Synthesis and structure-activity relationships of 16-modified analogs of 2-methoxyestradiol

Article abstract of DOI:10.1016/j.bmc.2007.09.011

A series of 16-modified 2-methoxyestradiol analogs were synthesized and evaluated for antiproliferative activity toward HUVEC and MDA-MB-231 cells, and for susceptibility to conjugation. In addition, the estrogenicity of these analogs was accessed by meas

Full text of DOI:10.1016/j.bmc.2007.09.011

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 15 (2007) 7524–7537
Synthesis and structure–activity relationships
of 16-modified analogs of 2-methoxyestradiol
Gregory E. Agoston,* Jamshed H. Shah, Theresa M. LaVallee, Xiaoguo Zhan,
Victor S. Pribluda and Anthony M. Treston
EntreMed, Inc., Discovery Research Department, 9640 Medical Center Drive, Rockville, MD 20850, USA
Received 13 July 2007; revised 10 September 2007; accepted 10 September 2007
Available online 14 September 2007
Abstract—A series of 16-modified 2-methoxyestradiol analogs were synthesized and evaluated for antiproliferative activity toward
HUVEC and MDA-MB-231 cells, and for susceptibility to conjugation. In addition, the estrogenicity of these analogs was accessed
by measuring cell proliferation of the estrogen-dependent cell line MCF7 in response to compound treatment. It was observed that
antiproliferative activity dropped as the size of the 16 substituent increased. Selected analogs tested in glucuronidation assays had
similar rates of clearance to 2-methoxyestradiol, but had enhanced clearance in sulfonate conjugation assays.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
OH
17
OH
16
C
D
2-Methoxyestradiol (2ME2, 1, Fig. 1—carbon numbering
and ring labels added for clarity), an endogenous metab-
olite of estradiol (2), is produced in vivo by hydroxyl-
ation of estradiol at the 2-position and subsequent
methylation of the 2-hydroxyestradiol by catechol-O-
methyltransferase (COMT). 2ME2 has low affinity for
the estrogen receptors and its physiological function is
unknown. Although this endogenous metabolite was
long thought to have no physiological function, several
studies during the past decade demonstrated that it
has antiproliferative activity in a wide range of cells
from varied origins. Furthermore, numerous studies
have demonstrated the efficacy of orally administered
2ME2 in animal oncology models.1 Currently, 2ME2
is undergoing clinical evaluation in oncology in a num-
ber of Phase 1 and 2 trials under the name Panzemꢁ
NCD.
2
3
H3CO
HO
B
A
HO
1
2
Figure 1.
on microtubule dynamics,3–5 effects on cell cycle kinases,
dysregulating hypoxia-inducible factor-1 (HIF-1),6 and
inhibition of superoxide dismutase (SOD) enzymatic
activity.7 Consistent with the low affinity of 2ME2 for
the estrogen receptors a and b, its antiproliferative activ-
ity is independent of interaction with the estrogen recep-
tors.8 Apoptosis induced by treatment with 2ME2
occurs irrespective of p53 status,9–11 and recent evidence
suggests this activity is mediated through activation of
the extrinsic pathway following upregulation of Death
Receptor 5 (DR5).12 2ME2 binds the colchicine binding
site of tubulin and destabilizes microtubules in tumor
cells in vitro and in vivo. Through its disruption of
microtubules, 2ME2 causes reduction in HIF-1a protein
levels.6 HIF-1a is a component of the dimeric transcrip-
tion factor HIF-1 which regulates over 70 genes
involved in cell survival, proliferation, metastasis, and
angiogenesis.13,14
While the angiogenic, pro-apoptotic and tumor inhibi-
tory effects of 2ME2 and its ability to induce apoptosis
are well documented,2 the mechanism of action at the
molecular level is still being investigated. Several targets
have been proposed, such as binding to tubulin, effects
Keywords: 2-Methoxyestradiol analogs; Structure–activity relation-
ships; Antiproliferative agent; Conjugation.
*
To elucidate the determinants of 2ME2 activity and to
prepare more efficacious analogs, structure–activity
Corresponding author. Tel.: +1 240 864 2600; fax: +1 301 294
1775; e-mail: grega@entremed.com
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.09.011

G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
7525
relationship (SAR) studies have been conducted by sev-
eral groups. In general, these SARs were evaluated in
terms of antiproliferative activity across the NCI 60
tumor cell line screen (GI50), and/or by binding to tubu-
lin and inhibition of tubulin polymerization. Most of
these SARs have concentrated on modifying or incorpo-
rating new substituents in the A, B, and D rings of
2ME2. In 1995, Cushman reported several 2-modified
analogs of estradiol.15 In this series of compounds, sub-
stituents that resembled the 2-methyl ether such as eth-
oxy and 20,20-difluoroethoxy as well as propynyl and
propene had similar activities or were more potent than
2ME2. In 1997, Cushman16 reported that modification
of position 6 in some 2-modified analogs of 2ME2
resulted in compounds with similar or improved GI50
values compared to 2ME2, when the position 6 substit-
uents were OH, NH2 or ketone. In 1997, Miller and
Macdonald reported that expansion of the A-ring to
give tropinones had better binding to the cholchicine site
of tubulin and inhibited tubulin polymerization more
effectively than 2ME2, with the best analog reported
to be approximately 5-fold more potent on tubulin.17,18
In 2000, Wang19 reported that expansion of the steroid
B-ring to a seven membered ring caused an increase in
the GI50 value for the NCI 60 tumor cell line screen,
indicating these substitutions did not result in better
antiproliferative activity. The exception in this series
was the 6-acetamide, which retained similar activity to
2ME2. In 2002, Hughes reported cell cycle analysis
(via inhibition of DNA synthesis) and ER affinity data
for a series of 2-methoxy and 2-ethoxyestradiol ana-
logs.20 In the same publication, the authors also pre-
sented a CoMFA model for ER affinity and inhibition
of DNA synthesis. In 2002, Cushman reported a series
of 2-modified analogs of 2ME2 that were cytotoxic,
and of which some were potent inhibitors of tubulin
polymerization.21 This work was continued in 2004 with
additional 2- and 17-modified 2ME2 analogs.22 Rao has
reported on 2ME2 analogs with multiple A, B, and D
ring modifications which have in vitro antiproliferative
activity.23 Several members of this series are ꢀ2- to
10-fold more potent inhibitors of MDA-MB-231 and
SK-OV-3 cell proliferation than 2ME2 and ꢀ25-fold
more potent at inhibition of tubulin polymerization. In
summary, a number of groups have demonstrated that
2ME2 can be modified to improve inhibition of tubulin
polymerization and antiproliferative activity.
that other D-ring modifications may yield analogs with
increased metabolic stability.
As 17-oxidation and conjugation are major pathways
for metabolism and deactivation of 2ME2, we hypothe-
sized that the addition of steric and/ or electronic bulk at
position 16 may prevent or slow oxidation and/or conju-
gation at position 17 and consequently improve the anti-
angiogenic and antitumor effects observed with 2ME2
in vivo. To the best of our knowledge, there has been
no structure–activity relationships established for
16-alkyl or 16-heteroalkyl substituted 2ME2 analogs.
In this paper, we report a novel series of 2ME2 analogs
with substitutions at position 16 of the D-ring and their
structure–activity relationships. The biological activities
assessed in vitro are antiproliferative activity toward
HUVEC and MDA-MD-231 cells, level of estrogenic
activity as reflected in MCF-7 cells, and glucuronide
and sulfate conjugation by human liver enzymes.
2. Results
2.1. Chemistry
All analogs were prepared using similar reaction condi-
tions reported for estradiol analogs. 2ME2 was first pro-
tected as a benzyl ether using K2CO3/benzyl bromide in
ethanol, then oxidized to 3-benzyl-2-methoxyestrone (3)
using Swern conditions.29 All synthetic schemes utilized
3-benzyloxy-2-methoxy estrone (3) as a starting material
to prepare the 16-substituted analogs.
Compound 3 gave access to multiple 16-substituted ana-
logs using enolate chemistry. It has been reported that
direct alkylation of estrone with unactivated alkyl ha-
lides gives a mix of products including both a and b
isomers as well as disubstituted products.30–32 Only acti-
vated electrophiles, such as allyl halides, are reported to
yield monosubstituted products by direct alkylation of
estrones. Based on this precedence, and established
chemistry for estradiol analogs, several of the 16-substi-
tuted analogs that had predominantly the a isomer were
prepared as in Scheme 1.31 Thus, 3-benzyloxy-2ME1 (3)
was treated with LDA, and alkylated with the appropri-
ate allylic halide to give 4a, b, and c. Subsequent low
temperature reduction with lithium aluminum hydride
gave 17b-alcohols 5a, b, and c from the corresponding
ketone. Concurrent debenzylation and olefin reduction
using H2 Pd/C gave the final products 6, 7, and 8.
The metabolism of hormonal steroids is well known. In
the case of estradiol (E2), oxidation of the 17-hydroxy
functionality to
a ketone by 17b-hydroxysteroid
dehydrogenase (type 2, 17b-HSD) is a major metabolic
route, which yields estrone (E1) from E2.24 Additionally,
conjugation of estradiols in the form of glucuronides
and sulfonate esters is a major metabolic pathway for
these compounds, with >95% of the E2 analog 17-ethy-
nylestradiol being converted to conjugates at steady
state.25 Recent evidence has revealed that similar trans-
formations occur in humans and rats following oral
administration of 2ME2.26,27 Recently, Newman
reported that 3,17-bissulfamate-2-methoxyestradiol
had antiproliferative activity and was not a substrate
for 17-b-hydroxysteroid dehydrogenase.28 This suggests
Stereochemical assignments were made by observing the
18-CH3 group and H-16 1H NMR signals. It is well
established that for 18-CH3 the 16a-stereoisomer is up-
field from the 16b isomer (generally about 0.02–
0.05 ppm and the chemical shift is around 0.9 ppm). In
the case of H-16, the 16b-stereoisomer is about 0.4–
0.5 ppm downfield from the 16a-stereoisomer and these
signals are centered around 3.5 ppm.31 The reason for
the preferential formation of the a-stereoisomer in syn-
thesis of 6 and 7 can be explained by the steric shielding
of the b-face of position 16 by the C18-axial methyl
group, which forces the electrophile to approach from

7526
G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
O
O
LiAlH4, THF
1.2 eq LDA, R-X
R
MeO
BnO
MeO
BnO
3
16α major isomer
4a allyl
4b crotyl
4c 2-iso-butenyl
OH
OH
R
Pd(C), H2,EtOAc
R
MeO
HO
MeO
BnO
6 propyl
7 butyl
5a allyl
8
iso-butyl (16β-isomer predominate)
5b crotyl
5c 2-iso-butenyl (16β-isomer predominate)
Scheme 1.
the less hindered alpha face. The exception to this prece-
dence in this series is the case of 16-iso-butylene 4c
which could be a result of isomerization.
reduction by catalytic hydrogenation gave final products
13, 14, and 15. 16-Hydroxymethyl-2ME2 (16, 17) was
prepared by reduction of carboxymethyl intermediate 9
with lithium aluminum hydride at a low temperature to
reduce the 17-ketone to the b-stereoisomer, followed by
warming to ambient temperature to reduce the ester func-
tionality giving 12d.34 Benzyl deprotection using catalytic
hydrogenation and separation by column chromatogra-
When the 16b isomer was the desired isomer, Scheme 2
was utilized.31 Thus, 3-benzyloxy-2-methoxyestrone (3)
was methoxycarbonylated with LDA and cyanomethyl
carboxylate to give 9, then alkylated with KH and the
appropriate alkyl halide giving 10a, b, and c. Subsequent
decarboxymethylation33 gave 11a, b, and c followed by
ketone reduction as above gave protected 16b-alkyl-3-
benzyloxy 2ME2 12a, b, and c. Deprotection and olefin
1
phy gave products 16 and 17. It was observed by H
NMR that the ester 9 was approximately a 3:1 mixture
of b:a stereoisomers. Purification by column chromatog-
raphy isomerized the majority to the more stable
O
1. LiAlH4, THF
2 .Pd(C), H2
EtOAc
O
LDA,
NCCOOMe,
COOMe
THF
MeO
BnO
MeO
BnO
3
9
O
O
COOMe
RX, KH, THF
MeO
BnO
R
MeO
BnO
R
LiCl, H2O, DMF
10a allyl
11a allyl
10bcrotyl
10c methyl
11bcrotyl
11c methyl
OH
OH
MeO
HO
Pd(C), H2
EtOAc
R
MeO
R
LiAlH4, THF
BnO
13 propyl
12a allyl
14 butyl
12bcrotyl
12c methyl
15 methyl
16 α-hydroxymethyl
17 β−hydroxymethyl
Scheme 2.

G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
7527
b-stereoisomer. However, upon LAH reduction of 9 and
subsequent purification by column chromatography, a
small amount of the a-stereoisomer was observed and
was isolated by column chromatography. The a and b
p2–p4) was employed as an in vitro surrogate for antian-
giogenic activity and MDA-MB-231 (human breast can-
cer cell line) was used for antitumor activity. The ability
to sustain the proliferation of estrogen-dependent MCF-
7 breast carcinoma cells in media depleted of other ste-
roidal estrogens was assessed as a measure of estrogenic-
ity. Selected analogs were tested as substrates for
glucuronide and sulfate conjugation in an in vitro assay
using human liver enzymes.
1
isomers’ H and 13C NMR data are in agreement with
Goto.35 The 17b,16b-stereoisomer H-16 had a doublet
at 3.94 ppm with a coupling constant of 9.8 Hz and the
17b, 16a-stereoisomer had the corresponding doublet at
3.47 ppm with coupling constant of 7.5 Hz.15
Two 16a-substituted analogs were prepared using
Scheme 3. The use of hydrazones to give 16a-substituted
steroids is well established.36–39 Thus, 3 was converted to
hydrazone 18 using N,N-dimethyl hydrazine, which was
alkylated using LDA and the appropriate alkyl halide
giving 19a and b. Hydrolysis of the hydrazone with
CuCl2 gave ketones 20a and b, which were reduced to
the b-alcohols 21a and b at low temperatures, with lith-
ium aluminum hydride. Removal of the benzyl ether was
accomplished by catalytic hydrogenation giving 22 and
23. Additionally, 16a-(dimethylaminomethyl)-2ME2
was prepared as shown in Scheme 4, thus alkylation of
3 with neat tert-butoxy-bis(dimethylamino)methane at
155 ꢂC followed by catalytic hydrogenation and ketone
reduction yielded 26.40
3. Discussion
2ME2 has been shown to have similar antiproliferative
activities across a wide variety of tumor and non-tumor
cell lines from various species.1 In the current study, we
also found similar IC50 values for 2ME2 in human endo-
thelial cells and MDA-MB-231 tumor cells. In contrast,
substitutions at the 16-position of the D-ring have differ-
ent effects on the activity of the resulting analogs and
several structure–activity relationships can be deter-
mined from the in vitro data in Table 1.
There is a general trend for the 16-substituted analogs
that as steric bulk increases, the IC50 value for MDA-
MB-231 proliferation increases. While the 16-methyl
(15, 22) and ethyl (23) analogs have approximately the
same activity as 2ME2, when propyl (6, 13), butyl (7,
14) or iso-butyl (8) are incorporated at position 16, there
is a 35- to 45-fold drop in the antiproliferative activity
assessed with MDA-MB-231 tumor cells.
2.2. Biology
Screening of these analogs for antiproliferative activity
was performed using two different cell types. Human
umbilical vein endothelial cell (HUVEC, early passage
O
N
N
NH2NMe2
MeO
BnO
MeO
BnO
3
18
O
N
R
N
Cu(II)
THF/H2O
MeO
BnO
LDA, RX
R
MeO
BnO
20a methyl
20b ethyl
19a methyl
19b ethyl
OH
OH
R
Pd(C), H2
EtOAc
R
MeO
HO
LiAlH4, THF
MeO
BnO
22 methyl
23 ethyl
21a methyl
21b ethyl
Scheme 3.

7528
G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
O
O
t-BuOCH(NMe2)
MeO
N
MeO
BnO
BnO
3
24
OH
O
N
LiAlH4, THF
MeO
HO
N
Pd(C), H2
EtOAc
MeO
HO
26
25
Scheme 4.
Table 1. 16-Substituted 2-methoxyestradiol analogs antiproliferative activity
Compound
Compound (a/b)
MDA-MB-231 IC50 (lM)
HUVEC IC50 (lM)
MCF7 SI (relative to 2-methoxyestradiol)
1
22
15
23
16
17
6
2-Methoxyestradiol
16a-Methyl
16b-Methyl (1:2)
16a-Ethyl
16a-Hydroxymethyl
16b-Hydroxymethyl
16a-Propyl (7.3:1)
16b-Propyl (2:1)
16b-Butyl (1:2.6)
16a-Butyl (2:1)
1.00 0.05
0.74 0.21
2.40 0.07
3.59 1.91
4.26 3.52
6.09 2.62
35.10 7.68
45.15 2.75
37.93 1.97
27.94 0.09
38.33 1.73
22.73 2.10
0.84 0.02
0.35 0.15
2.26 0.28
1.35 0.57
2.92 1.40
6.36 1.99
6.55 1.57
17.11 1.74
8.17 0.87
18.17 1.94
7.15 2.04
36.82 0.97
1.00
2.07 0.43
1.62 0.43
1.09 0.30
0.93 0.48
1.14 0.20
0.83 0.16
0.93 0.14
0.82 0.18
0.73 0.11
0.41 0.02
1.43 0.45
13
14
7
8
26
16b-iso-Butyl (1:2)
16a-Dimethylaminomethyl
Values are averages SD of at least two different experiments, in which the effect at each concentration of the analog was assessed by triplicate.
Structure–activity relationships for antiproliferative
activity toward endothelial cells (HUVEC) are different,
as the results show that larger 16-substituents are better
tolerated in these cells. Methyl (15, 22), ethyl (23),
a-propyl (6), b-butyl (14) and b-iso-butyl (7) retain
activities within one order of magnitude of the antipro-
liferative activity of 2ME2. In the case of the larger alkyl
substituents, b-propyl (13) and a-butyl (7), there is an
approximate 20-fold drop in activity.
serve as more specific antiangiogenic agents. However,
this would need to be investigated further by testing
the antiproliferative activity of these analogs with other
tumor cell lines in order to evaluate whether this is a
general structure–activity relationship or a specific prop-
erty directed only toward certain tumor cell lines.
When polar substituents such as hydroxymethyl (16, 17)
are incorporated at position 16, there is a drop in anti-
proliferative activity toward HUVEC ranging from
3- to 8-fold compared to 2ME2. There was little discrim-
ination between the a and b 16-hydroxymethyl moiety.
In the case of 16-dimethylaminomethyl (26), for which
the amine would have a positive charge at the pH of
the assay, there is a 43-fold drop in activity compared
to 2ME2. On the other hand, if we compare the
16-a-ethyl (23) to the 16-a-hydroxymethyl (16), which
are sterically similar but electronically dissimilar, there
is not a large difference in their activities in HUVEC
or MDA-MB-231 antiproliferative activity, or in their
estrogenic activity for MCF-7 cells. In contrast, while
in MDA-MB-231 cells, there is a less than 2-fold differ-
ence in activity between congener isomers 16-dimethy-
laminomethyl (26) and 16-iso-butyl (8), there is a
significant difference when these compounds are com-
No clear stereochemical structure–activity correlation
was observed for the 16-substituted 2ME2 derivatives
in the cell types tested. Although this observation was
made with the analogs prepared with only a moderate
degree of enantiomeric selectivity (up to 7.3:1), there is
not a large preference for a or b substitution in either
cell line.
Several 16-substituted analogs display disparity between
their activities in HUVEC and MDA-MB-231 tumor
cells. For example, 16a-propyl (6) has a ꢀ5-fold differ-
ence in activity between cell types, and similar effects
are observed with iso-butyl (8) and b-butyl (14). These
analogs are potentially selective for antiproliferative
activity against endothelial cells and, in principle, could

G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
7529
pared on HUVEC. There are two possible interpreta-
tions of these data. First, the positively charged nitrogen
of 26 may be detrimental for HUVEC antiproliferative
activity. Alternatively, there may be a stereochemical
component for positively charged heterosubstituents
(e.g. 26) whereas there is no such component for neutral
substituents since a large discrimination between 16-a
and b isomers was not observed, as discussed previously.
es. After 60 min of incubation, 85% of 2ME2 remained
unchanged. Consistent with these results, recent in vivo
data from humans and rats showed that conjugation of
2ME2 occurs mainly via glucuronidation and not sulfo-
nation.41 Under the same assay conditions, 24 was rap-
idly conjugated and only 8.3% of parent remained after
60-min incubation. These results indicate that this sub-
stitution, which had only very minor effects on the other
properties assessed, would probably enhance the meta-
bolic clearance of this analog by virtue of its enhanced
rate of sulfonation.
The 16-methyl substituent (both a and b stereochemis-
try) is significantly more estrogenic than 2ME2, as is
the 16-dimethylaminomethyl substituent. All other 16-
substituted 2ME2 analogs had about equal activity to
2ME2 in sustaining proliferation of estrogen-dependent
MCF-7 tumor cells. We have recently shown that the
estrogenic activity observed in this assay with 2ME2 is
related to its metabolism. We found that in the presence
of 2ME2 and a metabolic inhibitor, there was a decrease
in the proliferation of MCF-7 cells accompanied by an
increase in the antiproliferative activity of the 2ME2 (re-
duced IC50)6. It is unclear with the series of analogs pre-
sented here whether compounds such as 15, 22, and 26
are intrinsically estrogenic, or whether their metabolites
are responsible for this activity. More detailed studies on
metabolism and estrogen-like proliferative mechanism
for these compounds are outside the scope of this
investigation.
4. Conclusion
A novel series of 2ME2 analogs modified at position 16
in the D-ring have been synthesized, and various struc-
ture–activity relationships were determined. These ana-
logs were prepared by incorporating activated alkyl
halides via a reactive enolate. In general, the analogs
showed good antiproliferative activity toward HUVEC,
an assay in which a wide range of substituents were well
tolerated at this position. On the other hand, only
methyl and ethyl substituents were well tolerated for
antiproliferative activity against MDA-MB-231 breast
tumor cells. Most of these analogs have approximately
the same activity as 2ME2 in sustaining the proliferation
of the estrogen-dependant MCF-7 breast tumor cells
when assessed at levels below their antiproliferative
concentrations.
Human clinical PK, preclinical ADME, and in vitro
data indicate that in addition to 17-oxidation, conjuga-
tion is a major metabolic pathway for 2ME2. To explore
the effect of 16-substitutions on this process, 16a-butyl-
2ME2 (7), 16a-methyl 2ME2 (22), 16b-hydroxymethyl-
2ME2 (17), 16a-ethyl (22), and racemic 16-ethyl-2ME2
(27) were tested as substrates for UDP-glucuronosyl
transferase (UDPGT) activity in human liver microsome
preparations compared to 2ME2 as a control. 2ME2
was rapidly conjugated in this in vitro assay and, after
60 min, only 9.4% of parent remained unchanged. Sim-
ilar results were obtained with all the 16-modified ana-
logs assessed, suggesting that a similar extent of
conjugation could be expected in vivo for these analogs.
Steroids having larger 16 substitutents (e.g. isopropyl)
were not tested since they had a significant reduction
in in vitro activity and would not be considered as po-
tential drug candidates. The results are summarized in
Table 2.
Analogs that had similar antiangiogenic activity to
2ME2 were similarly good substrates for conjugation
by UDPGT. Interestingly, substitution of the 16 posi-
tion of 2ME2 with an ethyl group unveiled a modula-
tory role for this position, since this analog has the
propensity to form sulfonate esters much more effi-
ciently than 2ME2.
This series of analogs demonstrate that certain modifica-
tions can be made at position 16 of 2ME2 without
affecting the overall antiproliferative activity, and other
modifications result in agents with selective antiprolifer-
ative activity for endothelial cells. Future studies for
these molecules could include determination of the
activity of these analogs over a wide variety of tumor
cell types and in animal tumor models. Additionally,
other assays such as chorioallantoic membrane, cell
migration and matrigel plug assay could be used to ex-
plore the role these molecules have in inhibiting specific
stages in the angiogenic cascade.
Sulfonylation is another potential conjugation pathway
for 2ME2. Both 2ME2 and racemic 16-ethyl-2ME2 were
assessed as substrates for liver cytosolic sulfotransferas-
Table 2. Conjugation of 16-substituted 2ME2 analogs
Compound
16-Substitutent
% Parent remaining
at t = 0 min
Glucuronidation
(% remaining at t = 60 min)
Sulfonylation
(% remaining at t = 60 min)
1
7
H
100
100
100
100
100
100
9.4
0
85
—
—
—
—
8.3
a-Butyl
22
17
23
27
a-Methyl
b-Methanol
a-Ethyl
12
17
0
Ethyl (racemic)
—
All assays were performed by BD Biosciences, Woburn, MA. Each assay was performed in duplicate, reported results are the average value from each
determination.

7530
G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
5. Experimental
to give a white solid. The solid was crystallized with hot
ethanol to give white crystals (9.94 g, 25.5 mmol, 76%
overall yield from 2ME2). Selected spectral data: 1H
NMR (300 MHz, CDCl3) d 7.28–7.48 (m, 5H), 6.86 (s,
5.1. General
Chemicals and reagents of high purity were obtained
from Aldrich, Fluka, or Acros chemical companies
and were used without further purification. 2ME2 was
purchased from Tetrionics (now SAFC), Madison, WI.
Chemical reactions were monitored by thin layer chro-
matography using Merck precoated silica gel 60F254
plates. Flash columns packed manually with silica gel
1H), 6.66 (s, 1H), 3.88 (s, 3H), 0.94 (s, 3H). IR (cmꢁ1
neat) 2920, 1731, 1519, 1202, 1012.
,
5.3. 3-(Benzyloxy)-16a-(E-crotyl)-2-methoxyestra-
1,3,5(10)-triene-17-one (4b)
LDA (2M Aldrich, heptane/THF/ethylbenzene, 1.2 mL,
2.4 mmol) was added to anhydrous THF (10 mL) and
cooled to ꢁ78 ꢂC, and 3 (780 mg, 2.0 mmol) in THF
(10 mL) was added dropwise. Following addition, the
mixture was warmed to 0 ꢂC and stirred 1 h. The mix-
ture was then cooled to ꢁ78 ꢂC and DMPU (1 mL),
then E-crotyl bromide (205 lL, 2.0 mmol) were added
dropwise. The mixture was warmed to rt over 4 h. The
reaction was quenched by carefully adding water
(100 mL) and washing with ethyl acetate (2 · 100 mL).
The combined organics were washed with water
(100 mL) and saturated NaCl (100 mL). The solution
was dried with MgSO4, filtered, and solvent was re-
moved under reduced pressure. The crude product was
purified using hexane–ethyl acetate (9:1) with a SiO2
Biotage FLASH apparatus. Obtained 680 mg
(1.53 mmol) product and recovered 121 mg (0.31 mmol)
of starting material (90% yield based on recovered start-
ing material). Selected spectral data: 1H NMR
(300 MHz, CDCl3) d 7.28–7.48 (m, 5H), 6.86 (s, 1H),
6.66 (s, 1H), 5.34–5.59 (m, 2H), 5.13 (s, 2H), 3.88 (s,
3H), 0.87 & 0.97 (s, total 3H, ratio 1:2).
˚
(Merck, grade 9385, 230–400 mesh, 40 A) or Biotage
(FLASH 12TM or FLASH 40TM) columns were used for
flash chromatography. 1H and 13C nuclear magnetic
resonance (NMR) spectra were recorded on a Bruker
DPX 300 MHz spectrometer. Chemical shifts are
reported in parts per million (d) relative to tetramethyl-
silane as internal standard (0.0 d). IR spectra were
recorded on a Perkin-Elmer 783 spectrometer and wave
numbers are reported in cmꢁ1. Elemental analysis was
performed by Atlantic Micro Lab., Norcross, GA. Both
glucuronidation and sulfotransferase assays were
preformed by BD Gentest, Woburn, MA.
5.2. 3-(Benzyloxy)-2-methoxyestra-1,3,5(10)-triene-17-
one (3)
2ME2 (10.09 g, 33.4 mmol) and potassium carbonate
(22 g, 278 mmol) were suspended in anhydrous ethanol
and cooled to 0 ꢂC. Benzyl bromide (11.4 mL,
95.8 mmol) was added dropwise and, following the addi-
tion, the mixture was brought to reflux for 8 h. The solu-
tion was cooled to rt, and the solvent was removed under
reduced pressure. The resulting residue was diluted with
approximately 200 mL water, and washed with ethyl ace-
tate (3 · 200 mL). The combined organic phases were
washed with water (200 mL), saturated NaHCO3
(200 mL), and saturated NaCl (200 mL). The organic
layer was dried with Na2SO3, filtered, and solvent was re-
moved under reduced pressure. The product was dried in
vacuo with gentle heating with a heat gun to give a yel-
lowish glass (13.54 g, quantitative yield) and was used
without further purification. Selected spectral data:
1 H NMR (300 MHz, CDCl3) d 7.29–7.53 (m, 5H), 6.88
(s, 1H), 6.65 (s, 3H), 5.11 (s, 2H), 3.87 (s, 3H), 3.7 (t,
5.4. 16a-(Allyl)-3-(benzyloxy)-2-methoxyestra-1,3,5(10)-
triene-17-one (4a)
Using the same general procedure for 4b, obtained
822 mg 4a (64%) from 3-benzyloxy-2-methoxyestra-
1,3,5(10)-triene-17-one (1.17 g, 3.0 mmol). 1H NMR
(300 MHz, CDCl3) d 7.50–7.27 (m, 5H), 6.86 (s, 1H),
6.65 (s, 1H), 5.88–5.72 (m, 1H), 5.13 (s, 2H), 5.09–5.02
(m, 2H), 3.88 (s, 3H), 2.89–2.74 (m, 2H), 2.68–2.47 (m,
2H), 2.45–2.13 (m, 3H), 2.13-1.90 (m, 3H), 1.68–1.24
(m, 4H), 0.98 (s) and 0.89 (s, total 3H, ratio 2:1).
J = 8 Hz, 1H), 0.80 (s, 3H). FT-IR (neat, ATR, cmꢁ1
3341, 2920, 2864, 1605, 1513, 1453, 1254, 1211, 1117,
1022. 3-(benzyloxy)-2-methoxyestra-1,3,5(10)-triene-
)
5.5. 3-(Benzyloxy)-16a-(iso-butenyl) 2-methoxyestra-
1,3,5(10)-triene-17-one (4c)
17b-ol was converted to 3 using the Swern oxidation.
Oxalyl chloride (38 mmol, 19 mL, 2M, CH2Cl2) was
added to anhydrous CH2Cl2 (25 mL) and cooled to
ꢁ46 ꢂC. DMSO (5.40 mL, 76 mmol) was added dropwise
and stirred for 2 min. 3-Benzyloxy-2-methoxyestra-
1,4,5(10)-triene-3,17b-diol in CH2Cl2/DMSO (10 mL/
15 mL) and added within 5 min and the resulting mixture
was stirred for 1 h. Triethylamine (170 mmol, 23.5 mL)
was added dropwise, stirred 5 min and warmed to rt.
Water (200 mL) was added and the mixture was washed
with CH2Cl2 (3 · 200 mL). The combined organics were
washed with water (200 mL), dilute HCl (1% aq,
200 mL), saturated NaHCO3 (200 mL), and saturated
NaCl (200 mL). The organics were dried with MgSO4, fil-
tered, and solvent was removed under reduced pressure
Using the same general procedure for 4b, obtained
749 mg 4c (48%) from 3 (1.17 g, 3.0 mmol). H NMR
(300 MHz, CDCl3) d 7.54–7.26 (m, 5H), 6.87 (s, 1H),
6.60 (s, 1H), 5.13 (s, 2H), 4.88–4.69 m, 2H), 3.89 (s,
3H), 2.91–1.24 (m, 17 H), 1.00 (s) and 0.91 (s, total
3H, ratio 1:2.5).
1
5.6. 3-(Benzyloxy)-16b-(carbomethoxy)-2-methoxyestra-
1,3,5(10)-triene-17-one (9)
Compound 3 (1.6113 g, 2.978 mmol) was dissolved in
THF (15 mL), cooled to ꢁ 78 ꢂC and LDA (1.80 mL,
2M, Aldrich, Heptane/THF/ethylbenzene) was added
dropwise and stirred for 1 h. Methyl cyanoformate
(237 lL, 3 mmol) in DMPU (1 mL) was added and the

G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
7531
1
mixture warmed to rt over 18 h. Water (100 mL) was
carefully added, and the mixture was washed with ethyl
acetate (3 · 100 mL). The combined organics were washed
with saturated NaCl (100 mL), dried with Na2SO4, filtered
and solvent was removed under reduced pressure. Final
purification of product using hexane–ethyl acetate
(85:15) then switching to hexane–ethyl acetate (75:25) with
a SiO2 flash column yielded 806 mg product (1.8 mmol,
60%). Selected spectral data: 1H NMR (300 MHz, CDCl3)
d 7.28–7.48 (m, 5H), 6.85 (s, 1H), 6.66 (s, 1H), 5.13 s (2H),
3.88 (s, 3H), 3.78 (s, 3H), 3.23 (dd, J = 9, 10 Hz, 1H), 1.0 (s,
3H). FT-IR (neat, ATR, cmꢁ1) 2929, 2860, 1750, 1723,
1604, 1508, 1211, 1014.
tral data: H NMR (300 MHz, CDCl3) d 7.28–7.48 (m,
5H), 6.85 (s, 1H), 6.66 (s, 1H), 566–5.79 (m, 1H),
5.15–5.20 (m, 2H), 5.13 (s, 2H), 3.88 (s, 3H), 3.75 (s,
3H), 0.99 (s, 3H).
5.9. 3-(Benzyloxy)-16-(E-crotyl)-16-carbomethoxy-2-
methoxyestra-1, 3,5(10)-triene-17-one (10b)
Using the same general procedure as in 10a, obtained
10b (879 mg, 57%) from 9 (1.0892, 2.43 mmol). 1H
NMR (300 MHz, CDCl3) d7.50–7.26 (m, 5H), 6.86 (s,
1H), 6.67 (s, 1H), 5.71–5.50 (m, 1H), 5.40–5.25 (m,
1H), 5.13 (s, 2H), 3.88 (s, 3H), 3.75 (s, 3H), 2.87–2.71
(m, 3H), 2.46–2.20 (m, 4H), 2.17–1.93 (m, 3H), 1.80–
1.26 (m, 10H), 0.99 (s, 3H).
5.7. 3-(Benzyloxy)-16a/b-hydroxymethyl-2-methoxyes-
tra-1,3,5(10)-triene-17b-ol (12d)
5.10. 3-(Benzyloxy)-16-(carbomethoxy)-16-methyl-2-
methoxyestra-1,3,5(10)-triene-17-one (10c)
Compound 9 (826 mg, 1.84 mmol) was dissolved in anhy-
drous THF (120 mL) and cooled to ꢁ78 ꢂC, LAH (1M
THF, Aldrich, 3.68 mL) was added dropwise, stirred for
1 h at ꢁ78 ꢂC then warmed to rt for an additional h.
The reaction was quenched by careful addition of ethyl
acetate (3 mL) and water (3 mL). Additional water
(100 mL) and ethyl acetate (100 mL) were added, and
the aqueous layer was washed with ethyl acetate
(100 mL). The combined organics were washed with sat-
urated NaCl (100 mL), dried with NaSO4, filtered and
solvent was removed under reduced pressure. Product
was purified by Biotage FLASH apparatus using a
hexanes–ethyl acetate gradient elution (85:15 to 4:6
hexanes–ethyl acetate). Obtained 420 mg of product with
16b stereochemistry. This material was used for the prep-
Using the same general procedure as in 10a, obtained
10c (275 mg, 46%) from 9 (584, 1.3 mmol). H NMR
1
(300 MHz, CDCl3) d 7.50–7.24 (m, 5H), 6.84 (s, 1H),
6.65 (s, 1H), 5.13 (s, 2H), 3.88 (s, 3H), 3.74 (s, 3H),
2.90–2.74 (m, 2H), 2.58–2.24 (m, 3H), 2.09–1.89 (m,
4H), 1.71–0.84 (m, 9H), 1.03 (s, 3H).
5.11. 16b-(Allyl)-3-(benzyloxy)-2-methoxyestra-
1,3,5(10)-triene-17-one (11a)
16-Allyl-16-carbomethoxy-3-benzyloxy-2-methoxyestra-
1,3,5(10)-triene-17-one (697 mg, 1.42 mmol), LiCl
(1.15 g, 27 mmol), water (485 lL, 27 mmol) were dis-
solved in DMF (63 mL) and refluxed for 20 h. The reac-
tion was cooled to rt and 1 N HCl (100 mL) was added.
The layers were separated and the aqueous phase was
washed with ether (2 · 100 mL) the combined organics
were washed with water (100 mL), and saturated NaCl
(100 mL). The organics were dried with MgSO4, filtered,
and solvent was removed under reduced pressure. Purifi-
cation by 85:15 hexanes–ethyl acetate SiO2 Biotage
FLASH apparatus gave purified product 66% (401 mg,
0.93 mmol). Selected spectral data: 1H NMR (300 MHz,
CDCl3) d 7.28–7.48 (m, 5H), 6.85 (s, 1H), 6.65 (s, 1H),
5.69–5.88 (m, 1H), 5.13 (s, 2H), 5.00–5.08 (m, 2H), 5.88
(s, 3H), 0.98 and 0.88 (s, total 3H, ratio 1:1.4). FT-IR
(neat, ATR, cmꢁ1) 2925, 2855, 1726, 1514, 1214, 1103.
1
aration of 17. H NMR (300 MHz, CDCl3) d 7.51–7.27
(m, 5H), 6.87 (s, 1H), 6.65 (s, 1H), 5.13 (s, 2H), 3.97 (d,
J = 9.8 Hz, 1H), 3.88 (s, 3H),), 3.67 (dd, J = 3.8, 7 Hz,
1H), 2.90–2.65 (m, 2H), 2.60–1.00 (m, 15H), 0.87 (s,
3H). Also obtained 53 mg of product containing mixture
of a/b stereochemistry (ratio undetermined). This mate-
rial was used for the preparation of 16. 1H NMR
(300 MHz, CDCl3) d 7.50–7.27 (m, 5H), 6.87 (s, 1H),
6.64 (s, 1H), 5.13 (s, 3H), 3.88 (s, 3H), 3.87–3.79 (m)
and 3.69 (app t, J = 8.7, 9.8 Hz, total 1H), 3.55 (d,
J = 7.2 Hz, 1H), 2.88–2.66 (m, 2H), 2.46–1.11 (m, 15H),
0.88 (s, 3H).
5.8. 16-(Allyl)-3-(benzyloxy)-16-(carbomethoxy)-2-meth-
oxyestra-1,3,5(10)-triene-17-one (10a)
5.12. 16b-(E-Crotyl)-3-benzyl-2-methoxyestra-1,3,5(10)-
triene-17-one (11b)
Compound 9 (0.840 g, 1.87 mmol), potassium hydride
(1.5 g, 10.9 mmol, 30% mineral oil dispersion, washed
in hexanes), and 18-crown-6 (120 mg, 0.4 mmol) were
mixed in THF (40 mL) and refluxed for 1 h. The mixture
was cooled to rt, and allyl bromide (537 lL, 6.2 mmol)
was added, then refluxed for 18 h. After cooling to rt,
the reaction was quenched by carefully adding water
(2 mL), stirring, then adding additional water
(100 mL). This mixture was washed with ethyl acetate
(2 · 100 mL) and the combined organics were washed
with saturated NaCl (100 mL). The organics were dried
with MgSO4, filtered, and solvent was removed under
reduced pressure. Purification using 85:15 hexanes–ethyl
acetate using a SiO2 Biotage FLASH apparatus yielded
697 mg of product (1.42 mol, 76% yield). Selected spec-
Using the same general procedure as in 11a obtained 11b
(700 mg, 92%) from 10b (879 mg, 1.74 mmol). Product
was used as is without additional purification. 1H NMR
(300 MHz, CDCl3) d 7.50–7.26 (m, 5H), 6.85 (s, 1H),
6.65 (s, 1H), 5.60–5.34 (m, 2H), 5.13 (s, 2H), 3.88 (s,
3H), 2.86-2.74 (m, 2H), 2.62-1.19 (m, 18H), 0.86 (s, 3H).
5.13. 3-(Benzyloxy)-16b-methyl-2-methoxyestra-
1,3,5(10)-triene-17-one (11c)
Using the same general procedure as in 11a, obtained 11c
(180 mg, 44%) from 10c (482 mg, 1.0 mmol). Product was
purified using Biotage FLASH apparatus using 85:15

7532
G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
1
Hexanes–ethyl acetate as eluent. H NMR (300 MHz,
CDCl3) d 7.50–7.26 (m, 5H), 6.86 (s, 1H), 6.66 (s, 1H),
5.13 (s, 2H), 3.88 (s, 3H), 2.86–2.72 (m, 2H), 2.68–1.11
(m, 15H), 0.97 (s) and 0.90 (s, total 3H, ratio 1:2).
5.17. 3-(Benzyloxy)-2-(methoxy)-17-(N,N-dimethylhyd-
razone)-estra-1,3,5(10)-triene (18)
Compound (3) (4.6733 g, 11.99 mmol) was suspended in
absolute ethanol (80 mL) and 1,1-dimethylhydrazine
(9.492 g, 158 mmol), then DMF (2 mL) was added.
The mixture was refluxed for 20 h, cooled to rt, and
poured into water (200 mL) The mixture was washed
with ether (2 · 200 mL) and the combined organic layers
were washed with water (2 · 200 mL), then saturated
NaCl (1 · 200 mL). The combined organics were dried
with Na2SO4, filtered, and solvent was removed under
reduced pressure. Product was purified with Biotage
FLASH apparatus using 3:1 hexanes–ethyl acetate (1%
triethylamine) as eluent, starting material was recovered
(1.9037 g, 4.88 mmol) and 18 was obtained as a clear
colorless oil that crystallized in vacuo (1.5229 g, 49%
based on recovered starting material). Selected spectral
data: 1H NMR (300 MHz, CDCl3) d 7.50–7.16 (m,
5H), 6.86 (s, 1H), 6.65 (s, 1H), 5.13 (s, 2H), 3.89 (s,
3H), 2.86–2.74 (m, 2H), 2.59–1.22 (m, 11 H), 2.50 (s,
6H), 0.93 (s, 3H).
5.14. 3-(Benzyloxy)16a-(N,N-(dimethyl)aminomethyl)-2-
methoxyestra-1,3,5(10)-triene-17-one (24)
Compund 3 (1.51 g, 3.87 mmol) was suspended in
tert-butoxy bis(dimethylamino)methane (1.64 mL, 8.13
mmol) and heated in an oil bath (155 ꢂC) for 1.5 h,
during which time the steroid dissolved. The reaction
mixture was cooled to rt, and poured into ice water
(100 mL) and washed with CH2Cl2 (2 · 100 mL). The
organics were washed with brine (100 mL), dried with
magnesium sulfate, filtered, and solvent was removed
under reduced pressure gave product which was used
without further purification (1.82 g, quantitative yield).
Selected spectral data: 1H NMR (300 MHz, CDCl3)
d 7.23–7.47 (m, 5H), 6.87 (s, 1H), 6.64 (s, 1H), 5.12
(s, 2H), 3.88 (s, 3H), 3.07 (s, 6H), 0.91 (s, 3H).
5.15. 16a-(N,N-(Dimethyl)aminomethyl)-2-methoxyes-
tra-1,3,5(10)-triene-17-one-3-ol (25)
5.18. 3-(Benzyloxy)-16a-(methyl)-17-(N,N-dimethylhyd-
razone)estra-1,3,5(10)-triene (19a)
Compound 24 (473 mg, 1.06 mmol) was dissolved in
ethyl acetate (25 mL) in a Parr pressure bottle. The bottle
was flushed with Ar, and Pd/C (10%) (2.5 g) was added.
The bottle was fitted to a Parr hydrogenator, filled and
purged with H2 five times, and then pressurized with H2
to 50 psi. The reaction mixture was agitated for 48 h, then
filtered through celite and solvent was removed under re-
duced pressure. The crude product was purified using a
Biotage FLASH apparatus 99:1:1 CHCl3–CH3OH:
NH4 OH (30% aq) eluent. Product 25 was obtained
(70 mg, 18%). Selected spectral data: 1H NMR
(300 MHz, CDCl3) d 6.79 (s, 1H), 6.66 (s, 1H), 3.86 (s,
3H), 3.04–0.91 (m, 14H), 2.28 (s, 6H), 0.87 (s, 3H).
Compound 18 (1.555 g, 3.6 mmol) was dissolved in THF
(20 mL) and cooled to 0 ꢂC. n-Butyllithium (1.46 M, hex-
anes, Aldrich, 2.9 mL, 4.32 mmol) was added dropwise
and stirred 30 min, methyl iodide (263 lL, 4.22 mmol)
was then added and the mixture was warmed to rt over
30 min and stirred an additional 2 h. The reaction was
quenched by addition of saturated ammonium chloride
(100 mL) and the resulting solution was washed with
ether (2 · 100 mL) and the combined organics were
washed with saturated NaCl (1 · 100 mL), dried with
Na2SO4, filtered, and solvent was removed under reduced
pressure. The product was purified with a Biotage
FLASH apparatus using 6:1 hexanes–ethyl acetate (1%
triethylamine) as eluent. Obtained 19a (715 mg, 45.5%).
1H NMR (300 MHz, CDCl3) d 7.51–7.25 (m, 5H), 6.87
(s, 1H), 6.65 (s, 1H), 5.13 (s, 2H), 3.88 (s, 3H), 2.88–
2.69 (m, 2H), 2.61–1.21 (m, 13H), 2.44 (s, 6H), 1.32 (d,
J = 7.16 Hz, 3H), 0.89 (s, 3H).
5.16. 16a-(N,N-(Dimethyl)aminomethyl)-2-methoxyes-
tra-1,3,5(10)-triene-3,17b-diol (26)
Compound 25 (70 mg, 0.2 mmol) was dissolved in anhy-
drous THF (10 mL) and cooled to ꢁ78 ꢂC. LAH (17 mg,
0.45 mmol) was added and the mixture was stirred for
2.5 h. The mixture was quenched by careful addition of
ethyl acetate (3 mL) then water (3 mL), the mixture was
warmed to rt, then additional ethyl acetate (30 mL) and
water (30 mL) were added. The layers were separated,
and the aqueous layer was washed with ethyl acetate
(30 mL). The combined organic layers were washed with
water (30 mL) and saturated NaCl (30 mL), dried with
MgSO4, filtered, and solvent was removed under reduced
pressure. Product was purified using Biotage FLASH
apparatus with 99:1:1 CHCl3–CH3OH–NH4OH (30%
aq) as eluent. Obtained 26 (18 mg, 5%). Selected spectral
data: 1H NMR (300 MHz, CDCl3) d 6.81 (s, 1H), 6.65 (s,
1H), 3.88 (s) and 3.85 (obscured d) (total 4H), 2.28 (s,
6H), 0.87 (s, 3H). 13C NMR (75 MHz, CDCl3) d 145.1,
143.8, 132.2, 129.7, 115.1, 108.7, 83.1, 77.6, 63.6, 56.4,
49.5, 45.03, 44.98, 44.55, 38.6, 38.1, 34.9, 30.5, 29.3,
28.0, 27.0, 13.2. Anal. Calcd for C22H33O3N1/4H2O: C,
72.59; H, 9.28; N, 3.85 Found: C, 72.80; H, 9.17; N, 3.66.
5.19. 3-(Benzyloxy)-16a-(ethyl)-17-(N,N-dimethylhyd-
razone)estra-1,3,5(10)-triene (19b)
Using the same general procedure as in 19a, with ethyl
iodide (824 mg, 5.28 mmol) obtained 19b (1.334 g,
66%) from 18 (1.9258 g, 4.4 mmol). 1H NMR
(300 MHz, CDCl3) d 7.52–7.22 (m, 5H), 6.90 (s, 1H),
6.67 (s, 1H), 5.14 (s, 2H), 3.91 (s, 3H), 2.48 (s, 6H),
2.90–0.86 (m, 19H), 0.93 (s, 3H).
5.20. 3-(Benzyloxy)-16a-(methyl)estra-1,3,5(10)-triene-
17-one (20a)
Copper(II) chloride dihydrate (860 mg, 6.4 mmol) was
dissolved in water (10 mL) and (19a) (715 mg, 1.6 mmol)
in THF/H2 O (35 mL/7 mL) was added. The mixture was
stirred for 4 h, and was poured into water (200 mL). The
mixture was washed with ether (2 · 200 mL) and the

G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
7533
combined organics were washed with saturated NaCl
(1 · 200 mL). The organics were dried over Na2SO4, fil-
tered, and solvent was removed under reduced pressure.
The product was purified with Biotage FLASH appara-
tus using 8:1 hexanes–ethyl acetate as eluent. Obtained
20a (380 mg, 58%), which was used as is for ketone reduc-
1H NMR (300 MHz, CDCl3) d 7.51–7.26 (m, 5H), 6.88
(s, 1H), 6.65 (s, 1H), 5.13 s, 2H), 4.79 (s, 2H), 3.89 (s,
3H), 3.87–3.80 (m) and 3.37–3.28 (m, total 1H, ratio
undetermined), 3.88–2.66 (m, 2H), 2.53–1.00 (m, 15H),
0.86 (s) and 0.83 (s, total 3H, ratio 1:2.3).
1
tion. H NMR (300 MHz, CDCl3) d 7.50–7.27 (m, 5H),
5.25. 3-(Benzyloxy)-16b-(allyl)-2-methoxyestra-
1,3,5(10)-triene-17bol (12a)
6.87 (s, 1H), 6.67 (s,1H), 5.14 (s, 2H), 3.89 (s, 3H),
2.87–2.76 (m, 2H), 2.67–2.54 (m, 1H), 2.46–2.54 (m,
1H), 2.46–2.21 (m, 2H), 2.03–1.90 (m, 2H), 1.80–1.22
(m, 8H), 1.17 (d, J = 7.5 Hz, 3H), 0.98 (s, 3H).
Using the same general procedure for 5b, obtained 12a
(268 mg, 72%) from 11a (401 mg, 0.93 mmol). Product
was purified with Biotage FLASH apparatus using 4:1
1
5.21. 3-(Benzyloxy)-16a-(ethyl)estra-1,3,5(10)-triene-17-
one (20b)
hexanes–ethyl acetate as eluent. H NMR (300 MHz,
CDCl3) d 7.50–7.25 (m, 5H), 6.87 (s, 1H), 6.64 (s, 1H),
5.97–5.79 (m, 1H), 5.12 (s, 2H), 5.12–4.97 (m 2H), 3.88
(s, 3H), 3.82 (d, J = 10.2 Hz) and 3.43 (d, J = 7.5 Hz,
total 1 H, ratio ꢀ2:1), 2.85–2.67 (m, 2H), 2.57–1.03 (m,
15H), 0.84 (s) and 0.82 (s, total 3H, ratio ꢀ2:1).
Using the same general procedure as in 20a, obtained
20b (838 mg, 68%) from 19b (1.334 g, 2.9 mmol), which
was used as is for ketone reduction. 1H NMR
(300 MHz, CDCl3) d 7.50–7.25 (m, 5H), 6.86 (s, 1H),
6.66 (s, 1H), 5.13 (s, 2H), 3.88 (s, 3H), 2.91–2.66 (m,
2H), 2.49–1.18 (m, 15H), 0.97 (s, 3H).
5.26. 3-(Benzyloxy)-16b-(E-crotyl)-2-methoxyestra-
1,3,5(10)-triene-17b-ol (12b)
5.22. 3-(Benzyloxy)-16a-(E-crotyl)-2-methoxyestra-
1,3,5(10)-triene-17b-ol (5b)
Using the same general procedure for 5b, obtained 12b
(300 mg, 42%) from 11b (700 mg, 1.61 mmol). Product
was purified with Biotage FLASH apparatus using 4:1
1
Compound 4b (680 mg, 1.53 mmol) was dissolved in
anhydrous THF (10 mL), and cooled to ꢁ78 ꢂC. Lith-
ium aluminum hydride (3.06 mmol, 116 mg) was added
and the solution was stirred for 2 h. The reaction was
quenched by carefully adding water (2 mL) and warm-
ing to rt, then adding additional 50 mL portion of water.
The mixture was washed with ethyl acetate (2 · 50 mL)
and the combined organics were washed with water
(50 mL), saturated NaCl (50 mL), dried with MgSO4,
filtered, and solvent was removed under reduced
pressure. The mixture was purified with 3:1 hexane–
ethyl acetate using a SiO2 Biotage FLASH apparatus
hexanes–ethyl acetate as eluent. H NMR (300 MHz,
CDCl3) d 7.51–7.26 (m, 5H), 6.87 (s, 1H), 6.64 (s, 1H),
5.59–5.38 (m, 2H), 5.13 (s, 2H), 3.88 (s, 3H), 3.81 (dd,
J = 4, 7 Hz) and 3.33 (d, J = 7 Hz, total 1H, ratio
ꢀ4:1), 2.88–2.66 (m, 2H), 2.48–0.98 (m, 11H), 0.84 (d,
J = 3 Hz, 2H), 0.81 (s, 3H).
5.27. 3-(Benzyloxy)-16b-(methyl)-2-methoxyestra-
1,3,5(10)-triene-17b-ol (12c)
Using the same general procedure for 5b, obtained 12c
(158 mg, 88%) from 11c (180 mg, 0.44 mmol). Product
was purified with Biotage FLASH apparatus using 3:1
to give 500 mg purified product (5b, 1.12 mmol, 73%
yield). Selected spectral data:
1
1
H NMR (300 MHz,
hexanes–ethyl acetate as eluent. H NMR (300 MHz,
CDCl3) d 7.28–7.48 (m, 5H), 6.87 (s, 1H), 6.64 (s, 1H),
5.47–5.56 (m, 2H), 5.12 (s, 2H), 3.88 (s, 3H), 3.8 (d,
J = 9 Hz) and 3.33 (d, J = 8 Hz) total 1H, ratio 1:1.7,
0.84 and 0.81 (s, 3H total).
CDCl3) d 7.50–7.27 (m, 5H), 6.87 s, 1H), 6.64 (s, 1H),
5.13 (s, 2H), 3.88 (s, 3H), 3.73 (d, J = 9.8 Hz) and 3.23
(d, J = 7.5 Hz, total 1H, ratio ꢀ2:1), 2.86–2.68 (m,
2H), 2.40–1.22 (m, H), 1.16 (d, J = 7 Hz) and 1.05 (d,
J = 7.5 Hz, total 3H, ratio ꢀ2:1), 0.82 (s, and 0.80 (s,
total 3H, ratio ꢀ2:1).
5.23. 3-(Benzyloxy)-16a-(allyl)-2-methoxyestra-
1,3,5(10)-triene-17b-ol (5a)
5.28. 3-(Benzyloxy)-16a-(methyl)-2-methoxyestra-
1,3,5(10)-triene-17b-ol (21a)
Using the same general procedure for 5b, obtained 5a
(914 mg, quantitative) from 4a (822 mg, 1.91 mmol).
Product was used as is without additional purification.
1H NMR (300 MHz, CDCl3) d 7.50–7.25 (m, 5H), 6.86
(s, 1H), 6.64 (s, 1H), 5.89–5.79 (m, 1H), 5.12 (s, 2H),
5.10–4.97 (m, 2H), 3.87 (s, 3H), 3.87–3.76 (m) and
3.39–3.30 (m, total 1H, ratio not determined), 2.87 (m,
2H), 2.43 (m) and 1.03 (m, 15H), 0.83 (s) and 0.82 (s,
total 3H, ratio ꢀ3:1).
Using the same general procedure for 5b, obtained 21a
(116 mg, 73%) from 20a (156 mg, 0.39 mmol). Product
was purified with Biotage FLASH apparatus using 3:1
1
hexanes–ethyl acetate as eluent. H NMR (300 MHz,
CDCl3) d 7.50–7.27 (m, 5H), 6.88 (s, 1H), 6.65 (s, 1H),
5.13 (s, 2H), 3.88 (s, 3H), 3.23 (d, J = 7.5 Hz, 1H),
2.84–2.66 (m, 2H), 2.37–1.21 (m, 13H), 1.17 (d,
J = 6.8 Hz, 3H), 0.83 (s, 3H).
5.24. 3-(Benzyloxy)-16b-(isobutenyl)-2-methoxyestra-
1,3,5(10)-triene-17b-ol (5c)
5.29. 3-(Benzyloxy)-16a-(ethyl)-2-methoxyestra-
1,3,5(10)-triene-17b-ol (21b)
Using the same general procedure for 5b, obtained 5c
(841 mg, quantitative) from 4c (749 mg, 1.43 mmol).
Product was used as is without additional purification.
Using the same general procedure for 5b, obtained 21b
(873 mg, 43%) from 20b (838 mg, 4.83 mmol). Product

7534
G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
1
was purified with Biotage FLASH apparatus using 3:1
hexanes–ethyl acetate as eluent. H NMR (300 MHz,
spectral data: H NMR (300 MHz, CDCl3) d 6.81 (s,
1H), 6.66 (s, 1H), 3.87 (s, 3H), 3.76 (d, J = 10 Hz) and
3.29 (t, J = 7 Hz) (total 1H, ratio 2:1), 0.95 (t,
J = 7 Hz, 3H), 0.83 and 0.80 (s, total 3H). 13C NMR
(75 MHz, CDCl3) d 145.1, 143.8, 132.2, 129.9, 115.1,
108.7, 88.4 and 82.8, 56.5, 48.9 and 48.7, 44.7 and
44.5, 44.4 and 44.0, 40.1, 39.0 and 38.7, 38.4 and 38.1,
34.1, 32.7, 30.4 and 29.4, 27.5 and 27.4, 27.0 and 26.9
and 22.1 and 22.0, 14.7 and 14.67, 12.8 and 12.3. FT-
IR (neat, ATR, cmꢁ1) 3411, 2923, 1504,1446, 1267,
1202, 1118, 1024. Anal. Calcd for C22H32O31/4H2O: C,
75.71; H, 9.39. Found: C, 75.61; H, 9.33.
1
CDCl3) d 7.50–7.25 (m, 5H), 6.87 (s, 1H), 6.64 (s, 1H),
5.13 (s, 2H), 3.88 (s, 3H), 3.53–3.23 (m, 1H), 2.87–0.74
(m, 17H), 0.83 (s, 3H).
5.30. 16a-(n-Butyl)-2-methoxyestra-1,3,5(10)-triene-
3,17b-diol (7)
Compound 5b (500 mg, 1.12 mmol) was dissolved in
ethyl acetate (25 mL) in Parr reaction bottle. The bottle
was flushed with argon, and Pd/C (10%, 2.5 g) was
added. The bottle was fitted to a Parr hydrogenator,
filled and purged with hydrogen five times, pressurized
to 50 psi, and agitated for 24 h. The mixture was filtered
through a celite pad, the solvent was removed under re-
duced pressure and purified with a 3:1 hexane ethyl ace-
tate SiO2 flash column. Crystallization from ethyl acetate
hexanes yielded 358 mg product (1.0 mmol, 89%).
5.34. 16b-(n-Butyl)-2-methoxyestra-1,3,5(10)-triene-
3,17b-diol (14)
Using the same general procedure for 7, obtained 14
(170 mg, 70%) from 12b (300 mg, 0.67 mmol). Selected
1
spectral data: H NMR (300 MHz, CDCl3) d 6.81 (s,
1
Selected spectral data: H NMR (300 MHz, CDCl3) d
1H), 6.66 (s, 1H), 5.43 (s, 1H), 3.88 (s, 3H), 3.76 (d,
J = 10 Hz) 3.29 (d, J = 8 Hz) (total 1H, ratio 2.6:1),
0.83 and 0.80 (s, total 3H). 13 C NMR (75 MHz, CDCl3)
d 145.1, 143.8, 132.2, 129.9, 115.1, 108.7, 88.4 and 82.8,
56.5, 48.9 and 48.7, 47.7 and 44.5, 44.4 and 44.2, 40.4
and 38.1, 37.2 and 35.8, 32.8 and 31.6 and 31.3, 31.0
and 30.5, 29.4, 27.9 and 27.7, 27.0 and 26.9, 23.4 and
23.3, 14.54 and 14.50, 12.8 and 12.3. FT-IR (neat,
ATR, cmꢁ1) 3221, 2921, 1594, 1504, 1416, 1265, 1200,
1021. Anal. Calcd for C23H34O3: C, 77.04; H, 9.56.
Found: C, 77.06; H, 9.65.
6.81 (s, 1H), 6.66 (s, 1H), 3.87 (s, 3H), 3.76 (d,
J = 10 Hz) and 3.29 (d, J = 8 Hz) (total 1H, ratio 1:2),
0.82 and 0.79 (s, 3H). FT-IR (neat, ATR, cmꢁ1) 3245,
2914, 1606, 1523, 1414, 1258, 1028. Anal. Calcd for
C20H34O3: C, 77.44; H, 9.56. Found: C, 76.64; H, 9.51.
5.31. 16a-(n-Propyl)-2-methoxyestra-1,3,5(10)-triene-
3,17b-diol (6)
Using the same general procedure for 7, obtained 6
(577 mg, 78%) from 5a (914 mg, 2.12 mmol). Selected
1
spectral data: H NMR (300 MHz, CDCl3) d 6.81 (s,
5.35. 16b-(Methyl)-2-methoxyestra-1,3,5(10)-triene-
3,17b-diol (15)
1H), 6.66 (s, 1H), 5.43 (s, 1H), 3.87 (s, 3H), 3.29 (t,
J = 7 Hz, 1H), 0.95 (t, J = 7 Hz, 3H), 0.83 and 0.80 (s,
total 3H, ratio 7.3:1). 13C NMR (75 MHz, CDCl3) d
145.0, 143.8, 132.2, 130.0, 115.0, 108.5, 88.6, 56.6, 49.0
and 48.8, 44.7, 44.5, 44.2, 39.0 and 38.7, 38.4, 37.3,
30.5, 29.4, 27.7, 26.9, 22.1 and 22.0, 14.8, 12.3. Anal.
Calcd for C22H32O3: C, 76.69; H, 9.37. Found: C,
76.55; H, 9.44.
Using the same general procedure for 7, obtained 15
(65 mg, 52%) from 12c (157 mg, 0.38 mmol). Selected
1
spectral data: H NMR (300 MHz, CDCl3) d 6.82 (s,
1H), 6.66 (s, 1H), 3.88 (s, 3H), 3.73 (d, J = 9.8 Hz) and
3.23 (d, J = 7.5 Hz, total 1 H, ratio 2:1), 2.84–2.74 (m
2H), 2.41–1.22 (m, H), 1.17 (d, J = 6.8 Hz) and 1.05
(d, J = 7.5 Hz, total 3H, ratio 1:2), 0.83 (s) and 0.81 (s,
C NMR (75 MHz, CDCl3) d
13
5.32. 16b-(iso-Butyl)-2-methoxyestra-1,3,5(10)-triene-
3,17b-diol (8)
total 3H), ratio 2:1).
145.1, 143.8, 132.2, 129.9, 115.1, 108.7, 90.0 and 82.5,
56.4, 49.0 and 48.5, 44.7 and 44.5, 38.9 and 38.7, 38.2
and 37.2, 34.5, 32.4, 29.4, 27.9 and 27.6, 26.9, 20.8,
16.8, 12.8 and 12.2. FT-IR (neat, ATR, cmꢁ1) 3411,
2923, 1504, 1446, 1267, 1202, 1118, 1024. Anal. Calcd
for C20H28O31/4H2O: C, 74.85; H, 8.95. Found: C,
74.83; H, 8.86.
Using the same general procedure for 7, obtained 8
(427 mg, 63%) from 5c (841 mg, 1.89 mmol). Selected
1
spectral data: H NMR (300 MHz, CDCl3) d 6.81 (s,
1H), 6.66 (s, 1H), 5.43 (s, 1H), 3.88 (s, 3H), 3.77 (dd,
J = 9, 10 Hz) and 3.26 (t, J = 7 Hz) (total 1 H, ratio 2:1),
0.84 and 0.80 (s, total 3H). 13C NMR (75 MHz, CDCl3)
d 145.0, 143.9, 132.2, 129.9, 115.0, 108.5, 89.1 and 82.9,
56.5, 49.0 and 48.7, 44.7 and 44.6, 42.2 and 40.8, 39.0
and 38.7, 38.2, 37.8 and 37.3, 32.7, 30.6, 29.4, 27.9 and
27.7, 27.2 and 27.0, 26.9 and 26.7, 24.7 and 23.8, 22.7
and 21.7, 12.8 and 12.3. IR (neat, cmꢁ1, ATR) 3525,
2913, 1506, 1258, 1202, 1026. Anal. Calcd for C22
H30O3: C, 76.69; H, 9.37. Found: C, 76.82; H, 9.47.
5.36. 16a-(Hydroxymethyl)-2-methoxyestra-1,3,5(10)-
triene-3,17b-diol (16)
Using the same general procedure for 7, obtained 16a-
hydroxymethyl-2-methoxyestra-1,3,5(10)-triene-3,17b-diol
(16) (14 mg, 33%) after purification using Biotage FLASH
apparatus with 95:5 CHCl3–CH3OH as eluent from 12d.
1H NMR (300 MHz, CDCl3) d 6.78 (s, 1H), 6.62 (s, 1H),
3.84 (s, 3H), 3.75 (dd, J = 6,8 Hz, 1H), 3.61 (dd, J = 8.3,
9.2 Hz, 1H), 3.47 (d, J = 7.5 Hz, 1H), 2.80–2.66 m, 2H),
2.36–1.10 (m, 13H), 0.835 (s, 3H). 13C NMR (75 MHz,
CDCl3) d 145.0, 143.8, 132.0, 129.9, 115.0, 108.6, 86.2,
67.1, 56.4, 49.1, 45.9, 44.63, 44.56, 38.9, 37.0, 29.3, 27.7,
5.33. 16b-(n-Propyl)-2-methoxyestra-1,3,5(10)-triene-
3,17b-diol (13)
Using the same general procedure for 7, obtained 13
(111 mg, 51%) from 12a (268 mg, 0.665 mmol). Selected

G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
7535
27.1, 26.9, 12.4. Anal. Calcd for C20H28O41/4H2O: C,
71.28; H, 8.53. Found: C, 71.64; H, 8.42.
hydride (1.65 mmol, 63 mg) in THF using the same
procedure as in 5b. Obtained 463 mg (0.9 mmol, 82%
yield) of 16a/b-ethyl-2-methoxyestra-1,3,5(10)-triene-
17b-ol from 467 mg (1.1 mmol) from the corresponding
5.37. 16b-(Hydroxymethyl)-2-methoxyestra-1,3,5(10)-tri-
ene-3,17b-diol (17)
1
ketone. H NMR (300 MHz, CDCl3) d 7.51–7.24 (m,
5H), 6.87 (s, 1H), 6.65 (s, 1H), 5.13 (s, 2H), 3.88 (s,
3H), 3.81–3.72 (m, isomer) and 3.34–3.27 (m, isomer)
(total 1H, 1:1 ratio), 2.87–2.67 (m, 2H), 2.38–0.86 (m,
17 H), 0.83 (s, isomer) and 0.79 (s, isomer) (total 3H,
1:1 ratio).
Obtained 17 (229 mg, 66%) from 12d (420 mg,
1.0 mmol) after purification using Biotage Flash appara-
tus with 95:5 CHCl3–CH3OH as eluent.1H NMR
(300 MHz, CDCl3) d 6.79 (s, 1H), 6.64 (s, 1H), 3.94
(d, J = 9.8 Hz, 1H), 3.84 (s, 3H), 3.83 (d, J = 9.8 Hz,
1H), 3.65 (dd, J = 4.2, 7.4 Hz, 1H), 2.81–2.72 (m, 2H),
2.56–1.00 (m, 13H), 0.85 (s, 3H).13 C NMR (75 MHz,
CDCl3) d 145.0, 143.8, 132.2, 130.0, 115.0, 108.5,
88.6,56.5, 49.0 and 48.8, 44.7, 44.5, 44.2, 39.0 and
38.7, 38.4, 37.3, 30.5, 29.4, 27.7, 26.9, 22.1 and 22.0,
14.8, 12.3. FT-IR (neat, ATR, cmꢁ1) 3283, 3091, 2919,
1602, 1513, 1445, 1204, 1119, 1013. Anal. Calcd for
C20H28 O4: C, 72.25; H, 8.49. Found: C, 72.24; H, 8.48.
The benzyl protecting group was removed by dissolving
3-benzyoxy-16a/b-ethyl-2-methoxyestra-1,3,5(10)-tri-
ene-17b-ol (463 mg, 0.9 mmol) in THF (20 mL) in a
Parr reactor bottle. The mixture was then charged with
a catalytic amount of Pd(OH)2 purged five times with
H2 gas, then finally pressurized to 50 psi. The mixture
was agitated at rt for 24 h, filtered through Celite. The
filtrate was evaporated under reduced pressure and crys-
tallized with chloroform to give 50 mg final product
1
5.38. 16a-(Methyl)-2-methoxyestra-1,3,5(10)-triene-
3,17b-diol (22)
(0.15 mmol, 17% yield). H NMR (300 MHz, CDCl3)
d 6.81 (s, 1H), 6.66 (s, 1H), 5.44 (br s, 1H), 3.88 (s,
3H), 3.81–3.71 (m, b isomer), 3.34–3.26 (m, a isomer)
(total 1 H, 1:1 isomer ratio), 2.85–2.72 (m, 2H)
2.36–0.89 (m, 17 H), 0.83 (s) and 0.79 (s) (total 3H,
1:1 isomer ratio). Anal. Calcd for C21H30O3: C, 76.33;
H, 9.15. Found: C, 76.18; H, 9.16.
Using the same general procedure for 7, obtained 22
(142 mg, 44%) from 21a (404 mg, 1.01 mmol). Selected
1
spectral data: H NMR (300 MHz, CDCl3) d 6.81 (s,
1H), 6.66 (s, 1H), 3.87 (s, 3H), 3.23 (d, J = 7 Hz) (s,
13
1H), 0.81 (s, 3 H).
C NMR (75 MHz, CDCl3) d
145.0, 143.8, 132.2, 130.0, 115.0, 108.5, 90.1, 56.5,
48.6. 44.7, 39.0, 38.8, 37.2, 32.4, 29.4, 27.7, 26.9, 20.8,
12.2. Anal. Calcd for C20 H28O31/4H2O: C, 74.85; H,
8.95. Found: C, 74.98; H, 8.65.
6. Biology
6.1. Proliferation assays
5.39. 16a-(Ethyl)-2-methoxyestra-1,3,5(10)-triene-3,17b-
diol (23)
MDA-MB-231 human breast carcinoma cells were
grown in DMEM containing 10% fetal bovine serum
(FBS) (Hyclone Laboratories, Logan, UT) supple-
mented with 2 mM L-glutamine, 100 units/ml penicillin,
and 100 lg/ml streptomycin (Irvine Scientific, Santa
Anna, CA). Proliferation was assessed by detection of
DNA synthesis by use of the 5-bromo-20-deoxyuridine
(BrdU) cell proliferation colorimetric ELISA kit from
Roche (Indianapolis, IN) according to the manufac-
turer’s instructions. For BrdU assays, the cells were
seeded at 2000 cells/well in a 96-well plate, allowed to
attach overnight then exposed to increasing concentra-
tions of the different compounds for 48 h. Each condi-
tion was assessed in triplicate and the experiments
were carried out a minimum of two times. Results in
Table 1 are means of the data from each independent
experiment SD.
Using the same general procedure for 7, obtained 23
(156 mg, 24%) from 21b (873 mg, 2.09 mmol). Selected
1
spectral data: H NMR (300 MHz, CDCl3) d 6.79 (s,
1H), 6.62 (s, 1H), 3.85 (s, 3H), 3.26 (d, J = 7.2 Hz,
1H), 2.80–2.72 (m, 2H), 2.32–0.83 (m, 17H), 0.81 (s,
3H). 13C NMR (75 MHz, CDCl3) d 145.1, 143.8,
132.2, 129.9, 115.1, 108.7, 88.0, 56.4, 48.7, 45.9, 44.7,
44.5, 39.0, 37.2, 30.0, 29.3, 28.6, 27.7, 26.9, 13.1, 12.3
Anal. Calcd for C21H30O3: C, 76.33; H, 9.15. Found:
C, 76.55; H, 9.21.
5.40. 16a/b-(Ethyl)-2-methoxyestra-1,3,5(10)-triene-
3,17b-diol (27)
Using the same general procedure as for the preparation
of 4b. Obtain 467 mg (1.1 mmol, 38% yield) of 3-benzyl-
oxy-16a/b-ethyl-2-methoxyestra-1,3,5(10)-triene- 17-one
from ethyl iodide (700 mg, 4.4 mmol, 1.5 eq) and
1.1315 g (2.9 mmol) of 4b. 1H NMR (300 MHz, CDCl3)
d 7.49–7.30 (m, 5H), 6.86 (s, 1H), 6.66 (s, 1H), 5.13 (s,
2H), 3.88 (s, 3H), 2.86ꢁ2.75 (m, 2H), 2.46–1.24 (m,
17H), 0.97 (s) and 0.88 (s) (total 3H, 1:1 isomer ratio).
Di substituted product (306 mg, 0.68 mmol) was also
isolated.
HUVEC were grown in EGM (Clonetics). HUVEC
were seeded at 5000 cells/well in 96-well plates. After
being allowed to attach overnight, cells were then
treated with increasing concentrations of drug for 48 h
at 37 ꢂC. The preparation of the drugs and BrdU prolif-
eration assay were performed as indicated above.
MCF-7, an estrogen-dependent breast carcinoma cell
line, was maintained in DMEM/F12 (1:1) containing
10% (v/v) FBS (Hyclone Laboratories, Logan, UT)
and 1X antibiotic–antimycotic. MCF-7 cells, which were
used between passage 60 and passage 90, were the kind
Racemic 3-benzyloxy-16a/b-ethyl-2-methoxyestra-1,3,
5(10)-triene-17-one was reduced with lithium aluminum

7536
G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
gift of Dr. Dorraya El Ashry of Georgetown University.
For MCF-7 estrogen-dependent proliferation assays, the
cells were seeded in complete media at 20–30,000 cells/
well in 24-well plates. After allowing the cells to adhere
overnight, the seeding density was determined by cell
counts. Cells were then washed with PBS (37 ꢂC) and
starved by placing them in IMEM–phenol red free
media containing 2% charcoal–dextran fetal bovine-
stripped serum (Georgetown University) and 1X antibi-
otic–antimitotic. After 3 days of starvation, cells were
treated with or without increasing concentrations of
compounds, replacing the media every 2–3 days with
the same stripped serum medium with test compounds,
and counted after 8 or 10 days of treatment. Proliferation
was measured by cell counting using a Coulter Z1 cell
counter (Coulter Corporation, Hialeah, FL). Each
zcondition was done in triplicate in at least two
independent experiments. Results are presented as the
stimulation index (SI) relative to 2-methoxyestradiol
(defined as 1.00).
6.3. Sulfonyl transferase assays
Incubations with human liver cytosol were performed in
50 mM Tris buffer, pH 7.5 containing 0.1 mM 30-phos-
pho-adenosyl-50-phosphosulfate (PAPS) and 5 lM of
each test substance. During sample workup, the protein
was precipitated by acetonitrile containing an internal
standard (2-(20,20,20-trifluoroethoxy)estradiol) and the
supernatant was directly injected into an LC/MS system
at 0, 15, 30, and 60 min.
In the control samples lacking PAPS, incubations were
performed in 50 mM Tris buffer, pH 7.5, containing hu-
man liver cytosol and 5 lM of each test substance. Sam-
ples were worked up as above at 0 and 60 min.
For control samples lacking human liver cytosol, incu-
bations were performed in 50 mM Tris buffer, pH 7.5,
with 0.1 mM PAPS and 5 l M of each test substance.
Sample workup was performed as described above at 0
and 60 min.
6.2. Glucuronidation conjugation assays
A positive control used 7-hydroxycoumarin for human
liver cytosolic-mediated loss of parent compound medi-
ated by sulfonyl transferase. The substrate concentra-
tion was 5 lM.
Incubation with human liver microsomes was preformed
with 50 mM Tris, pH 7.5 containing 10 mM magnesium
chloride, 2 mM Uridine-50-dihydrophosphoglucuronic
acid (UDPGA), 25 lg alamethicin/mg liver microsomal
protein and 5 lM of each test substance. The sample
was worked up by precipitation of protein by acetoni-
trile containing internal standard (2-(20,20,20-trifluoro-
ethoxy)estradiol) and direct injection of the
supernatant onto an LC/MS at 0, 15, 30, and 60 min
(details below).
Analytical method parameters for HPLC/MS: analysis
of samples used a C18 Waters Symmetry column
2.1 · 150 mm with water and methanol as mobile phase.
The flow rate was 0.2 ml/min, 20 lL injection volume,
and the flow gradient initiated with 50% water/50%
methanol, over 5 min gradient changed to 100% metha-
nol, was held for 2 min, then returned to initial condi-
tions over 2 min. The mass spectrometer was a
Micromass Quattro LC and was operated in the nega-
tive ion mode using ESI. Selected ions were as follows:
2-(20,20,20-trifluorethoxy)estradiol was m/z 330, for 2-
methoxyestradiol m/z 286, and 16-ethyl-2-methoxyestra-
diol m/z 314.
The control incubations were as follows: For samples
lacking UDPGA, incubations were performed in
50 mM Tris, pH 7.5, containing 10 mM magnesium
chloride, 25 lg alamethicin/mg liver microsomal pro-
tein, 0.5 mg/ml human liver microsomes, and 5 lM of
each test substance. Sample workup was performed as
above at 0 and 60 min. For control samples lacking liver
microsomes, incubations were performed in 50 mM
Tris, pH 7.5, containing 10 mM magnesium chloride,
alamethicin (same amount as in other control sample),
2 mM UDPGA, 0.5 mg/ml BSA and 5 lM of each test
substance. Sample workup was performed as above at
0 and 60 min.
Acknowledgment
The authors thank Mr. Christopher Herbstritt for
expert technical assistance.
References and notes
A comparator (estradiol) for liver microsomal-mediated
loss of parent was tested at 1, 10, 30 and 60 min time
points similar to that for the test substances. The sub-
strate concentration was 5 lM.
1. Pribluda, V. S.; Gubish, E. R.; LaVallee, T. M.; Treston,
A.; Swartz, G. M.; Green, S. J. Cancer & Metastasis Rev.
2000, 19, 173–179.
2. Mooberry, S. L. Drug Resist. Updat 2003, 6, 355.
3. Hamel, E.; Lin, C. M.; Flynn, E.; D’Amato, R. J.
Biochemistry 1996, 35, 1304–1310.
4. D’Amato, R. J.; Lin, C. M.; Flynn, E.; Folkman, J.;
Hamel, E. Proc. Natl. Acad. Sci. 1994, 91, 3964–3969.
5. Attalla, H.; Westberg, J. A.; Andersson, L. C.; Adlercre-
utz, H.; Makela, T. P. Biochem. Biophys. Res. Commun.
1998, 247, 616–619.
6. Mabjeesh, N. J.; Escuin, D.; LaVallee, T. M.; Pribluda, V.
S.; Swartz, M. G.; Johnson, M. S.; Willard, M. T.; Zhong,
H.; Simmons, J. W.; Giannakakou, P. Cancer Cell 2003, 3,
363.
Analytical method parameters for HPLC/MS: analysis
of samples used C18 Waters Symmetry column
2.1 mm · 50 mm with water and methanol as mobile
phase. The flow rate was 2.0 mL/min and the flow gradi-
ent initiated with 50% water, 50% methanol. Over 5 min,
gradient changed to 100% methanol which was held for
2 min. The gradient was changed back to initial condi-
tions over 1.5 min and was held for another 2 min.
The mass spectrometer was a Micromass Quattro LC
operated in the negative ion mode using ESI.

G. E. Agoston et al. / Bioorg. Med. Chem. 15 (2007) 7524–7537
7537
7. Huang, P.; Feng, L.; Oldham, E. A.; Keating, M. J.;
Plunkett, W. Nature 2000, 407, 390–395.
8. LaVallee, T. M.; Zhan, X. H.; Herbstritt, C. J.; Kough, E.
C.; Green, S. J.; Pribluda, V. S. Cancer Res. 2002, 62,
3691–3697.
26. (a) For data in humans: Wilding, G.; Sweeney, C.; King,
D. M.; Horvath, D.; Staab, M. J.; Fife, K.; Armstrong, V.;
Soriano, J.; Pribluda, V. S.; Gubish, E. R.; Sidor, C. 38th
Annual Meeting, American Society of Clinical Oncology,
May 2002, Orlando, FL; (b) Sledge, G. W.; Miller, K. D.;
Haney, L. G.; Nguyen, D. D.; Storniole, A. M.; Philips, E.
N.; Pribluda, V. S.; Gubish, E. R. May 2002, Orlando, FL;
(c) Miller, K. D.; Murray, D. J.; Curry, E.; Haney, L. G.;
McClaskey, G. W; Storniolo, A. M.; Philips, E.; Pribluda,
V. S.; Gubish, E. R. 38th Annual Meeting, American
Society of Clinical Oncology, May 2002, Orlando, FL.
For data in rats, Unpublished data, EntreMed, Inc.
27. Lakani, N. J.; Sparreboom, A.; Xu, X.; Veestra, T. D.;
Venitz, J.; Dahut, W. L.; Figg, W. D. J. Pharm. Sci. 2007,
96, 1821.
28. Newman, S.; Ireson, C. R.; Tutill, H. J.; Parsons, J. M.;
Leese, M. F. C.; Potter, M. P.; Potter, B. V. L.; Reed, M.
J.; Purohit, M. J. Cancer Res. 2006, 66, 324.
29. Mancuso, A. J.; Huang, S.-L.; Swern, D. J. Org. Chem.
1978, 43, 2480.
30. Fevig, T. L.; Kazenellenbogen, J. A. J. Org. Chem. 1987,
52, 247–251.
31. Tremblay, M. R.; Auger, S.; Poirier, D. Bioorg. Med.
Chem. 1995, 3, 505–523.
9. Mukhopadhyay, T.; Roth, J. A. Ongogene 1997, 14, 379–
384.
10. Seegers, J. C.; Lottering, M. L.; Grobler, C. J.; van
Papendorp, D. H.; Habbersett, R. C.; Shou, Y.; Lehnert,
B. E. J. Steroid Biochem. Mol. Biol. 1997, 62, 253–267.
11. Schumacher, G.; Kataoka, M.; Roth, J. A.; Mukhopad-
hyay, T. Clin. Cancer Res. 1999, 5, 439–493.
12. LaVallee, T. M.; Hembrough, W. A.; Papathanassiu, A.;
Pribluda, V. S.; Swartz, G.; Willams, M.; Grene, S. J.
AACR-NCI-EORTC International Conferences 1999, 17.
13. Semenza, G. L. Nat. Rev. Cancer 2003, 3, 721.
14. Calzada, M. J.; Del Peso, L. Cin. Transl. Oncol. 2007, 9,
278.
15. Cushman, M.; He, M.-H.; Katzenellenbogen, J. A.; Lin,
C. M.; Hamel, E. J. Med. Chem. 1995, 38, 2041–2049.
16. Cushman, M.; He, M.-H.; Katzenellenbogen, J. A.;
Varma, R. K.; Hamel, E.; Lin, C. M.; Ram, S.; Sachdeva,
Y. P. J. Med. Chem. 1997, 40, 2323–2334.
17. Miller, T. A.; Bulman, A. L.; Thompson, C. D.;
Garst, M. E.; Macdonald, T. L. J. Med. Chem. 1997,
40, 3836–3841.
18. Miller, T. A.; Bulman, A. L.; Thompson, C. D.; Garst, M.
E.; Macdonald, T. L. Biooorg. Med. Chem. Lett. 1997, 7,
1851–1856.
32. Attempted alkylation of 3-benzyl-2-methoxyestrone using
LDA as base at –78 ꢂC and subsequent quench with ethyl
iodide yielded 38% of racemic 16-ethyl-3-benzyl-2-
methoxyestrone and 23% of 16-(bisethyl)-3-benzyl-2-
methoxyestrone.
19. Wang, Z.; Yang, D.; Mohanakrishnan, A. K.; Fanwick, P.
E.; Nampoothiri, P.; Hamel, E.; Cushman, M. J. Med.
Chem. 2000, 43, 2419–2429.
33. Krapcho, A. P.; Weimaster, J. F.; Eldrich, J. M.; Jahngen,
E. G. E., Jr.; Lovey, A. J.; Stephens, W. P. J. Org. Chem.
1978, 43, 138–147.
20. Hughes, R. A.; Harris, T.; Altman, E.; Mccallister, D.;
Vlahos, R.; Robertson, A.; Cushman, M.; Wang, Z.;
Stweart, A. G. Mol. Pharmacol. 2002, 61, 1053.
21. Cushman, M.; Mohanakrishnan, A. K.; Hollingshead, M.;
Hamel, E. J. Med. Chem. 2002, 45, 4748.
22. Edsall, A. B.; Mohanakrishhnan, A. K.; Yang, D.;
Fanwick, P. E.; Hamel, E.; Hanson, A. D.; Agoston, G.
E.; Cushman, M. J. Med. Chem. 2004, 47, 5126.
23. Rao, P. N.; Cessac, J. W.; Tinley, T. L.; Mooberry, S. L.
Steroids 2002, 67, 1079.
34. Tremblay, M. R.; Auger, S.; Poirier, D. Synth. Met. 1995,
25, 2483–2495.
35. Goto, G.; Yoshioka, K.; Higara, K.; Miki, T. Chem.
Pharm. Bull. 1977, 25, 1295–1301.
36. Corey, E. J.; Enders, D. Tetrahedron Lett. 1976, 41, 3–6.
37. Corey, E. J.; Knapp, S. Tetrahedron Lett. 1976, 41, 3667–
3668.
38. Neef, G.; Eder, U; Haffer, G.; Sauer, G.; Nickolson, R.
DE Patent 2757157, 1977.
39. Neef, G.; Eder, U.; Weichert, R. J. J. Org. Chem. 1978, 43,
4679–4680.
40. Gonzalez, F. B.; Neef, G.; Eder, U.; Weichert, R.;
Schillinger, E.; Nishino, Y. Steroids 1982, 40, 171–187.
41. Unpublished data, EntreMed, Inc.
24. Santner, S. J.; Santen, R. J. J. Steroid Biochem. Mol. Biol.
1993, 45, 383–390.
25. Helton, E. D.; Williams, M. C.; Goldzieher, J. W. Steroids
1976, 27, 851–867.