7α-iodo and 7α-fluoro steroids as androgen receptor-mediated imaging agents

Article abstract of DOI:10.1021/jm990064o

We have synthesized several 7α-fluoro (F) and 7α-iodo (I) analogues of 5α-dihydrotestosterone (5α-DHT) and 19-nor-5α-dihydrotestosterone (5α- NDHT) and tested them for binding to the androgen receptor and for their biological activity in an in vitro assay

Full text of DOI:10.1021/jm990064o

J . Med. Chem. 1999, 42, 2021-2034
2021
7r-Iod o a n d 7r-F lu or o Ster oid s a s An d r ogen Recep tor -Med ia ted Im a gin g Agen ts
David C. Labaree,† Robert M. Hoyte,‡ Lynne V. Nazareth,§ Nancy L. Weigel,§ and Richard B. Hochberg*,†
Department of Obstetrics and Gynecology, Yale University School of Medicine, New Haven, Connecticut 06510,
Department of Chemistry, State University of New York, Old Westbury, New York 11568, and Department of Cell Biology,
Baylor College of Medicine, Houston, Texas 77030
Received February 5, 1999
We have synthesized several 7R-fluoro (F) and 7R-iodo (I) analogues of 5R-dihydrotestosterone
(5R-DHT) and 19-nor-5R-dihydrotestosterone (5R-NDHT) and tested them for binding to the
androgen receptor and for their biological activity in an in vitro assay with cells that have
been engineered to respond to androgens. The relative binding affinity to the androgen receptor
determined in competition assays showed that in the androstane series the fluoro steroids
have the highest affinity and that F-17R-CH3-DHT (4) has a higher affinity than 5R-DHT. All
other steroids were somewhat less potent than 5R-DHT with F-DHT (2) ) I-17R-CH3-DHT (3)
g F-NDHT (6) > F-17R-CH3-NDHT (8) ) I-DHT (1) g I-NDHT (5) > I-17R-CH3-NDHT (7).
The relative biological activity in cells transfected with the androgen receptor and an androgen
responsive reporter gene is 4 . 5R-DHT > 2 > 6 > 3 g 1 g 8 g 5 > 7. The iodinated compound,
I-17R-CH3-DHT (3), with the highest binding activity was synthesized labeled with 125I and
was shown to bind with high affinity, Ka ) 1.9 × 1010 L/mol, and low nonspecific binding to the
androgen receptor in rat prostatic cytosol. However, when radiolabeled [125I]-17R-CH3-DHT
([125I]3) was injected into castrated male rats, it showed very poor androgen receptor-mediated
uptake into the rat prostate. This was unexpected in light of its superior receptor binding
properties and its protection by the 17R-methyl group from metabolic oxidation at C-17.
However, the biological potency of I-17R-CH3-DHT (3) was not as high as would have been
expected. When I-DHT (1) and I-17R-CH3-DHT (3) were incubated in aqueous media at 37 °C
they rapidly decomposed, but they were stable at 0 °C. The fluorinated analogue 4 treated
similarly at 37 °C was completely stable. The products of the decomposition reaction of I-DHT
(1) at 37 °C were identified as iodide and principally 17â-hydroxy-5R-androst-7-en-3-one. The
temperature dependence of this elimination reaction explains the inconsistency between the
high binding to the androgen receptor (measured at 0 °C) and the low biological activity, as
well as the poor androgen receptor mediated concentration in vivo. The fluorinated analogue
F-17R-CH3-DHT (4) has both high affinity for the androgen receptor and high stability in
aqueous media. Of the compounds tested, 4 has the highest affinity for the androgen receptor
as well as the highest androgenic activity. Thus it is likely that F-17R-CH3-DHT 4 labeled
with 18F will be an excellent receptor-mediated diagnostic imaging agent.
In tr od u ction
shown to bind with high affinity and specificity to the
androgen receptor.9-11 Several have very good androgen
target organ (prostate) uptake in vivo. To the contrary,
synthesis of androgens with radioiodine for single-
photon emission computed tomography (SPECT) has
met with only limited success. It has been noted that
this has been a difficult problem because the androgen
receptor has a more specific binding region which
discriminates to a much greater extent than the estro-
gen receptor.12
Although androgens and estrogens with obvious
exceptions share many of the same characteristics
necessary for high-affinity binding to their receptors,
the strategies that produced excellent iodinated ligands
for the estrogen receptor did not succeed with andro-
gens. For example, while C-16R radioiodine-labeled
estrogens have excellent receptor binding characteris-
tics, 5R-dihydrotestosterone13 as well as testosterone14
labeled with iodine at 16R bind very poorly to the
androgen receptor. However, as noted above, these same
compounds when substituted at C-16 with fluorine are
excellent ligands. Likewise, the 17R-iodovinyl group
leads to excellent ligands for the estrogen receptor4 as
The design and synthesis of ligands for steroid recep-
tors that are labeled with various radioactive halogens
has led to important imaging agents as well as probes
for the study of hormone action. Different strategies
have produced estrogens and progestins substituted
with radioactive halogens at various positions in the
steroid nucleus or at short side chains. Some haloge-
nated estrogens1-4 and progestins5-8 have been shown
to bind with high affinity to their receptors and con-
centrate in steroid target tissues in a receptor-depend-
ent manner, and thus they have the potential to act as
imaging agents for the detection of hormone responsive
cancer. For this reason, isotopically labeled androgens
have been sought to detect hormone responsive prostate
tumors and metastases. Androgens labeled with 18F at
C-11â, C-16R, C-16â, or C-20 (at a 17R-methyl substitu-
ent) for positron emission tomography (PET) have been
* Address correspondence to: Richard B. Hochberg. Tel: (203) 785-
4001. E-mail: richard.hochberg@yale.edu.
† Yale University School of Medicine.
‡ State University of New York.
§ Baylor College of Medicine.
10.1021/jm990064o CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/14/1999

2022 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
Labaree et al.
F igu r e 1. Structure of 7R-halosteroids 1-8.
Sch em e 1a
well as the progesterone receptor,5,6 but this same
group, (E- or Z- 2′-[125I]iodovinyl) linked to 5R-dihy-
drotestosterone leads to poor ligands for the androgen
receptor that do not concentrate in androgen target
organs in vivo.15a Similarly, 7R-methyl-17R-(E-2′-[125I]-
iodovinyl)-19-nortestosterone, an analogue of mibole-
rone, binds with high affinity to the androgen receptor,15b
but it and the Z-isomer concentrated only slightly in
androgen target organs in vivo.8 While, in general, the
major problem in obtaining an androgen receptor-
mediated SPECT ligand has been low binding affinity,
some have shown good binding to the androgen receptor
but none have been useful in vivo.
a
(a) nBu4NF, MEK, 90 °C.
Sch em e 2a
More recently we have utilized the 7R-postion as a
site for radioiodine substitution in order to synthesize
a SPECT agent that could be labeled with 123I.16,17 We
chose substitution at C-7 because androgenic action is
potentiated in steroids substituted with a methyl group
at the 7R-position.18 In addition, C-7 is sufficiently
distant from C-3 and C-17 that halogen substitution has
minimal steric or stereoelectronic effects which might
interfere with the important interactions between the
receptor and the carbonyl and hydroxyl groups of the
steroid. We had synthesized 7R-iodo-5R-dihydrotes-
tosterone (I-DHT) and found that it bound with high
affinity and specificity to the androgen receptor. Nev-
ertheless, this iodinated analogue concentrated only
poorly in target organs. These results suggested that
I-DHT was rapidly catabolized producing inactive me-
tabolites in vivo that were not capable of androgen
receptor-mediated localization and that other analogues
of 7R-halogenated steroids might be superior imaging
agents. Herein, we describe experiments in which we
synthesized several 7R-iodo and 7R-fluoro analogues of
5R-DHT and 19-nor-5R-DHT (estranes), some of which
contain a 17R-methyl group designed to protect them
from oxidation of the 17â-hydroxyl (Figure 1). We tested
these steroids as ligands for the androgen receptor and
for their androgenic activity. The most active iodinated
analogue, 7R-iodo-17R-methyl-5R-dihydrotestosterone
(3), was synthesized labeled with radioiodine, [125I]3,
and we determined its binding to the androgen receptor
in vitro and its organ distribution in vivo.
Ch em istr y
a
(a) Ethylene glycol, pTsOH, benzene, reflux (17R-methyltes-
tosterone f 10); (b) CrO3-3,5-dimethylpyrazole, CH2Cl2, -23 °C
(10 f 11); (c) 10% Pd/C, H2, MeOH (11 f 12); (d) NaBH4-CeCl3,
THF-H2O-MeOH (12 f 13a ,b); (e) pTsOH, acetone-H2O, (13a ,b
f 14a ,b); (f) pTsCl, pyridine, 4 °C (14b f 15); (g) NaI, CH3CN,
83 °C (15 f 3); (h) nBu4NF, MEK, 85 °C (15 f 4).
F-DHT (2) was prepared (Scheme 1) by fluorination
of tosylate 9 (obtained as previously described17) with
nBu4NF in MEK. The 17R-CH3 halosteroids 3 and 4
were synthesized from 17R-methyltestosterone as shown
in Scheme 2. Ketalization of 17R-methyltestosterone

7R-Iodo and 7R-Fluoro Steroids
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11 2023
Sch em e 3a
a
(a) TBDPSCl, imidazole, DMF (19-hydroxyandrostenedione f 16); (b) pTsOH, 2-ethyl-2-methyl-1,3-dioxolane, 83 °C (16 f 17); (c)
NaBH4, THF-H2O (17 f 18); (d) Ac2O, pyridine (18 f 19); (e) CrO3-3,5-dimethylpyrazole, CH2Cl2, -23 °C (19 f 20); (f) NaBH4-CeCl3,
THF-MeOH (20 f 21); (g) nBu4NF, THF (21 f 22); (h) CrO3-pyridine, CH2Cl2 (22 f 23); (i) KOH, MeOH, 4 °C (23 f 24-26).
followed by allylic oxidation of the resulting ∆5(6)-ketal
with CrO3-3,5-dimethylpyrazole complex19 gave enone
11. Hydrogenation followed by reduction of the 7-ketone
gave a 1:6 mixture of 7R- and 7â-epimeric C-7 alcohols.
After removal of the C-3 ketal with pTsOH/acetone, the
desired 7â-alcohol 14b was isolated by flash chroma-
tography. The 7â-tosylate 15 was prepared from 14b
by reaction with pTsCl in pyridine.20 The halosteroids
3 and 4 were synthesized by exchange with NaI or nBu4-
NF, respectively.
had to be reduced to the allylic alcohol 21 with NaBH4-
CeCl3.22 The silyl group was removed with nBu4NF
forming 22. Oxidation of both the 7- and 19-alcohols
with CrO3-pyridine gave 23. Deformylation with meth-
anolic KOH produced the desired enone 24 and the ∆5(10)
isomer 25 as an inseparable mixture as well as the
dienol ether 26. Attempts to isomerize 25 to 24 with
continued base treatment led to the formation of 26.
Hydrogenation of the mixture of 24 and 25 led to a
further mixture of saturated ketones (uncharacterized).
However, this route produced a standard of 24 which
was essential in the identification of this intermediate
in the complex reaction mixture produced in the allylic
oxidation reaction of the next scheme.
The second route (Scheme 4) to 5-8 involves the
allylic oxidation of 19-nortestosterone 3-ketal. Although
a complex mixture of oxidation products was obtained,
hence a low yield, the desired enone 24 was obtained
cleanly and from an inexpensive starting material.
Ketalization of 19-nortestosterone acetate gave a mix-
ture of the ∆5(6)- and ∆5(10)-isomers in 1/1.6 ratio. Allylic
oxidation of this mixture using CrO3-3,5-dimethylpyra-
zole complex gave a mixture of several different oxida-
tion products. Purification by flash chromatography in
two different systems followed by crystallization gave
24 in an 8% yield. Hydrogenation of 24 with 10% Pd/C
gave 29, and reduction with NaBH4-CeCl3 gave a 1/3.5
Two different routes were attempted for the synthesis
of the 19-norsteroids 5-8. In the first route (Scheme 3)
the commercially available 19-hydroxyandrost-4-ene-3,-
17-dione was used as starting material in order to direct
isomerization of the ∆4,5-double bond to ∆5,6 during the
ketalization reaction and avoid formation of the ∆5,10
-
steroid. Ordinarily, when a C-10 methyl group is
present, as it is in methyltestosterone (Scheme 2), only
the ∆5,6-double bond is obtained. The advantage of the
C-10 hydroxymethyl group is that it can be facilely
removed after conversion to an aldehyde.21 First the 19-
hydroxyl group was protected as the tert-butyldiphe-
nylsilyl ether 16, the 3-ketone was protected as the ketal
17, and the 17-ketone was reduced with NaBH4 to form
the 17â-alcohol 18. Acetylation of 18 led to 19, followed
by allylic oxidation gave 20. To remove the silyl group
without ketal opening to the dienol ether, the C-7 ketone

2024 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
Labaree et al.
Sch em e 4a
a
(a) pTsOH, 2-ethyl-2-methyl-1,3-dioxolane, 83 °C (19-nortestosterone acetate f 27, 28); (b) CrO3-3,5-dimethylpyrazole, CH2Cl2, -23
°C (27 f 24); (c) 10% Pd/C, H2, MeOH (24 f 29); (d) NaBH4-CeCl3, THF-H2O-MeOH (29 f 30a ,b); (e) pTsOH, acetone-H2O (30a ,b
f 34); (f) pTsCl, pyridine, 4 °C (34 f 35); (g) NaI, CH3CN, 83 °C (35 f 36); (h) NaCO3, MeOH-H2O (36 f 5); (i) nBu4NF, MEK, 85 °C
(35 f 37); (j) NaCO3, MeOH-H2O (37 f 6).
mixture of 7R- and 7â-epimeric C-7 alcohols 30a , 30b.
(Proof that the hydrogenation yielded the 5R-steroid was
obtained by converting 30b into the known 17â-hydroxy-
5R-estran-3-one (33) by tosylation, reductive cleavage
with LiEt3BH, and deketalization.) The mixture of 30a ,
30b was deketalized, and the resulting 3-ketones were
purified by chromatography yielding the pure 7â-alcohol
34. Tosylation followed by exchange with NaI or nBu4-
NF and deprotection produced 5 and 6, respectively.
The 17R-methyl 19-norsteroids 7 and 8 were synthe-
sized from the mixture of 30a , 30b as follows (Scheme
5): protection of the hydroxyl group as the benzyloxy-
methyl (BOM) ether 38, deprotection of the 17â-hy-
droxyl group 39, and oxidation with CrO3-pyridine
yielded the 17-ketone 40. Reaction with CH3MgI gave
the 17R-methylsteroid 41. Deprotection of the BOM
ether by hydrogenolysis followed by deketalization
yielded 43. Tosylation followed by exchange with NaI
or nBu4NF produced 7 and 8, respectively.
showed that one of these compounds, 1, is an excellent
ligand, binding with high affinity and specificity to the
androgen receptor.16,17 The 5R-reduced structure was
chosen because it was analogous to the natural andro-
gen, 5R-DHT. We also attempted to synthesize the
corresponding 7R-iodinated ring A unsaturated ∆4-3-
ketones, analogous to testosterone, using a scheme
similar to that described for the 5R-reduced steroids.
Thus, we developed a method for the stereospecific
synthesis of the required precursor, 7â-hydroxytest-
osterone.23 While we could convert this alcohol to the
tosylate, attempted halogen displacement of the latter
invariably led to elimination (not described). In all, eight
iodinated or fluorinated analogues of 5R-DHT and 19-
nor-5R-DHT (Figure 1) were synthesized and tested for
binding to the androgen receptor in rat prostate cytosol
by competition for the binding of [3H]R1881. As can be
seen in Table 1, all of the 7R-halogenated steroids
showed fairly good binding to the androgen receptor. Of
the iodinated series, both of the 19-nor compounds 5
(47%) and 7 (25%) had the lowest RBAs. The fluoro 19-
norsteroids 6 (63%) and 8 (55%) had a somewhat higher
affinity for the receptor. Of the 5R-DHT analogues,
Resu lts a n d Discu ssion
We have synthesized several different steroids (Figure
1) with either iodine or fluorine at C-7R. Previously, we

7R-Iodo and 7R-Fluoro Steroids
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11 2025
Sch em e 5a
similar manner, the 17R-methyl fluoro analogue, F-17R-
CH3-DHT (4) (123%), was a stronger competitor than
F-DHT (2) (78%). Fluoro analogue 4 proved to be the
highest affinity ligand of all the compounds tested.
To better understand the biological activity of these
halogenated androgens, their androgenic potency was
measured in monkey kidney CVI cells that were tran-
siently transfected with genes for the androgen receptor
and an androgen responsive reporter.24 In contrast to
the receptor experiments, the cells are metabolically
active. Consequently, the steroids can be metabolized,
and we expected that the protection afforded by the 17R-
methyl group would be markedly apparent. Thus, we
anticipated a large disparity between the androgen
receptor competition experiments and the androgen
response assay when comparing the same compounds
with and without the 17R-methyl group, but this was
only found for the fluoro androstane pair 4 (418%) and
2 (60%) (Table 1). The opposite was found for the fluoro
estranes, where 8 (16%) was weaker than 6 (44%). In
the iodo series the 17R-methyl group appeared not to
confer an advantage in androgenic activity: in the
androstane series, 3 (28%) versus 1 (20%), and in the
estrane series, 7 (8%) versus 5 (13%).
An analysis of the androgen receptor binding data and
the androgenic potency data allowed for the formulation
of some structure-activity correlations. Considering
that the order of decreasing receptor binding affinity
(Table 1) of the steroids was F-17R-CH3-DHT (4) > 5R-
DHT > F-DHT (2) ) I-17R-CH3-DHT (3) g F-NDHT (6)
> F-17R-CH3-NDHT (8) ) I-DHT (1) g I-NDHT (5) >
I-17R-CH3-DHT (7), the following correlations were
noted. (1) All fluoro steroids have higher affinity than
their corresponding iodo analogue. (2) For both fluoro
and iodo steroids in the androstane series, the presence
of a 17R-methyl group increased affinity, while in the
estrane series this group decreased receptor affinity. (3)
For both fluoro and iodo steroids with a 17R-methyl
group, androstanes were better ligands than estranes,
whereas for compounds without a 17R-methyl group,
androstanes and estranes had similar binding affinity.
In the biological assay with the CV1 cells only the fluoro
steroids can be compared (discussed below). The order
of decreasing stimulatory activity was F-17R-CH3-DHT
(4) . 5R-DHT > F-DHT (2) > F-NDHT (6) > F-17R-
CH3-NDHT (8). Overall, the same structure-activity
trends were observed in the bioassay as in the receptor
binding assay. Of all the steroids tested only F-CH3-
DHT (4) was significantly more potent than 5R-DHT.
a
(a) BOMCl, iPr2EtN, CH2Cl2 (30b f 38); (b) 1% KOH-MeOH,
50 °C (38 f 39); (c) CrO3-pyridine, CH2Cl2 (39 f 40); (d) CH3MgI,
THF (40 f 41); (e) 10% Pd/C, H2, THF-H2O (41 f 42); (f) pTsOH,
acetone-H2O (42 f 43); (g) pTsCl, pyridine, 4 °C (43 f 44); (h)
NaI, CH3CN, 83 °C (44 f 7); (i) nBu4NF, MEK, 85 °C (44 f 8).
Ta ble 1. Relative Binding Affinity (RBA) to the Androgen
Receptor and Relative Androgenic Activity (RSA)a
compounds
RBA (mean ( SD) RSA (mean ( SD)
100 100
54 ( 8
20 ( 8b
5R-DHT
I-DHT (1)
F-DHT (2)
I-17R-CH3-DHT (3)
F-17R-CH3-DHT (4)
I-NDHT (5)
F-NDHT (6)
I-17R-CH3-NDHT (7)
F-17R-CH3-NDHT (8)
78 ( 15
71 ( 6
60 ( 13
28 ( 11b
418 ( 272
13 ( 4b
44 ( 3
123 ( 29
47 ( 5
63 ( 7
25 ( 4
8 ( 4b
55 ( 7
16 ( 7
Since 3 was the most potent of the iodinated steroids,
we synthesized it labeled with 125I, to be able to measure
directly the binding of [125I]3 to the androgen receptor.
Varying concentrations of [125I]3 were incubated over-
night at ∼0 °C with cytosol from rat prostate gland. The
binding of the radiolabeled steroid in the presence or
absence of saturating amounts of 5R-DHT to determine
nonspecific binding is shown in Figure 2. As can be seen,
saturation of binding occurred at very low concentra-
tions of the steroid, and nonspecific binding was low.
Scatchard analysis (Figure 2, inset) indicated that 3
binds to the androgen receptor with high affinity, Kd )
0.7 nM. Comparison of the saturation analysis with that
for [3H]5R-DHT indicated that the [125I]3 was carrier-
free, approximately 2200 Ci/mmol. The characteristics
a
Competition binding analysis: relative binding affinity (RBA)
( SD, compared to 5R-DHT were determined in incubations of
rat prostate cytosol at 0-2 °C for 20 h with [3H]R1881. In vitro
androgen bioassay: relative stimulatory activity (RSA) ( SEM
compared to 5R-DHT was determined with monkey kidney CV1
cells transiently transfected with genes for the androgen receptor
and an androgen responsive reporter. Cells were treated with
various concentrations of steroids and incubated for 24 h at 37
b
°C. The RSA values for the iodinated steroids 1, 3, 5, and 7, do
not represent the biological potency of the specified steroids; rather
it is probable that the measured potencies are predominantly those
of elimination products formed by dehydrohalogenation (see text).
I-17R-CH3-DHT (3) (71%) had a higher RBA than I-DHT
(1) (54%), the original 7R-iodo steroid that we synthe-
sized as a ligand for the androgen receptor.16,17 In a

2026 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
Labaree et al.
and the radioactivity was counted. The amount of radio-
activity per weight of each tissue was calculated and
compared to the prostate. As can be seen in Table 2,
there was a somewhat greater amount of 125I in the
prostate of the animals injected with [125I]3 at 1 h, 0.39,
compared to the androgen-blocked animals 0.23 (%
injected dose/g of tissue). There were also higher ratios
of prostate/blood and prostate/control tissues in the cas-
trated animals compared to the animals that received
saturating amounts of 5R-DHT. Similar results were
seen at 2 and 4 h. Nevertheless, the amount of radio-
activity in the prostate was very low as was the
prostate-to-control tissues ratio. High levels of radioac-
tivity were found in the thyroid, indicative of free iodide
from degradation of the iodo steroid. In a similar experi-
ment we compared the distribution of [125I]3 to that of
[
125I]1. There was little difference between the two ster-
oids. There was evidence of receptor-mediated uptake
in the prostate for both steroids, but the concentration
in the target organ was low. In this experiment per-
formed at a single time point, 1 h, the prostate-to-blood
ratio of [125I]1 was 1.34 ( 0.06 (5R-DHT blocked ) 0.94
( 0.05) and that of [125I]3 was 0.99 ( 0.08 (R-DHT
blocked ) 0.75 ( 0.04). The prostate-to-control tissues
ratio (spleen, muscle, lung) of [125I]1 was 2.13 ( 0.06
(5R-DHT blocked 1.58 ( 0.09) and that of [125I]3 was
1.82 ( 0.15 (5R-DHT blocked 1.33 ( 0.05). Thus, as seen
in the bioassay where 3 was about equipotent with 1
(Table 1), the 17R-methyl group in 3 did not result in a
higher concentration in the prostate in the in vivo
experiments.
The low uptake of [125I]1 into the prostate seemed
likely to be caused by rapid catabolism of the steroid.
This was evident by the much lower biological activity
of 3 compared to 4. Although, 3 has a lower RBA for
the receptor than 4, it seemed unlikely that this
difference, 71% versus 123%, respectively, could be
sufficient by itself to account for the extremely large
difference in their androgenic potency. We suspected
that oxidation of the 17â-hydroxyl group might not be
the limiting factor in the metabolism of the C-7-
iodinated compounds. Thus, we designed an experiment
to compare the metabolism of 1 and 3, measuring their
disappearance when incubated with rat liver micro-
somes. We found that the 125I steroids were “metabo-
lized” even in heat-denatured microsomal preparations.
F igu r e 2. Saturation analysis of the binding of [125I]3 to the
androgen receptor in rat prostate cytosol. Bt, total binding; Bns,
nonspecific binding determined with a saturating amount, 1
mM, of 5R-DHT; Bs, specific binding calculated as Bs ) Bt -
B
ns. The graph in the insert is the Scatchard representation
of the Bs data.
of [125I]3shigh affinity for the androgen receptor, high
specific activity, and low nonspecific bindingsindicated
that the 125I steroid was an excellent analytical ligand
for the androgen receptor. However, more importantly,
because 3 was a better ligand for the androgen receptor
than 1 and because the 17R-methyl group affords
protection of the 17â-hydroxyl, it seemed likely that 3
would be a good androgen receptor-mediated imaging
agent and that it would be concentrated in vivo in an-
drogen target receptor organs. Consequently, we tested
the tissue distribution of [125I]3 injected into castrated
male rats. Animals that were injected with sufficient
5R-DHT to saturate the androgen receptor (blocked
animals) were included as controls. At different times
after the injection of the [125I]3 steroids, the animals
were killed, various tissues were removed and weighed,
Ta ble 2. Distribution of Tissue Radioactivity following Intravenous Injection of 7R-[125I]-17R-CH3-5R-DHT ([125I]3) into Castrated
Male Ratsa
% injected dose/g
tissue
prostate
blood
muscle
spleen
fat
lung
kidney
0.5 h
1 h
1 h (blocked)b
2 h
4 h
0.21 ( 0.006
0.30 ( 0.01
0.06 ( 0.006
0.13 ( 0.03
0.20 ( 0.02
0.22 ( 0.006
0.26 ( 0.004
0.31 ( 0.01
1.58 ( 0.16
0.70 ( 0.01
1.54 ( 0.03
0.39 ( 0.05
0.33 ( 0.03
0.07 ( 0.007
0.16 ( 0.01
0.25 ( 0.03
0.24 ( 0.02
0.27 ( 0.02
0.29 ( 0.007
4.07 ( 0.95
1.17 ( 0.09
2.45 ( 0.14
0.23 ( 0.01
0.35 ( 0.02
0.08 ( 0.01
0.18 ( 0.01
0.29 ( 0.04
0.28 ( 0.02
0.30 ( 0.03
0.36 ( 0.06
3.55 ( 0.70
0.66 ( 0.03
1.30 ( 0.08
0.44 ( 0.03
0.31 ( 0.03
0.07 ( 0.005
0.15 ( 0.008
0.20 ( 0.02
0.24 ( 0.02
0.25 ( 0.01
0.23 ( 0.006
3.55 ( 0.70
1.43 ( 0.06
2.93 ( 0.02
0.44 ( 0.07
0.37 ( 0.07
0.07 ( 0.008
0.17 ( 0.04
0.17 ( 0.02
0.28 ( 0.04
0.23 ( 0.03
0.24 ( 0.03
9.70 ( 1.14
1.19 ( 0.04
2.54 ( 0.08
liver
thyroidc
prostate/blood
prostate/ Sp.Mu.Lu.d
a
Castrated male rats were injected subcutaneously with [125I]3, and at the indicated times the animals were killed and the tissues
b
were dissected, weighed, and counted. Blocked animals were injected subcutaneously with 500 mg of 5R-DHT 15 min before the
administration of the 125I tracer. c Thyroid also contained adhering esophagus. Sp.Mu.Lu., average of the tissues spleen, muscle and
d
lung; n ) 5. Values are ( SEM.

7R-Iodo and 7R-Fluoro Steroids
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11 2027
component, 45c, of 84% and two minor components, 45a
1
(5%) and 45b (11%), all having the same m/z (288). H
NMR analysis indicated that the major component 45c
was the ∆7-steroid, 17â-hydroxy-5R-androst-7-en-3-one.
The two minor components were not isolated separately,
but presumably, they are isomeric elimination products.
Thus, it was concluded that the 7R-iodo-5R-androstanes
are labile under physiological conditions. The dramatic
temperature dependence of the elimination reaction
explained why the receptor binding studies which were
performed at 0 °C showed no hint of the elimination of
iodide. [125I]Iodide would have been detected as non-
specifically bound radioactivity. Of note, the RBA values
in Table 1, which are determined at 0 °C, very likely
reflect the affinity of the iodinated compounds for the
androgen receptor. On the other hand, since the gene
activation studies in transfected cells were performed
for 24 h at 37 °C, the temperature at which the
iodinated steroids rapidly eliminate, it is likely that the
measured potencies were predominantly those of the
elimination products. The rapid elimination of the 7R-
iodo group in aqueous solution at 37 °C explained why
[
125I]3, although it bound with high affinity to the
androgen receptor and was protected from facile enzy-
matic oxidation at C-17, was concentrated only poorly
in the prostate of castrated male rats. It also explained
the high uptake of radioactivity by the thyroid (Table
2) in the in vivo study. Although we recognized that the
axial cycloalkyl C-I bond is susceptible to substitution
and elimination, we did not presume that the 7R-iodo
group would eliminate so rapidly at 37 °C. It is obvious
that the 7R-iodo steroids are too labile for use as
receptor-based imaging agents. However, when labeled
with 125I, they can be excellent ligands for sensitive
receptor binding studies performed at 0 °C.
The fluoro steroids strongly resist elimination. This
was apparent in both the chemical and biological studies
(Table 1) of 4. If 4, like 3, had been eliminated, then
the same product, 45, would have been produced and
the relative stimulatory activity of the two would have
been similar. That the potency of 4 was more than 10
times greater than that of 3 indicated that rapid
elimination did not occur. Indeed, 2 and 4 were shown
to be stable. When incubated in aqueous media under
the same conditions in which 1 was completely elimi-
nated, 2 and 4 were totally unchanged. Thus, it appears
likely that the 7R-fluoro steroid 4 will be an extremely
useful imaging agent. Of all the 7R-halogens described
in this study, this compound has the highest receptor
affinity as well as androgenic potency (Table 1). The
C-F bond is considerably stronger than the C-I bond,
and many 18F-labeled steroids (or ligands for their
receptors) have been shown to have the requisite
properties of high affinity and stability for in vivo
studies.9-11,25,26 Thus, 4 labeled with 18F should be an
excellent PET probe for androgen receptor-mediated
imaging of prostate tumors and metastases.
F igu r e 3. Elimination of [125I]3 in aqueous solution. The 125I-
labeled analogue 3 was incubated in buffer at the temperatures
and times indicated. The graph shows the plot of the radio-
activity that is extracted with EtOAc from the reaction
mixture.
Subsequently, we incubated the steroids in buffer at 37,
22, and 0 °C(Figure 3). Both iodinated steroids decom-
posed in a temperature-dependent manner. Thus, at 37
°C the t1/2 for 1 is 134 min and for 3, 65 min. At 22 °C
the decompositions were slower with half-lives of 421
and 951 min, respectively. In contrast, at 0 °C both
steroids were stable indefinitely. These studies ex-
plained the contradiction between the high affinity and
low nonspecific binding of these compounds to the
androgen receptor and the poor receptor-mediated
uptake in vivo. The former experiments were all per-
formed at 0 °C at which temperature the steroids were
stable, in contrast to the in vivo physiological temper-
ature of 37 °C, at which the steroids rapidly decomposed.
To determine the chemistry of the apparent decom-
position of the 7-iodo steroids, experiments were carried
out to identify the products. The nature of the radioac-
tive [125I]product formed at 37 °C from [125I]1 in these
experiments was analyzed by TLC, and it was found to
be iodide ion, 125I-, indicating that dehydrohalogenation
had occurred. In separate experiments, the steroid
product formed during the incubation of 1 in aqueous
buffer at 37 °C was investigated. A single spot was
detected by TLC. It was isolated by column chromatog-
raphy and characterized by HRMS, GC-MS, and 1H
NMR spectroscopy. GC-MS analysis showed one major
Exp er im en ta l Section
Gen er a l. Melting points were obtained in a Mel-temp
apparatus or a Thomas-Hoover hot stage and are uncorrected.
IR spectra were recorded on a Perkin-Elmer model 1600 FT-
IR or Beckman AccuLab 4 instruments. 1H NMR spectra were
recorded with the following instruments: Bruker WP100SY

2028 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
Labaree et al.
(100 MHz), GE Omega-300 (300 MHz), or Bruker AM500 (500
MHz); chemical shifts are reported relative to residual CHCl3.
Purification by flash chromatography was performed according
to the procedure of Still27 using 230-400 mesh silica gel (EM
Science). Solvents for chromatography and extraction were
reagent or HPLC grade. GC-MS experiments were performed
with a Hewlett-Packard system consisting of a model 5890 GC
interfaced with a model 5972 mass selective detector at 70 eV.
The injector temperature was 250 °C, and the detector
temperature was 280 °C. The GC column was operating at a
flow rate of 1 mL/min (He) at an initial temperature of 200 °C
for 5 min, then a temperature gradient of 10 °C/min for 5 min
to 250 °C, and held. Low- and high-resolution fast atom
bombardment (FAB) mass spectra were obtained on a VG
instrument (FAB SE) using a matrix of m-nitrobenzyl alcohol
or Magic Bullet + PEG by DR. Walter J . McMurray at the
Yale University Comprehensive Cancer Center. Elemental
analyses were performed by Schwarzkopf Micro Analytical
Laboratory, Woodside, NY. The computer program Prism was
purchased from GraphPad Software Inc. (San Diego, CA).
3H, H-19), 3.71 (dd, 1H, J ) 8.5, 8.5 Hz, H-17R), 4.65 (br d,
1H, J HF ) 49.3 Hz, H-7â); HRMS (FAB+, glycerol + TFA) calcd
for C19H30O2F 309.222984, found 309.221640. HPLC analysis
in both system H-1 (285 nm), tR ) 10 min, and system H-2
(285 nm), tR ) 7.25 min, indicated a purity of >99%.
3,3-(Eth ylen edioxy)-17r-m eth yl-5-an dr osten -17â-ol (10).
A solution of 17R-methyltestosterone (5 g, 16.5 mmol), ethylene
glycol (25 mL, 448 mmol), pTsOH (100 mg, 0.5 mmol) in
anhydrous benzene (280 mL) was heated at reflux in a 500-
mL round-bottom flask equipped with a Dean-Stark trap and
a mechanical stirrer for 4.5 h. The reaction mixture was
allowed to cool to room temperature, poured into saturated
aqueous NaHCO3, and extracted with CH2Cl2 (2×, 300 mL).
Combined organic extracts were dried over Na2SO4 and
concentrated in vacuo giving a white solid. Purification by flash
chromatography on a 5- × 14-cm column of silica gel using
hexanes-EtOAc (3:1) as eluent gave 4.7 g (82%) of 10 as a
white solid. Data for 10: mp 110-115 °C (acetone/trace
pyridine); TLC system T-2, Rf 0.35; IR (KBr) 3450 cm-1, 1110;
1H NMR (100 MHz, CDCl3) δ 0.95 (s, 3H, H-18), 1.10 (s, 3H,
H-19), 1.20 (s, 3H, 17-CH3), 4.01 (m, 4H, 3-ketal).
The cell culture reagents were obtained from Gibco-BRL
(Grand Island, NH). Monkey kidney CV1 cells were from
American Type Culture Collection (Rockville, MD). pZ523
columns were from 5 Primef 3 Prime, Inc. (Boulder, CO).
Poly-L-lysine was purchased from Sigma (St Louis, MO).
Carrier-free Na125I, [17R-methyl-3H]methyltrienolnone (R1881),
83.3 Ci/mmol, [3H]5R-DHT, 81 Ci mmol, and [3H]chorampheni-
col were from DuPont New England Nuclear Co., North
Billerica, MA. 19-Hydroxyandrostene-3,17-dione was from
Fluka, Buchs, Switzerland; all other nonradioactive steroids
were from Steraloids Inc., Newport, RI. Tetrahydrofuran
(THF) was distilled from benzophenone ketyl under N2 im-
mediately before use. Other reagents were from Aldrich
(Milwaukee, WI) and were used without further purification.
Ch r om a togr a p h ic System s. Thin-layer chromatography
(TLC) was performed using Merck silica gel plates (F254) (EM
Science, Darmstadt, Germany) and visualized using phospho-
molybdic acid or UV illumination. TLC systems: T-1, hexanes/
EtOAc (1:1); T-2, hexanes/EtOAc (2:1); T-3, hexanes/EtOAc (1:
3); T-4, hexanes/EtOAc (1:2), T-5, hexanes/EtOAc (3:1); T-6,
toluene/EtOAc (8:1); T-7, isooctane/EtOAc (1.5:1); T-8, acetone/
butanol/30% aqueous NH4OH/H2O (65:20:10:5). Analytical
high-performance liquid chromatography (HPLC) was per-
formed isocratically on a Waters 600E system equipped with
an Omniscribe recorder and a Waters 484 variable wavelength
detector. Radiolabeled compounds were purified by HPLC
using a Waters modular system consisting of a U6K injector,
a M-45 pump, a model 440 detector set at 254 nm, and a
reverse-phase column (Altex Ultrasphere 5 µm ODS, 4.6 mm
i.d. × 25 cm). Analytical HPLC systems were all at 1 mL/min:
H-1, Diol column (LiChrosorb Diol, 10 mm, 4.6 mm × 25 cm,
EM Science, Darmstardt, Germany) isooctane/iPrOH (9:1);
H-2, RP-18 column (LiChrosorb RP-18, 5 µm, 4.6 mm × 25
cm, EM Science) CH3CN/H2O (1:1); H-3, Altex Ultrasphere
ODS (5 µm 4.6 mm × 25 cm) MeOH/H2O (65:35); H-4, Altex
Ultrasphere ODS (5 µm, 4.6 mm × 25 cm) MeOH/H2O (70:
30). Semipreparative HPLC (system H-p) was performed on a
Waters Protein I-60 column (7.8 mm × 30 cm), isooctane/
iPrOH (9:1), at 3 mL/min.
3,3-(Eth ylen ed ioxy)-17â-h ydr oxy-17r-m eth yl-5-a n d r os-
ten -7-on e (11). This procedure is based on the literature
method.19 A mixture of CH2Cl2 (171 mL) and anhydrous CrO3
(33.7 g, 337 mmol) was stirred at -23 °C (dry ice/CCl4 bath)
in a 500-mL round-bottom flask under N2 for 2 min. A portion
of 3,5-dimethylpyrazole (32.6 g, 339 mmol) was added, and the
mixture was stirred for 15 min at -23 °C. To this was added
compound 10 (3.6 g, 10.5 mmol), and stirring was continued
at -23 °C for 2.5 h. The mixture was allowed to warm to 0 °C,
and 5 N NaOH (171 mL) was added. After stirring at 0 °C for
1 h, the mixture was poured into Et2O (300 mL) and washed
with 250 mL each of H2O, cold 0.2 N HCl, 5% NaOH, H2O,
and brine. The organic phase was dried (Na2SO4) and concen-
trated in vacuo giving a brown solid. Purification by flash
chromatography using toluene-EtOAc (3:1) as eluent gave
1.32 g (34%) of 11 as a white solid. Data for 11: mp 207-209
°C (methanol/trace pyridine); TLC system T-1, Rf 0.23; IR
(KBr) 1650 cm-1 1H NMR (100 MHz, CDCl3) δ 0.89 (s, 3H,
;
H-18), 1.23 (s, 3H, H-19), 1.60 (s, 3H, 17-CH3), 3.94 (s, 4H,
3-ketal), 5.70 (s, 1H, H-6).
3,3-(Eth ylen edioxy)-17â-h ydr oxy-17r-m eth yl-5r-an dr os-
ta n -7-on e (12). Ketone 11 (245 mg, 686 mmol) in MeOH (22
mL) was hydrogenated using 10% Pd on charcoal (150 mg) for
3 h under 50 psi of H2 in a Parr low-pressure apparatus. The
mixture was filtered to remove catalyst and the catalyst
washed with MeOH. Concentration of the filtrate in vacuo gave
220 mg (90%) of 12 as a white solid. Data for 12: mp 224-
225 °C (CH2Cl2/petroleum ether); TLC system T-1, Rf 0.33; IR
(KBr) 1700 cm-1 1H NMR (100 MHz, CDCl3) δ 0.90 (s, 3H,
;
H-18), 1.22 (s, 3H, H-19), 1.60 (s, 3H, 17-CH3), 3.92 (s, 4H,
3-ketal).
3,3-(Eth ylen ed ioxy)-17r-m eth yl-5r-a n d r osta n e-7r,17â-
d iol (13a ) a n d 3,3-(E t h ylen ed ioxy)-17r-m et h yl-5r-a n -
d r osta n e-7â,17â-d iol (13b). A solution of ketone 12 (263 mg,
0.725 mmol), CeCl3 (3.78 g, 10.1 mmol), H2O (7.5 mL) in THF
(82 mL) and MeOH (83 mL) was stirred at room temperature
as NaBH4 (383 mg, 10.1 mmol) was added slowly over 5 min.
The mixture was stirred at room temperature for 0.5 h, poured
into saturated aqueous NH4Cl (150 mL), and extracted with
CH2Cl2 (3×, 150 mL). Combined organic extracts were dried
over Na2SO4 and concentrated in vacuo giving 270 mg (100%)
of 13 as a white foam. This material was an inseparable
mixture of 7R- and 7â-isomers 13a and 13b in a 1/6 ratio as
determined by inspection of the 1H NMR spectrum and was
used without further purification. Data for 13a and 13b: TLC
17â-Hyd r oxy-7r-flu or o-5r-a n d r osta n -3-on e (2). A solu-
tion of tosylate 917 (54.3 mg, 0.089 mmol), nBu4NF (1.32 mL
of a 1.0 M solution in THF, 1.32 mmol) in anhydrous methyl
ethyl ketone (7 mL) was stirred and heated at 85 °C for 4 h
under Ar. The reaction mixture was poured into H2O (30 mL)
and extracted with CH2Cl2 (3×, 50 mL). Combined organic
extracts were dried over Na2SO4 and concentrated in vacuo
giving an orange oil. Purification by flash chromatography on
a 2- × 15-cm column of silica gel using hexanes-EtOAc (2:1)
followed by flash chromatography on a 2- × 15-cm column of
silica gel using CHCl3-EtOAc (6.7:1) gave 9.8 mg of 2 as a
white solid containing 10% of a close migrating impurity.
Semipreparative HPLC in system H-p gave 6.2 mg (23%) of 2
as a white solid. Data for 2: mp 230-233 °C (acetone); T-1, Rf
0.43; 1H NMR (300 MHz, CDCl3) δ 0.77 (s, 3H, H-18), 1.03 (s,
1
system T-1, Rf 0.13; IR (KBr) 3455 cm-1; H NMR (100 MHz,
CDCl3) δ 3.43 (m, 1H, H-7R), 3.86 (m, 1H, H-7â).
7r,17â-Dih ydr oxy-17r-m eth yl-5r-an dr ostan -3-on e (14a)
a n d 7â,17â-d ih yd r oxy-17r-m et h yl-5r-a n d r ost a n -3-on e
(14b). A solution of the ketal mixture 13a and 13b (143 mg,
0.392 mmol), pTsOH (82 mg, 0.43 mmol) in acetone (21 mL)
with 40 drops of H2O was stirred and heated at 65 °C in a
flask equipped with a reflux condenser for 1 h 45 min under

7R-Iodo and 7R-Fluoro Steroids
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11 2029
N2. The reaction mixture was allowed to cool to room temper-
ature, poured into saturated aqueous NaHCO3, and extracted
with CH2Cl2 (3x, 50 mL). Combined organic extracts were dried
over Na2SO4 and concentrated in vacuo giving a yellow oil.
Purification by flash chromatography on a 2- × 17-cm column
of silica gel using EtOAc-isooctane (3:1) as eluent gave 5.9
mg (4.7%) of 14a , 5.9 mg (4.7%) of a 1:1 mixture of 14a and
14b, and 88 mg (70%) of 14b as a white solid. Data for 14a :
mp 114-116 °C (CH2Cl2-petroleum ether); TLC system T-3,
Rf 0.38; IR (KBr) 3460 cm-1, 1700; 1H NMR (100 MHz, CDCl3)
δ 0.88 (s, 3H, H-18), 1.03 (s, 3H, H-19), 1.25 (s, 3H, 17-CH3),
3.93 (m, 1H, H-7â). Data for 14b: mp 198-200 °C (CH2Cl2-
(266 mg, 0.878 mmol), imidazole (120 mg, 1.75 mmol), tert-
butyldiphenylsilyl chloride (297 µL, 1.14 mmol) in anhydrous
dimethylformamide (1.8 mL) was stirred at room temperature
under N2 for 17 h. The reaction mixture was poured into H2O
(100 mL) and extracted with CH2Cl2 (3x, 70 mL). Combined
organic extracts were washed with brine (30 mL), dried over
Na2SO4, and concentrated in vacuo giving a yellow oil.
Purification by flash chromatography on a 2- × 17 cm column
of silica gel using hexanes-EtOAc (2:1) as eluent gave 432
mg (91%) of 16 as a white foam. Data for 16: TLC system
1
T-1, Rf 0.55; H NMR (500 MHz, CDCl3) δ 0.76 (s, 3H, H-18),
1.03 (s, 9H, tBu), 3.84 and 3.93 (AB quartet, 2H, J ) 10.3 Hz,
H-19), 5.90 (s, 1H, H-4), 7.39 (m, 6H, ArH), 7.61 (m, 4H, ArH);
HRMS (FAB, mNBA + PEG + NaOAc) calcd for C35H45O3Si
541.313799, found 541.313633.
petroleum ether); TLC system T-3, Rf 0.31; IR (KBr) 3460 cm-1
,
1700; 1H NMR (100 MHz, CDCl3) δ 0.90 (s, 3H, H-18), 1.06 (s,
3H, H-19), 1.22 (s, 3H, 17-CH3), 3.42 (br m, 1H, half-width 25
Hz, H-7R). Anal. (C20H32O3) C, H.
19-(ter t-Bu t yld ip h en ylsiloxy)-3,3-(et h ylen ed ioxy)-5-
a n d r osten -17-on e (17). A solution of enone 16 (3.65 g, 6.75
mmol), p-toluenesulfonic acid (15 mg, 0.079 mmol) in 2-ethyl-
2-methyl-1,3-dioxolane (115.8 mL) was stirred and heated at
83 °C under Ar for 7.5 h. The reaction mixture was allowed to
cool to room temperature, poured into saturated aqueous
NaHCO3 (70 mL), and extracted with CH2Cl2 (3x, 90 mL).
Combined organic extracts were dried over Na2SO4 and
concentrated in vacuo giving an oil. Purification of the residue
by flash chromatography on a 3- × 21-cm column of silica gel
using hexanes-EtOAc (3:1) as eluent gave 1.79 g (45%) of 17
and 1.28 g of recovered starting material. Data for 17: mp
189-191 °C (EtOAc/petroleum ether); TLC system T-5, Rf 0.67;
1H NMR (500 MHz, CDCl3) δ 0.77 (s, 3H, H-18), 1.07 (s, 9H,
tBu), 3.70 and 3.71 (AB quartet, 2H, J ) 11.5 Hz, H-19), 3.90
(m, 4H, 3-ketal), 5.57 (s, 1H, H-6), 7.40 (m, 6H, ArH), 7.65 (m,
4H, ArH); HRMS (FAB, mNBA + PEG + NaOAc) calcd for
17â-Hyd r oxy-17r-m eth yl-7â-(p-tolu en esu lfon yloxy)-5r-
a n d r osta n -3-on e (15). A solution of 7â-hydroxysteroid 14b
(88 mg, 0.27 mmol), p-toluenesulfonyl chloride (1.04 g, 5.49
mmol) in anhydrous pyridine (36 mL) was allowed to stand at
4 °C for 5 days in a sealed flask.20 The reaction mixture was
poured into CH2Cl2 (100 mL) and washed with H2O (50 mL).
The aqueous layer was back-extracted with CH2Cl2 (2x, 50
mL). Combined organic extracts were dried over Na2SO4 and
concentrated in vacuo giving a yellow solid. Purification of the
residue by flash chromatography on a 2- × 15-cm column of
silica gel using hexanes-EtOAc (1:1) as eluent gave 85.3 mg
(65%) of 15. Crystallization from acetone/petroleum ether gave
the analytical sample as white needles. Data for 15: mp 150-
151 °C (acetone/petroleum ether); TLC system T-1, Rf 0.25;
1H NMR (500 MHz, CDCl3) δ 0.86 (s, 3H, H-18), 1.03 (s, 3H,
H-19), 1.21 (s, 3H, 17-CH3), 2.46 (s, 3H, ArCH3), 4.48 (ddd,
1H, J ) 10.4, 10.4, 5.3 Hz, H-7R), 7.33 (d, 2H, J ) 7.9 Hz,
ArH), 7.78 (d, 2H, J ) 8.1 Hz, ArH). Anal. (C27H38SO5) C, H, S.
C
37H49O4Si 585.340014, found 585.339600.
19-(ter t-Bu t yld ip h en ylsiloxy)-3,3-(et h ylen ed ioxy)-5-
17â-Hyd r oxy-7r-iod o-17r-m et h yl-5r-a n d r ost a n -3-on e
(3). A solution of tosylate 15 (57.8 mg, 0.122 mmol), NaI (182
mg, 1.22 mmol) in anhydrous acetonitrile was stirred and
heated at 83 °C for 1.75 h. The reaction mixture was allowed
to cool to room temperature, diluted with CH2Cl2 (75 mL), and
washed with 10% aqueous sodium thiosulfate solution (20 mL)
and H2O (20 mL). The organic phase was dried over Na2SO4
and concentrated in vacuo giving a yellow foam. Purification
of the residue by flash chromatography on a 2- × 17-cm column
of silica gel using hexanes-EtOAc (3:1) as eluent gave 47 mg
(89%) of 3 as a white solid. Data for 3: mp 121-122 °C dec
(acetone/petroleum ether); TLC system T-1, Rf 0.48; 1H NMR
(300 MHz, CDCl3) δ 0.93 (s, 3H, H-18), 1.09 (s, 3H, H-19), 1.26
(s, 3H, 17-CH3), 4.66 (apparent quartet, 1H, J ) 2.75 Hz,
H-7â); HRMS (FAB+) calcd for C20H32O2I 431.144708, found
431.144030. Anal. (C20H31O2I) C, H, I.
a n d r osten -17â-ol (18). A solution of ketone 17 (1.35 g, 2.31
mmol) in tetrahydrofuran (29.4 mL) and H2O (1 mL) was
stirred at room temperature as NaBH4 was added. The
reaction mixture was stirred for 6 h under Ar, poured into
saturated aqueous NH4Cl (70 mL), and extracted with CH2-
Cl2 (3x, 70 mL). Combined organic extracts were dried over
Na2SO4 and concentrated in vacuo. Purification of the residue
by flash chromatography on a 3- × 21-cm column of silica gel
using hexanes-EtOAc (2.5:1) as eluent gave 1.13 g (84%) of
18. Data for 18: TLC system T-5, Rf 0.19; HRMS (FAB+) calcd
for C37H51O4Si 587.355664, found 587.354990.
19-(ter t-Bu t yld ip h en ylsiloxy)-3,3-(et h ylen ed ioxy)-5-
a n d r osten -17â-yl Aceta te (19). A solution of alcohol 18 (808
mg, 1.38 mmol) and acetic anhydride (5.0 mL, 5.3 mmol) in
pyridine (10 mL) was stirred at room temperature for 24 h
under Ar. Reaction mixture was poured into saturated aqueous
CuSO4 (200 mL) and extracted with CH2Cl2 (4x, 100 mL).
Combined organic extracts were washed with brine (100 mL),
dried over Na2SO4, and concentrated in vacuo giving a yellow
solid. Purification of the residue by flash chromatography on
a 3- × 20-cm column of silica gel using hexanes-EtOAc (5:1)
as eluent gave 797 mg (92%) of 19. Data for 19: mp 165-166
°C (acetone/petroleum ether); TLC system T-5, Rf 0.42; 1H
NMR (300 MHz, CDCl3) δ 0.75 (s, 3H, H-18), 1.07 (s, 9H, tBu),
2.05 (s, 3H, OAc), 3.65 and 3.69 (AB quartet, 2H, J ) 11.1 Hz,
H-19), 3.88 (m, 4H, 3-ketal), 4.58 (dd, 1H, J ) 8.9, 7.7 Hz,
H-17R), 5.52 (br s, 1H, H-6), 7.44-7.34 (m, 6H, ArH), 7.65 (m,
4H, ArH); MS (M + H)+ 629; HRMS (FAB+) calcd for C39H53O5-
Si 629.366229, found 629.365480.
17â-H yd r oxy-7r-flu or o-17r-m e t h yl-5r-a n d r ost a n -3-
on e (4). A solution of tosylate 15 (178.8 mg, 0.377 mmol) in
anhydrous 2-butanone (47 mL) was stirred at room temper-
ature as a 1 M solution of nBu4NF in THF (5.27 mL, 5.27
mmol) was added. The reaction mixture was stirred at 85 °C
for 4 h, allowed to cool to room temperature, poured into H2O
(100 mL), and extracted with CH2Cl2 (3x, 100 mL). Combined
organic extracts were dried over Na2SO4 and concentrated in
vacuo giving a yellow oil. Purification by repeated flash
chromatography, first on a 3- × 20-cm column of silica gel
using hexanes-EtOAc (1:1) as eluent and then a 3- × 20-cm
column of silica gel using isooctane-EtOAc (1:1), followed by
a 3- × 20-cm column of silica gel using CHCl3-EtOAc (6.7:1),
gave 43.7 mg (36%) of 4 as a white solid. Crystallization from
acetone/petroleum ether gave the analytical sample as white
plates. Data for 4: mp 206-210 °C (acetone/petroleum ether);
19-(ter t-Bu t yld ip h en ylsiloxy)-3,3-(et h ylen ed ioxy)-7-
oxoa n d r ost-5-en -17â-yl Aceta te (20). Compound 19 (545
mg, 0.867 mmol) was oxidized using the procedure described
in the preparation of compound 11. Purification by flash
chromatography using hexanes-EtOAc (2.5:1) as eluent gave
470 mg (84%) of 20 as a clear colorless oil. Data for 20: TLC
system T-5, Rf 0.135; 1H NMR (500 MHz, CDCl3) δ 0.84 (s,
3H, H-18), 1.03 (s, 9H, tBu), 2.05 (s, 3H, OAc), 3.00 (dd, 1H, J
) 11.8, 11.8 Hz, H-8), 3.78 (d, 1 part of AB quartet, 1H, J )
10.5 Hz, H-19), 3.89 (m, 5H, H-19 & 3-ketal), 4.63 (dd, 1H, J
) 8.4, 8.4 Hz, H-17R), 5.85 (br s, 1H, H-6), 7.48-7.38 (m, 6H,
1
TLC system T-4, Rf 0.56; H NMR (300 MHz, CDCl3) δ 0.87
(s, 3H, H-18), 1.03 (s, 3H, H-19), 1.25 (s, 3H, 17-CH3), 4.64 (br
d, 1H J HF ) 48.7 Hz, H-7â); HRMS (FAB+) calcd for C20H32O2F
323.238634, found 323.237470. HPLC analysis in both system
H-1 (285 nm), tR ) 8.5 min, and H-2 (285 nm), tR ) 9 min,
indicated a purity of >99%. Anal. (C20H31O2F) C, H, F.
19-(ter t -Bu t yld ip h e n ylsiloxy)-4-a n d r ost e n e -3,17-d i-
on e (16). A solution of 19-hydroxyandrost-4-ene-3,17-dione

2030 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
ArH), 7.65-7.61 (m, 4H, ArH); HRMS (FAB+) calcd for
Labaree et al.
preparation of 11. Purification by flash chromatography using
hexanes-EtOAc (2:1) as eluent and then rechromatography
using toluene-EtOAc (4:1) as eluent followed by crystallization
of the product-containing fractions from MeOH gave 357 mg
(8%) of 24 as a white solid. Data for 24: mp 224-227 °C
(MeOH); TLC system T-1, Rf 0.49; 1H NMR (500 MHz, CDCl3)
δ 0.81 (s, 3H, H-18), 2.03 (s, 3H, OAc), 3.96 (m, 4H, 3-ketal),
4.64 (dd, 1H, J ) 8.3, 8.3 Hz, H-17R), 5.75 (s, 1H, H-6); HRMS
(FAB, mNBA + PEG + NaOAc) calcd for C22H31O5 375.21715,
found 375.21665.
C
39H51O6Si 643.345493, found 643.344360.
19-(ter t-Bu t yld ip h en ylsiloxy)-3,3-(et h ylen ed ioxy)-7â-
h yd r oxya n d r ost-5-en -17-yl Aceta te (21). Compound 21 was
prepared by reduction of of enone 20 (2.25 g, 3.50 mmol) as
described for the preparation of 13a and 13b. Flash chroma-
tography on a 5- × 14-cm column of silica gel using hexanes-
EtOAc (1.5:1) as eluent gave 1.84 g (82%) of 21 as a white
foam. Data for 21: TLC system T-1, Rf 0.43; 1H NMR (300
MHz, CDCl3) δ 0.79 (s, 3H, H-18), 1.09 (s, 9H, tBu), 2.05 (s,
3H, OAc), 3.66 and 3.72 (AB quartet, 2H, J ) 10.9 Hz, H-19),
3.89 (m, 4H, 3-ketal), 3.90 (m, 1H, H-7R), 4.59 (dd, 1H, J )
8.2 Hz, H-17R), 5.46 (br s, 1H, H-6), 7.36-7.46 (m, 6H, Ar-
H), 7.62-7.68 (m, 4H, Ar-H); HRMS (FAB+) calcd for
Meth od B. Defor m yla tion of 23: A solution of aldehyde
23 (78 mg, 0.19 mmol) in MeOH (1.58 mL) and THF (3.32 mL)
was stirred at 4 °C as a solution of methanolic KOH (22.8 mg
KOH in 1.74 mL of MeOH) was added. Mixture was allowed
to stand at 4 °C for 1 h, poured into saturated aqueous NH4Cl
(30 mL), and extracted with CH2Cl2 (3x, 30 mL) and EtOAc
(1x, 30 mL). Combined organic extracts were dried over Na2-
SO4 and concentrated in vacuo. Purification by flash chroma-
tography on a 2- × 16-cm column of silica gel using hexanes-
EtOAc (2:1) as eluent gave 32.5 mg (45%) of 24 and 25 as an
inseparable mixture of ∆5,6- and ∆5(10)-isomers in a 4/6 ratio
as a white solid and 36.2 mg (50%) of 26 as a yellow oil. Data
C
39H51O5Si 627.350579, found 627.349770.
7â,19-Dih yd r oxy-3,3-(eth ylen ed ioxy)-5-a n d r osten -17â-
yl Aceta te (22). A solution of allylic alcohol 21 (1.80 g, 2.79
mmol) and nBu4NF (8.37 mL of a 1 M solution in THF, 8.37
mmol) in THF (50 mL) was stirred at room temperature under
Ar for 24 h. The reaction mixture was poured into saturated
aqueous NH4Cl (150 mL) and extracted with CH2Cl2 (3x, 150
mL). Combined organic extracts were dried over Na2SO4 and
concentrated in vacuo giving a white solid. Purification by flash
chromatography on a 3- × 13-cm column of silica gel using
EtOAc as eluent gave 916 mg (81%) of 22 as a white solid.
1
for 25: TLC system T-4, Rf 0.68; H NMR (attributed to 25,
300 MHz, CDCl3) δ 0.79 (s, 3H, H-18), 2.03 (s, 3H, OAc), 2.95
(br d, 1H, J ) 21.0 Hz, H-6eq), 3.98 (m, 4H, 3-ketal), 4.62 (dd,
1H, J ) 7.4, 7.4 Hz, H-17R); HRMS (FAB+) calcd for C22H31O5
375.21715, found 375.21581. Data for 26: TLC system T-4, Rf
0.15; 1H NMR (300 MHz, CDCl3) δ 0.79 (s, 3H, H-18), 3.67
(dd, 1H, J ) 8.5, 8.5 Hz, H-17R), 3.92 (m, 4H, OCH2CH2O),
5.39 (s, 1H, H-6), 5.64 (s, 1H, H-4).
1
Data for 22: TLC system T-3, Rf 0.155; H NMR (500 MHz,
CDCl3) δ 0.88 (s, 3H, H-18), 2.05 (s, 3H, OAc), 3.67 and 3.85
(AB quartet, 2H, J ) 11.4 Hz, H-19), 3.95 (m, 4H, 3-ketal),
4.00 (m, 1H, H-7R), 4.61 (dd, 1H, J ) 7.2 Hz, H-17R), 5.64 (s,
1H, H-6); HRMS (FAB+) calcd for C23H33O5 389.232800, found
389.233560.
3,3-(Eth ylen edioxy)-7-oxo-5r-estr an -17â-yl Acetate (29).
A solution of enone 24 (23.9 mg, 0.0638 mmol) and 10% Pd on
carbon (16.9 mg) in MeOH (3 mL) was stirred under 1 atm of
H2 for 1.5 h. The reaction mixture was filtered through Celite,
and the solid was washed with MeOH (30 mL). The filtrate
was concentrated in vacuo and purified by flash chromatog-
raphy on a 2- × 18-cm column of silica gel using hexanes-
EtOAc (2:1) as eluent to give 19.6 mg (82%) of 29 as a white
solid. Data for 29: mp 132-135 °C (MeOH); TLC system T-1,
Rf 0.55; 1H NMR (500 MHz, CDCl3) δ 0.79 (s, 3H, H-18), 2.03
(s, 3H, OAc), 3.92 (m, 4H, 3-ketal), 4.64 (dd, 1H, J ) 8.4, 8.4
Hz, H-17R); HRMS (FAB, mNBA + PEG + NaOAc) calcd for
7,19-Dioxo-3,3-(eth ylen ed ioxy)-5-a n d r osten -17â-yl Ace-
ta te (23). Chromium trioxide (2.7 g, 27 mmol) was added to a
magnetically stirred solution of pyridine (4.37 mL, 54 mmol)
in CH2Cl2 (50 mL). The solution was stirred at room temper-
ature for 15 min, then diol 22 (916 mg, 2.25 mmol) in CH2Cl2
(15 mL) was added by syringe, and the reaction was stirred
at room temperature for 1 h under Ar. The reaction mixture
was diluted with 2-propanol (10 mL) and H2O (100 mL) and
extracted with CH2Cl2 (3x, 150 mL). Combined organic extracts
were dried over Na2SO4 and concentrated in vacuo. Purifica-
tion of the residue by flash chromatography on a 3- × 14-cm
column of silica gel using hexanes-EtOAc (1:1) gave 788 mg
(87%) of 23 as a white solid. Data for 23: mp 202-211 °C dec
(CH2Cl2/hexanes); TLC system T-3, Rf 0.67; 1H NMR (500
MHz, CDCl3) δ 0.76 (s, 3H, H-18), 2.05 (s, 3H, OAc), 3.98 (m,
4H, 3-ketal), 4.63 (dd, 1H, J ) 7.8, 7.8 Hz, H-17R), 6.02 (s,
1H, H-6), 9.88 (s, 1H, H-19); HRMS (FAB, mNBA + PEG +
NaOAc) calcd for C23H31O6 403.212064, found 403.212767.
C
22H33O5 377.232800, found 377.232100.
3,3-(Eth ylen ed ioxy)-7r-h yd r oxy-5r-estr a n -17â-yl Ace-
ta te (30a ) a n d 3,3-(Eth ylen ed ioxy)-7â-h yd r oxy-5r-estr a n -
17â-yl Aceta te (30b). Ketone 29 (308 mg, 0.812 mmol) was
reduced using the method described in the preparation of 13a
and 13b giving a 1:3.5 mixture of 7R- and 7â-isomers as a
white foam. Purification by flash column chromatography on
a 3- × 20-cm column of silica gel using isooctane-EtOAc (1.5:
1) as eluent gave 8 mg of pure 30a and 236 mg (79%) of 30a
and 30b as a 1:4 mixture of 7R- and 7â-epimers as seen by
inspection of the 1H NMR spectrum. Data for 30a : TLC system
3,3-(E t h ylen ed ioxy)-5-est r en -17â-yl Acet a t e (27) a n d
3,3-(Eth ylen ed ioxy)-5(10)-estr en -17â-yl Aceta te (28). Tes-
tosterone acetate (19 g, 60 mmol) was ketalized as described
in the preparation of 17. Purification on a 5- × 16-cm column
of flash silica gel using hexanes-EtOAc (4:1) as eluent gave
14.2 g (66%) of 27 as a mixture of ∆5,6- and ∆5(10)-isomers in a
1/1.6 ratio as determined by inspection of the 1H NMR spec-
trum. Crystallization of a 3.12-g portion of this mixture with
absolute EtOH gave 0.58 g of the ∆5,6-isomer 27 as fine needles.
A 569-mg portion of the mixture was chromatographed on a
2- × 20-cm column of flash silica gel using toluene-EtOAc (8:
1) as eluent and gave 55.8 mg of pure ∆5(10)-isomer 28, 6.7 mg
of pure ∆5,6-isomer 27, and 407 mg of a mixture of ∆5,6- and
1
T-7, Rf 0.28; H NMR (500 MHz, CDCl3) δ 0.82 (s, 3H, H-18),
2.05 (s, 3H, OAc), 3.89 (br s, 1H, H-7â), 3.94 (m, 4H, 3-ketal),
4.66 (dd, 1H, J ) 8.5, 8.5 Hz, H-17R); HRMS (FAB, mNBA +
PEG
+ NaOAc) calcd for C22H35O5 379.248450, found
379.247700. Data for 30b: TLC system T-7, Rf 0.24; 1H NMR
(500 MHz, CDCl3) δ 0.82 (s, 3H, H-18), 2.04 (s, 3H, OAc), 3.41
(ddd, 1H, J ) 9.9, 9.9, 4.4 Hz, H-7R), 3.93 (m, 4H, 3-ketal),
4.59 (dd, 1H, J ) 8.8, 8.8 Hz, H-17R); HRMS (FAB, mNBA
+ PEG + NaOAc) calcd for C22H35O5 379.248450, found
379.248200.
∆
5(10)-isomers. Data for 27: mp 164-168 °C (EtOH); TLC
system T-6, Rf 0.38; 1H NMR (500 MHz, CDCl3) δ 0.81 (s, 3H,
H-18), 2.03 (s, 3H, OAc), 3.95 (m, 4H, 3-ketal), 4.61 (dd, 1H, J
) 8.7, 8.7 Hz, H-17R), 5.46 (br d, 1H, J ) 5.4 Hz, H-6); HRMS
(FAB, mNBA + PEG + NaOAc) calcd for C22H33O4 361.2379,
found 361.2373. Data for 28: TLC system T-6, Rf 0.44; 1H
NMR (500 MHz, CDCl3) δ 0.79 (s, 3H, H-18), 2.03 (s, 3H,
OAc), 3.98 (m, 4H, 3-ketal), 4.63 (dd, 1H, J ) 8.6, 8.6 Hz,
H-17R).
3,3-(E t h ylen ed ioxy)-7â-(p -t olu en esu lfon yloxy)-5r-es-
tr a n -17â-yl Aceta te (31). Tosylation of 30b (46.8 mg, 0.124
mmol) was carried out as described for 15. Purification by flash
column chromatography on a 2- × 16-cm column of silica gel
using hexanes-EtOAc (2:1) as eluent gave 53 mg (80%) of
31 as a white foam. Data for 31: TLC system T-2, Rf 0.28;
1H NMR (500 MHz, CDCl3) δ 0.78 (s, 3H, H-18), 2.03 (s, 3H,
OAc), 2.45 (s, 3H, ArCH3), 3.91 (m, 4H, 3-ketal), 4.48 (ddd,
1H, J ) 10.6, 10.6, 4.4 Hz, H-7R), 4.57 (dd, 1H, J ) 8.4, 8.4
3,3-(Eth ylen edioxy)-7-oxoestr a-5-en -17â-yl Acetate (24).
Meth od A. Oxid a tion of 27: A mixture of a solution of 27
and 28 (4.12 g, 11.4 mmol) was oxidized as described for the

7R-Iodo and 7R-Fluoro Steroids
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11 2031
Hz, H-17R), 7.32 (d, 2H, J ) 8.0, Ar-H), 7.77 (d, 2H, J ) 8.0,
Ar-H); HRMS (FAB+) calcd for C29H41O7S 533.257301, found
533.256050.
Na2SO4 and concentrated in vacuo giving a yellow oil. Purifica-
tion by flash chromatography on a 1- × 13-cm column of silica
gel using hexanes-EtOAc (1:1) as eluent gave 7 mg (84%) of
5 as a white solid. Data for 5: mp 154-156 °C (acetone/
petroleum ether); TLC system T-1, Rf 0.29; 1H NMR (500 MHz,
CDCl3) δ 0.83 (s, 3H, H-18), 3.72 (dd, 1H, J ) 8.6, 8.6 Hz,
H-17R), 4.57 (br d, 1H, J ) 2.4 Hz, half width ) 9.0 Hz, H-7â);
HRMS (EI) calcd for C18H28O2I 403.113407, found 403.112350.
HPLC system H-1 (280 nm), tR ) 11 min, and system H-2 (280
nm), tR ) 15 min, >99% pure.
7r-F lu or o-3-oxo-5r-estr a n -17â-yl Aceta te (37). Tosylate
35 (5.1 mg, 0.0104 mmol) was fluorinated as described for the
preparation of 4. Purification by flash chromatography on a
1- × 17-cm column of silica gel using hexanes-EtOAc (3:1) as
eluent gave 1.7 mg (48%) of 37 as an oil. Data for 37: TLC
system T-1, Rf 0.64; 1H NMR (500 MHz, CDCl3) δ 0.83 (s, 3H,
H-18), 2.05 (s, 3H, OAc), 4.67 (dd, 1H, J ) 8.6, 7.9 Hz, H-17R),
4.70 (br d, 1H, J ) 49 Hz, H-7R).
7r-F lu or o-17â-h yd r oxy-5r-estr a n -3-on e (6). Compound
37 (10.8 mg, 0.0321 mmol) was saponified as described for the
preparation of 5. Purification by flash chromatography on a
1- × 13-cm column of silica gel using 1:1 hexanes-EtOAc gave
5.9 mg (63%) of 6 as a white solid. Data for 6: mp 105-106.5
°C (acetone/petroleum ether); TLC system T-1, Rf 0.26; 1H
NMR (500 MHz, CDCl3) δ 0.78 (s, 3H, H-18), 3.72 (dd, 1H, J
) 8.8, 8.8 Hz, H-17R), 4.70 (br d, 1H, J HF ) 48.6 Hz, H-7â);
HRMS (EI+) calcd for C18H27O2F 294.199509, found 294.199500.
HPLC system H-1 (285 nm), tR ) 10 min, and system H-2 (285
nm), tR ) 6.5 min, >99% pure.
3,3-(Eth ylen ed ioxy)-5r-estr a n -17â-ol (32). A solution of
tosylate 31 (30.1 mg, 0.0565 mmol) in anhydrous THF (650
mL) was stirred at room temperature as LiEt3BH (226 µL of
a 1 M solution in THF, 0.226 mmol) was added by syringe.
The reaction mixture was stirred and heated at 65 °C for 5 h,
cooled to room temperature, and quenched with water (60
drops). To this were added 3 N NaOH (125 µL) and 30% H2O2
(125 µL). The reaction mixture was allowed to stand at room
temperature for 5 min and extracted with CH2Cl2 (3x, 10 mL).
Combined organic extracts were washed with aqueous 10%
sodium thiosulfate (5 mL) and concentrated in vacuo. Purifica-
tion by flash column chromatography on a 2- × 16-cm column
of silica gel eluting with hexanes-EtOAc (2:1) gave 16.9 mg
(94%) of 32 as a white solid. Data for 32: TLC system T-2, Rf
0.31; 1H NMR (500 MHz, CDCl3) δ 0.75 (s, 3H, H-18), 3.65
(dd, 1H, J ) 8.6 Hz, H-17R), 3.94 (m, 4H, 3-ketal); HRMS (EI+)
calcd for C20H32O3 320.235145, found 320.235100.
17â-Hyd r oxy-5r-estr a n -3-on e (33). Deketalization of 32
(8.6 mg, 0.027 mmol) was carried out as described for 14a and
14b. Purification by flash column chromatography on a 1- ×
12-cm column of silica gel using hexanes-EtOAc (1.5:1) as
eluent gave 6.9 mg (93%) of 33 as a white solid. The physical
data as follows was identical to an authentic standard pur-
chased from Steraloids. Data for 33: TLC system T-2, Rf 0.19;
1H NMR (500 MHz, CDCl3) δ 0.78 (s, 3H, H-18), 3.66 (dd, 1H,
J ) 8.5 Hz, H-17R); 13C NMR (126 MHz, CDCl3) δ 11.1, 23.2,
25.8, 30.2, 30.5, 30.6, 33.9, 36.6, 41.1, 41.3, 43.1, 43.7, 45.8,
47.8, 48.7, 50.0, 81.9, 211.8; HRMS (EI+) calcd for C18H28O2
276.208930, found 276.208667.
7â-Hyd r oxy-3-oxo-5r-estr a n -17â-yl Aceta te (34). Deket-
alization of 30a and 30b (126 mg, 0.333 mmol) was carried
out as described for 14a and 14b. Purification by flash
chromatography on a 3- × 18-cm column of silica gel using
EtOAc-hexanes (2:1) as eluent gave 16.8 mg (15%) of the 7R-
hydroxy isomer, 76.2 mg (69%) of the 7â-hydroxy isomer 34,
and 4.6 mg of a 1:1 mixture of these isomers. Data for 34: mp
200.5-202 °C (acetone/petroleum ether); TLC system T-3, Rf
0.43; 1H NMR (500 MHz, CDCl3) δ 0.86 (s, 3H, H-18), 2.05 (s,
3H, OAc), 3.44 (m, 1H, halfwidth ) 21.9 Hz, H-7R), 4.61 (dd,
1H, J ) 8.4, 8.4 Hz, H-17R); 13C NMR (126 MHz, CDCl3) δ
12.1, 21.2, 25.6, 26.3, 27.7, 30.2, 36.7, 41.0, 41.2, 43.4, 43.5,
44.8, 45.5, 47.8, 48.0, 49.2, 73.8, 82.3, 171.2, 211.0; HRMS
(FAB, mNBA + PEG +NaOAc) calcd for C20H31O4 335.222235,
found 335.222733. Anal. (C20H30O4) C,H.
3-Oxo-7â-(p-tolu en esu lfon yloxy)-5r-estr a n -17â-yl Ace-
ta te (35). Compound 34 (61.6 mg, 0.184 mmol) was tosylated
as described for the preparation of 15. Purification by flash
chromatography on a 2- × 18-cm column of silica gel using
hexanes-EtOAc (2:1) as eluent gave 88.6 mg of 35 as a yellow
foam. Crystallization from acetone-petroleum ether gave 81
mg (90%) of 35 as white needles. Data for 35: mp 175-177
°C (acetone/petroleum ether); TLC system T-1, Rf 0.51; 1H
NMR (500 MHz, CDCl3) δ 0.81 (s, 3H, H-18), 2.04 (s, 3H, OAc),
2.46 (s, 3H, Ar-CH3), 4.47 (ddd, 1H, J ) 10.6, 10.6, 4.5 Hz,
H-7R), 4.58 (dd, 1H, J ) 9.1, 8.1 Hz, H-17R), 7.34 (d, 2H, J )
8.1 Hz, Ar-H), 7.78 (d, 2H, J ) 8.7 Hz, Ar-H); HRMS (FAB+)
calcd for C27H37O6S 489.231086, found 489.230080.
3,3-(Eth ylen ed ioxy)-7â-[(p h en ylm eth oxy)m eth oxy]-5r-
estr a n -17â-yl Aceta te (38). A solution of the 1:4 mixure of
epimers 30a , 30b (75.6 mg, 0.200 mmol), benzyl chloromethyl
ether (555 µL, 3.99 mmol), diisopropylethylamine (696 µL, 3.99
mmol) in anhydrous CH2Cl2 (4 mL) was stirred at room
temperature for 2 h under Ar. The reaction mixture was
poured into H2O (10 mL) and extracted with CH2Cl2 (3x, 15
mL). Combined organic extracts were dried over Na2SO4 and
concentrated in vacuo giving an orange oil. Purification by
flash chromatography on a 1- × 17-cm column of silica gel
using hexanes-EtOAc (5:1) as eluent gave 100 mg (100%) of
38 as a clear colorless oil; 1H NMR analysis showed no evidence
1
of the 7R-epimer. Data for 38: TLC system T-1, Rf 0.74; H
NMR (500 MHz, CDCl3) δ 0.83 (s, 3H, H-18), 2.04 (s, 3H, OAc),
3.27 (ddd, 1H, J ) 10.3, 10.3, 4.2 Hz, H-7R), 3.93 (m, 4H,
3-ketal), 4.51 and 4.69 (AB quartet, 2H, J ) 11.8 Hz, PhCH2O),
4.58 (dd, 1H, J ) 8.7, 8.7 Hz, H-17R), 4.74 and 4.84 (AB
quartet, 2H, J ) 7.0 Hz, OCH2O), 7.35 (m, 5H, Ar-H); HRMS
(FAB, mNBA + PEG +NaOAc) calcd for C30H43O6 499.305965,
found 499.304967.
3,3-(Eth ylen ed ioxy)-7â-[(p h en ylm eth oxy)m eth oxy]-5r-
estr a n -17â-ol (39). A solution of acetate 38 (277 mg, 0.556
mmol) in 1% KOH-MeOH (50 mL) was stirred and heated at
50 °C for 3 h under N2. The reaction mixture was allowed to
cool to room temperature, poured into H2O (50 mL), and
extracted with CH2Cl2 (3x, 50 mL). Combined organic extracts
were dried over Na2SO4 and concentrated in vacuo giving a
white wax. Purification by flash chromatography on a 2- ×
20-cm column of silica gel using hexanes-EtOAc (1.5:1) as
eluent gave 173 mg (75%) of 39 as a white solid. Data for 39:
mp 107-108 °C (acetone/petroleum ether); TLC system T-1,
Rf 0.54; 1H NMR (500 MHz, CDCl3) δ 0.78 (s, 3H, H-18), 3.25
(ddd, 1H, J ) 10.2, 10.2, 3.8 Hz, H-7R), 3.60 (dd, 1H, J ) 8.7,
8.7 H-17R), 3.93 (m, 4H, 3-ketal), 4.68 and 4.50 (AB quartet,
2H, J ) 11.7 Hz, PhCH2O), 4.83 and 4.73 (AB quartet, 2H, J
) 6.9 Hz, OCH2O), 7.33 (m, 5H, Ar-H); HRMS (FAB, mNBA
+ PEG +NaOAc) calcd for C28H41O5 457.2954, found 457.2962.
3,3-(Eth ylen ed ioxy)-7â-[(p h en ylm eth oxy)m eth oxy]-5r-
estr a n -17-on e (40). A solution of pyridine (315 µL, 3.9 mmol)
in anhydrous CH2Cl2 (12 mL) was stirred at room temperature
as CrO3 (195 mg, 1.95 mmol) was added. The reaction mixture
was stirred for 15 min, and a solution of alcohol 39 (148 mg,
325 mmol) in CH2Cl2 (28 mL) was added. The reaction mixture
was stirred at room temperature for 45 min under N2,
quenched with iPrOH (10 mL), poured into H2O (50 mL), and
7r-Iod o-3-oxo-5r-estr a n -17â-yl Aceta te (36). Tosylate 35
(24.8 mg, 0.051 mmol), was iodinated as described for the
preparation of 3. Purification by flash chromatography on a
2- × 17-cm column of silica gel using hexanes-EtOAc (3:1) as
eluent gave 19.4 mg (84%) of 36. Data for 36: TLC system
1
T-1, Rf 0.76; H NMR (500 MHz, CDCl3) δ 0.88 (s, 3H, H-18),
2.05 (s, 3H, OAc), 4.55 (ddd, 1H, J ) 3.1, 2.4, 2.4 Hz, H-7â),
4.66 (dd, 1H, J ) 8.7, 8.7 Hz, H-17R).
7r-Iod o-17â-h yd r oxy-5r-estr a n -3-on e (5). A solution of
acetate 36 (9.2 mg, 0.021 mmol) in saturated aqueous Na2-
CO3 (0.3 mL) and MeOH (2.3 mL) was stirred at room
temperature for 42 h. The reaction mixture was poured into
H2O (20 mL) and extracted with CH2Cl2 (2x, 30 mL) and then
EtOAc (2x, 30 mL). Combined organic extracts were dried over

2032 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
Labaree et al.
extracted with CH2Cl2 (3x, 75 mL). Combined organic extracts
were dried over Na2SO4 and concentrated in vacuo. Purifica-
tion of the residue by flash column chromatography on a 2- ×
18-cm column of silica gel using hexanes-EtOAc (3:1) as
eluent gave 134 mg (88%) of 40 as a white solid. Data for 40:
mp 159-161 °C (acetone/petroleum ether); TLC system T-1,
Rf 0.47; 1H NMR (500 MHz, CDCl3) δ 0.92 (s, 3H, H-18), 2.42
(dd, 1H, J ) 18.7, 9.0 Hz, H-16â), 3.39 (m, 1H, halfwidth 21.5
Hz, H-7R), 3.94 (m, 4H, 3-ketal), 4.71 and 4.54 (AB quartet,
2H, J ) 11.8 Hz, PhCH2O), 4.89 and 4.80 (AB quartet 2H, J
) 6.6 Hz, OCH2O), 7.35 (m, 5H, Ar-H).
3,3-(Eth ylen edioxy)-7â-[(ph en ylm eth oxy)m eth oxy]-17r-
m eth yl-5r-estr a n -17â-ol (41). A solution of ketone 40 (55.6
mg, 0.122 mmol) and methylmagnesium iodide (5.2 mL of a
3.0 M solution in Et2O, 15.6 mmol) in anhydrous THF (1.7
mL) was stirred at room temperature for 48 h under N2. The
reaction mixture was poured into saturated aqueous NH4Cl
(20 mL) and extracted with CH2Cl2 (2x, 20 mL) and EtOAc
(2x, 20 mL). Combined organic extracts were dried over Na2-
SO4 and concentrated in vacuo giving a yellow oil. Purification
by flash chromatography on a 2- × 15-cm column of silica gel
using hexanes-EtOAc (2:1) as eluent gave 36 mg (62%) of 41
as a white solid. Data for 41: TLC system T-1, Rf 0.54; 1H
NMR (500 MHz, CDCl3) δ 0.89 (s, 3H, H-18), 1.20 (s, 3H, 17-
CH3), 3.25 (m, 1H, halfwidth ) 23.4 Hz, H-7R), 3.92 (m, 4H,
3-ketal), 4.68 and 4.50 (AB quartet, 2H, J ) 11.8 Hz, PhCH2O),
4.82 and 4.73 (AB quartet, 2H, J ) 6.9 Hz, OCH2O), 7.33 (m,
5H, Ar-H).
417.129650. HPLC system H-1 (280 nm), tR ) 9.5 min, and
system H-2 (280 nm), tR ) 17 min, >99% pure.
7r-Flu or o-17â-h yd r oxy-17r-m eth yl-5r-estr a n -3-on e (8).
Tosylate 44 (10 mg, 0.022 mmol) was fluorinated as described
in the preparation of 4. Purification by flash chromatography
on a 1- × 19-cm column of silica gel eluting with hexanes-
EtOAc (1:1) gave 3.9 mg (58%) of 8. Data for 8: mp 205-209
°C (acetone/hexanes); TLC system T-1, Rf 0.26; 1H NMR (500
MHz, CDCl3) δ 0.89 (s, 3H, H-18), 1.26 (s, 3H, 17R-CH3), 4.71
(br d, 1H, J HF ) 49.1 Hz, H-7â); HRMS (EI) calcd for C19H30O2F
309.222984, found 309.221920. HPLC system H-1 (285 nm),
tR ) 10 min, and system H-2 (285 nm), tR ) 8 min >99% pure.
Radiosyn th esis of 17â-Hydr oxy-7r-[125I]iodo-5r-an dr os-
tan -3-on e ([125I]1) an d 17â-Hydr oxy-7r-[125I]iodo-17r-m eth -
yl-5r-a n d r osta n -3-on e ([125I]3). The synthesis of [125I]1 by
substitution reaction with Na125I and tosylate 9 followed by
saponification with 1% KOH/MeOH was performed as we have
previously described17 except that the purification of the
intermediate p-nitrobenzoyl ester was omitted; e.g., the crude
iodination product was directly hydrolyzed and then purified
by HPLC in system H-3. Radiolabeled [125I]3 was synthesized
in a similar manner by the exchange of tosylate 15 with Na125I.
A solution of Na125I (5 mCi, 2.47 nmol, low pH) in H2O was
mixed with 20 µL of a solution containing 1 mg/mL sodium
thiosulfate in acetonitrile/H2O (9:1) and evaporated to dryness
under vacuum. To ensure removal of H2O, 20 µL of acetonitrile
was added to the vial and evaporated again; 20 µL of a solution
containing 23 µg of tosylate 15 in acetonitrile was added; the
vial was sealed and heated at 83 °C for 110 min. The reaction
was allowed to cool to room temperature, diluted with 1 mg/
mL aqueous Na2S2O3 solution (20 µL), and vented through a
charcoal filter. The mixture was diluted with 100 µL of MeOH/
H2O (70/30) and purified by HPLC in system H-4. Material
eluting as a peak containing the bulk of the radioactivity at
13-15 min was collected and concentrated in vacuo to give
3,3-(E t h ylen ed ioxy)-17r-m et h yl-5r-est r a n e-7â,17â-d i-
ol (42). To a suspension of 10% palladium on carbon (293 mg)
in H2O (4.5 mL) and THF (10 mL) was added a solution of
alcohol 41 (38 mg, 0.081 mmol) in THF (8 mL). The reaction
mixture was stirred under an atm of H2 for 3 h and poured
through Celite. The solid was washed with MeOH, and the
filtrate was concentrated in vacuo giving a white solid.
Purification by flash chromatography on a 1- × 15-cm column
of silica gel using EtOAc-hexanes (2:1) as eluent gave 27 mg
(96%) of 42 as a white solid. Data for 42: TLC system T-1, Rf
0.14; 1H NMR (500 MHz, CDCl3) δ 0.89 (s, 3H, H-18), 1.22 (s,
3H, 17R-CH3), 3.40 (ddd, 1H, J ) 10.1, 10.1, 4.0 Hz, H-7R),
[
125I]3 (65% radiochemical yield). The dried residue was
dissolved in benzene:ethanol (4/1) and stored at 4 °C under
argon where it was stable for months. This material comi-
grated with authentic 3 by TLC (system T-1), Rf 0.48.
An d r ogen Recep tor An a lysis. Cytosol prepared from rat
prostate were used to compare the binding of compounds 1-8
to 5R-DHT in competition experiments and for saturation
binding studies of [125I]3 using the methods previously de-
scribed.16,17 Prostate glands were obtained from Sprague-
Dawley rats that had been castrated 48 h before sacrifice. The
glands were suspended and homogenized in ice-cold TEGDMo
buffer (10 mM Tris, 1.5 mM Na2-EDTA, 10% (v/v) glycerol,
1.0 mM dithiothreitol, 25 mM sodium molybdate, pH 7.4 at 4
°C) at 1.5 mL/prostate and centrifuged at 105 000g for 45 min
at 4 °C. The supernatant (cytosol) was frozen on dry ice and
stored at -80 °C until assay. For assay, the cytosol was
defrosted on ice and diluted with TEGDMo buffer to a
concentration of 6-10 mg of protein/mL.
Com p etitive Bin d in g. Binding affinities relative to 5R-
DHT were determined by incubating aliquots of rat prostate
cytosol at 4 °C for 20 h with [3H]R1881 in the presence or
absence of compounds 1-8 or 5R-DHT.16 The final concentra-
tion of [3H]R1881 was 2-4 nM, and the competitor concentra-
tions ranged from 0.005 nM to 3 µM. All incubates contained
1 µM triamcinolone acetonide to prevent potential binding to
the progesterone receptor. Bound radioligand was separated
from free by LH-20 Sephadex gel filtration and quantified by
counting. Displacement curves were analyzed by a curve-fitting
method with the use of the computer program Prism (Graph-
Pad Software Inc., San Diego, CA). The results of three
experiments performed in duplicate are presented in Table 1.
3.93 (m, 4H, 3-ketal); HRMS (EI+) calcd for
350.245710, found 350.246000.
C21H34O4
7â,17â-Hyd r oxy-17r-m eth yl-5r-estr a n -3-on e (43). De-
ketalization of 42 (17.5 mg, 0.0499 mmol) was carried out as
described for the preparation of 14a , 14b. Purification by flash
chromatography on a 1- × 17-cm column of silica gel using
EtOAc-hexanes (3:1) as eluent gave 15 mg (98%) of 43 as a
white solid. Data for 43: mp 179-181 °C (acetone/petroleum
ether); TLC system T-3, Rf 0.23; 1H NMR (500 MHz, CDCl3) δ
0.92 (s, 3H, H-18), 1.23 (s, 3H, 17R-CH3), 3.43 (ddd, 1H, J )
10.8, 9.3, 4.5 Hz, H-7R); 13C NMR (125.8 MHz, CDCl3) δ 14.1,
25.7, 25.9, 26.0, 30.3, 31.4, 39.2, 41.1, 41.4, 43.6, 45.0, 45.6,
46.4, 48.0, 49.0, 49.1, 74.3, 80.6, 210.9; HRMS (EI+) calcd for
C
19H30O3 306.219495, found 306.219400. Anal. (C19H30O3) C,
H.
17â-Hyd r oxy-17r-m eth yl-7â-(p-tolu en esu lfon yloxy)-5r-
estr a n -3-on e (44). Tosylation of 43 (13.6 mg, 0.0444 mmol)
was achieved as described in the preparation of 15. Purification
by flash chromatography on a 1- × 20-cm column of silica gel
using hexanes-EtOAc (1:1) as eluent gave 18 mg (88%) of 44
as a white solid. Data for 44: mp 149-152 °C (acetone/
petroleum ether); TLC system T-3, Rf 0.57; 1H NMR (500 MHz,
CDCl3) δ 0.88 (s, 3H, H-18), 1.21 (s, 3H, 17-CH3), 2.46 (s, 3H,
Ar-CH3), 4.49 (ddd, 1H, J ) 10.5, 10.5, 4.3 Hz, H-7R), 7.34
(d, 2H, J ) 8.0, Ar-H), 7.78 (d, 2H, J ) 8.0, Ar-H).
7r-Iod o-17â-h yd r oxy-17r-m eth yl-5r-estr a n -3-on e (7).
Tosylate 44 (6.7 mg, 0.014 mmol) was iodinated as described
for the preparation of 3. Purification by flash chromatography
on a 1- × 18-cm column of silica gel using hexanes-EtOAc
(1:1) gave 5.2 mg (87%) of 7 as a white solid. Data for 7: mp
74-76 °C (acetone/petroleum ether); TLC system T-4, Rf 0.5;
1H NMR (500 MHz, CDCl3) δ 0.95 (s, 3H, H-18), 1.26 (s, 3H,
17-CH3), 4.58 (ddd, 1H, J ) 3.2, 3.2, 2.8 Hz, H-7â); HRMS
(FAB+, glycerol + TFA) calcd for C19H30O2I 417.129057, found
Sa tu r a tion Bin d in g An a lysis. Androgen receptor binding
of [125I]3 was performed with rat prostate cytosols as we
previously described.17 Aliquots of the cytosol were incubated
for 20 h at 4 °C with a range of concentrations of [125I]3.
Nonspecific binding was assessed in parallel incubations
containing 1 µM radioinert 5R-DHT. As above all incubates
also contained 1 µM triamcinolone acetonide. Bound [125I]3 was
separated from free by gel filtration on LH-20 Sephadex

7R-Iodo and 7R-Fluoro Steroids
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11 2033
columns. The column eluents containing the receptor-bound
steroid were counted and specific binding was calculated as
the difference between total binding and nonspecific binding
(measured in the presence of excess radioinert 5R-DHT). The
resulting data were analyzed by the method of Scatchard using
a computer-assisted nonlinear curve-fitting method (Prism).
Comparison of the saturation analysis with that for [3H]5R-
DHT performed simultaneously as we have previously de-
scribed17 indicated that the 125I product was carrier-free; e.g.
the specific activity was within error of theoretical, ap-
proximately 2200 Ci/mmol (not shown).
bution of [125I]3 was compared to that of [125I]1. Animals were
killed, and tissues were removed, weighed, and counted at a
single time point, 1 h. The results are presented in the Results
section.
Sta bility of 7r-Iod o a n d 7r-F lu or o Ster oid s in Aqu e-
ou s Solu tion . A 10-µL aliquot of a 1 × 105 cpm/µL solution
of [125I]1 or [125I]3 in EtOH was added to 1 mL of 0.1 M Tris-
HCl buffer at pH 7.4 and incubated for 0, 15, 30, 60, 120, or
180 min at 0, 22, and 37 °C. Aliquots of the incubation
mixtures were extracted with EtOAc (1x, 3 mL) and (1x, 2 mL).
The combined organic extracts were evaporated under a N2
stream and counted. Incubations were performed in triplicate,
and the results for [125I]3 are presented in Figure 3. The plot
of the fraction of initial radioactivity extracted in the organic
phase after incubation versus time was analyzed using a one-
phase exponential fit with Prism software (GraphPad Software
Inc., San Diego, CA) to determine the half-life.
In Vitr o An d r ogen Bioa ssa y. Steroids were assayed for
androgenic activity in a bioassay using cells transfected with
the androgen receptor and an androgen responsive reporter
gene. The human androgen receptor construct, an expression
vector driven by the cytomegalovirus promoter, was provided
by Dr. Michael McPhaul.28 The GRE2E1bCAT reporter con-
struct was provided by Dr. J ohn Cidlowski.29 It has two
glucocorticoid/androgen response elements from the tyrosine
aminotransferase promoter, followed by the TATA box of the
adenovirus E1b gene and the chloramphenicol acetyl trans-
ferase (CAT) gene. Plasmid DNA was purified using pZ523
columns. Monkey kidney CV1 cells were grown in Dulbecco’s
modified Eagle medium (DMEM) supplemented with 10% fetal
bovine serum, 100 µg/mL penicillin, and 100 µg/mL strepto-
mycin in incubators with humidified air and 5% carbon dioxide
at 37 °C. The day before transfection, cells were washed with
Hanks buffered saline solution, trypsinized, and plated at a
density of 2 × 105 cells/well in six-well plates in DMEM with
10% charcoal-stripped serum. At the time of transfection, the
medium was replaced with DMEM devoid of serum. The CV1
cells were transfected by the nonrecombinant adenoviral-
mediated DNA transfer procedure as described earlier.24,30 For
each well, 1 ng of androgen receptor DNA and 0.5 µg of GRE2-
E1bCAT reporter plasmid were incubated with 5 × 107 virus
particles. The virus-DNA complex was allowed to infect cells
for 2 h in DMEM devoid of serum, following which the medium
was supplemented with charcoal-stripped serum to a final
concentration of 5% and the cells and virus-DNA mix were
incubated for 24 h.
Twenty-four hours following infection, the cells were treated
separately with compounds 1-8, 5R-DHT, in ethanol (from
10-12 to 10-8 M) or ethanol vehicle alone (final ethanol
concentration ) 0.05%) and incubated for 24 h. (After the
incubation had begun, the ethanolic stock solutions of the
steroids were evaporated. Several days later they were ana-
lyzed by TLC. There was no evidence of decomposition of the
steroids.) Following incubation the cells were harvested by
scraping in TEN buffer (40 mM Tris, 1 mM EDTA, 150 mM
NaCl, pH 8.0), collected by centrifugation, resuspended in 0.25
M Tris, pH 7.5, and lysed by three rounds of freeze-thawing.
The CAT activity was determined by incubating 2.5-10 µg of
total protein31with [3H]chloramphenicol (20 µCi/µmol) and
butyryl CoA substrate as described earlier.32 A 2:1 mixture of
tetramethylpentadecane:xylenes was used to extract acylated
chloramphenicol, which was then counted. The data were
analyzed using a computer-assisted nonlinear curve-fitting
method (Prism) to determine the ED50 for each compound. The
results of three experiments performed in triplicate are pre-
sented in Table 1. The relative stimulatory activity (RSA) of
each steroid is shown as compared to the potency of 5R-DHT.
An aliquot of the aqueous phase from a 1 h incubation of
[
125I]3 at 37 °C was spotted on a TLC plate with a standard of
NaI and developed with system T-8. The radioactivity was
contained in the band migrating with carrier iodide (Rf 0.7)
which was visualized using bromocresol green indicator. Thus,
the radioactivity left in the aqueous solution after incubation
at 37 °C is in the form of free iodide. The identity of the
dehalogenated steroid resulting from the elimination reaction
in the aqueous incubation was determined as follows. A
solution of 3 mg (0.007 mmol) of 1 in 30 mL of 0.017 M sodium
phosphate buffer, pH 7.4, containing 42% (v/v) EtOH was
stirred and heated at 37 °C for 48 h. The reaction mixture
was poured into H2O (50 mL), extracted with CH2Cl2 (3x, 50
mL), dried over Na2SO4, and concentrated in vacuo. TLC
analysis in system T-2 showed a single product with Rf ) 0.27.
Purification of the residue by flash column chromatography
on a 1- × 20-cm column of silica gel using 1.5/1 hexanes/EtOAc
as eluent gave 0.5 mg of compound 45 as a white solid. Data
for 45: 1H NMR (500 MHz, CDCl3) δ 0.64 (s, 3H, H-18), 1.04
(s, 3H, H-19), 3.77 (dd, 1H, J ) 9.0, 7.9 Hz, H-17R), 5.20 (m,
1H, H-7); HRMS (FAB, NH4 + PEG) calcd for C19H29O2
289.2168, found 289.2161. GC-MS analysis (baseline resolu-
tion) showed three products: 45a ,b,c (5:11:84). The GC-MS
analysis established the presumptive molecular ion at m/z 288
for all three components, consistent with the expected
C
19H28O2: 45a , 10.89 min, m/z 288 (100); 45b, 11.16 min, m/z
288 (98); 45c, 11.72 min, m/z 288 (98). The fragmentation
patterns of all three were very similar and did not allow for
definitive assignments of structure.
Fluorinated steroids 2 and 4 were subjected to the same
aqueous conditions as described above for 1. Compounds 2 (132
µg, 0.43 µmol) in 540 µL of EtOH, 1.25 mL of 0.017 M sodium
phosphate buffer, pH 7.4; 4 (161 µg, 0.5 µmol) in 620 µL of
EtOH, 1.46 mL of buffer were stirred at 37 °C for 48 h. The
reaction mixtures were extracted with CH2Cl2 (3x, 2 mL). The
combined organic extracts were evaporated under N2 and
analyzed by TLC in system T-2. In both cases only unchanged
starting material was detected; i.e., no elimination product 45
was found.
Ack n ow led gm en t. We wish to acknowledge the
technical assistance of Donna Raucci, Toni Reynolds,
Hoang Nguyen, and William E. Bingman III. This work
was supported in part by NIH Grants GM08180 (to
R.M.H.), CA37799 (to R.B.H.), and CA68615 (to N.L.W.)
and NIH Training Grant T32-HD07165 (to L.V.N.).
Distr ibu tion of [125I]3 in Vivo. Male Sprague-Dawley
rats, approximately 250 g, were castrated and on the following
day injected in the tail vein with 5.4 µCi (2.4 pmol) of [125I]3
in 100 µL of ethanol/saline (1:4). Blocked animals were injected
subcutaneously with 500 µg of 5R-DHT in 250 µL of propylene
glycol/saline (3:2) 15 min before the administration of the 125I
tracer. Each group contained 5-6 animals. At 0.5, 1, 2, and 4
h after the injection of the tracer, the animals were anesthe-
tized with ether and killed by exsanguination. Immediately
afterward, tissues were extirpated, weighed, and counted. The
entire thyroid was taken, which contained some adhering
esophagus. The results are presented in Table 2. In another
experiment performed in a similar manner, the tissue distri-
Refer en ces
(1) Hochberg, R. B. 16R-iodo-17â-estradiol: A Highly specific ra-
dioiodinated probe for the estrogen receptor. In Cytotoxic
Estrogens in Hormone Receptive Tumors; Raus, J ., Martens, H.,
Leclercq, G., Eds.; Academic Press: New York, 1980; pp 123-
140.
(2) Hochberg, R. B.; Rosner, W. The Interaction of 16R-[125I]-
Iodoestradiol With Estrogen Receptor and Other Binding Pro-
teins. Proc. Natl. Acad. Sci. U.S.A. 1980, 77, 328-332.

2034 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 11
Labaree et al.
(3) Zielinski, J . E.; Yabuki, H.; Pahuja, S. L.; Larner, J . M.;
Dihydrotestosterone As a Potential Radioligand for Androgen
Receptor. J . Med. Chem. 1994, 37, 1224-1230.
Hochberg, R. B. 16R-[125I]Iodo-11â-Methoxy-17â-Estradiol:
A
Radiochemical Probe for Estrogen Sensitive Tissues. Endocrinol-
ogy 1986, 119, 130-139.
(4) Hanson, R. N.; Franke, L. A. Preparation and Evaluation of 17R-
(17) Labaree, D. C.; Brown, T. J .; Hoyte, R. M.; Hochberg, R. B. 7R-
[
125I]Iodo-5R-Dihydrotestosterone: a Radiolabeled Ligand for the
Androgen Receptor. J . Nucl. Med. 1997, 38, 402-409.
(18) Segaloff, A. The Enhanced Local Androgenic Activity of 19-nor
Steroids and Stabilization of Their Structure by 7R and 17R-
Methyl Substituents to Highly Potent Androgens by Any Route
of Administration. Steroids 1963, 1, 299-315.
[
125I]Iodovinyl-11â-Methoxyestradiol As a Highly Selective Ra-
dioligand for Tissues Containing Estrogen Receptors. J . Nucl.
Med. 1984, 25, 998-1002.
(5) Hochberg, R. B.; Hoyte, R. M.; Rosner, W. E-17R-(2-[125I]-
Iodovinyl)-19-Nortestosterone: The Synthesis of a Gamma-
Emitting Ligand for the Progesterone Receptor. Endocrinology
1985, 117, 2550-2552.
(19) Salmond, W. G.; Barta, M. A.; Havens, J . L. Allylic Oxidation
With 3,5-Dimethylpyrazole. Chromium Trioxide Complex. Ste-
roidal R5-7-Ketones. J . Org. Chem. 1978, 43, 2057-2059.
(20) Cragg, G. M. L.; Davey, C. W.; Hall, D. N.; Meakins, G. D.;
Richards, E. E.; Whateley, T. L. Hydroxy-Steroids. Part IV. The
Preparation and Spectra of Steroid Olefins. J . Chem. Soc. 1966,
1266-1276.
(21) Hagiwara, H.; Noguchi, S.; Nishikawa, M. Syntheses of 19-Nor-
3-Ketosteroids. Chem. Pharm. Bull. (J pn) 1960, 8, 84-85.
(22) Kumar, V.; Amann, A.; Ourisson, G.; Luu, B. Stereospecific
Syntheses of 7â- and 7R-Hydroxycholesterols. Synth. Commun.
1987, 17, 1279-1286.
(23) Labaree, D. C.; Hoyte, R. M.; Hochberg, R. B. A Direct Stereo-
selective Synthesis of 7-â-Hydroxytestosterone. Steroids 1997,
62, 482-486.
(24) Nazareth, L. V.; Weigel, N. L. Activation of the Human Androgen
Receptor Through a Protein Kinase A Signaling Pathway. J .
Biol. Chem. 1996, 271, 19900-19907.
(25) Kiesewetter, D. O.; Kilbourn, M. R.; Landvatter, S. W.; Heiman,
D. F.; Katzenellenbogen, J . A.; Welch, M. J . Preparation of Four
Fluorine- 18-Labeled Estrogens and Their Selective Uptakes in
Target Tissues of Immature Rats. J . Nucl. Med. 1984, 25, 1212-
1221.
(26) French, A. N.; Wilson, S. R.; Welch, M. J .; Katzenellenbogen, J .
A. A Synthesis of 7 Alpha-Substituted Estradiols: Synthesis and
Biological Evaluation of a 7 Alpha-Pentyl-Substituted BODIPY
(6) Grill, H.-J .; Pollow, K.; Heubner, A.; Laurent, H.; Hoffmeister,
H.; Schulze, P. E.; Elger, W. Iodine 125-Labeleled Vinyl Nor-
testosterone:
a New High Affinity Ligand for Progesterone
Receptor Determination. Acta Endocrinol. 1985, 108, 125-126.
(7) Pomper, M. G.; Katzenellenbogen, J . A.; Welch, M.; Brodack, J .
W.; Mathias, C. J . 21-[18F]Fluoro-16 Alpha-Ethyl-19-Norproges-
terone: Synthesis and Target Tissue Selective Uptake of a
Progestin Receptor Based Radiotracer for Positron Emission
Tomography. J . Med. Chem. 1988, 31, 1360-1363.
(8) Ali, H.; Rousseau, J .; Van Lier, J . E. Synthesis of (17R,20E/Z)
Iodovinyl Testosterone and 19-Nortestosterone Derivatives As
Potential Radioligands for Androgen and Progesterone Recep-
tors. J . Steroid Biochem. Mol. Biol. 1994, 49, 15-29.
(9) Choe, Y. S.; Lidstrom, P. J .; Chi, D. Y.; Bonasera, T. A.; Welch,
M. J .; Katzenellenbogen, J . A. Synthesis of 11â-[18F]Fluoro-5R-
Dihydrotestosterone and 11â-[18F]Fluoro-19-nor-5R-Dihydrotes-
tosterone: Preparation Via Halofluorination-Reduction, Receptor
Binding, and Tissue Distribution. J . Med. Chem. 1995, 38, 816-
825.
(10) Liu, A.; Katzenellenbogen, J . A.; VanBrocklin, H. F.; Mathias,
C. J .; Welch, M. J . 20-[18F]Fluoromibolerone, a Positron-Emitting
Radiotracer for Androgen Receptors: Synthesis and Tissue
Distribution Studies. J . Nucl. Med. 1991, 32, 81-88.
(11) Liu, A.; Dence, C. S.; Welch, M. J .; Katzenellenbogen, J . A.
Fluorine-18-Labeled Androgens: Radiochemical Synthesis and
Tissue Distribution Studies on Six Fluorine-Substituted Andro-
gens, Potential Imaging Agents for Prostatic Cancer. J . Nucl.
Med. 1992, 33, 724-734.
Fluorescent Conjugate and a Fluorine-18-Labeled 7 Alpha-
Pentylestradiol Analogue. Steroids 1993, 58, 157-169.
(27) Still, C. W.; Kahn, M.; Mitra, A. Rapid Chomatographic Tech-
nique for Preparative Separations With Moderate Resolution.
J . Org. Chem. 1978, 43, 2923-2925.
(12) Welch, M. J .; Downer, J . B.; Katzenellenbogen, J . A. Steroid
hormone receptors as targets for diagnostic imaging. In Current
Directions in Radiopharmaceutical Research and Development;
Matias, J . R., Ed.; Kluwer Academic Publishers: Dordrecht,
Netherlands, 1996; pp 137-156.
(28) Tilley, W. D.; Wilson, C. M.; Marcelli, M.; McPhaul, M. J .
Androgen Receptor Gene Expression in Human Prostate Car-
cinoma Cell Lines. Cancer Res. 1990, 50, 5382-5386.
(29) Allgood, V. E.; Oakley, R. H.; Cidlowski, J . A. Modulation by
Vitamin B6 of Glucocorticoid Receptor-Mediated Gene Expres-
sion Requires Transcription Factors in Addition to the Gluco-
corticoid Receptor. J . Biol. Chem. 1993, 268, 20870-20876.
(30) Allgood, V. E.; Zhang, Y.; O’Malley, B. W.; Weigel, N. L. Analysis
of Chicken Progesterone Receptor Function and Phosphorylation
Using an Adenovirus-Mediated Procedure for High-Efficiency
DNA Transfer. Biochemistry 1997, 36, 224-232.
(31) Bradford, M. M. A Rapid and Sensitive Method for the Quan-
titation of Microgram Quantities of Protein Utilizing the Prin-
ciple of Protein-Dye Binding. Anal. Biochem. 1976, 72, 248-54,
248-254.
(32) Zhang, Y.; Bai, W.; Allgood, V. E.; Weigel, N. L. Multiple
Signaling Pathways Activate the Chicken Progesterone Recep-
tor. Mol. Endocrinol. 1994, 8, 577-584.
(13) Hoyte, R. M.; Rosner, W.; Hochberg, R. B. Synthesis of [16R-125I]-
Iodo-5R-Dihydrotestosterone and Evaluation of Its Affinity for
the Androgen Receptor. J . Steroid Biochem. 1982, 16, 621-628.
(14) Symes, E. K.; Coulson, W. F.; Farrant, R. D.; Milroy, E. J . G.
16R-Iodo-Testosterone: Chemical Synthesis and Evaluation As
a Potential Radiopharmaceutical. Biochem. Pharmacol. 1985,
34, 3173-3178.
(15) (a) Hoyte, R. M.; MacLusky, N. J .; Hochberg, R. B. The Synthesis
and Testing of E-17R-(2-Iodovinyl)-5R-Dihydrotestosterone and
Z-17R-(2-Iodovinyl)-5R-Dihydrotestosterone As γ-Emitting Ligands
for the Androgen Receptor. J . Steroid Biochem. 1990, 36, 125-
132. (b) Hoyte, R. M.; Brown, T. J .; MacLusky, N. J .; Hochberg,
R. B. 7R-Methyl-17R-(E-2′-[125I]Iodovinyl)-19-Nortestosterone: a
New Radioligand for the Detection of Androgen Receptor.
Steroids 1993, 58, 13-23.
(16) Hoyte, R. M.; Borderon, K.; Bryson, K.; Allen, R.; Hochberg, R.
B.; Brown, T. J . Synthesis and Evaluation of 7R-Iodo-5R-
J M990064O