Synthesis and antimitotic activity of novel 2-methoxyestradiol analogs-Part II

Article abstract of DOI:10.1016/j.steroids.2007.11.003

The syntheses and antimitotic activity of several novel 18a-homo-analogs of 2-methoxyestradiol are described. Structural modifications of the parent 2-methoxy-18a-homoestradiol include introduction of unsaturation in the D-ring and methylation of the 17-OH. Of seven analogs synthesized, one has demonstrated superior biological activities compared to 2-methoxyestradiol. The relationship between biological activity and the conformational preference of the 13-ethyl group as determined by computational analysis is discussed.

Full text of DOI:10.1016/j.steroids.2007.11.003

steroids 7 3 ( 2 0 0 8 ) 158–170
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/steroids
Synthesis and antimitotic activity of novel
2-methoxyestradiol analogs—Part II
Pemmaraju N. Raoa,∗, James W. Cessaca, James W. Boyda,
Arthur D. Hansonb, Jamshed Shahb
a
Department of Organic Chemistry, Southwest Foundation for Biomedical Research, P.O. Box 760549, San Antonio,
TX 78245-0549, USA
b
EntreMed Inc., 9640 Medical Center Drive, Rockville, MD 20850, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The syntheses and antimitotic activity of several novel 18a-homo-analogs of 2-
methoxyestradiol are described. Structural modifications of the parent 2-methoxy-18a-
homoestradiol include introduction of unsaturation in the D-ring and methylation of the
17-OH. Of seven analogs synthesized, one has demonstrated superior biological activities
compared to 2-methoxyestradiol. The relationship between biological activity and the con-
formational preference of the 13-ethyl group as determined by computational analysis is
discussed.
Received 20 June 2007
Received in revised form
5 October 2007
Accepted 12 November 2007
Published on line 22 November 2007
© 2007 Elsevier Inc. All rights reserved.
Keywords:
Steroids
Estrogens
18a-Homosteroids
Estrogens
Antiangiogenesis
Antimitotic activity
1.
Introduction
oxo counterparts. It is not known whether these structural
modifications enhance the binding affinity towards tubulin or
The nonestrogenic endogenous metabolite of estradiol,
namely 2-methoxyestradiol (2ME2) has been shown to pos-
sess antiangiogenic and antimitotic properties [1,2]. Because
of the exciting biological activities of 2ME2, several analogs of
this compound have been synthesized and tested for various
biological activities. We have earlier reported [3,4] the syn-
thesis of several ring-D modified 2ME2 analogs and observed
that by introducing an additional unsaturation in ring-D such
as a 14-dehydro or a 15-dehydro derivative, the in vitro anti-
proliferative activity could be considerably enhanced relative
to the parent 2ME2 compound. We also observed that the
17-hydroxy analogs are more active in vitro than their 17-
decrease the rate of metabolic inactivation. In order to explore
further the structure-activity relationship in 2ME series of
compounds, we have now synthesized additional derivatives
of 2ME2 and evaluated their biological activity.
In a recent phase I trial of 2ME2 in patients with metastatic
breast cancer, extensive metabolic oxidation of 2ME2 to 2-
methoxyestrone (2ME1) was observed [5]. Since it has also
been observed that the 17␤-hydroxy derivatives of 2ME2
are more active in vitro than their corresponding 17-oxo
derivatives, it would be desirable to minimize any metabolic
oxidation of the 17␤-hydroxy group of potential 2ME2 analogs
to the corresponding 17-oxo compounds. We reasoned that
∗
Corresponding author. Tel.: +1 210 258 9414; fax: +1 210 670 3321.
E-mail address: pnrao@sfbr.org (P.N. Rao).
0039-128X/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2007.11.003

steroids 7 3 ( 2 0 0 8 ) 158–170
159
this could possibly be achieved by two basic structural
modifications. First by increasing the steric bulk of the 13␤-
substituent next to the 17␤-hydroxy function and second by
protecting the 17␤-hydroxy group as a methyl ether. 2ME2
has a methyl group at C-13 position adjacent to the 17␤-
hydroxyl group. If we could replace the C-13 methyl group
with a sterically bulkier ethyl group, the 17␤-hydroxyl func-
tion might be protected from metabolic deactivation. Along
similar lines, conversion of the 17␤-hydroxy group to the
corresponding methyl ether should also decrease metabolic
oxidation. Incorporating these structural modifications along
with the introduction of additional unsaturation in ring-D, we
synthesized seven new 2ME2 derivatives and evaluated their
cytotoxic activity in a variety of cell types. In this communi-
cation, we now present the details of our investigations.
18a), 3.73 (t, J = 7.1 Hz, 17-H), 6.65 (t, J = 3.1 Hz, 4-H), 6.63 (dd,
J1 = 8.5 Hz, J2 = 2.6 Hz, 2-H), 7.15 (d, J = 8.7 Hz, 1-H). Analysis cal-
culated for C19H26O2·2/3MeOH: C, 76.77; H, 9.38. Found: C,
76.77; H, 9.35.
2.1.2. 18a-Homoestra-1,3,5(10)-trien-3,17ˇ-diyl-diacetate
(3)
Under nitrogen, the 3,17-diol (2, 9.41 g, 33 mmol) was dissolved
in 100 ml of pyridine. Acetic anhydride (11.5 ml, 0.12 mol) was
added and the mixture stirred over the weekend. Analysis
by TLC confirmed complete reaction (5% acetone/CH2Cl2).
The reaction was quenched with methanol and the sol-
vent removed in vacuo. This residue was crystallized from
methanol to give the pure 3,17-diacetate (3, 9.7 g, 80%) as a
white powder. mp = 124–126 ◦C; FTIR (ATR) ꢀmax: 2930, 1758,
1727, 1604, 1576 cm−1
. NMR (300 MHz, CDCl3), ıH (ppm):
0.96 (t, J = 7.6 Hz, 18a), 2.05 (s, 17-OAc), 2.78 (s, 3-OAc),
4.75 (t, J = 8.5 Hz, 17-H), 6.79 (t, J = 2.6 Hz, 4-H), 6.83 (dd,
J1 = 5.7 Hz, J2 = 2.7 Hz, 2-H), 7.27 (d, J = 5.7 Hz, 1-H). Analysis
calculated for C23H30O4: C, 74.56; H, 8.16. Found: C, 74.63;
H, 8.11.
2.
Experimental
2.1.
Chemistry
Melting points were determined on
a Thomas-Hoover
apparatus and are uncorrected. Nuclear magnetic reso-
nance spectra were recorded on a General Electric GE-300
(300 MHz) spectrometer as deuterochloroform (CHCl3) solu-
tions using tetramethylsilane (TMS) as an internal standard
(ı = 0) unless noted otherwise. Infrared spectra were recorded
on Thermo-Nicolet model 370 FT-IR instrument equipped
with an attenuated reflectance (ATR) accessory. Combus-
tion analyses were performed by Midwest Microlabs Ltd.
(Indianapolis, IN). ‘Flash column’ chromatography was per-
formed on 32–64 ␮M silica gel obtained from EM Science,
Gibbstown, NJ. ‘Dry column’ chromatography was performed
on 70–230 mesh silica gel, also obtained from EM Science.
Thin-layer chromatography (TLC) analyses were carried out
on silica gel GF (Analtech) glass plates (2.5 cm × 10 cm with
250 ␮M layer and prescored). Most chemicals and solvents
were analytical grade and used without further purification.
Commercial reagents were purchased from Aldrich Chemi-
cal Company (Milwaukee, WI). Levonorgestrel was provided
as a gift by Wyeth-Ayerst Research, CN 8000, Princeton, NJ
08534-8000.
2.1.3. 2-Acetyl-3-hydroxy-18a-homoestra-1,3,5(10)-
trien-17ˇ-yl-acetate (4)
Under nitrogen, the 3,17-diacetate (3, 8.91 g, 24 mmol) was
dissolved in 650 ml of CH2Cl2. Zirconium tetrachloride (26 g,
0.11 mol) was added and the mixture stirred 48 h at room
temperature. Analysis by TLC confirmed complete reaction
(5% acetone/CH2Cl2). The mixture was chilled to 0 ◦C and
quenched with 300 ml of water. The product extracted with
CH2Cl2 (3×), the organic phase washed with water (3×), and
brine. The organic fractions were dried over sodium sulfate,
filtered, and concentrated in vacuo. This material was crys-
tallized from methanol to give the pure 2-acetyl compound
(4, 5.18 g, 63%) as a white powder. mp = 240–242 ◦C; FTIR (ATR)
ꢀmax: 2943, 1714, 1636, 1614, and 1573 cm−1. NMR (300 MHz,
CDCl3), ıH (ppm): 0.96 (t, J = 7.4 Hz, 18a), 2.05 (s, 17-OAc), 2.62 (s,
3-Ac), 4.76 (t, J = 8.4 Hz, 17-H), 6.69 (s, 4-H), 7.59 (s, 1-H), 12.05
(s, 3-OH). Analysis calculated for C23H30O4: C, 74.56; H, 8.16.
Found: C, 74.36; H, 8.15.
2.1.4. 2-Acetyl-3-benzyloxy-18a-homoestra-1,3,5(10)-
trien-17ˇ-yl-acetate (5)
2.1.1. 3,17ˇ-Dihydroxy-18a-homoestra-1,3,5(10)-triene
(2)
Under nitrogen, the 2-acetyl compound (4, 5.18 g, 14 mmol)
was dissolved in 165 ml of DMF. Potassium carbonate (16.2 g,
0.12 mol) and benzyl chloride (7.17 ml, 87 mmol) were added
and the mixture stirred at 60 ◦C for 48 h. Analysis by TLC (5%
acetone/CH2Cl2) confirmed complete reaction. The reaction
was quenched with 500 ml of water whereby a white flocky
precipitate formed. The precipitate was filtered, washed with
water until neutral, and dried in vacuo. This material was crys-
tallized from methanol/CH2Cl2 to give the pure 3-benzyl ether
(5, 5.59 g, 87%) as a white powder. mp = 213–215 ◦C; FTIR (ATR)
ꢀmax: 2939, 2869, 1734, 1657, and 1603 cm−1. NMR (300 MHz,
CDCl3), ıH (ppm): 0.93 (t, J = 7.4 Hz, 18a), 2.05 (s, 17-OAc), 2.57 (s,
2-Ac), 4.74 (t, J = 8.3 Hz, 17-H), 5.13 (s, benzyl CH2), 6.73 (s, 4-H),
7.70 (s, 1-H), 7.34-7.45 (m, benzyl aromatic). Analysis calcu-
lated for C30H36O4·1/6CH3OH: C, 77.76; H, 7.93. Found: C, 77.89;
H, 7.72.
3-Hydroxy-18a-homoestra-1,3,5(10)-trien-17-one (1) was pre-
pared from levonorgestrel as described earlier [6]. Under
nitrogen, the 3-phenol (1, 9.4 g, 33 mmol) was dissolved in
50 ml of 1:1 EtOH/H2O. Sodium borohydride (2.5 g, 66 mmol)
was dissolved in 450 ml of 1:1 EtOH/H2O. The sodium borohy-
dride solution was added to the steroid solution drop wise over
2 h and stirred overnight. Analysis by TLC confirmed complete
reaction (5% acetone/CH2Cl2). Excess sodium borohydride was
decomposed with acetic acid, the solvent evaporated, and
the residue extracted with ethyl acetate. The ethyl acetate
was washed with water, brine, dried over sodium sulfate and
evaporated. The residue was crystallized from methanol/H2O
to give the pure 3,17-diol (2, 9.31 g, 98%) as a white pow-
der. mp = 183 ◦C; FTIR (ATR) ꢀmax: 3377, 3270, 2930, 1614, and
1582 cm−1. NMR (300 MHz, CDCl3), ıH (ppm): 0.92 (t, J = 7.4 Hz,

160
steroids 7 3 ( 2 0 0 8 ) 158–170
CDCl3), ıH (ppm): 0.99 (t, J = 7.4 Hz, 18a), 3.87 (s, 2-OCH3), 3.81
(t, J = 8.4 Hz, 17-H), 5.10 (s, benzyl CH2), 6.63 (s, 4-H), 6.48
(s, 1-H), 7.30–7.48 (m, benzyl aromatic). Analysis calculated
for C27H34O3·2/3H2O C, 77.48; H, 8.51. Found: C, 77.92; H,
8.26.
2.1.5. 3-Benzyloxy-18a-homoestra-1,3,5(10)-trien-
2,17ˇ-diyl-diacetate (6)
Under nitrogen, the 3-benzyl ether (5, 5.59 g, 12 mmol) was
dissolved in 160 ml of CH2Cl2. Sodium phosphate (4.48 g,
32 mmol) and 3-chloroperoxybenzoic acid (4.17 g, 24 mmol)
were added to the steroid solution and the suspension stirred
for 24 h at room temperature. Water was added and the
mixture extracted with CH2Cl2 (3×). The organic phase was
washed with water (1×), 10% sodium sulfite (1×), and half
saturated sodium bicarbonate (1×). The organic fractions
were dried over sodium sulfate, filtered, and concentrated
in vacuo. The residue was purified by flash chromatogra-
phy to give the pure 2,17-diacetate (6, 4.57 g, 79%) as a
white foam. mp = 98–100 ◦C; FTIR (ATR) ꢀmax: 2933, 1759,
2.1.8. 2-Methoxy-3,17ˇ-dihydroxy-18a-homoestra-
1,3,5(10)-triene (9)
A mixture of the 2-methoxy compound (8, 200 mg, 0.5 mmol)
and 5% palladium on carbon (200 mg) in ethanol (50 ml) was
hydrogenated in a Parr shaker apparatus at 40 psi hydrogen
pressure overnight. Analysis by TLC (3% acetone/CH2Cl2) con-
firmed complete reaction. The mixture was filtered through
Celite while rinsing with dichloromethane. The solvent was
removed in vacuo. This material was crystallized from
dichloromethane/hexanes to give the pure 3,17-diol (9, 126 mg,
1729, and 1612 cm−1
. NMR (300 MHz, CDCl3), ıH (ppm):
0.92 (t, J = 7.4 Hz, 18a-H), 2.05 (s, 17-OAc), 2.25 (s, 2-OAc),
4.74 (t, J = 8.3 Hz, 17-H), 5.05 (s, benzyl CH2), 6.72 (s, 4-H),
6.95 (s, 1-H), 7.26–7.38 (m, benzyl aromatic). Analysis cal-
culated for C30H36O5: C, 75.60; H, 7.61. Found: C, 75.85;
H, 7.69.
81%) as a white powder. mp = 114–115 ◦C; FTIR (ATR) ꢀmax
:
3842, 3359, 2921, 2850, 1592, 1505 cm−1. NMR (300 MHz, CDCl3),
ıH (ppm): 1.00 (t, J = 7.5 Hz, 18a), 3.86 (s, 2-OCH3), 3.77-3.821
(m, 17-H), 6.65 (s, 4-H), 6.79 (s, 1-H). Analysis calculated for
C20H28O3·1/6hexanes: C, 76.25; H, 9.24. Found: C, 76.20; H, 9.17
2.1.6. 2,17ˇ-Dihydroxy-3-benzyloxy-18a-homoestra-
1,3,5(10)-triene (7)
2.1.9. 2,17ˇ-Dimethoxy-3-benzyloxy-18a-homoestra-
1,3,5(10)-triene (10)
Under nitrogen, the 2,17-diacetate (6, 4.50 g, 9.5 mmol) was dis-
solved in 250 ml of methanol. To this solution 50 ml of 1M
NaOH was added and the mixture stirred at 60 ◦C for 2 h.
Analysis by TLC (5% acetone/CH2Cl2) indicated an incomplete
reaction. An additional 25 ml of 1 M NaOH was added and
the mixture stirred an additional hour. Analysis by TLC at
that time indicated a complete reaction. The reaction mix-
ture was cooled to room temperature and glacial acetic acid
was slowly added with stirring until the mixture was neu-
tral. The mixture was diluted with 300 ml of water whereby
a precipitate formed. Methanol was removed in vacuo. The
residue was extracted with ethyl acetate (3×), the organic
phase washed with water (3×), and brine. The organic frac-
tions were dried over sodium sulfate, filtered and concentrated
in vacuo. The solid residue was crystallized from ethyl acetate
to give the pure 2,17-diol (7, 2.61 g, 71%) as a white powder.
Under nitrogen, the 2-methoxy compound (8, 400 mg,
0.95 mmol) was dissolved in 20 ml of THF. This solution was
added drop wise to 0.08 g (∼2 mmol) of 60% NaH in 10 ml of
THF. The mixture was refluxed for 1 h and cooled to room
temperature. Methyl iodide (0.27 ml, 4.3 mmol) was added and
the mixture stirred for 2 h at room temperature. At this point
TLC (5% acetone/CH2Cl2) showed no reaction. An additional
methyl iodide (6 ml, 9.6 mmol) was added and the mixture
stirred overnight at room temperature. Analysis by TLC con-
firmed complete reaction. The reaction was quenched with
water and extracted with ether (3×), the ether fraction was
washed with water (3×), and brine. The organic fractions
were dried over sodium sulfate, filtered and concentrated in
vacuo. This material was crystallized from acetone/hexanes
to give the pure 2,17-dimethoxy-3-benzyl ether (10, 266 mg,
64%). mp = 113–114 ◦C; FTIR (ATR) ꢀmax: 2922, 2860, 1604, and
1514 cm−1. NMR (300 MHz, CDCl3), ıH (ppm): 0.98 (t, J = 7.5 Hz,
18a), 3.37 (s, 17-OCH3), 3.87 (s, 2-OCH3), 3.35 (t, J = 7.75 Hz, 17-
H), 5.11 (s, benzyl CH2), 6.62 (s, 4-H), 6.85 (s, 1-H), 7.27-7.46 (m,
benzyl aromatic). Analysis calculated for C28H36O3: C, 79.96;
H, 8.63. Found: C, 80.03; H, 8.68.
mp = 178 ◦C; FTIR (ATR) ꢀmax: 3531, 3261, 2922, and 1604 cm−1
.
NMR (300 MHz, CDCl3), ıH (ppm): 0.98 (t, J = 7.55 Hz, 18a),
3.80 (t, J = 8.35 Hz, 17-H), 5.04 (s, benzyl CH2), 6.66 (s, 4-H),
6.92 (s, 1-H), 7.35–7.45 (m, benzyl aromatic). Analysis cal-
culated for C26H32O3: C, 79.56; H, 8.22. Found: C, 79.38; H,
8.24.
2.1.7. 2-Methoxy-3-benzyloxy-18a-homoestra-1,3,5(10)-
trien-17ˇ-ol (8)
2.1.10. 2,17ˇ-Dimethoxy-3-hydroxy-18a-homoestra-
1,3,5(10)-triene (11)
Under nitrogen, the 2,17-diol (7, 2.057 g, 5.2 mmol) was dis-
solved in 30 ml of THF. To the solution lithium hydroxide
(0.55 g, 1.3 mmol) and dimethyl sulfate (0.55 ml, 7.7 mmol)
were added and the mixture refluxed at 60 ◦C overnight. Anal-
ysis by TLC (5% acetone/CH2Cl2) confirmed complete reaction
and the mixture cooled to room temperature. The solvent
was removed in vacuo and the residue purified by flash chro-
matography (5% acetone/CH2Cl2). Crystallization attempts
from different solvent systems failed and the compound (8)
was obtained as a viscous oil which was dried under vacuum
to obtain a white foam which melted 240–242 ◦C; FTIR (ATR)
ꢀmax: 3522, 2927, 2861, 1601, and 1506 cm−1. NMR (300 MHz,
A mixture of the 2,17-dimethoxy-3-benzyl ether (10, 150 mg,
0.35 mmol) and 5% palladium on carbon (150 mg) in
ethanol (60 ml) was hydrogenated in a Parr shaker appa-
ratus at 40 psi hydrogen pressure overnight. Analysis by
TLC (3% acetone/CH2Cl2) confirmed complete reaction. The
mixture was filtered through Celite while rinsing with
dichloromethane. The solvent was removed in vacuo and the
remaining reside purified by flash chromatography using the
TLC solvent system. This material was triturated from cold
ether, rinsed with pentane and dried in vacuo to yield the pure
2,17-dimethoxy compound (11, 110 mg, 94%). mp = 161–162 ◦C;
FTIR (ATR) ꢀmax: 3392, 2933, 2868, 1591, 1506 cm−1. NMR

steroids 7 3 ( 2 0 0 8 ) 158–170
161
Na2SO4, and concentrated in vacuo to give the 2-methoxy
derivative (14, 53.7 g, >100%) as a yellow oil which still
contained some solvent. FTIR (ATR) ꢀmax: 2930, 1608, and
1507 cm−1. NMR (300 MHz, CDCl3), ıH (ppm): 0.98 (t, J = 7.5 Hz,
18a), 3.37 (s, 17-OCH2OCH3), 3.51 (s, 3-OCH2OCH3), 3.86 (s,
2-OCH3), 4.63 (dd, J1 = 9.0 Hz, J2 = 6.3 Hz, 17-OCH2OCH3), 5.19
(s, 3-OCH2OCH3), 6.84 (s, 4-H), 6.86 (s, 1-H). The material was
used without further purification.
(300 MHz, CDCl3), ıH (ppm): 0.97 (t, J = 7.4 Hz, 18a), 3.37 (s, 17-
OCH3), 3.85 (s, 2-OCH3), 3.49 (t, J = 7.55 Hz, 17-H), 6.65 (s, 4-H),
6.80 (s, 1-H). Analysis calculated for C21H30O3·1/6H2O: C, 75.64;
H, 9.17. Found: C, 75.54; H, 8.91.
2.1.11. 3,17ˇ-Bis(methoxymethoxy)-18a-homoestra-
1,3,5(10)-triene (12)
Under nitrogen, the diol (2, 47.2 g, 0.165 mol) was dissolved in
1.2 l of dry THF. Chloromethyl methyl ether (62 ml, 5 equiv.)
and diisopropylethylamine (170 ml, 6 equiv.) were added and
the reaction heated to 65 ◦C overnight. The reaction was
cooled to room temperature, quenched with 20% NH4Cl, and
extracted with EtOAc (3×). The organic phase washed with
water (3×), brine, dried over Na2SO4, and concentrated in
vacuo to give the pure 3,17␤-dimethoxymethyl ether (12,
48.2 g, 78 %) as a pale yellow oil. The material was used with-
out further purification. FTIR (ATR) ꢀmax: 2929, 2876, 1608, and
1496 cm−1. NMR (300 MHz, CDCl3), ıH (ppm): 0.98 (t, J = 7.5 Hz,
18a), 3.38 (s, 17-OCH2OCH3), 3.48 (s, 3-OCH2OCH3), 4.64 (dd,
J1 = 9.0 Hz, J2 = 6.45 Hz, 17-OCH2OCH3), 5.15 (s, 3-OCH2OCH3),
6.78 (d, J = 2.4 Hz, 4-H), 6.84 (dd, J1 = 8..4 Hz, J2 = 2.4 Hz, 2-H), 7.21
(d, J = 8.7 Hz, 1-H). Analysis calculated for C23H34O4: C, 73.76; H,
9.15. Found: C, 73.34; H, 8.86.
2.1.14. 2-Methoxy-3,17ˇ-dihydroxy-18a-homoestra-
1,3,5(10)-triene (9)
Under nitrogen, 2-methoxy compound (14, 53.7 g, 0.132 mol)
was hydrolyzed with a mixture of THF (300 ml) and 6M HCl
(500 ml) at room temperature overnight. Analysis by TLC (2%
Acetone/CH2Cl2) indicated a complete reaction. The reaction
was diluted with water, concentrated in vacuo and extracted
with EtOAc (3×). The organic phase was washed with water,
brine, dried over Na2SO4, and concentrated in vacuo. The
residue was purified by flash column (6% Acetone/CH2Cl2).
This material was crystallized from dichloromethane/hexanes
to give the pure 3,17-diol (9, 30.3 g, 72%) as yellow solid.
mp = 114–115 ◦C; FTIR (ATR) ꢀmax: 3484, 3362, 2921, 2872, and
1706 cm−1. NMR (300 MHz, CDCl3), ıH (ppm): 1.00 (t, J = 7.5 Hz,
18a), 3.84 (s, 2-OCH3), 6.64 (s, 4-H), 6.79 (s, 1-H). Analysis cal-
2.1.12. 2-Hydroxy-3,17ˇ-bis(methoxymethoxy)-18a-
homoestra-1,3,5(10)-triene (13)
culated for C20
76.20; H, 9.17
H28O3·1/6hexanes: C, 76.25; H, 9.24. Found: C,
Under nitrogen, the dimethoxymethyl ether (12, 56.4 g,
0.151 mol) was dissolved in 1 l of dry THF and chilled to
−78 ◦C. To this was added sec-BuLi (215 ml, 1.4 M, 2 equiv.) at
such a rate the temperature did not exceed −65 ◦C and the
reaction mixture stirred at −78 ◦C for 3 h. Trimethyl borate
(68 ml, 4 equiv.) was then added maintaining the temperature
below −65 ◦C and the reaction stirred for 15 min at −78 ◦C
and then warmed to 0 ◦C. The reaction was quenched with
1 l of 20% NH4Cl, then allowed to come to room temper-
ature and stirred for 1 h. Sodium perborate (93 g, 4 equiv.)
was then added at such a rate that the temperature did not
exceed 35 ◦C and the reaction carried out at room temperature
overnight. The reaction mixture was concentrated in vacuo
and extracted with EtOAc (3×). The organic phase was washed
with water, brine, dried over Na2SO4, and concentrated in
vacuo. The residue was purified by flash choromatography
(2% acetone/CH2Cl2) to give (13, 60.4 g, 84 %) as a yellow
oil. FTIR (ATR) ꢀmax: 3407, 2930, 1589, and 1504 cm−1. NMR
(300 MHz, CDCl3), ıH (ppm): 0.99 (t, J = 7.5 Hz, 18a), 3.36 (s,
17-OCH2OCH3), 3.52 (s, 3-OCH2OCH3), 4.63 (dd, J1 = 9.0 Hz,
J2 = 6.6 Hz, 17-OCH2OCH3), 5.16 (s, 3-OCH2OCH3), 6.80 (s, 4-
H), 6.90 (s, 1-H). The material was used without further
purification.
2.1.15. 2-Methoxy-3-acetoxy-18a-homoestra-1,3,5(10)-
trien-17ˇ-ol (15)
Under nitrogen, the diol (9, 20 g, 62.3 mmol) was selec-
tively acetylated in isopropanol with acetic anhydride (18 ml,
3 equiv.) in the presence of NaOH (95 ml, 2 M, 3 equiv.) at room
temperature for 90 min. Analysis by TLC (5% acetone/CH2Cl2)
indicated
a complete reaction. The reaction was slowly
quenched with methanol, diluted with water, concentrated
in vacuo and extracted with EtOAc (3×). The organic phase
was washed with water, brine, dried over Na2SO4, and con-
centrated in vacuo to give the pure 3-acetate (15, 20.56 g, 91%)
as a yellow foam which resisted crystallization. mp = 84 ◦C;
FTIR (ATR) ꢀmax: 3523, 2931, 2870, 1751, and 1644 cm−1. NMR
(300 MHz, CDCl3), ıH (ppm): 0.93 (t, J = 7.5 Hz, 18a), 2.29 (s, 3-
OAc), 3.80 (s, 2-OCH3), 6.73 (s, 4-H), 6.89 (s, 1-H). Analysis
calculated for C22H30O4·2/3H2O: C, 71.32; H, 8.52. Found: C,
71.15; H, 8.16.
2.1.16. 2-Methoxy-3-acetoxy-18a-homoestra-
1,3,5(10)-trien-17-one (16)
Under nitrogen, the 17-hydroxy compound (15, 20 g,
55.8 mmol) was dissolved in 400 ml of acetone and chilled
to 0 ◦C. Jones’s Reagent was slowly added with stirring until
the yellow-orange color persisted (∼40 ml). the reaction
was stirred an additional 5 min then slowly quenched with
isopropanol. The solution was concentrated in vacuo, diluted
with water, and extracted with EtOAc (3×). The organic phase
was washed with water, brine, dried over Na2SO4, and con-
centrated in vacuo to give the 17-ketone (16, 19 g, 95.5%). The
material was used without further purification. mp = 159 ◦C;
FTIR (ATR) ꢀmax: 2932, 1762, 1731, 1614, and 1508 cm−1. NMR
(300 MHz, CDCl3), ıH (ppm): 0.81 (t, J = 7.5 Hz, 18a), 2.31 (s,
3-OAc), 3.81 (s, 2-OCH3), 6.76 (s, 4-H), 6.89 (s, 1-H). Analysis
2.1.13. 2-Methoxy-3,17ˇ-bis(methoxymethoxy)-18a-
homoestra-1,3,5(10)-triene (14)
Under nitrogen, the 2-hydroxy compound (13, 44.7 g,
0.115 mol) was dissolved in 500 ml of dry THF. Lithium
hydroxide (6 g, 1.25 equiv.) and dimethyl sulfate (12.2 ml,
1.12 equiv.) were added and the reaction was refluxed for
4 h. Analysis by TLC (CH2Cl2) indicated a complete reaction.
The reaction was cooled to room temperature, diluted with
water, concentrated in vacuo, and extracted with EtOAc (3×).
The organic phase was washed with water, brine, dried over

162
steroids 7 3 ( 2 0 0 8 ) 158–170
calculated for C22H28O4·1/6H2O: C, 73.51; H, 7.94. Found: C,
was purified via flash chromatography (1.5% acetone/CH2Cl2)
to give a mixture of the double bond isomers (19 and 20, 6.92 g,
49%) as a dark brown solid. The material was used without
further purification.
73.73; H, 7.84.
2.1.17. 17,17-Ethylenedioxy-2-methoxy-18a-homoestra-
1,3,5(10)-trien-3-yl-acetate (17)
Under nitrogen, the 17-ketone (16, 19 g, 53 mmol) was dis-
solved in 300 ml of CH2Cl2. To this solution were added
triethylorthoformate (28.8 ml, 3.25 equiv.), ethylene glycol
(19 ml, 6.4 equiv.) and tosic acid (0.77 g, 0.076 equiv.) and the
reaction mixture stirred at room temperature overnight. Anal-
ysis by TLC (2% acetone/CH2Cl2) indicated a partial reaction.
Additional triethylorthoformate (35 ml) and ethylene glycol
(10 ml) were added and refluxed for an additional 3 h. The
reaction was cooled to room temperature, quenched with sat-
urated NaHCO3 and extracted with CH2Cl2 (3×). The organic
phase was washed with water, brine, dried over Na2SO4, and
concentrated in vacuo to give a dark yellow oil. Analysis
showed an incomplete reaction. The material was re-reacted
over the weekend using 60 ml of triethylorthoformate, 40 ml of
ethylene glycol and 0.8 g of tosic acid. Subsequent purification
and re-acetylation afforded the ketal (17, 20 g, 94%) as a yellow
foam which resisted crystallization. mp = 97–98 ◦C; FTIR (ATR)
ꢀmax: 2936, 2875, 1763, 1614, and 1508 cm−1. NMR (300 MHz,
CDCl3), ıH (ppm): 0.99 (t, J = 7.35 Hz, 18a), 2.31 (s, 3-OAc), 3.82 (s,
2-OCH3), 3.90–3.95 (m, 17-ketal CH2’s), 6.74 (s, 4-H), 6.90 (s, 1-
H). Analysis calculated for C24H32O5·3/4H2O: C, 69.62; H, 8.16.
Found: C, 69.51; H, 8.00.
2.1.20. 2-Methoxy-3-hydroxy-18a-homoestra-1,3,5(10)14-
tetraen-17-one (21) and 2-methoxy-3-hydroxy-18a-
homoestra-1,3,5(10)15-tetraen-17-one (22)
Under nitrogen, the mixture of the double bond isomers
(19 and 20, 8 g, 22.4 mmol) was dissolved in 400 ml of 6:1
acetone/water. Tosic acid (0.41 g 0.1 equiv.) was added and
the reaction mixture stirred at room temperature for 2.5 h.
Analysis by TLC (2% acetone/CH2Cl2) indicated the reac-
tion was complete. The reaction mixture was evaporated
to 1/3 volume and extracted with CH2Cl2 (3×). The organic
phase was washed with water, brine, dried over Na2SO4, and
concentrated in vacuo. The residue was purified by flash chro-
matography (2.5% acetone/CH2Cl2) to give a mixture of the
ꢂ14 and ꢂ15 isomers (21 and 22, 2.7 g) as well as the pure
ꢂ15 derivative (22, 0.46 g, 45%). The ꢂ15 compound (22) was
crystallized from ether/hexanes. mp = 163 ◦C; FTIR (ATR) ꢀmax
:
3236, 2959, 2929, 1686, 1609, and 1523 cm−1. NMR (300 MHz,
CDCl3), ıH (ppm): 0.79 (t, J = 7.5 Hz, 18a), 3.86 (s, 2-OCH3), 6.06
(dd, J1 = 6.0 Hz, J2 = 3.3 Hz, 16-H), 5.72 (s, 3-OH), 6.66 (s, 4-H), 6.77
(s, 1-H), 7.81 (dd, J1 = 6.0 Hz, J2 = 2.7 Hz, 15-H). Analysis calcu-
lated for C20H24O3: C, 76.89; H, 7.74. Found: C, 76.36; H, 7.71.
2.1.21. 2-Methoxy-18a-homoestra-1,3,5(10),14,16-
2.1.18. 17,17-Ethylenedioxy-16ꢁ-bromo-2-methoxy-18a-
homoestra-1,3,5(10)-trien-3-yl-acetate (18)
pentaen-3-17ˇ-diyl-diacetate (23)
Under nitrogen, a mixture of ꢂ14 and ꢂ15 compounds (21 and
22, 2.7 g, 8.64 mmol) was dissolved in 51 ml of acetic anhy-
dride. Isopropenyl acetate (50 ml, 53 equiv.) and Tosic acid
(1.0 g, 0.6 equiv.) were added and the reaction mixture was
refluxed overnight. Analysis by TLC (2% acetone/CH2Cl2) indi-
cated a complete reaction. The reaction was cooled to room
temperature, quenched with water and stirred for 1 h. The
mixture was then concentrated in vacuo and extracted with
EtOAc (3×). The organic phase was washed with water, brine,
dried over Na2SO4, and concentrated in vacuo. The residue
was purified by flash chromatography (2.5% acetone/CH2Cl2)
and subsequently crystallized from ether/hexanes to give the
pure pentaene derivative (23, 1.6 g, 46.6%) as a white solid.
Under nitrogen, the 17-ketal (17, 20 g, 50 mmol) was dissolved
in 400 ml of THF and chilled to −5 ◦C. Phenyltrimethylammo-
nium tribromide (22.5 g, 1.2 equiv.) was added in 5 g portions
over 0.5 h. Once the addition was complete, the reaction mix-
ture was stirred at −5 ◦C overnight. The reaction was quenched
with cold, saturated NaHCO3 and extracted with EtOAc (3×).
The organic phase was washed with saturated NaHCO3 (2×),
10% sodium thiosulfate, cold water (2×), and brine, dried
over Na2SO4, and concentrated in vacuo. The residue was
purified by flash column and subsequently crystallized from
CH2Cl2/hexanes to give the pure 16-bromo compound (12,
22.94 g, 95%) mp = 194 ◦C; FTIR (ATR) ꢀmax: 2936, 2878, 1763,
1614, and 1508 cm−1. NMR (300 MHz, CDCl3), ıH (ppm): 1.00 (t,
J = 7.050 Hz, 18a), 2.31 (s, 3-OAc), 3.81 (s, 2-OCH3), 3.96 – 4.21
(m, 17-ketal CH2’s), 4.69 (dd, J1 = 10.350 Hz, J2 = 3.450 Hz, 16-H),
6.74 (s, 4-H), 6.90 (s, 1-H). Analysis calculated for C24H31O5Br:
C, 60.13; H, 6.52; Br, 16.67. Found: C, 59.96; H, 6.35; Br,
16.85.
mp = 133 ◦C; FTIR (ATR) ꢀmax: 2933, 2865, 1763, and 1615 cm−1
.
NMR (300 MHz, CDCl3), ıH (ppm): 0.53 (t, J = 7.350 Hz, 18a), 2.23
(s, 17-OAc), 2.31 (s, 3-OAc), 3.81 (s, 2-OCH3), 5.93 (t, J = 1.95 Hz,
15-H), 6.22 (d, J = 2.7 Hz, 16-H), 6.79 (s, 4-H), 6.89 (s, 1-H). Anal-
ysis calculated for C24H28O5: C, 72.71; H, 7.12. Found: C, 72.64;
H, 7.14.
2.1.19. 17,17-Ethylenedioxy-2-methoxy-18a-homoestra-
1,3,5(10),14-tetraen-3-ol (19) and 17,17-ethylenedioxy-
2.1.22. 2-Methoxy-3,17ˇ-dihydroxy-18a-homoestra-
1,3,5(10),14-tetraene (24)
2-methoxy-18a-homoestra-1,3,5(10),15-tetraen-3-ol (20)
Under nitrogen, the 16-bromo compound (18, 19 g, 19.8 mmol)
was dissolved in 100 ml of dry xylenes. This solution was
added to freshly prepared potassium tert-butoxide (43 g,
19.3 equiv.) in 500 ml of dry xylenes and refluxed overnight.
The reaction was cooled to room temperature, quenched with
water and extracted with EtOAc (3×). The organic phase was
washed with water, brine, dried over Na2SO4, and concen-
trated in vacuo to give a dark brown residue. The material
Under nitrogen, a solution of the enolacetate (23, 1.5 g,
3.8 mmol) in ethanol (20 ml) and THF (20 ml) was chilled to
0 ◦C in an ice bath. An ice cold solution of sodium borohy-
dride (0.9 g, 6.3 equiv.) in ethanol/water (1:1, 50 ml) was added
to the steroid solution. The reaction mixture was stirred for
1 h, allowed to warm to room temperature, and then stirred
overnight. Analysis by TLC (2% acetone/CH2Cl2) indicated a
complete reaction. The reaction was quenched with acetic
acid, concentrated to a small volume, and extracted with

steroids 7 3 ( 2 0 0 8 ) 158–170
163
by flash chromatography and subsequent crystallization from
ether/hexanes afforded the pure 3-hydroxy (27, 0.2 g, 75%) as
a white solid. mp = 202 ◦C; FTIR (ATR) ꢀmax: 3383, 2935, 2834,
1618 and 1507 cm−1. NMR (300 MHz, CDCl3), ıH (ppm): 0.85 (t,
J = 7.2 Hz, 18a), 3.42 (s, 17-OMe), 3.85 (t, J = 8.25 Hz, 17-H), 3.86 (s,
2-OCH3), 5.34 (m, 15-H), 6.66 (s, 4-H), 6.79 (s, 1-H). Analysis cal-
culated for C21H28O3·1/6Et2O: C, 76.36; H, 8.77. Found: C, 76.14;
H, 8.55.
EtOAc (3×). The organic phase was washed with water, brine,
dried over Na2SO4, and concentrated in vacuo. The residue
was purified by flash column to give the pure diol (24, 0.93 g,
69%) as an amorphous powder. mp = 65 ◦C; FTIR (ATR) ꢀmax
:
3409, 2920, 2853, 1592, and 1506 cm−−1. NMR (300 MHz, CDCl3),
ıH (ppm): 0.89 (t, J = 7.5 Hz, 18a), 3.87 (s, 2-OCH3), 4.21 (t,
J = 8.4 Hz, 17-H), 5.21 (m, 15-H), 6.67 (s, 4-H), 6.80 (s, 1-H). Anal-
ysis calculated for C20H26O3·1/2H2O: C, 74.27; H, 8.41. Found:
C, 74.33; H, 8.21.
2.1.26. 2-Methoxy-3,17ˇ-dihydroxy-18a-homoestra-
1,3,5(10),15-tetraene (28)
2.1.23. 2-Methoxy-3-acetoxy-18a-homoestra-1,3,5(10),
14-tetraen-17ˇ-ol (25)
To a solution of the 17-ketone (22, 0.4 g, 1.28 mmol), in
ether (380 ml), chilled to −5 ◦C, was added solid LiAlH4 (1.0 g,
19.75 equiv.) in small portions, and the reaction mixture stirred
for 1 h. Analysis by TLC (5% acetone/CH2Cl2) indicated a
complete reaction. The reaction was quenched with the drop-
wise addition of 5% H2SO4 and extracted with EtOAc (3×).
The organic phase was washed with water, brine, dried over
Na2SO4, and concentrated in vacuo. Purification by flash
chromatography (5% acetone/CH2Cl2) and subsequent crys-
tallization from methanol afforded the pure diol (28, 0.15 g,
37%) as a white solid. mp = 164 ◦C; FTIR (ATR) ꢀmax: 3519, 3240,
2919, 1589, and 1506 cm−1. NMR (300 MHz, CDCl3), ıH (ppm):
0.91 (t, J = 7.5 Hz, 18a), 3.87 (s, 2-OCH3), 5.71 (m, 16-H), 5.98 (dd,
J1 = 7.8 Hz, J2 = 1.8 Hz, 15-H), 6.67 (s, 4-H), 6.79 (s, 1-H). Analy-
sis calculated for C20H26O3·1/2H2O: C, 74.27; H, 8.41. Found: C,
74.38; H, 8.15.
Under nitrogen, the diol (24, 0.7 g, 2.2 mmol) was dissolved
in 20 ml of isopropanol. Acetic anhydride (5 ml, 52 mmol) and
2 M NaOH (20 ml) were added and the reaction stirred at room
temperature for 2 h. Analysis by TLC (2% acetone/CH2Cl2) indi-
cated a complete reaction. The reaction was quenched with
methanol, concentrated to a small volume and extracted with
EtOAc (3×). The organic phase was washed with water, brine,
dried over Na2SO4, and concentrated in vacuo. Crystallization
from ether/hexanes afforded the pure 3-acetate (25, 0.61 g,
77%) as a white solid. mp = 139 ◦C; FTIR (ATR) ꢀmax: 3510, 2922,
2854, 1755, 1615, and 1508 cm−1. NMR (300 MHz, CDCl3), ıH
(ppm): 0.88 (t, J = 7.35 Hz, 18a), 2.31 (s, 3-OAc), 3.80 (s, 2-OCH3),
4.19 (t, J = 8.55 Hz, 17-H), 5.34 (m, 15-H), 6.76 (s, 4-H), 6.90 (s, 1-
H). Analysis calculated for C22H28O4·1/2H2O: C, 72.30; H, 8.00.
Found: C, 72.47; H, 7.70.
2.1.24. 2,17ˇ-Dimethoxy-18a-homoestra-1,3,5(10),
14-tetraen-3-yl-acetate (26)
2.1.27. 2-Methoxy-3-acetoxy-18a-homoestra-1,3,5(10),
15-tetraen-17ˇ-ol (29)
Under nitrogen, fluoroboric acid (48%, 2.5 ml, 13 equiv.) was
added to a solution of the 3-acetate (25, 0.5 g, 1.4 mmol)
in CH2Cl2 (20 ml) chilled to 0 ◦C, and stirred for 10 min.
Trimethylsilyl diazomethane (1 ml, 1.4 equiv.) was slowly
added and the reaction mixture stirred for 20 min. An addi-
tional amount of trimethylsilyl diazomethane was added
every 20 min. After the final addition, the reaction was stirred
for an additional 0.5 h. Analysis by TLC (2% acetone/CH2Cl2)
indicated a complete reaction. The reaction was quenched
with water and extracted with CH2Cl2 (3×). The organic
phase was washed with water, brine, dried over Na2SO4,
and concentrated in vacuo. Purification by flash column (5%
acetone/CH2Cl2) afforded the pure 17-methoxy derivative (26,
0.37 g, 71%) as a yellow foam. The material was used with-
out further purification. FTIR (ATR) ꢀmax: 2934, 2859, 1763,
1613, and 1508 cm−1. NMR (300 MHz, CDCl3), ıH (ppm): 0.85
(t, J = 7.5 Hz, 18a), 2.30 (s, 2-OAc), 3.41 (s, 17-OMe), 3.74 (t,
J = 8.25 Hz, 17-H), 3.80 (s, 2-OCH3), 5.34 (m, 15-H), 6.75 (s, 4-H),
6.89 (s, 1-H).
Under nitrogen, the 3-hydroxy compound (28, 0.25 g,
0.79 mmol) was dissolved in 20 ml of isopropanol. Acetic
anhydride (5 ml, 52 mmol) and 2 M NaOH (20 ml) were added
and the reaction mixture was stirred at room temperature
for 3 h. Analysis by TLC (2% acetone/CH2Cl2) indicated a com-
plete reaction. The reaction was quenched with methanol,
concentrated to a small volume, and extracted with EtOAc
(3×). The organic phase was washed with water, brine, dried
over Na2SO4, and concentrated in vacuo. The residue was
purified by flash column (2% acetone/CH2Cl2) to give the pure
3-acetate (29, 0.17 g, 61%) as a white foam. The material was
used without further purification. FTIR (ATR) ꢀmax: 3427, 2923,
2854, 1751, 1614, and 1508 cm-1. NMR (300 MHz, CDCl3), ıH
(ppm): 0.90 (t, J = 7.5 Hz, 18a), 2.30 (s, 3-OAc), 3.81 (s, 2-OCH3),
5.71 (m, 16-H), 5.97 (m, 15-H), 6.75 (s, 4-H), 6.89 (s, 1-H).
2.1.28. 2,17ˇ-Dimethoxy-18a-homoestra-1,3,5(10),
15-tetraen-3-yl-acetate (30)
Under nitrogen, 48% fluoroboric acid (1.2 ml, 13 equiv.) was
added to a solution of the 3-acetate (29, 0.175 g, 0.49 mmol)
in CH2Cl2 (10 ml) chilled to 0 ◦C, and stirred for 10 min.
Trimethylsilyl diazomethane (8 ml, 32.5 equiv.) was added
drop wise over 2.5 h. Once the addition was complete, the
reaction was stirred an additional 90 min. Analysis by TLC
(2% acetone/CH2Cl2) indicated a complete reaction. The reac-
tion was quenched with water and extracted with CH2Cl2 (3×).
The organic phase was washed with water, brine, dried over
Na2SO4, and concentrated in vacuo. Purification by flash col-
umn (5% acetone/CH2Cl2) afforded the pure 17-methoxy (30,
0.13 g, 71%) as a white foam which melted at 142 ◦C; FTIR (ATR)
2.1.25. 2,17ˇ-Dimethoxy-3-hydroxy-18a-homoestra-
1,3,5(10),14-tetraene (27)
Under nitrogen, the 3-acetate (26, 0.3 g, 0.81 mmol) was dis-
solved in 60 ml of 3:1 MeOH/H2O. Potassium carbonate (0.44 g,
4 equiv.) was added and the reaction mixture stirred at room
temperature overnight. Analysis by TLC (5% acetone/CH2Cl2)
indicated a complete reaction. The reaction was quenched
with water, concentrated, and extracted with EtOAc (3×).
The organic phase was washed with water, brine, dried over
Na2SO4, and concentrated in vacuo. Purification of the residue

164
steroids 7 3 ( 2 0 0 8 ) 158–170
ꢀmax: 2921, 1764, 1614 and 1511 cm−1. NMR (300 MHz, CDCl3),
ıH (ppm): 0.88 (t, J = 7.5 Hz, 18a), 2.30 (s, 3-OAc), 3.47 (s, 17-
OCH3), 3.81 (s, 2-OCH3), 5.80 (m, 16-H), 5.95 (dd, J1 = 6.0 Hz,
J2 = 2.1 Hz, 15-H), 6.75 (s, 4-H), 6.89 (s, 1-H). Analysis calculated
for C23H30O4: C, 74.56; H, 8.16. Found: C, 74.20; H, 8.30.
vacuo. Purification of the residue by flash chromatography
gave the pure tetraene (33, 0.120 g, 19%) as a viscous oil.
FTIR (ATR) ꢀmax: 3549, 2923, 2853, 1591 and 1504 cm−1. NMR
(300 MHz, CDCl3), ıH (ppm): 0.79 (t, J = 7.5 Hz, 18a), 3.85 (s, 2-
OCH3), 5.82 (m, 16-H), 6.04 (m, 17-H), 6.64 (s, 4-H), 6.78 (s,
1-H). Analysis calculated for C20H26O2·1/2H2O: C, 79.70; H, 8.81.
Found: C, 79.78; H, 8.87.
2.1.29. 2,17ˇ-Dimethoxy-3-hydroxy-18a-homoestra-
1,3,5(10),15-tetraene (31)
Under nitrogen, the 3-acetate derivative (30, 0.1 g, 0.27 mmol)
was dissolved in 20 ml of 3:1 MeOH/H2O. Potassium carbon-
ate (1.0 g, 27 equiv.) was added and the reaction mixture
stirred at room temperature for 2 h. Analysis by TLC (2%
acetone/CH2Cl2) indicated a complete reaction. The reac-
tion was quenched with water, and extracted with EtOAc
(3×). The organic phase was washed with water, brine, dried
over Na2SO4, and concentrated in vacuo. Purification of the
residue by flash chromatography and subsequent crystalliza-
tion from CH2Cl2/hexanes gave the pure 3-hydroxy compound
2.2.
Biological assays
Biological assays were carried out by EntreMed Inc., Rockville,
MD.
2.3.
Cell cultures
Human umbilical vein endothelial cells (HUVEC) were
obtained from Clonetics (San Diego, CA), MDA-MB-231, and
U87-MG cells were obtained from ATCC (American Type
Culture Collection, Manassas, VA). HUVEC cultures were main-
tained for up to five passages in EGM (Endothelial Growth
Medium) containing bovine brain extract (Clonetics) and 1×
antibiotic–animycotic (BioWhittaker, Walkersville, MD). MDA-
MB-231 and U87-MG cells were maintained in DMEM/F12 (1:1)
containing 10% (v/v) fetal bovine serum (Hyclone Laboratories,
Logan, UT) and 1× antibiotic–antimycotic.
(31, 0.067 g, 76%) as a white solid. mp = 180 ◦C; FTIR (ATR) ꢀmax
:
3378, 2930, 1617, and 1504 cm−1. NMR (300 MHz, CDCl3), ıH
(ppm): 0.87 (t, J = 7.5 Hz, 18a), 3.46 (s, 17-OCH3), 3.87 (s, 2-OCH3),
5.80 (m, 16-H), 5.96 (m, 15-H), 6.65 (s, 4-H), 6.78 (s, 1-H). Anal-
ysis calculated for C21H28O3·1/10H2O: C, 76.38; H, 8.61. Found:
C, 76.19; H, 8.52.
2.1.30. 2-Methoxy-3-hydroxy-18a-homoestra-
1,3,5(10)-trien-17-one p-toluenesulfonylhydrazone (32)
2.4.
In vitro inhibition of proliferation
To a solution of the ketone (16, 1.0 g, 3.1 mmol) in MeOH (20 ml)
was added p-toluene sulfonhydrazide (0.72 g, 1.25 equiv.) and
the reaction mixture refluxed overnight. Analysis by TLC (2%
acetone/CH2Cl2) indicated an incomplete reaction. Hydrochlo-
ric acid (4 drops) was added and the reaction mixture refluxed
for an additional 4 h. Analysis by TLC (2% acetone/CH2Cl2)
indicated an incomplete reaction. Additional HCl was added
(four drops) and the reflux continued an additional 2 h. Anal-
ysis by TLC (2% acetone/CH2Cl2) indicated approximately
90% completion. The reaction was cooled to room temper-
ature and stored in the refrigerator over the weekend. The
reaction mixture was diluted with water and extracted with
EtOAc (3×). The organic phase was washed with water, brine,
dried over Na2SO4, and concentrated in vacuo to give the p-
toluenesulfonylhydrazone (32, 1.65 g) as a green/brown solid.
The material was used without further purification. FTIR (ATR)
ꢀmax: 3519, 3225, 2928, 1597, and 1508 cm−1. NMR (300 MHz,
CDCl3), ıH (ppm): 0.50 (t, J = 7.2 Hz, 18a), 2.44 (d, J = 3.9 Hz, ben-
zyl CH3), 3.87 (s, 2-OCH3), 6.64 (s, 4-H), 6.78 (s, 1-H), 7.316 (d,
J = 7.5 Hz, benzyl CH2’s), 7.86 (d, J = 7.8 Hz, benzyl CH2’s).
Proliferation assays were performed by evaluating detection
of DNA synthesis by the use of the 5-bromo-2ꢀ-deoxyuridine
(BrdU) cell proliferation colorimetric ELISA kit from Roche
(Indianapolis, IN) according to the manufacturer’s instruc-
tions. The cells were seeded at 1000 cells/well (MDA-MB-231
and U87-MG cells, anti-tumor activity) or 3000 cells/well
(HUVEC, anti-angiogenic activity) in a 96 well plate, allowed
to attach overnight and then exposed to the compound to be
tested for 48 h. IC50 values and standard deviations (S.D.) were
calculated for each assay using Sigma Plot (Systat Software
Inc., San Jose, CA). Reported IC50 values are the mean and SD
of a minimum of three separate assays.
2.5.
Computational analysis
All calculations were carried out using the MM3 forcefield
[14] as implemented by PCMODEL (Version 9.1, Serena Soft-
ware, Bloominton, IN). After initial input of basic starting
geometries, the conformational space of each compound
was explored by means of the GMMX routine (Global Energy
Minimization) in PCMODEL using the MM3 forcefield, and
an energy window of 20 kcal/mol. All degrees of conforma-
tional freedom were considered including ring conformations
and rotatable bond orientations. The global minima thus
obtained was then used as a starting point for the con-
formational profile calculations presented in Figs. 7 and 8
whereby the C14–C13–C18–C18a torsional angle was fixed
in incremental 5◦ steps as the remainder of the molecule
was minimized. The results were plotted and interpolated
(cubic spline) using CurveExpert 1.3 (Hyams Development,
Hixson TN).
2.1.31. 2-Methoxy-3-hydroxy-18a-homoestra-
1,3,5(10),16-tetraene (33)
Under nitrogen, the 17-p-toluenesulfonylhydrazone (32, 1.0 g,
2.1 mmol) was dissolved in dry THF (130 ml) and chilled to
0 ◦C in an ice/salt bath. To this solution was added chilled
butyl lithium solution (3.3 ml, 4 equiv.). Once the addition was
complete, the reaction mixture was stirred at room tempera-
ture for 72 h. Analysis by TLC (2% acetone/CH2Cl2) indicated
a complete reaction. The reaction mixture was chilled to 0 ◦C,
quenched with 20% NH4Cl, concentrated to a small volume,
and extracted with EtOAc (3×). The organic phase was washed
with water, brine, dried over Na2SO4, and concentrated in

steroids 7 3 ( 2 0 0 8 ) 158–170
165
Fig. 1 – Synthesis of compounds (2–11).
methanol/water gave the 2,17␤-diol-3-benzyl ether compound
(7) in 71% yield. Methylation of this material in THF with
dimethyl sulfate in the presence of lithium hydroxide mono-
hydrate at 60 ◦C overnight gave the 2-methoxy derivative (8)
in 92% yield. Removal of the benzyl ether from compound (8)
via catalytic hydrogenation with 5% palladium on carbon in
ethanol at 40 psi gave compound (9) in 81% yield. Compound
(8) was also converted to the 2,17␤-dimethoxy derivative (10)
in 64% yield by refluxing with solid sodium hydride in THF,
followed by the addition of iodomethane. Finally, catalytic
hydrogenation of compound (10) with 10% palladium on car-
bon in ethanol at 40 psi gave compound (11) in 94% yield.
An alternate synthesis of compound (9) is illustrated in
Fig. 2 and is based on the procedure described by Paaren
and Duff [8]. Treatment of 2,17␤-diol derivative (2) with
chloromethyl methyl ether and N,N-diisopropylethylamine at
65 ◦C in THF overnight afforded the dimethoxymethyl ether
(12) in 78% yield. This material was then reacted with n-BuLi
in THF at −65 ◦C followed by the addition of trimethyl borate
and sodium perborate to give the 2-hydroxy compound (13) in
84% yield. Methylation of this material in THF with dimethyl
sulfate in the presence of lithium hydroxide monohydrate at
60 ◦C overnight gave the 2-methoxy derivative (13) in quanti-
tative yield. This material was then hydrolyzed with 6M HCl
in THF to give the 3,17␤-diol derivative (9) in 72% yield.
3.
Results and discussion
3.1.
Chemistry
The 3-hydroxy-17-ketone derivative (1) was prepared from
Norgestrel as previously described [5]. The syntheses of com-
pounds (2–11) are outlined in Fig. 1 and were based on
the procedures we previously developed for the 13-methyl
analogs [7]. Reduction of the ketone via sodium borohydride
in ethanol/THF at room temperature overnight gave the diol
compound (2) in 98% yield. Simple acetylation of this material
with acetic anhydride in pyridine at room temperature for 48 h
gave the 3,17␤-diacetate compound (3) in 80% yield. Treatment
of the diacetate derivative (3) with zirconium (IV) chloride in
dichloromethane at room temperature for 48 h gave the 2-
acetyl-3-hydroxy-17␤-acetate compound (4) in 63% yield. This
compound was then converted to the 3-benzyl ether deriva-
tive (5) in 87% yield by reaction with benzyl chloride in the
presence of potassium carbonate in DMF at 65 ◦C overnight.
Conversion of this material to the diacetate derivative (6)
was accomplished by reaction with 3-chloroperoxybenzoic
acid and sodium phosphate in dichloromethane at room
temperature for 48 h in 79% yield. Simple base hydroly-
sis of the diacetate material (6) with sodium hydroxide in
Fig. 2 – Alternative synthesis of 2-methoxy-18a-homoestra-1,3,5(10)-trien-17␤-ol (9).

166
steroids 7 3 ( 2 0 0 8 ) 158–170
Fig. 3 – Synthesis of ring-D-unsaturated 2-methoxy-18a-homoestra-1,3,5(10)-trien-3-ol derivatives.
Fig. 4 – Synthesis of 2-methoxy-3-hydroxy-18a-homoestra-1,3,5(10),16-tetraene.
For the synthesis of ring-D-unsaturated compounds
(15–31), the general procedure we developed earlier [3] was
followed and the synthetic scheme is presented in Fig. 3. The
3,17␤-diol derivative (9) was then selectively acetylated in iso-
propanol with acetic anhydride in the presence of 2 M KOH
to give the 3-acetate derivative (15) in 91% yield. Oxidation
of this material with Jones’s reagent in acetone at 0 ◦C gave
the 17-ketone compound (16) in 95% yield. Treating this mate-
rial with triethylorthoformate, ethylene glycol and tosic acid
in methylene chloride at room temperature overnight gave
the 17-ketal (17) in 94% yield. Reaction of the 17-ketal (17)
with phenyltrimethylammonium tribromide in THF at −5 ◦C
afforded the 16␰-bromo-17,17-ethylenedioxy compound (18)
in 95% yield. Dehydrobromination by refluxing with potas-
sium t-butoxide in xylene gave a mixture of the ꢂ14 and ꢂ15
compounds (19 and 20) in 49% total yield. The mixture of
unsaturated compounds (19 and 20) was reacted with tosic
acid in 6:1 acetone/water at room temperature to give a mix-
ture of ꢂ14 and ꢂ15 ketone derivatives (21 and 22) as well as the
pure ꢂ15 compound (22), with a total yield of 49%. The mixture
Fig. 5 – Compounds observed with 13-ethyl gauche g+ conformation.

steroids 7 3 ( 2 0 0 8 ) 158–170
167
in 37% yield. This material was then selectively acetylated
in isopropanol with acetic anhydride in the presence of 2 M
NaOH to give the 3-acetate derivative (29) in 61% yield. Methy-
lation of this material with trimethylsilyl diazomethane in
methylene chloride in the presence of aqueous fluoroboric
acid (48%) at 0 ◦C gave the 17␤-methoxy compound (30) in 71%
yield. Hydrolysis of this material with potassium carbonate in
3:1 methanol/water at room temperature gave the 3-hydroxy
compound (31) in 76% yield.
The synthesis of the tetraene derivative (33) is presented
in Fig. 4 and is based on the procedure of Berliner et al. [10].
Compound (16) was refluxed with p-toluenesulfonhydrazide
in methanol in the presence of HCl for 16 h to give the 3-
hydroxy-17-p-toluenesulfonylhydrazone derivative (32). This
material was then treated with n-BuLi in THF at 0 ◦C for 72 h
to give the ꢂ16 material (33) in 19% yield.
of isomers (21 and 22) was refluxed with isopropenyl acetate
and tosic acid in acetic anhydride to give the pure ꢂ14,16-3,17-
diacetate compound (23) in 46% yield. This material was then
reduced with sodium borohydride in 1:1 methanol/water at
0 ◦C to give the ꢂ14-3,17␤-diol (24) in 69% yield.
Selective acetylation of compound (24) in isopropanol with
acetic anhydride in the presence of 2 M NaOH to gave the
3-acetate material (25) in 77% yield. The 3-acetate material
(25) was then methylated with trimethylsilyl diazomethane
in the presence of aqueous fluoroboric acid according to the
procedure of Aoyama and Shioiri [9] to give the 17␤-methoxy
compound (26) in 71% yield. Hydrolysis of this material with
potassium carbonate in 3:1 methanol/water at room temper-
ature gave the 3-hydroxy compound (27) in 75% yield.
Reduction of the ꢂ15 compound (22) with lithium alu-
minum hydride in ether at 0 ◦C gave the ꢂ15-3,17␤-diol (28)
Table 2 – MM3 relative energies of trans- and gauche g+ conformations of 18a-homo steroids
Compound no.
Structure
MM3 minimized conformations
gauche g+
Dihedrala
Trans
MM3 ERel
Dihedrala
MM3 ERel
9
11
24
27
28
31
0.0
191.4
191.7
189.8
190.2
191.1
2.50
2.50
0.0
59.3
60.3
54.5
54.9
56.0
0.0
1.47
1.55
0.09
0.0
0.0
0.10
0.0
191.5
172.8
0.0
56.6
55.6
33
2.24
a
Dihedral angle for the C14–C13–C18–C18a bond.

168
steroids 7 3 ( 2 0 0 8 ) 158–170
tent biological activity (0.7- to 1.6-fold). Methylation of the
17-OH to give compound (11) results in a substantial increase
in biological activity (3- to 5-fold more active). Introduction
of the ꢂ14-bond in compound (24) significantly lowers bio-
logical activity with the exception of the U87-MG cell line
where it is about twice as active. Unlike compound (9), 17-
O-methylation of (24) to give (27) further decreases biological
activity in all cell lines tested. Introduction of the ꢂ15-bond
in compound (28) also leads to a considerable decrease in
activity, but in this case, O-methylation to give compound
(31) results in a 2- to 4-fold increase in activity compared to
compound (28). However, compound (31) is still about half as
active as either (9) or the parental 2ME2 (34). Finally, intro-
duction of the ꢂ16-bond with removal of the 17-OH group in
(33) provides 2- to 6-fold less biological activity compared to
2ME2.
Fig. 6 – Observed conformations of 18a-homosteroids.
3.2.
Biological activity
The compounds tested for biological activity are shown. 2-
Methoxyestradiol (34) and the ꢂ-14 analog (35) are included
here for comparison.
The IC50 values for inhibition of proliferation were gen-
erated in three cell types: human umbilical vein endothelial
cells (HUVEC), human breast carcinoma cells (MDA-MB-231)
and human gliomablastoma cells (U87-MG). The data are pre-
sented in Table 1. The IC50 value is the concentration at which
cell proliferation is inhibited by 50%.
4.
Discussion
The decrease in biological activity for compound (33) could
be explained by the absence of oxygen functionality at C-
17, but the results for compounds (24), (27), (28) and (31) are
somewhat puzzling in light of our earlier observation that
introduction of unsaturation in ring-D leads to increased bio-
logical activity [4]. As can be seen from Table 1, introduction
of a ꢂ14-bond in 2ME2 (34) to give (35) gives a 4- to 37-fold
increase in biological activity.
Replacement of the 13-methyl group of 2-methylestradiol
with an ethyl group in compound (9) gives about equipo-
One possible explanation for these results could lie in
the observation of crystal structural data reported for cer-
tain 18a-homosteroids containing unsaturation in ring-D.
van Geerestein et al. [11,12] and Eckle et al. [13] have
observed an unusual 18a-methyl orientation in the crys-
tal structure of Gestodene (36) and 11-methylenegestodene
(37). Normally, the observed orientation of the ethyl
group in 18a-homosteroids with saturated D-rings is trans
with respect to the C/D ring junction. However, crys-
tal structures of compounds (36) and (37) were observed
with the C14–C13–C18–C18a bond in the gauche g+ con-
figuration with the 18a-methyl group directly over the
D-ring.
A similar observation was made by Chekhlov
Fig. 7 – MM3 Energy profile for rotation of the 13-ethyl
group of compounds 9 and 24.
et al. [14] for rac-3-methoxy-18a-homoestra-1,3,5(10),8,14-
pentaen-17-one (38). These compounds along with an
illustration of the conformational difference are shown in
Figs. 5 and 6.
In order to investigate this possibility for compound (24)
and (28), conformational energy profiles were calculated for
these compounds along with the parental compound (9) using
the MM3 forcefield [15] implemented in PC-Model [16]. The
results are shown in Figs. 7 and 8.
While the expected trans conformation for compound
(9) is favored by ∼2.5 kcal/mol, these calculations indicate
the gauche conformation is favored for the ꢂ14 com-
pound (24) by about 1.5 kcal/mol and both conformations
are about equally populated for the ꢂ15-compound (28).
Table 2 summarizes the results of similar calculations car-
ried out for all the 18a-homo derivatives tested in this
study.
Experimental evidence in support of these calculations
can be seen in the chemical shift data for the C18a methyl
group for the 18a-homosteroids in this investigation. Table 3
Fig. 8 – MM3 Energy profile for rotation of the 13-ethyl
group of compounds 9 and 28.

steroids 7 3 ( 2 0 0 8 ) 158–170
169
compares these values along with the data for the corre-
sponding 13-methyl compounds (34), (39), and (35). It can
be seen from these data that there is an increase in upfield
shifts in the 18a-methyl for the sequences (9)–(28)–(24) and
(11)–(31)–(27) which is opposite of the trend observed for the
13-methyl compounds (34)–(39)–(35). This could be explained
by increased shielding of the methyl group by the double bond
caused by an increased population of the gauche g+ confor-
mation as predicted by the calculations in Figs. 7 and 8 and
Table 2.
The net effect of an increased population of the gauche g+
conformations for compounds (24), (27), (28) and (31) would
be a decrease in accessibility of the beta-face of the D-ring
for interaction with tubulin. This could explain the decrease
in biological activities observed for these compounds com-
pared to (34) and (35). This hypothesis is in agreement with
Table 1 – IC50 values for inhibition of proliferation
Compound no.
Structure
Inhibition of proliferation IC50 (␮M) S.D.
HUVEC
MDA-MB-231
U87-MG
34
0.68 0.15
0.51 0.09
0.17 0.07
0.03 0.04
1.28 0.64
2.76 0.16
2.08 1.86
0.69 0.14
0.75 0.17
0.17 0.05
0.19 0.13
1.11 0.16
1.35 0.62
5.14 3.93
1.48 0.62
0.92 0.03
0.51 0.23
0.04 0.04
0.72 0.01
2.13 0.04
4.98 1.16
9
11
35
24
27
28
31
33
0.97 0.33
1.26 0.64
2.39 0.01
2.32 0.30
3.60 1.58
2.69 0.01

170
steroids 7 3 ( 2 0 0 8 ) 158–170
Table 3 – 18-Methyl and 18a-methyl chemical shifts
Compound
18- or 18a-Methyl chemical shifts
Compound
Shift
0.79
1.00
0.97
Compound
Shift
Compound
Shift
R1 = Me, R2 = OHa
R1 = Et, R2 = OH
R1 = Et, R2 = OMe
34
9
11
39
28
31
0.89
0.91
0.87
35
24
27
1.02
0.89
0.85
a
Data taken from Refs. [3] and [7].
[7] Rao P, Cessac JW. A new, practical synthesis of
2-methoxyestradiols. Steroids 2002;67:1065–70.
[8] Paaren HE, Duff SR. Synthesis of 2-hydroxyestradiol
derivatives. US Patent No. 648,841,9B1; 2002.
the observation that for 16-substituted or 15,16-disubstituted
2ME2 analogs, the ␤-isomer is usually less active than the ␣-
isomer [3,17].
[9] Aoyama T, Shioiri T. Trimethylsilyldiazomethane: a
convenient reagent for the o-methylation of alcohols.
Tetrahedron Lett 1990;38:5507–8.
Acknowledgements
[10] Berliner D, Adams NW, Jennings-White CL. Method of
altering hypothalamic function by nasal administration of
estrene steroids. US Patent No. 5,793,571; 1998.
Financial support from Southwest Foundation for Biomedical
Research and EntreMed Inc. is gratefully acknowledged.
[11] van Geerestein VJ, Duisenberg AJM, Duits MHG, Kanters JA,
Kroon J. Structure of modifications (I) and (II) of 13-ethyl-17␤-
hydroxy-18,19-dinor-17␣-pregna-4,15-diene-20-yn-3-one
(Gestodene). Acta Cryst C 1987;43:2402–5.
r e f e r e n c e s
[12] van Geerestein VJ, Kanters JA, Kroon J. Structure of
13-ethyl-17␤-hydroxy-11-methylene-18,19-dinor-17␣-
pregna-4,15-diene-20-yn-3-one (11-methylenegestodene).
Acta Cryst C 1988;44:329–32.
[1] Seegers JC, Aveling M-L, van Aswgan CH, Cross M, Koch F,
Joubert WS. The Cytotoxic effects of estradiol-17␤,
catecholestradiols and methoxyestradiols on dividing MCF-7
and HeLa Cells. J Steroid Biochem 1989;32:797–809.
[2] Fotsis T, Zhang Y, Pepper MS, Aldercreutz H, Montesano R,
Nawroth PP, et al. The endogenous oestrogen metabolite
2-methoxyestradiol inhibits angiogenesis and suppresses
tumor growth. Nature 1994;368:237–9.
[13] Eckle E, Hoyer G-A, Hofmeister H, Laurent H, Schomburg D,
Wiechert R. Die Kristall- und Molekulstruktur von Gestoden.
¨
Liebigs Ann Chem 1988:199–202.
[14] Chekhlov AN, Ionov SP, Dodonov MV, Ananchenko SN.
Synthesis and X-ray study of rac-3-methoxy-18-methyl-
1,3,5(10),8,14-estrapentaen-17-one and
[3] Rao PN, Cessac JW, Tinley TL, Mooberry SL. Synthesis and
antimitotic activity of novel 2-methoxyestradiol analogs.
Steroids 2002;67:1079–89.
rac-3-methoxy-18-methyl-1,3, 5(10)-estratrien-17-one.
Bioorg Khim 1983;9:978–85.
[4] Tinley TL, Leal RM, Randall-Hlubek DA, Cessac JW, Wilkins
LR, Rao PN, et al. Novel 2-methoxyestradiol analogs with
antitumor activity. Cancer Res 2003;63:1538–49.
[5] James J, Murry DJ, Treston AM, Storniolo AM, Sledge GW,
Sidor C, et al. Phas I safety, pharmokinetic and
pharmacodynamic studies of 2-methoxyestradiol alone or in
combination with docetaxel in patients with locally
recurrent or metastatic breast cancer. Invest New Drugs
2006;25:41–8.
[15] Lii JH, Allinger NH. Directional hydrogen bonding in the
MM3 force field. II. J Comput Chem 1998;19:1001–16 [and
references therein].
[16] Gilbert KE. PC Model, version 9.1. Serena Software,
Bloomington, IN.
[17] Agoston GE, Shah JH, Hunsucker KA, Pribluda VS, LaVallee
TM, Green SJ, Herbstritt CJ, Zhan XH, Treston AM.
Antiangiogenic agents. US Patent No. 7,135,581;
2006.
[6] Rao PN, Cessac JW, Kim HK. Preparative chemical methods
for aromatization of 19-nor-ꢂ4-3-oxosteroids. Steroids
1994;59:621–7.