound to the solid phase. To
increase the ratio of the Z-isomer, triethylborane was used as
a radical initiator and the reaction was run at low tempera-
ture. The proportion of the Z-isomer (Z/Eˆ1:10) increased,
however, the reaction required a longer time and the loading
yield for the hydrostannylation was slightly lower than at
higher temperature (1.44 mmol gϪ1 by the dry weight
difference method) because of more unreacted 17a-ethynyl
estradiol in the reaction mixture.
The resin-bound hydrostannnylated estradiol was subjected
to the Stille coupling reaction22 using a variety of substituted
aryl halides to generate the target compounds (Table 1). As
shown in Scheme 1, Pd(PPh3)4 was used as the catalyst for
the reaction and 3,5-di-t-butyl-4-hydroxytoluene (BHT)
was added as a scavenger. The use of Pd(PPh3)4 generated
an insoluble by-product that caused coloration of the resin,
however, it was easily removed by rinsing it through
the built-in filter (50–70 mm). After completion of all the
reaction steps, the product was cleaved from the
resin by saponification with 5 N NaOH dissolved in
CH3OH–Dioxane (1:3).
Compound
R1 (ortho)
R2 (meta)
R3 (para)
Yield (%)
4a:E
5a:E
6a:E
6b:Z
7a:E
8a:E
8b:Z
9a:E
CF3
H
H
H
CH3
H
H
CF3
H
H
H
CH3
CH3
H
H
H
CF3
CF3
H
H
H
OCH3
38
33
49
17
38
75
54
36
H
H
assay confirmed the presence of the steroids on the resins.16–18
The absence of color change with bromocresol green
suggested that no free carboxylic acid groups remained on
the resin.19 The appearance of a peak at 3301 cmϪ1 in the IR
spectrum, corresponding to the C–H stretch of the ethynyl
group, also confirmed the reaction and a shift of carbonyl
As shown in Table 1, the unoptimized yields of the Stille
reactions on solid phase ranged from 17–75%, comparable
to those observed for solution phase synthesis.23
Compounds 5a (para-trifluoromethylphenyl, E-isomer)
and 5b (para-trifluoromethylphenyl, Z-isomer) were
isolated from the Stille reaction in a ratio of 98:2.
Compound 7a (meta-methylphenyl, E-isomer) and 7b
(meta-methylphenyl, Z-isomer) were also obtained in a
ratio of 96:4. Although the Z-tri-n-butylstannyl vinyl
estradiol was initially present on the resin, no Z-isomers
of compound 3a, 4a, 6a or 8a were isolated from the Stille
coupling, instead, 17a-vinyl estradiol, resulting from
protiodestannylation was recovered as a side product.
Because an excess of reagent was used to drive the reaction
to completion, unreacted hydrostannylated 17a-E/Z-(tri-
n-butylstannyl)-vinyl estradiol was not detected after the
Stille reaction. It is possible that the Z-isomers either
isomerized to thermodynamically more stable E-isomers
under the conditions required for the Stille reaction or
underwent protiodestannylation. As previously observed,
the Z-isomer is much more susceptible to protiodestannyl-
ation than the E-isomer and the appearance of the side
product under either solid phase or solution phase synthesis
was approximately the same.
absorption to higher frequency (from 1690 to 1734 cmϪ1
was also observed.
)
The subsequent hydrostannylation step incorporated either
the use of hydrostannylation of bound ethynyl estradiol
(Method A) or hydrostannylation of ethynyl estradiol in
solution phase synthesis followed by coupling to the resin
(Method B). The resin-bound 17a-ethynyl estradiol was
hydrostannylated with tributyltin hydride using triethyl-
borane as a radical initiator20 to afford a mixture of the
17a-E/Z-tri-n-butylstannylvinyl estradiol in 20–30%
(0.12 mmol gϪ1 of E, 0.01 mmol gϪ1 of Z) loading yields.
Varying the reaction conditions, e.g. different solvents,
temperatures, or reaction times, did not improve the yields.
Therefore, a direct coupling of 17a-E/Z-tri-n-butylstannyl-
vinyl estradiols used to overcome the low efficiency of this
step. 17a-Ethynyl estradiol was hydrostannylated at 60ЊC
and the crude mixture was directly transferred to the resin
slurry in CH2Cl2. The mixture was treated with a 2–3 fold
excess of DCC and a catalytic amount of DMAP was added.
The loading yield for the coupling reaction was
0.59 mmol gϪ1 with a Z/E ratioˆ1:20. The low loading
yield was due to use of the acetic acid for the protonation
of phenoxide ion after cleavage, subjecting the products to
protiodestannylation and reducing the expected loading
yield. Because the cleavage after hydrostannylation did
not provide a precise loading yield, we subsequently used
the dry weight difference between pre- and post-reaction to
determine the loading yield. Using the dry weight difference
method, the yield for the hydrostannylation reaction was
1.55 mmol gϪ1 for both E- and Z-isomers. Because hydro-
stannylation on the resin did not afford satisfactory yields,
Method B was the method of choice. As we have previously
reported21 the ratio of E and Z isomers is a function of the
reaction temperature, time and stoichiometric ratio of
The isolated product were characterized by standard spec-
troscopic methods (FT-IR, 1H and 13C NMR) and analytical
methods. The data were consistent with the proposed struc-
tures. Stereochemical assignments for compounds 5a and
5b were based on the C20, C21 olefinic proton coupling
constants for which Eˆ16 Hz and Zˆ12.9 Hz, respectively.
For compounds 7a and 7b, the observed coupling constants
were 18.2 Hz for the C20 E-vinyl proton and 13.1 Hz for the
C20 Z-vinyl proton. In 13C NMR, long range couplings were
observed for the compounds 3a–5a and 5b containing the
trifluoromethyl group. Coupling with strongly electronega-
tive fluorine was found at the carbon directly attached to the
fluorine (1JC–F) and one (2JC–F) and two carbons distant (3JC–F).
The carbons appeared as quartets and the coupling constants

1626
C. Y. Lee, R. N. Hanson / Tetrahedron 56 (2000) 1623–1629
1
2
3
were approximately JC–Fˆ270 Hz, JC–Fˆ32 Hz, JC–F
ˆ3–5 Hz, respectively.
Atlantic Microlab, Inc. (Norcross, GA). As on-resin reaction
monitoring methods, color tests and FT-IR methods were
used. Bomocresol green (0.5% in ethanol, pHˆ8) was used
to assay for free carboxylic acids.18 The color of the stock
solution was dark blue and changed to yellow in the
presence of free carboxy groups. Antimony (III) chloride
solution (25% in CCl4) was also used to determine whether
the steroid (17a-ethynyl estradiol) was coupled to the resin
and a positive test result for the presence of estradiol was
indicated by the color purple.16–18 In addition, a spectro-
scopic method (FT-IR) was facilitated to detect chromo-
phore change by reaction.
Initial biological evaluation of these compounds indicates
that they retain substantial affinity for the ER-LBD (results
to be published elsewhere). Because both the properties of
the aryl substitutent and its position on the ring (o/m/p)
appear to affect the receptor binding, a more extensive
evaluation of the derivatives is required.
In conclusion, we have applied the Solid Phase Synthesis
methodology using carboxylated resins to generate a series
of novel ER-LBD ligands. The initial objectives of this
study, the simplification of the purification steps and the
simultaneous production of both E- and Z-isomers, were
largely achieved. The products were in acceptable yields,
however no attempt had been made at this point to optimize
conditions and clearly the yields could be improved.
Analysis of the products indicated that the initial method
provided only the E-isomer for most of the target
compounds even though both E and Z-isomers were present
after hydrostannylation reaction. We anticipate that modifi-
cations in both the coupling and cleavage steps would
improve the yields for the chemically more sensitive
Z-isomers. Nevertheless, this study has demonstrated the
feasibility of solid phase synthesis for generating a variety
of functionalized estradiol derivatives. Based on our pre-
liminary biological results, we anticipate that further
modifications of the phenyl group will yield promising
results and we intend to adapt these methods for use in a
combinatorial approach to generate diverse target
compounds as ER-LBD ligands.
Preparation of the carboxylated resin
(Method A). The Wang resins (1 g, 0.75 mmol) were
swelled in the CH2Cl2 overnight and rinsed twice with
THF, CH3OH, CH2Cl2 and acetone. Acetone (5 mL) was
added to the swelled resins. To the slurry was added 1 mL
of Jones reagent13 in a dropwise manner. The mixture was
allowed to stand at room temperature for 24 h. The resin
mixture was rinsed twice with water–acetone (1:1),
CH3OH, DMF, DMSO and CH2Cl2 and dried in vacuo.
The loading capacity after the carboxylation reaction was
0.4–0.6 mmol gϪ1, which was determined with the coupling
of 17a-ethynyl estradiol to the resin. The aliquot of the
resins was characterized by FT-IR. FT-IR (KBr) n: 3000–
3500 (OH, broad), 1690 (CyO, broad), 1603, 1492, 1452
(aromatic ring), 1279 (C–O).
(Method B). The carboxylation of a polystyrene resin was
accomplished using the method described by Farrall et al.14
FT-IR (KBr) n: 3420 (OH, broad), 1630 (CyO, broad),
1200–1400 (C–O, broad). Loading capacity: 1.5–1.9
Experimental
Materials
mmol gϪ1
.
Coupling 17a-ethynyl estradiol to the resins
Reagents and solvents were obtained from commercial
sources (Aldrich and Sigma) and were used without further
purification. Wang resins and carboxylated polystyrene
resins were obtained from Novabiochem. The loading capa-
cities of the resins, 0.75 mmol gϪ1 for the Wang resin and
2.47 mmol gϪ1 for the polystyrene resin, were determined
by the manufacturer.
The carboxylated Wang resin (2.3 g) or polystyrene resin
(2.5 g) was placed in the reactor equipped with a magnetic
stirrer. The resin was swelled in the CH2Cl2 for 5 h and
washed sequentially with THF, DMF, CH3OH, THF and
CH2Cl2. To the resin was added 0.23 g (1.1 mmol) of dicy-
clohexylcarbodiimide (DCC) and 5 mL of CH2Cl2 and the
mixture was mildly stirred for 10 min. To the slurry was
added 0.75 g (2.6 mmol) of 17a-ethynyl estradiol dissolved
in 10 mL of CH2Cl2–DMF (9:1) solvent and catalytic
amount of 4-dimethylaminopyridine (DMAP). The reaction
mixture was stirred for 5 min and then allowed to stand at
room temperature for 24 h. The resin was washed three
times with CH2Cl2, CH3OH, IPA (60ЊC), THF and DMF
(60ЊC).24 The rinsed resin was dried under vacuum for
5 h. The actual loading of the resin was determined by
quantitative measurement of the material by cleavage
from known weight of resin using 5 N-NaOH in CH3OH–
dioxane (1:3). The resin-bound steroids were characterized
by FT-IR and the cleaved compounds by 1H and 13C NMR
before proceeding to the next step. The loading capacity of
each resin was shown in Method A and B; FT-IR (KBr) n:
3437 (17b-OH), 3301 (17a-CxC–H), 1735 (CyO), 1607,
1493, 1452 (aromatic ring), 1216(C–O).
General methods
A specially designed flask which had a glass frit, through
which the reaction mixture could be filtered by applying
pressure, was used for the solid phase synthesis. Purifi-
cations for the intermediates were done by rinsing resins
three times with the following solvents: CH2Cl2, THF,
DMF, MeOH, CH2Cl2. The cleaved products were purified
on a silica gel column chromatography using the appropriate
solvents and were characterized by melting point, NMR, IR
and elemental analysis. Melting points were determined in
open capillary on an Electrothermal Melting Point Appa-
ratus and were uncorrected. IR spectra were recorded on a
Perkin–Elmer Model 1600 FT-IR spectrometer. 1H and 13
C
NMR spectra were obtained with a Varian XL-300 NMR
spectrometer at 300 MHz in CDCl3, acetone-d6, or DMSO-
d6 as a solvent. Elemental analyses were performed by

C. Y. Lee, R. N. Hanson / Tetrahedron 56 (2000) 1623–1629
1627
Hydrostannylation
Cleavage
The resin was swelled in CH2Cl2 (10 mL) containing 3 mL
of 5 N-NaOH in CH3OH–Dioxane (1:3), and stirred for 1 h.
This cleavage step was repeated three times. Most of the
product was collected from the first attempt, a small amount
by second hydrolysis and almost none from the third trial.
The fractions were combined, evaporated to dryness and
partitioned between ethyl acetate and water. Acetic acid
(1 mL, 5%) was added. The organic phase was washed
with 10% aqueous NaHCO3 to remove the residual acetic
acid, dried over MgSO4, filtered and evaporated to dryness.
The crude product was purified by silica gel column
chromatography or by recrystallization from the appropriate
solvent.
(Method A). The 17a-ethynyl estradiol coupled to the resin
(0.49 g, 0.57 mmol gϪ1) was placed in a dry 25 mL reaction
flask equipped with a reflux condenser and a magnetic
stirrer and was swelled in THF for 1 h. To the slurry in
the dry THF were treated triethylborane (0.7 mL) and
tributyltin hydride (1 mL).20 The mixture was allowed to
stand at 60–70ЊC for 48 h under a nitrogen atmosphere.
The reaction mixture was washed three times each with
CH2Cl2, CH3OH, DMF, CH2Cl2 and ethyl acetate and the
resultant resin was dried in vacuo. An aliquot of the resins
was cleaved with 5 N NaOH in CH3OH–CH2Cl2 (1:2) to
afford a mixture of E- and Z-isomers. The mixture was
separated by chromatography on the silica gel to give a
23% (0.13 mmol gϪ1) yield of products, consisting of 21%
(0.12 mmol gϪ1) of the E-isomer and 2% (0.01 mmol gϪ1
)
17a-20E-21-(2-Trifluoromethylphenyl)-19-norpregna-
1,3,5(10),20-tetraene-3,17b-diol (17a-E-(2-trifluoro
methylphenyl)-vinyl estradiol) (4a). Yieldˆ38%; Rfˆ0.19
(hexane–ethyl acetate, 4:1); mp 224–225ЊC; 1H NMR
(300 MHz, Acetone-d6, d) 1.02 (s, 3H, C18 methyl-H),
1.2–2.4 (m, steroid envelope), 2.7–2.9 (m, 2H, C6-H),
3.98(s, 1H, 17b hydroxyl-H), 6.53 (d, 1H, Jˆ2.3 Hz, C4-
H), 6.58 (dd, 1H, Jˆ2.6, 8.5 Hz, C2-H), 6.64 (d, 1H,
Jˆ15.7 Hz, C20 vinyl-H), 7.0 (dd, 1H, Jˆ2.5, 15.8 Hz, C21
vinyl-H), 7.07 (d, 1H, Jˆ8.7 Hz, C1-H), 7.42 (t, 1H,
Jˆ7.8 Hz, C26-H), 7.60 (t, 1H, Jˆ7.3 Hz, C25-H), 7.69 (d,
1H, Jˆ7.8 Hz, C27-H), 7.81 (d, 1H, Jˆ8.3 Hz, C24-H), 7.98
(s, C3 hydroxy-H); 13C NMR (75.4 MHz, Acetone-d6, d)
14.7 (C18), 24.1 (C15), 27.2 (C11), 28.3 (C7), (C6), 33.4
(C12), 37.5 (C16), 40.7 (C8), 44.6 (C9), 48.4 (C13), 50.0
(C14), 84.3 (C17), 113.5 (C2), 115.9 (C4), 123.4 (C21),
125.6 (q, Jˆ273.2 Hz, C28:CF3), 126.4 (q, Jˆ5.8 Hz, C24),
127.0 (C1), 127.4 (q, Jˆ29.4 Hz, C23), 127.8 (C26), 128.6
(C27), 132.0 (C25), 133.2 (C10), 137.9 (C22), 139.1 (C5),
142.4 (C20), 155.9 (C3); Anal. Calcd for C27H29O2F3: C,
73.30; H, 6.56. Found: C, 73.04; H, 6.68.
of the Z-isomer. Rf (Z-isomer)ˆ0.58 (hexane–ethyl acetate,
4:1); Rf (E-isomer)ˆ0.44 (hexane–ethyl acetate, 4:1);
1
Amorphous; H NMR (CDCl3, 300 MHz, d), 0.88 (s, 3H,
C18-methyl-H), 1.2–2.4 (m, steroid envelope and tributyl-
stannyl-H), 2.7–2.9 (m, 2H, C6-H), 6.06 (d, 1H, Jˆ19.4 Hz,
C21 vinyl-H), 6.22 (d, 1H, Jˆ19.4 Hz, C20 vinyl-H), 6.79 (d,
1H, Jˆ2.4 Hz, C4-H), 6.84 (dd, 1H, Jˆ2.6, 8.4 Hz, C2-H),
7.28 (d, 1H, Jˆ8.8 Hz, C1-H); 13C NMR (CDCl3), 9.6 (C22,
4C), 13.7 (C24, 4C), 14.2 (C18), 23.4 (C15), 26.4 (C11), 27.3
(C25, 4C), 27.4 (C7), 29.2 (C23, 4C), 29.6 (C6), 32.4
(C12), 35.9 (C16), 39.4 (C8), 43.8 (C9), 46.7 (C13), 49.0
(C14), 85.6 (C17), 112.6 (C2), 115.2 (C4), 124.6 (C21),
126.5 (C1), 132.7 (C10), 138.3 (C5), 152.4 (C20), 153.3
(C3); FT-IR (KBr) n: 3445 (17b-OH, broad), 1719
(CyO), 1653 (CyC), 1607, 1493, 1451 (aromatic ring),
1217 (C–O).
(Method B). The 17a-ethynyl estradiol (3 g, 10 mmol) was
dissolved in THF and treated with triethylborane (2 mL,
17 mmol) and tributyltin hydride (3 g, 11 mmol). The
mixture was stirred with a magnetic stirrer at 60ЊC for
16 h. The crude mixture (7.73 g) was evaporated to dryness,
redissolved in the CH2Cl2, and transferred to the swelled
resin (5 g) in CH2Cl2 in the presence of DCC. A catalytic
amount of DMAP was added to the mixture, which was
allowed to stand for 24 h. The resultant functionalized
resin was treated as previously described. The total loading
for both E- and Z-isomers was 0.59 mmol gϪ1 with
0.56 mmol gϪ1 of E-isomer and 0.03 mmol gϪ1 of Z-isomer,
however, by the dry weight difference between pre- and
post-reaction, the loading for both E- and Z-isomers was
17a-20E-21-(3-Trifluoromethylphenyl)-19-norpregna-
1,3,5(10),20-tetraene-3,17b-diol (17a-E-(3-trifluoro
methylphenyl)-vinyl estradiol) (5a). Yieldˆ33%; Rf
(E-isomer)ˆ0.19 (hexane–ethyl acetate, 4:1); mp 244–
1
246ЊC; H NMR (300 MHz, Acetone-d6, d), 1.01 (s, 3H,
C18-methyl), 1.2–2.4 (m, steroid envelope), 2.7–2.9 (m,
2H, C6-H), 3.98 (s, 1H, 17b hydroxyl-H), 6.53 (d, 1H,
Jˆ2.6 Hz, C4-H), 6.58 (dd, 1H, Jˆ2.6, 8.3 Hz, C2-H),
6.74 (d, 1H, Jˆ16 Hz, C21 vinyl-H), 6.84 (d, 1H,
Jˆ16 Hz, C20 vinyl-H), 7.06 (d, 1H, Jˆ8.3 Hz, C1-H),
7.54–7.56 (m, 2H, C25, C27-H), 7.75–7.79 (m, 2H, C23,
C26-H), 7.93 (s, C3-hydroxy-H); 13C NMR (75.4 MHz,
Acetone-d6, d), 14.7 (C18), 24.1 (C15), 27.3 (C11), 28.3
(C7), (C6), 33.5 (C12), 37.5 (C16), 40.7 (C8), 44.6 (C9), 48.4
(C13), 50.1 (C14), 84.2 (C17), 113.5 (C2), 115.9 (C4), 123.6 (q,
Jˆ5.6 Hz, C25), 124.1 (q, Jˆ3.7 Hz, C23), 125.4 (q,
Jˆ271 Hz, C28:CF3), 126.0 (C26), 127.0 (C1), 130.2 (C21),
130.7 (C27), 131.2 (q, Jˆ32 Hz, C24), 132.0 (C10), 138.4
(C5), 139.7 (C20), 139.9 (C22), 155.9 (C3); Anal. Calcd for
C27H29O2F3: C, 73.30; H, 6.56. Found: C, 73.42; H, 6.68.
1.55 mmol gϪ1
.
Electrophilic destannylation on the resin
The Stille reaction was used to couple the anchored E- and
Z-stannylvinyl estradiol to aryl halides. The resin was added
to the reaction flask, swelled in the CH2Cl2, subsequently
treated with 10 mL of anhydrous toluene. To the resultant
slurry was added a 3–4 fold excess of the functionalized
aryl halide, 1–2 crystals of 3,5-di-t-butyl-4-hydroxytoluene
(BHT), and Pd(PPh3)4.13–15 The reaction was allowed to
proceed at 90–100ЊC for 24 h. After cooling, the resin
was washed as previously described, dried in vacuo and
weighed.
17a-20E-21-(4-Trifluoromethylphenyl)-19-norpregna-
1,3,5(10),20-tetraene-3,17b-diol (17a-E-(4-trifluoro
methylphenyl)-vinyl estradiol) (6a). Yieldˆ49%; Rfˆ
0.15 (hexane–ethyl acetate, 4:1); mp 215–217ЊC; 1H

1628
C. Y. Lee, R. N. Hanson / Tetrahedron 56 (2000) 1623–1629
NMR (Acetone-d6, 300 MHz, d), 1.02 (s, 3H, C18 methyl-
H), 1.2–2.4 (m, steroid envelope), 2.7–2.9 (m, 2H, C6-H),
3.90 (s, 1H, 17b hydroxyl-H), 6.53 (d, 1H, Jˆ2.6 Hz, C4-H),
6.58 (dd, 1H, Jˆ2.6, 8.4 Hz, C2-H), 6.73 (d, 1H, Jˆ16 Hz,
C21 vinyl-H), 6.85 (d, 1H, Jˆ16 Hz, C20 vinyl-H), 7.07 (d,
1H, Jˆ8.3 Hz, C1-H), 7.64 (d, 2H, Jˆ8.7 Hz, C23, C27-H),
7.70 (d, 2H, Jˆ8.6 Hz, C24, C26-H), 8.0 (s, C3-hydroxy-H);
13C NMR (75.4 MHz, Acetone-d6, d) 14.7 (C18), 24.1 (C15),
27.3 (C11), 28.3 (C7), (C6), 33.5 (C12), 37.6 (C16), 40.7 (C8),
44.6 (C9), 48.5 (C13), 50.2 (C14), 84.2 (C17), 113.5 (C2),
115.9 (C4), 125.4 (q, Jˆ270.6 Hz, C28:CF3), 126.0 (C21),
126.2 (q, Jˆ3.5 Hz, C26), 126.2 (q, Jˆ3.5 Hz, C24), 127.0
(C1), 127.6 (C23, C27), 128.9 (q, Jˆ32 Hz, C25), 132.0 (C10),
138.4 (C5), 140.6 (C20), 142.7 (C22), 155.9 (C3); Anal.
Calcd for C27H29O2F3: C, 73.30; H, 6.56. Found: C,
73.36; H, 6.79.
138.4 (C5), 138.7 (C22, C24), 155.9 (C3); Anal. Calcd for
C27H32O2: C, 83.51; H, 8.25. Found: C, 83.23; H, 8.42.
17a-20Z-21-(3-Methylphenyl)-19-norpregna-1,3,5(10),
20-tetraene-3,17b-diol (17a-Z-(3-methylphenyl)-vinyl
estradiol) (8b). Yieldˆ54% (0.01 g); Rfˆ0.25 (hexane–
1
acetone, 4:1); H NMR (300 MHz, Acetone-d6, d) 0.95 (s,
3H, C18 methyl-H), 1.2–2.4 (m, steroid envelope), 2.31 (s,
3H, C28 methyl-H), 2.7–2.9 (m, 2H, C6-H), 3.27 (s, 1H, 17b
hydroxyl-H), 5.96 (d, 1H, Jˆ13.1 Hz, C21 vinyl-H), 6.44 (d,
1H, Jˆ13.1 Hz, C20 vinyl-H), 6.53 (d, 1H, Jˆ2.6 Hz, C4-H),
6.60 (dd, 1H, Jˆ2.6, 8.3 Hz, C2-H), 7.03 (d, 1H, Jˆ7.3 Hz,
C25-H), 7.11 (d, 1H, Jˆ8.3 Hz, C1-H), 7.17 (t, 1H,
Jˆ7.6 Hz, C26-H), 7.38–7.43 (m, 2H, C23, C27-H), 7.95 (s,
1H, C3 hydroxy-H); 13C NMR (75.4 MHz, Acetone-d6, d)
14.58 (C18), 21.42 (C28:methyl), 23.85 (C15), 27.40 (C11),
28.30 (C7), (C6), 32.97 (C12), 38.4 (C16), 40.9 (C8), 44.7
(C9), 48.8 (C13), 50.1 (C14), 84.3 (C17), 113.6 (C2), 116.0
(C4), 127.1 (C1), 127.8 (C27), 128.1 (C25), 128.3 (C26), 129.7
(C21), 131.4 (C23), 132.0 (C10), 137.1 (C20), 137.6 (C24),
138.45 (C5) 138.5 (C22), 155.9 (C3); Anal. Calcd for
C29H36O3: C, 80.55; H, 8.33. Found: C, 80.00; H, 8.41.
17a-20Z-21-(4-Trifluoromethylphenyl)-19-norpregna-
1,3,5(10),20-tetraene-3,17b-diol (17a-Z-(4-trifluoro
methylphenyl)-vinyl estradiol) (6b). Yieldˆ17%; Rfˆ
0.29 (hexane–ethyl acetate, 4:1); 1H NMR (300 MHz,
Acetone-d6, d) 0.97 (s, 3H, C18 methyl-H), 1.2–2.4 (m,
steroid envelope), 2.7–2.9 (m, 2H, C6-H), 3.89 (s, 1H,
17b hydroxyl-H), 6.12 (d, 1H, Jˆ12.9 Hz, C21 vinyl-H),
17a-20E-21-(4-Methoxyphenyl)-19-norpregna-1,3,5,(10),
20-tetraene-3,17b-diol (17a-E-(4-methoxyphenyl)-vinyl
estradiol) (9a). Yieldˆ36%; Rfˆ0.23 (CHCl3–CH3OH,
6.48–6.62 (m, 3H, C2, C , C20 vinyl-H), 7.11 (d, 1H,
4
Jˆ8.1 Hz, C1-H), 7.59 (d, 2H, Jˆ8.4 Hz, C23, C27-H), 7.80
1
(d, 2H, Jˆ8.4 Hz, C24, C26-H), 7.95 (s, C3 hydroxy-H).
99:1); H NMR (300 MHz, Acetone-d6, d) 0.99 (s, 3H,
C18 methyl-H), 3.68 (s, 1H, 17b hydroxy-H), 3.78 (s, 3H,
C28:methoxy-H), 6.46 (d, 1H, Jˆ16.1 Hz, C21-H), 6.51–
6.59 (m, 3H, C2, C4, C20-H), 6.88 (d, 2H, Jˆ8.8 Hz, C24,
C26-H); 7.07 (d, 1H, Jˆ8.3 Hz, C1-H); 7.39 (d, 2H,
Jˆ8.8 Hz, C23, C27-H), 7.95 (s, 1H, C3 hydroxy-H); 13C
NMR (75.4 MHz, Acetone-d6, d) 14.7 (C18), 24.1 (C15),
27.3 (C11), 28.3 (C7), (C6), 33.4 (C12), 37.3 (C16), 40.7
(C8), 44.7 (C9), 48.2 (C13), 50.0 (C14), 55.5 (C28:methoxy),
84.1 (C17), 113.5 (C2), 114.7 (C24, C26), 115.9 (C4), 127.0
(C1), 127.0 (C21), 128.3 (C23, C27), 131.4 (C22), 132.1 (C10),
134.9 (C20), 138.4 (C5), 155.9 (C3), 159.9 (C25).
17a-20E-21-(2-Methylphenyl)-19-norpregna-1,3,5(10),
20-tetraene-3,17b-diol (17a-E-(2-methylphenyl)-vinyl
estradiol) (7a). Yieldˆ38%; Rfˆ0.18 (hexane–acetone,
1
4:1); mp 199–200ЊC; H NMR (Acetone-d6, 300 MHz, d),
1.01 (s, 3H, C18 methyl-H), 1.2–2.4 (steroid envelope), 2.34
(s, 3H, C28 methyl-H), 2.7–2.9 (m, 2H, C6-H), 3.84 (s, 1H,
17b hydroxyl-H), 6.44 (d, 1H, Jˆ16 Hz, C21 vinyl-H),
6.52–6.63 (m, 2H, C2, C4-H), 6.83 (d, 1H, Jˆ16 Hz, C20
vinyl-H), 7.07 (d, 1H, Jˆ8.3 Hz, C1-H), 7.10–7.15 (m,
3H, C24, C25, C26-H), 7.48 (d, 1H, Jˆ6.8 Hz, C27-H), 7.97
(s, C3 hydroxy-H); 13C NMR (75.4 MHz, Acetone-d6, d)
14.7 (C18), 19.9 (C28: methyl), 24.1 (C15), 27.3 (C11), 28.3
(C7), (C6), 33.5 (C12), 37.5 (C16), 40.7 (C8), 44.7 (C9), 48.2
(C13), 50.1 (C14), 84.2 (C17), 113.5 (C2), 115.9 (C4),
125.4(C26), 126.5 (C25), 126.9 (C24), 127.0 (C1), 127.7
(C21), 130.8 (C27), 132.0 (C10), 135.9 (C20), 137.9
(C22), 138.4 (C5), 138.8 (C23), 155.9 (C3); Anal. Calcd
for C27H32O2: C, 83.51; H, 8.25. Found: C, 83.79; H,
8.65.
Acknowledgements
We are grateful to Dr David A. Forsyth at Northeastern
University for assisting the NMR data assignment and Drs
Lee W. Herman and Christopher Blackburn at Millenium
Pharmaceuticals for valuable comments on the preparation
of the carboxylated resins. This work has been supported in
part by grants from the Department of the Army, DAMD17-
99-1-9333 and the Public Health Service, 1-R01-CA81049-
01.
17a-20E-21-(3-Methylphenyl)-19-norpregna-1,3,5(10),
20-tetraene-3,17b-diol (17a-E-(3-methylphenyl)-vinyl
estradiol) (8a). Yieldˆ75%; Rfˆ0.17 (hexane–acetone,
1
4:1); mp 204–205ЊC; H NMR (300 MHz, Acetone-d6, d),
1.00 (s, 3H, C18 methyl-H), 1.2–2.4 (m, steroid envelope),
2.31 (s, 3H, C28 methyl-H), 2.7–2.9 (m, 2H, C6-H), 3.74 (s,
1H, 17b hydroxyl-H), 6.52–6.63 (m, 4H, C4, C2, C21 vinyl,
C20 vinyl-H), 7.03 (d, 1H, Jˆ7.3 Hz, C25-H), 7.07 (d, 1H,
Jˆ8.7 Hz, C1-H), 7.16–7.31 (m, 3H, Jˆ7.4 Hz, C23, C26,
C27-H), 7.93 (s, 1H, C3 hydroxy-H); 13C NMR (75.4 MHz,
Acetone-d6, d) 14.8 (C18), 21.4 (C28: methyl), 24.1 (C15),
27.3 (C11), 28.4 (C7), (C6), 33.5 (C12), 37.4 (C16), 40.8 (C8),
44.7 (C9), 48.3 (C13), 50.2 (C14), 84.2 (C17), 113.6 (C2),
116.0 (C4), 124.4 (C27), 127.0 (C1), 127.7 (C25), 127.8
(C26), 128.5 (C21), 129.2 (C23), 132.2 (C10), 137.0 (C20),
References
1. Napolitano, E.; Fiaschi, R.; Hanson, R. N. J. Med. Chem. 1991,
34, 2754–2759.
2. Hanson, R. N.; Napolitano, E.; Fiaschi, R.; Herman, L. W.
Steroids 1996, 61, 384–389.
3. Brzozowski, A. M.; Pike, A. C. W.; Dauter, Z.; Hubbard, R. E.;
¨
¨
Bonn, T.; Engstrom, O.; Ohman, L.; Greene, G. L.; Gustafsson,
˚
J.A.; Carlquist, M. Nature 1997, 389, 753–758.
4. Tremblay, M. R.; Poirier, D. Tetrahedron Lett. 1999, 40, 1277–
1280.

C. Y. Lee, R. N. Hanson / Tetrahedron 56 (2000) 1623–1629
1629
´
5. Wess, G.; Bock, K.; Kleine, H.; Kurz, M.; Guba, W.;
Hemmerle, H.; Lopez-Calle, E.; Baringhaus, K. H.; Glombik, H.;
Enhsen, A.; Kramer, W. Angew. Chem., Int. Ed. Engl. 1996, 35,
2222–2224.
14. Farrall, M. J.; Frechet, J. M. J. J. Org. Chem. 1976, 41, 3877–3882.
15. The loading capacities for the functionalized resins were
expressed in mmol gϪ1 and for the following steps were expressed
in % or mmol gϪ1
.
6. Cheng, Y.; Suenaga, T.; Stille, W. C. J. Am. Chem. Soc. 1996,
118, 1813–1814.
7. Blossey, E. C.; Cannon, R. G.; Ford, W. T.; Periyasamy, M.;
Mohanraj, S. J. Org. Chem. 1990, 55, 4664–4668.
8. Hodge, P.; Kemp, J.; Khoshdel, E.; Perry, G. M. React. Polym.
1985, 3, 299–313.
16. Carr, F. H.; Price, E. A. Biochem. J. 1926, 20, 497–501.
17. Blatz, P. E.; Estrada, A. Anal. Chem. 1972, 44, 570–573.
18. Jork, H.; Funk, W.; Fischer, W.; Wimmer, H. Thin-Layer
Chromatography, Reagent and Detection Method, VCH: New
York, 1990; Vol. 1a.
19. Gordon, A. J.; Ford, R. A. The Chemist’s Companion—A
Handbook of Practical Data, Techniques and References, Wiley:
New York, 1972 p 378.
9. Hodge, P.; Waterhouse, J. J. Chem. Soc., Perkin Trans. 1 1983,
2319–2323.
10. Chamoin, S.; Houldsworth, S.; Snieckus, V. Tetrahedron Lett.
1998, 39, 4175–4178.
20. Nozaki, K.; Oshima, K.; Utimoto, K. Tetrahedron 1989, 45,
923–933.
11. Plunkett, M. J.; Ellman, J. A. J. Am. Chem. Soc. 1995, 117,
3306–3307.
21. Hanson, R. N.; Napolitano, E.; Fiaschi, R. J. Med. Chem.
1998, 41, 4686–4692.
12. Bunin, B. A.; Ellman, J. A. J. Am. Chem. Soc. 1992, 114,
10997–10998.
22. Stille, J. K. Pure Appl.Chem. 1985, 57, 1771–1780.
23. Unpublished results.
13. Bowden, K.; Heilbron, I. M.; Jones, E. R. H.; Weedon, B. C. L.
J. Chem. Soc. 1946, 39–45.
24. Morales, G. A.; Corbett, J. W.; Degrado, W. F. J. Org. Chem.
1998, 63, 1172–1177.