C19-Steroids as androgen receptor modulators: Design, discovery, and structure-activity relationship of new steroidal androgen receptor antagonists

Article abstract of DOI:10.1016/j.bmc.2006.05.022

Dehydroepiandrosterone (DHEA), the most abundant steroid in human circulating blood, is metabolized to sex hormones and other C₁₉-steroids. Our previous collaborative study demonstrated that androst-5-ene-3β,17β-diol (Adiol) and androst-4-ene-3,17-dione (Adione), metabolites of DHEA, can activate androgen receptor (AR) target genes. Adiol is maintained at a high concentration in prostate cancer tissue; even after androgen deprivation therapy and its androgen activity is not inhibited by the antiandrogens currently used to treat prostate cancer patients. We have synthesized possible metabolites of DHEA and several synthetic analogues and evaluated their role in androgen receptor transactivation to identify AR modulators. Steroids with low androgenic potential in PC-3 cell lines were evaluated for anti-dihydrotestosterone (DHT) and anti-Adiol activity. We discovered three potent antiandrogens: 3β-acetoxyandrosta-1,5-diene-17-one 17-ethylene ketal (ADEK), androsta-1,4-diene-3,17-dione 17-ethylene ketal (OAK), and 3β-hydroxyandrosta-5,16-diene (HAD) that antagonized the effects of DHT as well as of Adiol on the growth of LNCaP cells and on the expression of prostate-specific antigen (PSA). In vivo tests of these compounds will reveal their potential as potent antiandrogens for the treatment of prostate cancer.

Full text of DOI:10.1016/j.bmc.2006.05.022

Bioorganic & Medicinal Chemistry 14 (2006) 5933–5947
C19-Steroids as androgen receptor modulators: Design,
discovery, and structure-activity relationship of new
steroidal androgen receptor antagonists
Padma Marwah,a Ashok Marwah,a Henry A. Lardy,a,*
Hiroshi Miyamotob and Chawnshang Changb
aDepartment of Biochemistry-Enzyme Institute, University of Wisconsin-Madison, 1710 University Avenue, Madison, WI 53726, USA
bGeorge Whipple Laboratory for Cancer Research, Departments of Pathology, Urology, and Radiation Oncology,
601 Elmwood Avenue, University of Rochester Medical Center, Rochester, NY 14642, USA
Received 28 February 2006; revised 11 May 2006; accepted 15 May 2006
Available online 8 June 2006
Abstract—Dehydroepiandrosterone (DHEA), the most abundant steroid in human circulating blood, is metabolized to sex hor-
mones and other C19-steroids. Our previous collaborative study demonstrated that androst-5-ene-3b,17b-diol (Adiol) and and-
rost-4-ene-3,17-dione (Adione), metabolites of DHEA, can activate androgen receptor (AR) target genes. Adiol is maintained at
a high concentration in prostate cancer tissue; even after androgen deprivation therapy and its androgen activity is not inhibited
by the antiandrogens currently used to treat prostate cancer patients. We have synthesized possible metabolites of DHEA and sev-
eral synthetic analogues and evaluated their role in androgen receptor transactivation to identify AR modulators. Steroids with low
androgenic potential in PC-3 cell lines were evaluated for anti-dihydrotestosterone (DHT) and anti-Adiol activity. We discovered
three potent antiandrogens: 3b-acetoxyandrosta-1,5-diene-17-one 17-ethylene ketal (ADEK), androsta-1,4-diene-3,17-dione 17-eth-
ylene ketal (OAK), and 3b-hydroxyandrosta-5,16-diene (HAD) that antagonized the effects of DHT as well as of Adiol on the
growth of LNCaP cells and on the expression of prostate-specific antigen (PSA). In vivo tests of these compounds will reveal their
potential as potent antiandrogens for the treatment of prostate cancer.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
The AR has high affinity for its physiological steroidal
ligands testosterone (T) and dihydrotestosterone
(DHT), the active metabolite of T. HF and BC, potent
non-steroidal antiandrogens that inhibit the binding of
DHT and T to the AR, are effective temporary treat-
ments for prostate cancer. But following a period of
remission, prostate cancer becomes resistant to antian-
drogens and growth is termed ‘androgen-independent.’
This resistance to therapy has been attributed to a mod-
est increase in androgen receptor mRNA expression or
to mutations of the androgen receptor gene.5 It has also
been reported that the AR must be capable of binding
its ligand to maintain the hormone-refractory stage,
and possibly that a modest increase in receptor concen-
tration permits the receptor to function despite the lower
levels of androgens in castrated patients.5 In addition,
increased AR levels confer responsiveness to non-canon-
ical ligands such as estrogen, hydrocortisone, or even
AR antagonists such as Flutamide, to behave as ago-
nists.6 Growth of an ‘androgen-independent’ cell line
was suppressed when the AR protein was ablated by
Both normal and malignant growth of the prostate
gland is completely dependent on available androgens.1
Therefore, suppression of androgen stimulation of the
prostate gland remains a cornerstone of the manage-
ment of locally advanced or metastatic prostate cancer.
Androgen deprivation can be achieved either by sup-
pressing the secretion of testicular androgens by means
of surgical or medical castration, or by inhibiting the ac-
tion of androgens using androgen antagonists such as
hydroxyflutamide (HF) or bicalutamide (BC). Androgen
antagonists can efficiently block androgen receptor
(AR)-mediated gene expression, and therefore offer a
potentially useful treatment option for androgen-depen-
dent prostate cancer.2–4
Keywords: C19-Steroids; Adiol; Androgen receptor; Antiandrogens;
Androstadienes.
*
Corresponding author. Tel.: +1 608 262 3372; fax: +1 608 265
2904; e-mail: halardy@wisc.edu
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.05.022

5934
P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
siRNA.7 This leads to an important question: are there
steroids other than DHT and T, and their metabolites,
which could bind to AR and be responsible for the pro-
gression of prostate cancer?
related to DHEA or its metabolites,8 but in which some
metabolically susceptible sites were rendered inactive by
appropriate structural modifications. We evaluated their
AR ligand specificity with the anticipation that they
might yield antiandrogenic response toward the AR,
while imparting increased in vivo metabolic stability
and pharmacological activity.
In rat whole liver homogenate dehydroepiandrosterone
(DHEA) is metabolized to androst-5-ene-3b,17b-diol
(Adiol), androst-4-ene-3,17-dione (Adione), and 7-oxy-
genated DHEAs, diols, triols, and testosterone (T).8
DHT, estrone, and estradiol are subsequent products
in metabolism and their biosynthetic pathway is depict-
ed in Figure 1. Our previous collaborative studies9,10
have demonstrated that Adiol and Adione could acti-
vate AR target genes in the presence of AR, and that
AR coactivator (ARA70) could further enhance Adiol’s
AR-transcriptional activity. Therefore, Adiol is an
androgen with intrinsic androgenic activity in human
prostate cancer cells.9,11 The androgenic effects of Adiol
were not readily inhibited by HF and BC.9 Adiol and its
sulfate ester are synthesized and secreted from the adre-
nals. They are also formed in several other organs by
reduction of DHEA at position 17. A high concentra-
tion of Adiol is maintained in the prostate cancer tissue,
even after androgen deprivation therapy,11 and is thus
likely to be the activator of growth in the ‘androgen-in-
dependent’ stage of prostate cancer. The daily produc-
tion of this androgen in man is estimated at ca.
1.4 mg. It is found in blood plasma at ꢀ1 ng/ml
(ꢀ3 · 10ꢁ9 M), and is a major metabolite of DHEA in
human prostate homogenate.11
2. Chemistry
DHEA (1), the base of a D5 series, was selected for its
antiandrogenic potential in prostate cancer cell line
studies9 and because of its beneficial effects in various
pathological conditions.12–16 DHEA metabolites (2, 10,
and 14) as well as some of its synthetic derivatives (5,
6, 12, and 16) were prepared from 1. Their synthesis,
as depicted in Scheme 1, is described in Section 6. Adiol
(2), a major metabolite of 1, is produced physiologically
by the reduction of the metabolically susceptible 17-car-
bonyl site by 17b-hydroxydehydrogenase (Fig. 1). Adiol,
but not DHEA, activates AR target genes in presence of
AR.9 For our biological assays, 2 was synthesized in the
laboratory in high purity (100%) and its purity/composi-
tion was checked by liquid chromatography-mass spec-
trometry (LC–MS) in electrospray ionization (ESI)
mode. 16-Oxygenated metabolites of DHEA,8 10 and
14 (Scheme 1), occur in humans17 and a function for
10 has been postulated.18 Metabolite 10 was prepared
in four steps from 1. The other metabolite, 14, was pre-
pared following the known procedure19 in three steps
from 1 and in 99.9% purity. A new compound 11, the
bis-methyl carbonate of 10, was synthesized in anticipa-
tion of a longer half-life and used for cell line assays.
Among synthetic derivatives of 1, compounds 3b-Acet-
oxy-17a-oxa-D-homo-androst-5-en-17-one (5),20 17a-
ethynylandrost-5-ene-3b,17b-diol (6),21 3b,16b-diacet-
oxyandrost-5-en-17-one (12),19 and androst-5-ene-3,17-
dione 3,17-diethylene ketal (16)22 (Scheme 1) were all
synthesized from 1 and their structure identification
and isomeric purity were established based on NMR,
LC–MS data, and reported melting points.20–22
The finding that AR function remains ligand dependent
in the normal and upregulated state5,6 prompted us to
synthesize and identify AR modulators. The goal was
to find new steroid antagonists with high-binding affini-
ty for the AR, and which will block the activity of ‘other
antagonists’, for example, Adiol on AR transactivation.
We focused our attention on synthesizing and evaluat-
ing C19-steroidal compounds that are structurally
To synthesize more diverse analogues, the metabolically
susceptible 7 position in the DHEA molecule was
exploited and various 7-oxygenated metabolites of 1 as
well as synthetic 7-oxygenated derivatives were prepared
as shown in Scheme 2. 7-Oxo derivatives (17–21, 23–25,
28, and 33) were prepared by our patented procedures
involving either the use of common household bleach23
as an oxidant or by allylic oxidation using chromi-
um(VI) compounds with N-hydroxyphthalimide24,25 in
organic solvents at room temperature. Stereoselective
reduction of 7-oxo derivatives using sodium borohy-
dride/cerium trichloride heptahydrate provided pure
androst-5-ene-3b,7b,17b-triol (26) whereas, K-selectride
afforded the corresponding 3b,7a,17b-triol (27).26 Com-
pound 28 was synthesized from 16 using our green
oxidation procedure.27 Steroid 28 was reduced at the
7-position by sodium borohydride to afford the mixture
of 7a- and 7b-hydroxy diastereomers (b/a = 6:4). The
7a-diastereomer (29) was separated from the mixture
by column chromatography and was isolated in pure
Figure 1. Biosynthetic pathway of C19-steroids and estrogens.

P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
5935
Scheme 1. DHEA and D5-derivatives. Reagents and conditions: (a) NaBH4, MeOH, rt; (b) Br2, CCl4, 0–5 ꢁC; (c) MMPP, DCM–MeOH, rt; (d) NaI,
THF; (e) sodium acetylide in xylene, DMSO; (f) isopropenyl acetate, p-TSA; (g) Br2, CCl4, ꢁ10 ꢁC; (h) MeOH, NaOMe; (i) AcOH–H2O, rt;
(j) pyridine, methylchloroformate; (k) Pb(OAc)4, AcOH; (l) 2 N NaOH, MeOH.
Scheme 2. 7-Oxygenated C19-steroids. Reagents and conditions: (a) PDC, NOH-phthalimide, Acetone; (b) i—Ac2O, p-TSA, microwave, 0.67 min; (c)
clorox bleach, tert-butyl hydroperoxide, ethyl acetate; (d) MeOH, NaOH; (e) NaBH4, CeCl3 7H2O, MeOH (f) K-selectride; THF (g) NaBH4,
MeOH–DCM; (h) diethylene glycol, toluene, p-TSA; (i) N–OH–phthalimide, benzoyl peroxide, air, acetone.
isomeric form, which was further confirmed by LC–MS
analysis and spectral data. 7-Oxygenated derivative 33
was prepared from testosterone propionate 31 in two
steps. Ketalization of the 3-ketone of 31 afforded D5-
ketal 32, which was oxidized under mild conditions
involving N-hydroxyphthalimide and air to produce
compound 33.
Compounds selected from the D4 series of C19-steroids
are shown in Scheme 3. Steroids 36 and 37 were pur-
chased from Steraloids Inc., to be evaluated in our pros-
tate cancer cell line studies. 17b-Acetoxyandrost-4-ene-
3,6-dione (35) was synthesized from testosterone-17b-
acetate (34) using household bleach and tert-butyl
hydroperoxide.27

5936
P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
5a-H steroids (Scheme 4): 3b-hydroxy-5a-androstan-
17-one (38), 16a-bromo-3b-hydroxy-5a-androstan-17-
one (39), 5a-androstane-3a,17b-diol (40), 17b-hydroxy-
17a-ethyl-5a-androstan-3-one (41), and 17b-hy-
droxy17a-ethynyl-5a-androstan-3-one (42) were thus
selected and purchased from commercial sources for
their AR-transcriptional activity evaluation.
Scheme 5 shows the sequence of steps for the synthesis
of conjugated/unconjugated homo- and heteroannular
androstadienes. Compound 3b-hydroxyandrosta-1,5-
dien-17-one 17-ethylene ketal (45) was prepared from
androsta-1,4-diene-3,17-dione (43) following a reported
procedure.28 Compound 45 on acetylation gave product
46 (ADEK), which on deketalization produced com-
pound 47. Compound 47 on alkaline hydrolysis yielded
compound 48, but on borohydride reduction afforded
steroid 49. The diene 51, a 17a-ethynyl derivative of
47, was synthesized using sodium acetylide in xylene
and subsequent acetylation of diene 50 in 1 min using
our microwave procedure.29 The compound androsta-
5,16-dien-3b-ol (53, HAD) was synthesized from 1 in
two simple steps following the procedure of Caglioti.30
The tosylhydrazone of 1 was prepared by refluxing 1
and tosylhydrazine in alcohol, which on subsequent
reduction with lithium aluminum hydride produced
D5,16 diene 5331 in good yield. 7-Oxygenated conjugated
diene 54 was prepared by selective dehydration proce-
dure developed for the detection and quantitation of
free 7-oxygenated steroids from their bio-conjugates32
in biological matrices, in nanogram quantities by LC–
MS. Selective dehydration of the 3-hydroxyl group of
3b,17b-dihydroxyandrost-5-en-7-one (24) using per-
chloric acid at room temperature afforded diene 54 in
excellent yield and purity. Compound 54 on subsequent
acetylation in a conventional microwave oven, in 40 s,
afforded 55 in excellent yield and purity.
Scheme 3. D4, C19-steroids. Reagents and conditions: (a) Ac2O, p-
TSA, microwave, 2 min; (b) clorox bleach, tert-butyl hydroperoxide.
Knowing the potent androgenic response of DHT, a 5a-
H metabolite of T, prompted us to examine the response
of 5a-H metabolites and/or synthetic derivatives of 1
and 30 in prostate cancer cell line assays. A variety of
Scheme 4. 5a-H, C19-steroids.
Scheme 5. C19-Androstadienes. Reagents and conditions: (a) diethylene glycol, p-TSA, benzene; (b) t-BuOK, DMSO, 5–10 ꢁC; (c) NaBH4, MeOH;
(d) Ac2O, pyridine, DMAP, rt, 1 h; (e) p-TSA, acetone, H2O, rt; (f) K2CO3, MeOH, H2O (g) sodium acetylide, DMSO; (h) Ac2O, p-TSA, microwave,
3.0 min; (i) TsNHNH2, MeOH, reflux; (j) LiAlH4, THF; (k) HClO4 (70%), MeOH, rt; (l) Ac2O, p-TSA, microwave, 0.67 min.

P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
5937
(wtAR) expression plasmid pSG5-AR and androgen
response element-reporter plasmid mouse mammalian
tumor virus (MMTV). After transfection, cells were cul-
tured for 24 h with 1 nM DHT or steroidal compounds
at 1000 nM. Chloramphenicol acetyltransferase (CAT)
assays or luciferase (Luc) assays were performed. AR-
transcriptional activity (% relative induction) data of
these 43 steroids from Schemes 1–6 (1000 nM) have
been shown in Table 1, and Figure 2 graphically exhibits
the activity profile (% relative CAT or Luc activity) of
these steroids, which are placed in Groups I–VI (shown
on the X-axis) together with AR agonist DHT (1 nM,
Group 0).
Some of the compounds 6, 14, 19, 27, 44, 46, 47, 48, 53,
and 58, from Schemes 1, 2, 5, and 6, that showed low
induction in the AR-transcriptional activity in PC-3
cells, were further screened for their ability to modulate
DHT- and Adiol-induced AR transactivation in tran-
siently transfected PC-3 cells with wtAR expression
plasmid (pSG5-AR) and androgen response element-re-
porter plasmid (MMTV) in the presence of DHT (1 nM)
or Adiol (2.5 or 50 nM). The transcription-antagonistic
activity of these steroidal derivatives has been report-
ed.36–38 Anti-DHT and anti-Adiol activity (% relative
induction) of these 10 steroids (1000 nM) has been
shown in Table 2, and Figure 3 graphically exhibits anti-
androgenic activity profile of these selected steroids at
1000 nM (OAK at 500 nM), and the response of HF,
and BC, on the 1 nM DHT- or 50 or 2.5 nM Adiol-
induced transcription in PC-3 cells.
Scheme 6. Other C19-steroids. Reagents and condition: (a) Iodine,
KOH, MeOH, H2O; (b) 5 N H2SO4; (c) 1 N NaOH, MeOH, reflux.
Scheme 6 shows C19-steroid derivatives with more diver-
sity. 11-Oxygenated steroids in a D5-series such as 56
and 57, 17a-ethynyl 19-nor steroid with an aromatic
A-ring 58, and steroid 63 with a 5–10 double bond were
purchased from Steraloids Inc. The 16,17-seco steroids
61 and 62 were synthesized in the laboratory. Lactone
61 is similar to lactone quassolidane, a parent substance
of many natural biological active products isolated from
the species Simarubaceae.33–35 It was synthesized from
3b,7b-dihydroxyandrost-5-en-17-one (59), a reduced
product of compound 17 (Scheme 2). The oxidative
D-ring opening of the corresponding 7b-hydroxy com-
pound with potassium hypoiodite produced the 16,17
diacid (60), which on subsequent acidification with
5N-sulfuric acid underwent lactonization to produce
the 7,16-lactone 61. The 3b,13a-dihydroxy-17,13-seco-
androst-5-ene-17-carboxylic acid (62) was prepared by
D-ring cleavage of D-homo-13,17-lactone (5, Scheme
1) in refluxing methanol containing 1 N sodium
hydroxide.
Androstadienes (Scheme 5) 44 (ADEK, 1 lM), 46
(OAK, 0.05–1 lM), and 53 (HAD, 1 lM) with dou-
ble bonds in ring A (46), A and B (44) or B and
D (53) exhibited more potent AR-antagonistic activ-
ity than compounds obtained by blocking metaboli-
cally susceptible sites in DHEA (Figs. 2 and 3,
and Tables 1 and 2). Compounds 44, 46, and 53
were further investigated and compared with HF
and BC for their anti-DHT and anti-Adiol activity
in various prostate cancer cell lines and the biologi-
cal data have been recently communicated.38 A 3- to
6-fold increase over mock treatment, in AR tran-
scription in PC-3, LNCaP, and CWR22R cell lines
by Adiol (2.5 nM) was impressively repressed (ꢀ25–
50%) by compounds 44, 46, and 53, whereas HF
and BC failed to inhibit Adiol-induced effect signifi-
cantly.38 These compounds were also tested for
DHT-induced PSA expression using Western blot
analysis and showed a suppressive effect. The effect
of these steroids on the induction/suppression of cell
growth was studied in LNCaP cells, with or without
DHT,38 incubated with steroidal compounds, HF
and BC. HAD, OAK, and ADEK as well as BC
showed no significant growth induction in the ab-
sence of DHT, but suppressed the stimulation by
DHT of LNCaP cell growth. Ligand-binding assays
were performed in LNCaP or COS-1 cells transfected
with wtAR in the presence of 1 nM [3H]-R1881 and
various concentrations (1–10,000 nM) of DHT, HF,
HAD, OAK, and ADEK. These compounds inhibit-
ed PSA expression and growth of prostate cancer
3. Biology
C19-Steroids starting from DHEA (1, Scheme 1), its pos-
sible metabolites 2, 14 (Scheme 1); 17, 24, 26, 27
(Scheme 2); 36 (Scheme 3); 38 (Scheme 4), and synthetic
C-19 steroids 5, 6, 11, 12, 15, 16 (Scheme 1); 19, 21, 22,
23, 25, 29, 33 (Scheme 2); 35, 37 (Scheme 3); 39–42
(Scheme 4); 43, 44, 46–51, 53–55 (Scheme 5); and 56–
58, 61–63 (Scheme 6) were first investigated for their
ability to induce AR-transcriptional activity. Biological
activities of some of the DHEA metabolites and synthet-
ic steroidal derivatives have been reported.36–38 The
reporter assays were developed to investigate androgenic
activity in PC-3 cells transfected with wild type AR

5938
P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
Table 1. Effects of Steroids on the induction of AR-transcriptional
activity in PC-3 Cells
4. Results and discussion
Compounda
Transcriptional activity
(% induction)
A ligand sensitive AR is involved in the regulation of
prostate growth, muscle and bone mass, and spermato-
genesis in males. Whereas AR agonists are therapeuti-
cally useful in the treatment of osteoporosis, cachexia,
contraception, and androgen deficiency,39 the antago-
nists are required for the treatment of prostate cancer.
During our quest for identifying AR modulators from
the array of natural and synthetic C19-steroids, it was
found9 that DHEA (1) and 7-oxo-DHEA (17) did not
activate AR target genes in the absence or presence of
coactivator (ARA70) in DU145 cells cotransfected with
reporter gene and expression plasmid (wtAR, mtAR877,
or mtAR708), whereas, Adiol (2) and T (30) did show
intrinsic activity.9 It was discovered that Adiol was a
strong activator of AR, and that HF (1 lM) and BC
(1 lM), the non-steroidal antiandrogens presently used
in prostate cancer therapy, failed to inhibit this Adiol
(2 nM)-induced response.9 The polar hydroxyl group
at position 17 of the DHT has been proposed to form
a hydrogen bond to the Thr-877 and Asn-705 residues
of the AR ligand-binding domain (LBD).40,41 It was ob-
served that various steroidal compounds with hydroxyl
groups at positions 3 and 17, such as 15, 36, and 40 (Ta-
ble 1 and Fig. 2), induced AR-transcriptional activity
(115–150%) more strongly than the known AR-agonist
DHT (100%), probably because of the similar bonding
affinity of the 17-hydroxyl group to the LBD. Protection
of the 17-hydroxyl group in the compounds 33 and 35
brought down the induction to 90%. Whereas, addition-
al oxo-group substitution at position 7 or 16 seemed to
lessen AR-transactivation drastically as was observed in
the case of compounds 26, 27 and 11, 12, 14 (8–25%,
Table 1 and Fig. 2). Disubstitution at position 17 in
the compounds 6, 16, 29, 37, 42, and 44 also resulted
in lower induction (2–50%, Table 1 and Fig. 2) possibly
because of steric hindrance. Among these compounds,
40 (Group IV, Scheme 4 and Fig. 2) from the 5a-H series
was identified as the most potent agonist (150% induc-
tion as compared to DHT 100%). Androsta-1,5-diene
compounds (Scheme 5) were designed and most of them
(44, 46, 47, 48, 49, and 50) were synthesized starting
from androsta-1,4-diene-3,17-dione (43, Scheme 5).
Among this group of compounds, those with 17-keto
or ketal group (46–48, Group 5, Fig. 2) have shown less
than 20% induction in the AR-transcription activity.
Compounds such as 49 and 50 in the same group, with
a free hydroxyl group at position 17, exhibited higher
induction (ꢀ40%, Table 1 and Fig. 2), which was much
lower than the effect of androstenediol compounds (15,
36, and 40, Table 1 and Fig. 2). Also androsta-5,16-di-
ene (53, Table 1) showed 11% and androsta-3,5-dienes
(54 and 55, Table 1) showed ꢀ20% induction in the
AR-transcription activity in the PC-3 cell line assays.
Compounds selected from Scheme 6 had diverse struc-
tures and they also exhibited interesting activity profiles
as shown in Table 1 and Figure 2.
CAT
Luc
10
5
6
—
2
11
12
14
15
16
17
19
21
22
23
24
25
26
27
29
33
35
36
37
38
39
40
41
42
43
44
46
47
48
49
50
51
53
54
55
56
57
58
61
62
63
—
—
—
3.6
16
15
—
37
—
—
—
—
56
—
—
25
—
44
90
90
—
—
78
74
—
—
—
62
44
8
115
—
8
3
5
10
—
70
8
—
8
—
—
—
115
50
—
—
150
75
35
—
—
—
—
—
—
—
—
—
20
18
10
18
2
12
15
38
43
13
11
—
—
—
—
—
11
6
—
—
90
—
a Steroids from Schemes 1–6. PC-3 cells were transfected with wtAR
expression plasmid pSG5-AR and MMTV-CAT or MMTV-Luc.
Twenty-four hours after transfection, cells were cultured without
hormone or with 1 nM DHT or 1000 nM individual steroids.36,37
CAT or Luc activity is % response relative to DHT (100%). Values
are the mean of at least three determinations.
cells, and have sufficient binding affinity to compete
with androgens. Active compounds (44, 46, and 53)
were also screened for their affinity for other steroid
receptors such as estrogen, progesterone, and gluco-
corticoid receptors and except for a little estrogenic
activity no other receptor-mediated activity was
found. Details of these assays, antiandrogenic activity
profile, PSA expression, binding affinity, and receptor
specificity of the active compounds are described in
our recent publication.38
From our first PC-3 cell line assays, some of the com-
pounds such as 6, 14, 19, 27, 46, 47, 48, and 53, and
58, that showed only marginal induction (2–15%, at
1000 nM) of AR transcription36–38 (Table 1 and

P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
5939
Figure 2. Activity profile of the Steroids on the transcriptional activity of AR. PC-3 cells were transfected with wtAR expression plasmid pSG5-AR
and MMTV-CAT (for compounds 6, 15, 17, 19, 21, 22, 24, 25, 27, 36, 37, 40–42, 54–58, and 63) or MMTV-Luc. Twenty-four hours after
transfection, cells were cultured without hormone or with 1 nM DHT or with 1000 nM individual steroids. CAT or Luc activity was determined.36,37
CAT or Luc activity expressed in %. Induction on the Y-axis. X-axis is divided into DHT (Group 0) and Groups I–VI. Each group on the X-axis
represents a group of steroids from respective Schemes (i.e., Schemes 1–6). The left-most bar on the X-axis represents DHT as 100%, and every other
single bar represents activity of a single steroid in the left to right sequence listed below the chart. Values are means of at least three determinations.
Table 2. Effects of selected steroids on DHT- and Adiol-induced
transcriptional activity of AR
14, 19, and 27, which showed 2–15% induction in the
AR transcription in PC-3 cell line assays (Table 1 and
Fig. 2), were not able to inhibit DHT-induced AR trans-
Compounda
Relative CATb or Luc activity (%)
anti-DHTc
anti-Adiol
activation (72–90% induction, Table 2). Although, they
did show better response in inhibiting Adiol-induced ef-
fect (25–60% induction, Table 2 and Fig. 3). Androsta-
dienes 44, 46, and 53 exhibited better anti-DHT (22–
30% induction, Table 2) and anti-Adiol effect (30–50%
induction, Table 2). It is noteworthy that despite the
structural differences in both ring A and ring D among
44, 46, and 53, the bioactivities were very similar for
these three compounds. Anti-DHT and anti-Adiol activ-
ity of compounds 44 and 46 can probably be partially
attributed to the steric hindrance created by the presence
of bulky ethylene ketal substituent at position 17. This
possible explanation is further supported by the fact that
the compound 58 showed consistent suppression (induc-
tion ꢀ40%, Table 2) of DHT/Adiol-stimulation (Fig. 3),
probably because of the disubstitution at position 17,
which creates steric hindrance at that position, and also
to the presence of aromatic A-ring. However, further de-
tailed studies are needed to completely understand this
phenomenon and to grasp the intricacies and details of
the structure-activity relationship.
HF
BC
6b
25
30
80
85
90
72
25
30
72
65
22
38
90
87
25d
30d
60d
—
14
19b
27b
44
40
30
46
47
—
—
48
53
58b
50
40d
a Selected steroid’s relative antiandrogenic response on the DHT- or
Adiol-induced AR-transcriptional activity.
b CAT or Luc activity was determined in PC-3 cells transiently
cotransfected with pSG5-AR and MMTV-CATb or MMTV-Luc.
c For anti-DHT activity, after 24 h cells were cultured with 1000 nM
individual steroid along with 1 nM DHT.
d For anti-Adiol activity 1000 nM of individual steroid and 50 nM
Adiol36 in PC-3 cells were cultured. HF (1 lM), BC (1 lM) or steroid
44 (0.5 lM), 46 (1 lM), 53 (1 lM), were cultured along with 2.5 nM
Adiol.38 Induction caused by DHT or Adiol set as 100%. The values
represent means of at least three determinations.
Blocking of possible metabolically susceptible sites in
the base DHEA molecule (positions 6, 7, 16, 17, and
11) or opening the steroidal D-ring in the molecule pro-
duced mixed results. But the steroidal compounds de-
signed and synthesized with a little distortion in the A,
B, and/or D-ring by introducing double bonds that
may or may not be in conjugation seemed more suitable
for suppressing DHT- or Adiol-induced AR-transacti-
vation in cell line assays (Table 2 and Fig. 3). It seems
that this small amount of rigidity in the molecules fixes
Fig. 2) were selected for further evaluation of their abil-
ity to inhibit DHT- and Adiol-induced36–38 AR-tran-
scriptional activity in the PC-3 cell line assays and
their relative antiandrogenic activity profile (CAT or
Luc) is shown in Figure 3. Compound 44 was also select-
ed to evaluate its antiandrogenic potential since it pro-
duced the 20–25% induction at varied (50–1000 nM)
concentrations38. It was observed that compounds 6,

5940
P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
Figure 3. Antiandrogenic activity profile of selected steroids on the DHT- and Adiol-induced AR-transcriptional activity. CAT or Luc activity was
determined in PC-3 cells transiently cotransfected with pSG5-AR and MMTV-CAT (compounds 6, 19, 27, and 58) or MMTV-Luc. For anti-DHT
activity, after 24 h cells were cultured with 1000 nM of individual steroid along with 1 nM DHT. For anti-Adiol activity cells were cultured with
1000 nM of steroids 6, 14, 19, 27, and 58 along with 50 nM Adiol,36 or compound HF (1 lM), BC (1 lM), 44 (0.5 lM), 46 (1 lM), and 53 (1 lM),
along with 2.5 nM Adiol. On the X-axis the left-most bars indicate induction caused by DHT and Adiol set as 100% followed by the effects of HF,
BC, and selected steroids on the anti-DHT and anti-Adiol activity. The values represent means of at least three determinations.
them in the appropriate position and location that AR
prefers to interact with them and not the co-regulators.
We identified three antiandrogens 44, 46, and 53 (OAK,
ADEK, HAD) from our reporter assays that were able
to interrupt androgen binding to the AR, suppress the
DHT- and Adiol-induced transactivation of wild type
and mutant ARs in prostate cancer cells, and inhibit
PSA expression in LNCaP cancer cells better than HF
or BC38. Their detailed activity on cell lines in compar-
ison with HF and BC, as well as their PSA expression
and binding affinity, have been reported in a recent
communication.38
AR transactivation on both wt and mutant ARs with
or without co-activators. They did show some estrogen
activity but no progesterone or glucocorticoid receptor
activity, which gives us further impetus to design and
synthesize better antiandrogens with no estrogenic
activity.
6. Experimental
Reaction chemicals and organic solvents were pur-
chased from Aldrich Chemical Company. Deionized
water (18.2 MX-cm) was used for the reactions and
for LC–MS detection and analysis. Steroids were pur-
chased either from Steraloids Inc., New Port, USA, or
were synthesized in this laboratory. Melting points
(ꢁC) were determined in open capillary tubes in an
electrically heated and stirred Thiele-type bath and
are uncorrected. Chemical structure of each individual
compound, previously known or unknown, was estab-
lished based on NMR spectral studies and the litera-
ture references, and confirmed by molecular mass
measurements and their fragmentation patterns in
the positive and/or negative ion mode using an online
LC–MS instrument. Reaction mixtures and final prod-
ucts were analyzed for identification, selectivity, %
conversion, and % purity by LC resolution, followed
by mass measurement. NMR spectra were recorded
at 200 and 300 MHz on a Bruker spectrometer and
400 MHz on a Varian FT-Unity-Innova (ui400) instru-
ment. Chemical shifts are given in parts per million (d)
downfield from tetramethylsilane (TMS = 0) as inter-
nal standard. Coupling constants are given in Hertz
(Hz). The following abbreviations are used: s (singlet),
d (doublet), dd (double doublet), t (triplet), q (quar-
tet), and m (multiplet). Flash column chromatography
was performed on silica gel (70–230 mesh).
5. Conclusion
Since the discovery of androgenic activity of naturally
occurring steroids, Adiol and Adione, in human pros-
tate cancer cells9 our interest was stimulated toward
identifying steroidal AR modulators. DHT, T, and Adi-
one, all belong to the family of C19-steroids and all are
produced metabolically from DHEA. The inability of
standard antiandrogens to prevent the activation of
AR by Adiol indicates some differences between the
combinations of AR-Adiol and AR-DHT. This was
born out by finding differences in the respective proteol-
ysis products of these two combinations.9 We therefore
screened DHEA metabolites and synthetic derivatives in
the group of C19-steroids, to obtain information con-
cerning structural requirements for binding to the AR.
The results also indicated that a little distortion in the
steroid skeleton could produce improved antagonist
activity that can thwart the effect of DHT as well as that
of Adiol on AR. The present study led to the discovery
of three potent antiandrogens, 44, 46, and 53, which
showed better antiandrogenic potential than HF and
BC, and could block DHT- as well as Adiol-induced

P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
5941
6.1. LC–MS–ESI method for analyzing steroidal
substrates
ed. After 24 h and 36 h, additional quantities of MMPP
(4.0 g and 2.0 g) were added. After stirring for 48 h, sol-
vent was evaporated and the mixture was extracted with
DCM. The organic phase was washed with water and
brine, dried over anhydrous magnesium sulfate, and sol-
vent was removed in vacuo. The residue was crystallized
from acetone–diethyl ether to afford product 4 (1.87 g,
91%) as a white solid, mp 167–70 ꢁC (decomp.) lit.20
Chromatography of all the reaction mixtures and the
isolated products was carried out on an Agilent 1100
LC–MS system comprised of
a capillary pump
(G1376A) operated in normal mode, a quaternary pump
(G1311A), column oven (G1313A), autosampler
(G1315A), diode array detector (G1315A), and a single
quadrupole mass detector (G1946A). Data were
acquired and processed using LC/MSD Chemstation
version A.09.03 software. LC was performed on a Zor-
bax-XDB C8 analytical column (2.1 · 100 mm, 3.5 lm),
protected with a Zorbax XDB-C8 guard column, and
maintained at 40 ꢁC. The column flow rate was set at
0.3 ml/min and the eluent was monitored at 205, 240,
and 280 nm with a reference wavelength of 360 nm.
An acetonitrile–water (ACN/W) linear gradient (%)
ACN/W: 30:70 at t = 0, 90:10 at t = 15 min, and 30:70
at t = 16 min followed by a 10 min post-run time was
used for the analyses. Formic acid (0.1%) at 0.04 ml/
min was added post-column. Compounds were analyzed
for their molecular mass and fragmentation using elec-
tro-spray ionization (ESI) in positive or negative mode.
Operating conditions were: drying gas (N2) 10 L/min;
drying gas temperature 350 ꢁC; nebulizer pressure
30 psi; capillary voltage ꢁ4500 V (for positive mode)
and +3000 V (for negative mode); gain 2, and fragmen-
tor 80 V. The samples were run in scan mode.
1
170.5–70.9ꢁ (decomp.). H NMR and LC–MS showed
two isomeric peaks (a- and b-) in the ratio of 1:2. LC
(k205): 6a-Br and 6b-Br. MSD (both isomers fragmented
similarly) (ESI, +ve): m/z 527, 529, 531 (t, [M+Na]+),
447, 449 (d, [M+NaꢁHBr]+), 367 [M+Naꢁ2· HBr]+,
345 [M+Hꢁ2· HBr]+, 307 [M+Naꢁ2· HBrꢁAcOH]+,
285 [M+Hꢁ2· HBrꢁAcOH]+, 267 [M+Hꢁ2· HBrꢁ
AcOHꢁH2O]+; 1H NMR (200 MHz, CDCl3): Major
isomer (6b-Br) d 5.5 (m, 1H, 3a-H), 4.84 (dd, J = 4.4,
2.2 Hz, 1H, 6a-H), 2.06 (s, 3H, OAc), 1.43 (s, 3H, 19-
CH3), 1.36 (s, 3H, 18-CH3), Minor isomer (6a-Br) d
5.2 (m, 1H, 3a-H), 4.82 (overlapped 6b-H), 2.08 (s,
3H, OAc), 1.30 (s, 3H, 19-CH3), 1.22 (s, 3H, 18-CH3).
Compound 4 was debrominated with sodium iodide in
tetrahydrofuran, and product 3b-acetoxy-17a-oxa-D-ho-
mo-androst-5-en-17-one (5) was purified by column
chromatography on silica gel (eluent-15% acetone in
petroleum ether). Yield 85%, mp 184–85 ꢁC (lit.20 183–
85 ꢁC), LC (k205) purity 99%. MSD (ESI, +ve): m/z
369 [M+Na]+, 287 [M+HꢁAcOH]+, 269 [M+Hꢁ
AcOHꢁH2O]+; 1H NMR (400 MHz, CDCl3): d 5.38
(t, J = 2.4 Hz, 1H, 6-H), 4.6 (m, 1H, 3a-H), 2.04 (s,
3H, OAc), 1.33 (s, 3H, 19-CH3), 1.00 (s, 3H, 18-CH3);
13C NMR d 171.6 (CO-17), 170.7 (CO-Ac), 139.8
(C-5), 121.7 (C-6), 83.3 (C-13), 73.7 (C-3), 49.08,
46.83, 34.58 (CH’s), 39.05, 38.0, 36.9, 31.2, 29.0, 27.8,
22.08, 20.3 (CH2’s), 21.6, 20.1, 19.4 (CH3’s).
6.2. Androst-5-en-3b,17b-diol (Adiol, 2)
The compound was synthesized in the laboratory by
sodium borohydride (1.5 mol. equiv) reduction of
DHEA in methanol (MeOH) at 20 ꢁC. Mp 181–82 ꢁC
(lit.42 180–81 ꢁC).
6.3. 3b-Acetoxy-17a-oxa-D-homo-androst-5-en-17-one (5)
6.4. 17a-Ethynylandrost-5-ene-3b,17b-diol (6)
Levy and Jacobson20 described its synthesis in 1947 uti-
lizing peracetic acid as the oxidizing agent. Our synthetic
process involved inexpensive and safe water-soluble
monoperoxyphthalic acid magnesium salt hexahydrate
(MMPP), which provides relatively faster reaction, easy
work-up, and pure product in excellent yield.
To a stirred solution of DHEA (1, 0.15 g, 0.5 mmol) in
dimethylsulfoxide (DMSO, 4.0 ml), a solution of sodium
acetylide (18 w% slurry in xylene/mineral oil, 2 ml) was
added slowly under nitrogen atmosphere. After being
stirred for 2 h at room temperature, the mixture was
cooled and quenched with a cold saturated solution of
ammonium chloride and extracted with ethyl acetate
containing 10% petroleum ether. The organic layer
was washed with water and brine, dried over anhydrous
magnesium sulfate, and concentrated in vacuum. The
residue on trituration with cold petroleum ether formed
an off-white solid, which was purified by crystallization
from acetone–petroleum ether. Yield 80% (0.13 g), mp
239–41 ꢁC (lit.21 240–42 ꢁC).
DHEA 3-acetate (5.0 g, 0.015 mol) was employed as the
starting material, which was first cooled to 0–5 ꢁC in
carbon tetrachloride (CCl4) and a solution of bromine
(0.9 ml, 0.0175 mol) in 10 ml CCl4 was added slowly to
the reaction mixture. It was stirred for 30 min at ambi-
ent temperature and then concentrated to 10 ml volume.
The 5a,6b-dibromo-3b-acetoxyandrostan-17-one (3)
was crystallized out by diluting the content in the flask
with cold petroleum ether (bp 35–60 ꢁC). Yield 97%
(7.15 g), mp 161–62 ꢁC (decomp.) (lit.20 162.9–163.2ꢁ).
1H NMR (200 MHz, CDCl3): d 5.5 (m, 1H, 3a-H),
4.85 (dd, 1H, J = 4.64, 1.95 Hz, 6a-H), 2.06 (s, 3H,
OAc), 1.49 (s, 3H, 19-CH3), 0.91 (s, 3H, 18-CH3).
6.5. 3b,16a-Dicarbomethoxyandrost-5-en-17-one (11)
Synthesis of this new compound was accomplished in
five steps from DHEA.
DHEA (22.0 g) was dissolved in a mixture of isopropen-
yl acetate (300 ml) and toluene-p-sulfonic acid (p-TSA,
1.6 g), and the solution was gently refluxed with contin-
uous but slow distillation. A steady stream of vapors
To a stirred solution of compound 3 (2.0 g, 4.0 mmol) in
dichloromethane (DCM)–MeOH (1:4, 100 ml,) at room
temperature, water (10 ml) and MMPP (8.0 g) were add-

5942
P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
was collected by upward distillation using a pressure-
equalizing funnel. Volume in the flask was maintained
at 200 ml by the addition of fresh isopropenyl acetate
for a period of 16 h. Solvent was evaporated and the res-
idue was dissolved in a mixture of ethyl acetate and tol-
uene (4:1), washed thoroughly with cold saturated
sodium bicarbonate solution, and with water and brine.
The dried solution was concentrated to 10 ml and dilut-
ed with ether. Compound was crystallized from metha-
nol into a white shiny solid. Mother liquor on further
concentration and crystallization afforded additional
quantities of the desired 3b,17-diacetoxyandrosta-5,16-
diene (7). Conversion 90%, yield 79% (20.3 g), mp
146–48 ꢁC (lit.43 146–47 ꢁC). 1H NMR (200 MHz,
CDCl3): d 5.48 (dd, J = 2.1, 1.7 Hz, 1H, 6-H), 5.4 (d,
J = 5 Hz, 1H, 16-H), 4.62 (m, 1H, 3a-H), 2.2 (s, 3H,
17-OAc), 2.04 (s, 3H, 3-OAc), 1.05 (s, 3H, 19-CH3),
0.92 (s, 3H, 18-CH3).
Compound 9 was used as such in the next step, it was
taken up in a mixture of glacial acetic acid (9.0 ml)
and water (1.0 ml), and the reaction mixture was stirred
at room temperature for 3 h. Acetic acid was removed
completely under vacuum and the residue was taken
up in cold water and the precipitated white solid was fil-
tered, washed with sodium bicarbonate and water. The
product 3b,16a-dihydroxy-androst-5-en-17-one (10)
was obtained in 89% yield (0.67 g) and was crystallized
from methanol, mp 180–84 ꢁC (lit.45 182–86 ꢁC). LC
(k205): purity 98%. MSD (ESI, +ve): m/z 327.2
[M+Na]+, 305.2 [M+H]+, 287.3 [M+HꢁH2O]+, 269.2
[M+Hꢁ2· H2O]+, 251.1 [M+Hꢁ3· H2O]+; 1H NMR
(200 MHz, CDCl3): d 5.4 (d, J = 5.12 Hz, 1H, 6-H),
4.4 (dd, J = 7.6, 2.2 Hz, 1H, 16b-H), 3.52 (m, 1H, 3a-
H), 1.04 (s, 3H, 19-CH3), 0.99 (s, 3H, 18-CH3); 13C
NMR (300 MHz, CDCl3) d 220.0 (CO-17), 141.0 (C-
5), 121.2 (C-6), 71.28, 71.19 (CH’s-3, 16), 50.1, 48.6
(CH’s), 42.1, 37.1, 31.4, 31.2, 30.9, 30.4, 20.0 (CH2’s),
19.4, 13.9 (CH3’s).
Compound 7 (2.0 g) was dissolved in dry carbon tetra-
chloride (60 ml) and the solution was cooled to
ꢁ10 ꢁC. To this vigorously stirred solution, a solution
of bromine (0.25 ml) in carbon tetrachloride (10 ml)
was added in 2 min. The solution was stirred for a fur-
ther 15 min, and then an aqueous solution of sodium
hydrogen sulfite was added. The bromo compound
was extracted with methylene chloride, which was then
washed with sodium bicarbonate solution and brine.
The solution was dried over anhydrous magnesium sul-
fate and concentrated in vacuo to 10 ml. 3b-Acetoxy-
16a-bromoandrost-5-en-17-one (8) was precipitated as
a white solid by the addition of ether. Yield 87%
(1.9 g), mp 172–73 ꢁC (decomp.) (lit.44 173–75 ꢁC). LC
(k205): purity 96%. MSD (ESI, +ve): m/z 431, 433
[M+Na]+, 349, 351 [M+HꢁAcOH]+; 1H NMR
(200 MHz, CDCl3): d 5.4 (d, J = 5 Hz, 1H, 6-H), 5.4
(dd, J = 5.4, 2.9 Hz, 1H, 16b-H), 4.62 (m, 1H, 3a-H),
2.04 (s, 3H, 3-OAc), 1.05 (s, 3H, 19-CH3), 0.93 (s, 3H,
18-CH3); 13C NMR (300 MHz, CDCl3) d 213.3 (CO-
17), 170.5 (CO-Ac), 140.0 (C-5), 121.6 (C-6), 73.6 (C-
3), 49.9, 48.2, 46.2, 30.7 (CH’s), 38.0, 36.8, 34.2, 32.2,
30.6, 27.7, 20.2 (CH2’s), 21.4, 19.3, 13.9 (CH3’s).
To a cooled (0 ꢁC) and stirred solution of 16a-hydroxy-
DHEA (10) (0.4 g, 1.3 mmol) in dry pyridine (5.0 ml), a
solution of methyl chloroformate (1.0 ml) in dichloro-
methane (5.0 ml) was added slowly in 1 h. The reaction
mixture was stirred for 4 h, diluted with some more
dichloromethane and then quenched with ice-cold
water. Organic layer was separated, washed in sequence
with dilute hydrochloric acid, saturated sodium bicar-
bonate, and water, and dried over anhydrous magne-
sium sulfate. Solvent was removed completely under
vacuum and the residue was stirred with cold diethyl
ether to yield 3b,16a-dicarbomethoxyandrost-5-en-17-
one (11) as a white solid in 71% yield (0.39 g). Crystal-
lized from methanol, mp 195–97 ꢁC, LC (k205): purity
99.5%. MSD (ESI, +ve): m/z 443.2 [M+Na]+, 367.3
[M+NaꢁMeOHꢁCO2]+, 269.2 [M+Hꢁ2· MeOHꢁ
2· CO2]+, 251.3 [M+Hꢁ2· MeOHꢁ2· CO2ꢁ H2O]+;
1H NMR (200 MHz, CDCl3): d 5.4 (d, J = 4.88 Hz,
1H, 6-H), 5.3 (dd, J = 8.8, 1.8 Hz, 1H, 16b-H), 4.5 (m,
1H, 3a-H), 3.82, (s, 3H, OCH3), 3.77 (s, 3H, OCH3),
1.04 (s, 3H, 19-CH3), 0.99 (s, 3H, 18-CH3).
Compound 8 (1.0 g, 2.4 mmol) was dissolved in hot
methanol (20 ml) and the solution was added to a hot
solution of sodium methoxide (methanol 20 ml and
sodium 1.0 g). The solution was stirred at reflux temper-
ature for 15 min, concentrated to half in vacuo, and
poured into cold water (300 ml). The product was
extracted with ethyl acetate–petroleum ether mixture
(4:1) and the organic solution was washed successively
with dilute hydrochloric acid, sodium bicarbonate, and
water. The dried solution was concentrated and cooled
to give 3b,16a-dihydroxy-17,17-dimethoxyandrost-5-
ene (9) as a white solid. LC (k205), purity 98%. MSD
(ESI, +ve): m/z 373.3 [M+Na]+, 351.3 [M+H]+, 333.3
6.6. 3b,16b-Diacetoxyandrost-5-en-17-one (12)
The compound was prepared as described in the lit.19 It
was obtained mixed with 3b-acetoxy,16b-(acetoxy)acet-
oxyandrost-5-en-17-one (13), and was separated by col-
umn chromatography on silica gel using acetone–
petroleum ether (1:9) and characterized using LC–MS
for purity and mass measurement and NMR for struc-
ture assignments. Compound 12, mp 167–69 ꢁC (lit.19
167–70 ꢁC), LC (k205): purity 99.2%. MSD (ESI, +ve):
m/z 411.2 [M+Na]+, 389.2 [M+H]+, 329.2 [M+Hꢁ
AcOH]+, 311.2 [M+HꢁAcOHꢁH2O]+, 269.1 [M+Hꢁ
2· AcOH]+, 251.1 [M+Hꢁ2· AcOHꢁH2O]+; H NMR
1
[M+HꢁH2O]+,
319.2
[M+HꢁMeOH]+,
301.2
(400 MHz, CDCl3): d 5.41 (d, J = 5.6 Hz, 1 H, 6-H),
4.99 (t, J = 8.8 Hz, 1H, 16b-H), 4.6 (m, 1H, 3a-H),
2.12 (s, 3H, OAc), 2.04 (s, 3H, OAc), 1.06 (s, 3H, 19-
CH3), 0.98 (s, 3H, 18-CH3); 13C NMR (400 MHz,
CDCl3) d 214.7 (CO-17), 170.7, 170.5 (CO-Ac), 140.2
(C-5), 121.8 (C-6), 74.8, 73.8 (CH’s-3, 16), 50.4, 46.3,
30.9 (CH’s), 38.2, 37.1, 31.8, 31.0, 29.6, 27.9, 20.3
[M+HꢁH2OꢁMeOH]+, 287.3 [M+Hꢁ2· MeOH]+,
269.2 [M+Hꢁ2· MeOHꢁH2O]+, 251.1 [M+Hꢁ2·
1
MeOHꢁ2· H2O]+; H NMR (200 MHz, CDCl3): d 5.3
(d, J = 5 Hz, 1H, 6-H), 4.3 (t, J = 8.9, 1H, 16b-H), 3.5
(m, 1H, 3a-H), 3.46, (s, 3H, 17-OCH3), 3.37 (s, 3H,
17-OCH3), 0.99 (s, 3H, 19-CH3), 0.79 (s, 3H, 18-CH3).

P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
5943
(CH2’s), 21.6, 21.0, 19.5, 14.4 (CH3’s). Minor product
13, mp 194–96 ꢁC (lit.19 195–97 ꢁC). LC (k205): purity
99.0%. MSD (ESI, +ve): m/z 469.2 [M+Na]+, 447.2
[M+H]+, 387.2 [M+HꢁAcOH]+, 369.2 [M+HꢁAcOHꢁ
H2O]+, 269.1 [M+HꢁAcOHꢁHOCOCH2OCOCH3]+,
251.1 [M+HꢁAcOHꢁH2OꢁHOCOCH2OCOCH3ꢁH2O]+.
5.76 (d, J = 1.7 Hz, 1H, 6-H), 5.3, 5.26 (dd, J = 9.2,
1.5 Hz, 1H, 16a-H), 4.6 (m, 1H, 3a-H), 3.82 (s, 3H,
OCH3), 3.80 (s, 3H, OCH3), 1.24 (s, 3H, 19-CH3), 1.0
(s, 3H, 18-CH3); 13C NMR (300 MHz, CDCl3) d 212.5
(CO-17), 199.8 (CO-7), 164.6 (C-5), 155.4, 154.8
(2· CO), 126.5 (C-6), 75.6, 75.3 (CH’s-3, 16), 55.2,
54.8 (2· OCH3), 49.7, 44.0, 43.2 (CH’s), 42.1, 37.7,
35.6, 31.8, 30.327.1, 20.1 (CH2’s), 17.4, 14.1 (CH3’s).
6.7. 3b,17b-Dihydroxyandrost-5-en-16-one (14)
The compound was prepared by saponification of com-
pound 12 (4.0 g) in methanol (60 ml) with 2 N sodium
hydroxide (10 ml) at room temperature in 2 h. Methanol
was partially evaporated and the contents were poured
into ice water and neutralized with acetic acid. The white
solid was filtered and washed with cold water, sucked dry,
and then crystallized from methanol to give compound 14
(2.3 g, 72%). Mp 203–5 ꢁC (lit.19 201–4 ꢁC). LC (k205):
purity 99.9%. MSD (ESI, +ve): m/z 327.2 [M+Na]+,
6.11. 3b,17b-Dihydroxyandrost-5-en-7-one (24)
Compound 2 was acetylated29 and subsequently oxidized
at position 7, following the sodium hypochlorite proce-
dure,27 to obtain 3b,17b-diacetoxyandrost-5-en-7-one
(25). Saponification of 25 yielded compound 24. Crystal-
lized from acetone–petroleum ether, mp 202–4 ꢁC
(lit.47 201–4 ꢁC), yield 62%. LC (k242): purity 99.6%.
MSD (ESI, +ve): m/z 327.2 [M+Na]+, 305.2 [M+H]+,
287.1 [M+HꢁH2O]+, 269.1 [M+Hꢁ2· H2O]+, 251.1
305.2
[M+H]+,
287.3
[M+HꢁH2O]+,
269.1
[M+Hꢁ2· H2O]+, 251.1 [M+Hꢁ3· H2O]+; 1H NMR
(400 MHz, CDCl3): d 5.4 (d, J = 5.2 Hz, 1H, 6-H), 3.8
(s, 1H, 17a-H), 3.5 (m, 1H, 3a-H), 1.05 (s, 3H, 19-CH3),
0.76 (s, 3H, 18-CH3); 13C NMR (400 MHz, CDCl3) d
218.0 (CO-17), 141.4 (C-5), 121.0 (C-6), 86.4, 71.5
(CH’s-3, 17), 50.3, 45.4, 31.1 (CH’s), 42.1, 37.2, 36.5,
35.9, 31.9, 31.4, 20.5 (CH2’s), 19.6, 11.5 (CH3’s).
[M+Hꢁ 3· H2O]+; H NMR (200 MHz, CDCl3): d 5.7
1
(d, J = 1.7 Hz, 1H, 6-H), 3.6 (t, J = 5.0 Hz, 1H, 17a-H),
3.6 (m, 1H, 3a-H), 1.22 (s, 3H, 19-CH3), 0.77 (s, 3H,
18-CH3).
6.12. Androst-5-ene-3b,7b,17b-triol (26)
The compound was prepared by reduction of 3b-
hydroxyandrost-5-ene-7,17-dione (17, 0.4 g, 1.32 mmol)
in methanol (15 ml) and DCM (5 ml) using sodium
borohydride (0.2 g, 5.3 mmol) and cerium trichloride
heptahydrate (0.48 g, 1.32 mmol) at 0–5 ꢁC in 30 min.
Yield 88%, mp 235–37 ꢁC (lit.26 247–48 ꢁC). LC (k205):
purity 99.2%. MSD (ESI, +ve): m/z 329.2 [M+Na]+,
289.2 [M+HꢁH2O]+, 271.2 [M+Hꢁ2· H2O]+, 253.2
[M+Hꢁ3· H2O]+; 1H NMR (200 MHz, DMSO +
D2O): d 5.18 (s, 1H, 6-H), 3.60 (d, J = 7.08 Hz, 7a-H),
3.5 (t, J = 7.81 Hz, 1H, 17a-H), 3.35 (m, 1H, 3a-H),
1.02 (s, 3H, 19-CH3), 0.68 (s, 3H, 18-CH3).
6.8. Androst-5-ene-3,17-dione 3,17-diethylene diketal (16)
The compound was prepared by the ketalization of and-
rost-5-ene-3,17-dione (2.0 g, 6.9 mmol) in refluxing tolu-
ene (100 ml) using ethylene glycol (14 ml) and toluene-p-
sulfonic acid (0.1 g). The reaction mixture was worked-
up after 5 h and the product was crystallized from meth-
anol, yield 2.4 g (92%), mp 172–74 ꢁC (lit.22 171–
72.5 ꢁC). LC (k205): purity 99.2%. MSD (ESI, +ve):
m/z 397.2 [M+Na]+, 375.2 [M+H]+, 313.2 [M+Hꢁ
HOCH2CH2OH]+, 251.1 [M+Hꢁ2· HOCH2CH2OH]+;
1H NMR (200 MHz, CDCl3): d 5.4 (t, J = 2.5 Hz, 1H,
6-H), 3.7 (m, OCH2–CH2O), 4.0 (m, OCH2–CH2O),
1.06 (s, 3H, 19-CH3), 0.86 (s, 3H, 18-CH3).
6.13. Androst-5-ene-3b,7a,17b-triol (27)
The compound was prepared by reduction of 3b,17b-
dihydroxyandrost-5-en-7-one (24, 0.5 g) in anhydrous
tetrahydrofuran (THF, 20 ml) using a 1.0 M solution
of potassium-tri-sec-butylborohydride (K-selectride) in
THF (10 ml) at ꢁ76 ꢁC for 2 h. Yield 58%, mp 266–
67 ꢁC (lit.26 271–73 ꢁC). LC (k205): purity 98.3%. MSD
(ESI, +ve): m/z 329.2 [M+Na]+, 289.2 [M+HꢁH2O]+,
271.2 [M+Hꢁ2· H2O]+, 253.2 [M+Hꢁ3· H2O]+; 1H
NMR (200 MHz, DMSO): d 5.46 (d, J = 5.0 Hz, 1H,
6-H), 3.65 (bs, 7b-H), 3.4 (m, 2H, 3a-H, 17a-H), 0.96
(s, 3H, 19-CH3), 0.68 (s, 3H, 18-CH3).
6.9. Synthesis of compounds 17–22 and 26 is reported
elsewhere.23,27,46
6.10. 3b,16a-Dicarbomethoxyandrost-5-ene-7,17-dione (23)
A mixture of 3b,16a-dicarbomethoxyandrost-5-en-17-
one (11, Scheme 2, 0.58 g) and N-hydroxyphthalimide
(0.5 g) in acetone was treated with pyridinium dichro-
mate (0.75 g). The oxidant was added in two portions
at 8 h interval and the reaction mixture was stirred at
room temperature for 24 h. Solvent was removed com-
pletely in vacuo and the solid residue was stirred with
ethyl acetate containing 10% hexane for 15 min, filtered
and washed twice with the same solvent. The combined
organic layer was washed in sequence with water, satu-
rated sodium bicarbonate solution, and brine, and dried.
Compound 23 was crystallized from acetone–diethyl
ether, yield 68%, mp 132–34 ꢁC. LC (k236): purity
98.2%. MSD (ESI, +ve): m/z 359.2 [M+HꢁCH3OCOOH]+,
283.3 [M+Hꢁ2· CH3OCOOH]+, 265.1 (M+Hꢁ2·
6.14. 7a-Hydroxyandrost-5-ene-3,17-dione 3,17-diethyl-
ene diketal (29)
Sodium borohydride (0.4 g, 10.5 mmol) was added to
the mixture of androst-5-ene-3,7,17-trione 3,17-diethyl-
ene diketal27 (2.0 g, 5.1 mmol) in a mixture of methanol
and DCM (2:3, 50 ml). The mixture was stirred at room
temperature for 30 min, poured into ice water and the
product was extracted with DCM. LC–MS analysis
indicated the presence of 7b- and 7a-hydroxy com-
1
CH3OCOOHꢁH2O)+; H NMR (200 MHz, CDCl3): d

5944
P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
pounds in 60:40 ratio. The 7a-isomer 29 was separated
by column chromatography on silica gel using 15% ace-
tone in petroleum ether as eluent, and was crystallized
from methanol. LC (k205): purity 95.4%. MSD (ESI,
+ve): m/z 413.2 [M+Na]+, 373.1 [M+HꢁH2O]+, 311.1
[M+HꢁH2OꢁHOCH2CH2OH]+, 249.1 [M+HꢁH2Oꢁ
(0.010 g) was added and the mixture was stirred at room
temperature for 1 h. The reaction mixture was poured
into ice water and the compound was extracted with
ether. The organic layer was washed with water several
times followed by brine and dried over anhydrous mag-
nesium sulfate. Evaporation of the solvent under vacu-
um below 35 ꢁC gave the crude solid. Crystallization
from methanol gave 4.25 g (85%) ADEK (46), mp
105–6 ꢁC. LC (k205): purity 99.7%. MSD (ESI, +ve):
1
2· HOCH2CH2OH]+; H NMR (200 MHz, CDCl3): d
5.57 (dd, J = 5.13, 2.0 Hz, 1H, 6-H), 3.95 (m, 4H,
OCH2–CH2O), 3.9 (m, 5H, OCH2–CH2O + 7b-H), 1.0
(s, 3H, 19-CH3), 0.87 (s, 3H, 18-CH3).
m/z
395.2
[M+Na]+,
373.2
[M+H]+,
313.2
[M+HꢁAcOH]+, 311.2 (M+HꢁHOCH2CH2OH)+,
251.2 (M+HꢁAcOHꢁHOCH2CH2OH)+; 1H NMR
(400 MHz, CDCl3): d 5.86 (dd, J = 10.4, 2.0 Hz, 1-H,
1-H), 5.5 (m, 2H, 2-H+6-H), 5.25 (m, 1H, 3a-H), 3.90
(m, 4H, OCH2–CH2O), 2.06 (s, 3H, OAc), 1.11 (s, 3H,
19-CH3), 0.88 (s, 3H, 18-CH3); 13C NMR (400 MHz,
CDCl3) d 170.9 (CO-Ac), 137.99 (C-5), 138.0, 125.5,
122.9 (CH’s-olefinic), 72.2 (C-3), 119.5 (C-17), 65.4,
64.8 (CH2-ketal), 50.8, 46.7,32.2 (CH’s), 36.1, 34.4,
30.9, 30.7, 22.9, 20.7 (CH2’s), 21.9, 21.6, 14.5 (CH3’s).
6.15. 17b-Propionyloxyandrost-5-ene-3,7-dione 3-ethyl-
ene ketal (33)
The compound was prepared from testosterone 17b-
propionate (31), which was ketalized at position 3 using
ethylene glycol, toluene, and toluene-p-sulfonic acid to
give 17b-propionyloxyandrost-5-en-3-one 3-ethylene
ketal (32). 1H NMR (200 MHz, CDCl3): d 5.34 (t,
J = 2.0 Hz, 6-H), 4.6 (t, J = 7.8 Hz, 17a-H), 3.9 (m,
4H, OCH2–CH2O), 2.32 (q, J = 7.2 Hz, 2H, CH2), 1.12
(t, J = 7.2 Hz, 3H, CH3), 1.02 (s, 3H, 19-CH3), 0.80 (s,
3H, 18-CH3).
6.19. 3b-Acetoxyandrosta-1,5-dien-17-one (47)
To a solution of 46 (0.44 g) in acetone–water (40 ml, 9:1)
p-TSA (0.12 g) was added and the mixture was stirred at
room temperature for 16 h. Acetone was removed and
the residue was diluted with some ice water and neutral-
ized to pH 6 using 5% aqueous sodium bicarbonate
solution. On further cooling, the separated white solid
was collected, washed with water, and dried (0.38 g,
98%). Product 47 crystallized from methanol, mp 185–
87 ꢁC. LC (k205): purity 99.8%. MSD (ESI, +ve): m/z
351.2 [M+Na]+, 329.2 [M+H]+, 269.2 [M+HꢁAcOH]+,
251.1 (M+HꢁAcOHꢁH2O)+; 1H NMR (400 MHz,
CDCl3): d 5.86 (dd, J = 10.0, 2.0 Hz, 1H, 1-H), 5.5 (m,
J = 10.0, 1.6 Hz, 2H, 2-H+6-H), 5.25 (m, 1H, 3a-H),
2.07 (s, 3H, OAc), 1.13 (s, 3H, 19-CH3), 0.91 (s, 3H,
18-CH3); 13C NMR (400 MHz, CDCl3) d 137.5, 125.9,
122.4 (CH’s-olefinic), 72.1 (C-3), 52.0, 47.0, 31.5
(CH’s), 36.1, 36.0, 31.6, 30.4, 22.0, 20.6 (CH2’s), 21.9,
21.6, 13.9 (CH3’s).
A stirred mixture of 32 (2.0 g, 5.2 mmol), N-hydroxyph-
thalimide (0.84 g, 5.2 mmol), and benzoyl peroxide
(0.02 g) in acetone (50 ml) was refluxed. Air was bubbled
into the refluxing solution for 8 h. Acetone was removed
and residue was taken up in toluene. After cooling at
10 ꢁC for 20 min, phthalimide was filtered out and the
organic solution was washed twice with saturated sodium
bicarbonate and subsequently with water and brine.
Compound 33 was crystallized from ethyl acetate-hexane,
yield 58%, mp 258–9 ꢁC. LC (k240): purity 99.8%. MSD
(ESI, +ve): m/z 425.2 [M+Na]+, 403.3 [M+H]+, 329.2
[M+HꢁC2H5COOH]+, 267.0 [M+HꢁC2H5COOHꢁ
HOCH2CH2OH]+; 1H NMR (200 MHz, CDCl3): d 5.67
(d, J = 1.7 Hz, 1H, 6-H), 4.6 (dd, J = 7.1, 1.7 Hz, 1H,
17a-H), 3.96 (m, 4H, OCH2–CH2O), 2.32 (q, J =
7.5 Hz, 2H, CH2), 1.14 (t, J = 7.5 Hz, 3H, CH3), 1.21 (s,
3H, 19-CH3), 1.1 (s, 3H, 18-CH3); 13C NMR (300 MHz,
CDCl3) d 201.1 (CO-17), 175.0 (CO-Propionyl), 165.0
(C-5), 126.4 (C-6), 108.8 (C-3), 81.7 (C-17), 64.6, 64.5
(CH2-ketal), 49.6, 44.9, 44.87 (CH’s), 41.8, 35.9,
35.6, 31.0, 27.8, 27.6, 25.9, 20.7 (CH2’s), 17.0, 12.1, 9.3
(CH3’s).
6.20. 3b-Hydroxyandrosta-1,5-dien-17-one (48)
Potassium carbonate (0.3 g) was added to a solution
of 47 (0.25 g) in methanol–water (15 ml, 9:1) and the
mixture was stirred at room temperature for 16 h.
Methanol was removed and the residue was diluted
with ice water and neutralized to pH 6 using 1 N
HCl solution. The solution on further cooling gave a
white solid, which was filtered, washed with water,
dried (0.18 g, 83%), and the product was crystallized
from aqueous methanol. Mp 138–40 ꢁC. LC (k205):
purity 99.8%. MSD (ESI, +ve): m/z 309.2 [M+Na]+,
6.16. 17b-Acetoxyandrost-4-ene-3,6-dione (35)
The compound was synthesized by the oxidation of tes-
tosterone acetate with tert-butyl hydroperoxide and
household bleach solution as reported27 earlier.
6.17. Steroidal compounds 36–42 were purchased from
Steraloids Inc.
287.2
[M+H]+,
269.1
[M+HꢁH2O]+,
251.2
[M+Hꢁ2· H2O]+; 1H NMR (400 MHz, CDCl3): d
5.78 (dd, J = 10.0, 2.0 Hz, 1H, 1-H), 5.57 (dt,
J = 10.0 Hz, 1H, 2-H), 5.45 (t, J = 2.0 Hz, 1H, 6-H),
4.2 (bt, 1 H, 3a-H), 1.12 (s, 3H, 19-CH3), 0.91 (s,
3H, 18-CH3); 13C NMR (400 MHz, CDCl3) d 221.0
(C-17), 139.4 (C-5), 136.0, 130.0, 121.3 (CH’s-olefinic),
69.9 (C-3), 52.1, 47.2, 31.6 (CH’s), 40.6, 36.0, 31.6,
30.4, 22.0, 20.6 (CH2’s), 22.1, 13.9 (CH3’s).
6.18. 3b-Acetoxyandrosta-1,5-dien-17-one 17-ethylene
ketal (46, ADEK)
3b-Hydroxyandrosta-1,5-dien-17-one 17-ethylene ketal
(45) was synthesized as described.28 To a cooled mixture
of crude 45 (4.5 g, 13.6 mmol) in pyridine (10 ml) and
acetic anhydride (4.0 ml), dimethylaminopyridine

P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
5945
6.21. 3b-Acetoxyandrosta-1,5-dien-17b-ol (49)
finic), 84.7 (Cꢂ), 83.5 (C-17), 75.1 (ꢂCH), 72.2 (C-3),
49.4, 46.4, 32.3 (CH’s), 37.6, 36.1, 33.1, 31.0, 23.8, 20.9
(CH2’s), 21.88, 21.69, 21.59, 13.7 (CH3’s).
To a solution of 47 (0.1 g) in methanol (20 ml) sodium
borohydride (0.05 g) was added and the mixture was
stirred at room temperature for 0.5 h. The solution
was diluted with ice water and neutralized to pH 7 with
dilute acetic acid. Methanol was evaporated and the
compound was extracted with ethyl acetate–petroleum
ether mixture (8:2), washed with water, and dried over
magnesium sulfate. Solvent was removed completely
and the residue was crystallized from methanol to ob-
tained pure product 49 (0.08 g, 80%). 1H NMR
(400 MHz, CDCl3): d 5.87 (dd, J = 10.0, 2.0 Hz, 1H,
1-H), 5.5 (m, 2H, 2-H+6-H), 5.25 (m, 1H, 3a-H), 3.65
(t, J = 8.4 Hz, 1H, 17a-H), 2.07 (s, 3H, 3-OAc), 1.12
(s, 3H, 19-CH3), 0.78 (s, 3H, 18-CH3); 13C NMR
(400 MHz, CDCl3) d 171.0 (CO-Ac), 138.1 (C-5),
138.0, 125.6, 122.9 (CH’s-olefinic), 82.0 (C-17), 72.2
(C-3), 51.6, 47.0, 32.0 (CH’s), 36.7, 36.1, 31.1, 30.7,
23.6, 20.9 (CH2’s), 21.9, 21.6, 11.3 (CH3’s).
6.24. Androsta-5,16-dien-3b-ol (HAD, 53)
A mixture of DHEA (7.0 g, 24 mmol) and tosylhydrazine
(0.3 g, 50 mmol) in methanol (300 ml) was refluxed for
24 h, concentrated to half, cooled, and the product 52
crystallized out. The white crystalline material was fil-
tered, washed with cold methanol, and dried to obtain
pure product in 73% yield (8.5 g). Additional quantities
were obtained from the mother liquor by crystallization
1
from aqueous methanol. H NMR (200 MHz, CDCl3):
d 7.83 (d, J = 8.4 Hz, 2H, Ar-H), 7.28 (d, J = 8.5 Hz,
2H, Ar-H), 5.33 (d, J = 4.5 Hz, 1H, 6-H), 3.51 (m, 1H,
3a-H), 2.43 (s, 3H, CH3), 1.01 (s, 3H, 19-CH3), 0.8 (s,
3H, 18-CH3).
To a solution of 52 (1.0 g, 2.2 mmol) in tetrahydrofuran
(30 ml), lithium aluminum hydride (1.5 g) was added
and the mixture was refluxed for 20 h. The mixture
was cooled and the excess of hydride was decomposed
with moist ether and water. The organic solution was
washed with water, brine and dried over MgSO4. Crys-
tallization from methanol afforded 53 in 80% yield
(0.48 g), mp 135–37 ꢁC (lit. 137 ꢁC31). LC (k205): purity
99.8%. MSD (ESI, +ve): m/z 273.3 [M+H]+, 255.2
[M+HꢁH2O]+; 1H NMR (200 MHz, CDCl3): d 5.86
(m, 1H, 17-H), 5.72 (m, 1H, 16-H), 5.38 (d, J =
5.1 Hz, 1H, 6-H), 3.5 (m, 1H, 3a-H), 1.05 (s, 3H, 19-
CH3), 0.80 (s, 3H, 18-CH3).
6.22. 3b,17b-Dihydroxy-17a-ethynylandrosta-1,5-diene (50)
A mixture of 3b-acetoxyandrosta-1,5-dien-17-one (47,
0.95 g, 2.9 mmol) and DMSO (20 ml) was stirred at
room temperature under nitrogen and treated with an
18% suspension of sodium acetylide in xylene (12 ml).
The mixture was stirred for 2 h then poured into cold
brine and extracted with ether (3 · 30 ml). The com-
bined extracts were washed with water, dried and con-
centrated to 5 ml. The crude product 50 was obtained
by diluting with petroleum ether. It was redissolved in
DCM, filtered, and concentrated to give pure product
50 (0.86 g, 96%), mp 215–17 ꢁC. LC (k205): purity
99.4%. MSD (ESI, +ve): m/z 335.2 [M+Na]+, 313.2
[M+H]+, 295.2 [M+HꢁH2O]+, 277.2 [M+Hꢁ2· H2O]+;
1H NMR (400 MHz, CDCl3 + DMSO): d 5.56 (d,
J = 10.0 Hz, 1H, 1-H), 5.37 (dd, J = 10.0 Hz, 1H, 2-H),
5.19 (d, J = 2.8 Hz, 1H, 6-H), 3.95 (bt, 1H, 3a-H), 2.36
(d, J = 1.6 Hz, 1H, acetylenic-H), 0.91 (s, 3H, 19-CH3),
0.67 (s, 3H, 18-CH3); 13C NMR (400 MHz, CDCl3 +
DMSO) d 139.5 (C-5), 135.5, 130.4, 121.1 (CH’s-olefin-
ic), 88.5 (Cꢂ), 79.3 (ꢂCH), 73,4 (C-17), 69.3 (C-3), 50.9,
46.6, 32.5 (CH’s), 40.4, 39.1, 32.6, 30.9, 23.2, 20.8
(CH2’s), 22.1, 12.9 (CH3’s).
6.25. 17b-Hydroxyandrosta-3,5-dien-7-one (54)
A solution of 3b,17b-dihydroxyandrost-5-en-7-one (24,
0.25 g, 0.82 mmol) in methanol (10 ml) was stirred with
perchloric acid (70% solution, 0.5 ml) at room tempera-
ture for 10 h. The mixture was diluted with cold water
and neutralized with sodium bicarbonate. The precipi-
tated solid was extracted with ethyl acetate–petroleum
ether (9:1), washed with water, and dried. Solvent was
evaporated to dryness and the residue was stirred with
cold ether. The precipitated solid was filtered, washed
with water, and dried to give 0.19 g (83%) of 54, mp
170–72 ꢁC (lit.48 170–72 ꢁC). LC (k285): purity 97.3%.
MSD (ESI, +ve): m/z 309.1 [M+Na]+, 287.2 [M+H]+;
1H NMR (200 MHz, CDCl3): d 7.26 (m, 2H, 3-H+4-
H), 5.61 (s, 1H, 6-H), 3.66 (t, J = 8.2 Hz, 1H, 17a-H),
1.14 (s, 3H, 19-CH3), 0.80 (s, 3H, 18-CH3).
6.23. 3b,17b-Diacetoxy-17a-ethynylandrosta-1,5-diene
(51)
A mixture of compound 50 (0.12 g, 0.38 mmol) and ace-
tic anhydride (0.8 ml) containing 1.0 mol% toluene-p-
sulfonic acid was heated in the conventional microwave
oven (925 W) for 3 min at high power setting.29 A usual
work-up yielded the product 51 in quantitative yield
(0.15 g), mp 155–57 ꢁC. LC (k205): purity 95.4%. MSD
(ESI, +ve): m/z 419.2 [M+Na]+, 359.2 [M+Naꢁ
AcOH]+, 337.2 [M+HꢁAcOH]+, 277.2 [M+Hꢁ2·
AcOH]+; 1H NMR (400 MHz, CDCl3): d 5.88 (dd,
J = 10.0, 2.0 Hz, 1H, 1-H), 5.48 (m, 2H, 2-H+6-H), 5.24
(m, 1H, 3a-H), 2.59 (s, 1H, acetylenic-H), 2.07 (s, 3H,
OAc), 2.05 (s, 3H, OAc), 1.12 (s, 3H, 19-CH3), 0.9 (s,
3H, 18-CH3); 13C NMR (400 MHz, CDCl3) d 170.9,
169.8 (CO-Ac), 138.0 (C-5), 137.9, 125.7, 122.8 (CH’s-ole-
6.26. 17b-Acetoxyandrosta-3,5-dien-7-one (55)
Acetylation of compound 54 was performed in a con-
ventional microwave oven (925 W) with acetic anhy-
dride (3 mol equiv) containing 0.5 mol% p-TSA, in
40 s at high power setting. Yield 90%, mp 217–19 ꢁC.
LC (k280): purity 99.6%. MSD (ESI, +ve): m/z 351.2
[M+Na]+, 329.2 [M+H]+, 269.2 [M+HꢁAcOH]+, 251.1
1
[M+HꢁAcOHꢁH2O]+; H NMR (200 MHz, CDCl3):
d 6.15 (m, 2H, 3-H+4-H), 5.61 (s, 1H, 6-H), 4.62 (dd,
J = 9.0 Hz, 1H, 17a-H), 2.05 (s, 3H, OAc), 1.13 (s, 3H,
19-CH3), 0.85 (s, 3H, 18-CH3).

5946
P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
6.27. Steroidal compounds 56–58 and 63 were purchased
from Steraloids Inc.
References and notes
1. Huggins, C. Physiol. Rev. 1945, 25, 281.
2. Hellerstedt, B. A.; Pienta, K. J. CA: A Cancer J. Clinicians
2002, 52, 154.
3. Singh, S. M.; Gauthier, S.; Labrie, F. Curr. Med. Chem.
2000, 7, 211.
6.28. 3b,7b-Dihydroxy-16,17-seco-androst-5-ene-16,17-
dioic acid (16 ! 7) lactone (61)
A solution of iodine (3.5 g) in methanol (80 ml) and a
solution of potassium hydroxide (2.54 g) in 50% aqueous
methanol (80 ml) were added to a solution of 3b,7b-
dihydroxyandrost-5-en-17-one49 (59, 0.85 g, 2.8 mmol)
in methanol (80 ml) under vigorous stirring. Both solu-
tions were added from separate funnels over a period of
5 h in such a manner that a slight excess of iodine was
maintained. The mixture was stirred at room temperature
for another 16 h. Methanol was removed and the residue
was diluted with ice water and extracted twice with ether.
The cool aqueous layer was acidified with 5 N sulfuric
acid and once again extracted with ether. The title com-
pound (61) was isolated as the major compound (0.42 g,
45%) from the aqueous layer on cooling, mp 235–45 ꢁC
(decomp.). The ether extracts (containing free acid and
mono-methyl ester) were combined and stirred with
potassium hydroxide in methanol. The residue on similar
work-up and acidification with 5 N sulfuric acid provided
additional quantities of the lactone 61. LC (k205): purity
99.0%. MSD (ESI, ꢁve): m/z 333.1 [M-H]ꢁ, 287.1 [M-
HꢁHCOOH]ꢁ; MSD (ESI, +ve): m/z 357.2 [M+Na]+,
4. Lunglmayr, G. Anti-Cancer Drugs 1995, 6, 508.
5. Chen, C. D.; Welsbie, D. S.; Tran, C.; Baek, S. H.; Chen,
R.; Vessella, R.; Rosenfeld, M. G.; Sawyers, C. L. Nat.
Med. 2004, 10, 33.
6. Veldscholte, J.; Ris-Stalpers, C.; Kuiper, G. G. J. M.;
Jenster, G.; Berrevoets, C.; Claassen, E.; van Rooij, H. C.
J.; Trapman, J.; Brinkmann, A. O.; Mulder, E. Biochem.
Biophys. Res. Commun. 1990, 173, 534.
7. Furutani, T.; Takeyama, K.; Koutoku, H.; Ito, S.; Tanig-
uchi, N.; Suzuki, E.; Kudoh, M.; Shibasaki, M.; Shikama,
H.; Kato, S. Biosci. Biotechnol. Biochem. 2005, 69, 2236.
8. Marwah, A.; Marwah, P.; Lardy, H. J. Chromatogr. B
2002, 767, 285.
9. Miyamoto, H.; Yeh, S.; Lardy, H.; Messing, E.; Chang, C.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 11083.
10. Miyamoto, H.; Chang, C. Int. J. Urol. 2000, 7, 32.
11. Mizokami, A.; Koh, E.; Fijita, H.; Maeda, Y.; Egawa, M.;
Koshida, K.; Honma, S.; Keller, E. T.; Namiki, M. Cancer
Res. 2004, 64, 765.
12. Coleman, D. L.; Schwizer, R. W.; Leiter, E. H. Diabetes
1984, 33, 26.
13. Ben-David, M.; Dikstein, S.; Bismuth, G.; Sulman, F. G.
Proc. Soc. Exp. Biol. Med. 1967, 125, 1136.
14. Loria, R. M.; Inge, T. H.; Cook, S. S.; Szakal, A. K.;
Regelson, W. J. Med. Virol. 1988, 26, 301.
15. Schwartz, A. G. Cancer Res. 1979, 39, 1129.
16. Shi, J.; Schulze, S.; Lardy, H. Steroids 2000, 65, 124.
17. Lardy, H.; Marwah, A.; Marwah, P. Vitam. Horm. 2005,
71, 263.
335.2
[M+H]+,
317.3
[M+HꢁH2O]+,
289.2
[M+HꢁHCOOH]+, 271.3 [M+HꢁH2OꢁHCOOH]+; 1H
NMR (200 MHz, CD3OD): d 5.4 (s, 1H, 6-H), 4.6 (d,
J = 9.1 Hz, 1H, 7a-H), 3.46 (m, 1H, 3a-H), 1.14 (s, 3H,
19-CH3), 1.1 (s, 3H, 18-CH3).
18. Lardy, H. Med. Hypotheses 2006, 66, 107.
19. Numazawa, M.; Shelangouski, M.; Nakakoshi, M.
Steroids 2001, 66, 743–748.
6.29. 3b,13a-Dihydroxy-17,13-seco-androst-5-ene-17-
carboxylic acid (62)
20. Levy, H.; Jacobson, R. P. J. Biol. Chem. 1947, 171, 71.
21. Stavely, H. E. J. Am. Chem. Soc. 1939, 61, 79.
22. Herzog, H. L.; Jevnik, M. A.; Tully, M. E.; Hershberg, E.
B. J. Am. Chem. Soc. 1953, 75, 4425.
23. Marwah, P.; Lardy, H. A.; Marwah, A. K. US Patent
6,274,746, 2001.
24. Marwah, P.; Lardy, H. A. PCT Int. Application WO 99/
47485, 1999.
25. Marwah, P.; Lardy, H. A. US Patent 6,384,251,
2002.
26. Dictionary of Steroids; Hill, R. A., Kirk, D. N., Makin, H.
L. J., Murphy, G. M., Eds.; Chapman & Hall: New York,
1991; p 58.
27. Marwah, P.; Marwah, A.; Lardy, H. A. Green Chem.
2004, 6, 570.
28. Kaneko, C.; Sugimoto, A.; Yamada, S.; Ishikawa,
M.; Sasaki, S.; Suda, T. Chem. Pharm. Bull. 1974,
22, 2101.
29. Marwah, P.; Marwah, A.; Lardy, H. A. Tetrahedron 2003,
59, 2273.
3b-acetoxy-17a-oxa-D-homo-androst-5-en-17-one
(5,
0.2 g, 0.58 mmol) in methanol (20 ml) was refluxed with
1 N sodium hydroxide solution (1 ml) for 1 h. Methanol
was evaporated under vacuum and the residue was
diluted with brine. The solution was adjusted to pH 5
with acetic acid and the solid was extracted with dichlo-
romethane–methanol (9:1), washed with brine, and
dried. The organic solvent was evaporated and the
residue taken up in ether and the white solid was
collected on a filter. Mp 245–47 ꢁC, LC (k205): purity
90.4%. MSD (ESI, +ve): m/z 345.2 [M+Na]+, 327.2
[M+NaꢁH2O]+, 305.2 [M+HꢁH2O]+, 287.3 [M+Hꢁ2·
H2O]+, 269.1 [M+Hꢁ3· H2O]+, 251.2 [M+HꢁCH2@CH
1
COOH]+; H NMR (400 MHz, CDCl3): d 5.3 (s, 1H,
6-H), 3.45 (m, 1H, 3a-H), 1.09 (s, 3H, 19-CH3), 0.95
(s, 3H, 18-CH3); 13C NMR (400 MHz, CDCl3 + DMSO,
3:1) d 181.6 (CO), 145.7 (C-5), 125.6 (C-6), 75.8
(C-3), 77.5 (C-13), 57.8, 54.1, 41.8 (CH’s), 46.9, 46.8,
42.0, 39.7, 37.4, 36.3, 28.7, 27.6 (CH2’s), 25.9, 24.2
(CH3’s).
30. Caglioti, L.; Magi, M. Tetrahedron 1963, 19, 1127.
31. Ohloff, G.; Maurer, B.; Winter, B.; Giersch, W. Helv.
Chim. Acta 1983, 66, 192.
32. Marwah, A.; Marwah, P.; Lardy, H. A. Bioorg. Chem.
2002, 30, 233.
33. Mitchell, R. E.; Stocklin, W.; Stefanovic, M.; Geissman,
T. A. Phytochemistry 1971, 10, 411.
Acknowledgment
34. Druey, J. Angew. Chem. 1960, 72, 677.
35. Belkin, M.; Fitzgerald, D. B. J. Natl. Cancer Inst. 1953,
13, 889.
This work was supported in part by Hollis-Eden Phar-
maceuticals, Inc.

P. Marwah et al. / Bioorg. Med. Chem. 14 (2006) 5933–5947
5947
36. Chang, H.-C.; Miyamoto, H.; Marwah, P.; Lardy, H.;
Yeh, S.; Huang, K.-E.; Chang, C. Proc. Natl. Acad. Sci.
U.S.A. 1999, 96, 11173.
37. Miyamoto, H.; Marwah, P.; Marwah, A.; Lardy, H.;
Chang, C. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
4440.
38. Miyamoto, H.; Marwah, P.; Marwah, A.; Yang, Z.;
Chung, C.-Y.; Altuwaijri, S.; Chang, C.; Lardy, H. Int. J.
Cancer 2005, 117, 866.
39. Negro-Vilar, A. J. Clin. Endocrinol. Metab. 1999, 84,
3459.
40. Matias, P. M.; Donner, P.; Coelho, R.; Thomaz, M.;
Peixoto, C.; Macedo, S.; Otto, N.; Joschko, S.; Scholz, P.;
Wegg, A.; Basler, S.; Schafer, M.; Egner, U.; Carrondo,
M. A. J. Biol. Chem. 2000, 275, 26164.
41. Sack, J. S.; Kish, K. F.; Wang, C.; Attar, R. M.; Kiefer, S.
E.; An, Y.; Wu, G. Y.; Scheffler, J. E.; Salvati, M. E.;
Krystek, S. R.; Weinmann, R.; Einspahr, H. M. Proc.
Natl. Acad. Sci. U.S.A. 2001, 98, 4904.
42. Schneider, J. J. J. Biol. Chem. 1970, 245, 5505.
43. Moffett, R. B.; Weisblat, D. I. J. Am. Chem. Soc. 1952, 74,
2183.
44. Glazier, E. R. J. Org. Chem. 1962, 27, 4397.
45. Okada, M.; Fukushima, D. K.; Gallagher, T. F. J. Biol.
Chem. 1959, 234, 1688.
46. Marwah, P.; Marwah, A.; Kneer, N.; Lardy, H. A.
Steroids 2001, 66, 581.
47. Zeng, Y.; Li, Y. J. Org. Chem. 2003, 68, 1603.
48. Weintraub, P. M.; Tiernan, P. L.; Benson, H. D.; Grunwell,
J. F.; Johnston, J. O’N.; Petrow, V. J. Med. Chem. 1976, 19,
1395.
49. Matsuzaki, Y.; Yoshida, S.; Honda, A.; Miyazaki, T.;
Tanaka, N.; Takagiwa, A.; Fujimoto, Y.; Miyazaki, H.
Steroids 2004, 69, 817.