A versatile synthesis of 17-heteroaryl androstenes via palladium-mediated Suzuki cross-coupling with heteroaryl boronic acids

Article abstract of DOI:10.1016/j.steroids.2006.02.004

Suzuki coupling of 17-iodoandrosta-5,16-dien-3β-ol (1) and 17-iodoandrosta-4,16-dien-3-one (2) with nine heteroaryl boronic acids (mainly 2- or 3-furanyl, thienyl, benzofuranyl and benzothienyl boronic acid derivatives) were carried out under normal Suzuki condition (Pd(PPh₃)₄, 2 M Na₂CO₃ and MeOH), generally yielded C₁₇-heteroaryl steroids in moderate (10-60%) yields, but furanyl-2- and 5-chlorothienyl-2-boronic acid did not give any coupling product.

Full text of DOI:10.1016/j.steroids.2006.02.004

steroids 7 1 ( 2 0 0 6 ) 585–590
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/steroids
A versatile synthesis of 17-heteroaryl androstenes via
palladium-mediated Suzuki cross-coupling with
heteroaryl boronic acids
a
b,∗
c
d
e
Jianping Chao , Yangzhi Ling , Xiaofeng Liu , Xuande Luo , A.M.H. Brodie
a
School of Materials and Chemical Engineering, Beijing Institute of Petrochemical Technology, Beijing 102600, China
Department of Medicinal Chemistry, School of Pharmaceutical Science, Peking University, Beijing 100083, China
School of Pharmaceutical Science, Peking University, Beijing 100083, China
National Institutes of Pharmaceutical Research and Development, Beijing 100226, China
b
c
d
e
Department of Pharmacology& Experimental Therapeutics, School of Medicine, University of Maryland, Baltimore, MD 21201, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Suzuki coupling of 17-iodoandrosta-5,16-dien-3␤-ol (1) and 17-iodoandrosta-4,16-dien-3-
one (2) with nine heteroaryl boronic acids (mainly 2- or 3-furanyl, thienyl, benzofuranyl
and benzothienyl boronic acid derivatives) were carried out under normal Suzuki condition
(Pd(PPh ) , 2 M Na CO and MeOH), generally yielded C17-heteroaryl steroids in moderate
Received 17 January 2006
Received in revised form 14
February 2006
3
4
2
3
Accepted 16 February 2006
Published on line 29 March 2006
(10–60%) yields, but furanyl-2- and 5-chlorothienyl-2-boronic acid did not give any coupling
product.
©
2006 Elsevier Inc. All rights reserved.
Keywords:
Suzuki coupling
17-Iodoandrosta-5,16-dien-3␤-ol or
4,16-dien-3-one
Furanyl or thienyl boronic acids
derivatives
1
7-Heteroaryl
androsta-5,16-diene-3␤-ol or
,16-dien-3-one
4
1
.
Introduction
gen, dihydrotestosterone (DHT), in the prostate. Inhibitors of
these enzymes have uses in the treatment of prostatic dis-
eases [1]. To date, only ketoconazole [2,3], an imidazole anti-
fungal agent, has been used for this purpose to treat patients
with advanced prostatic cancer. However, this agent is nei-
ther selective nor very potent and has a number of significant
side effects. Finasteride, an inhibitor of 5␣-R, has been recently
used as a treatment for BPH [4]. Finasteride only reduces DHT
levels by −70% in these patients, but testosterone levels are
often increased [5]. Although this is not a problem for BPH
Androgens have been implicated in the development and pro-
gression of common disorders of the prostate, most notably,
benign prostatic hyperplasia (BPH) and prostatic cancer. Two
important enzymes in the biosynthesis of androgens are 17␣-
hydroxylase/C17,20-lyase (P45017␣) which regulates an early
step in the biosynthesis of testosterone (T) and other andro-
gens in both the testes and adrenal gland, and 5␣-reductase
(
5␣-R), which converts testosterone to the more potent andro-
∗
Corresponding author.
E-mail address: lingyzhi@126.com (Y. Ling).
0039-128X/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2006.02.004

5
86
steroids 7 1 ( 2 0 0 6 ) 585–590
patients, it could result in growth stimulation of prostate can-
cers, since testosterone may bind to the androgen receptor, in
the absence of DHT. Therefore, the development of new types
of potent enzyme inhibitors that inhibit P45017␣, as well as the
pyridyl)boronate or diethyl(3-pyridyl)borane to synthesize 17-
pyridyl androstenes. But these boronates have to be prepared
in the lab [12,13]. Now, more and more heteroaryl boronic
acids are becoming commercially available and have been
established as an important reagent in the Suzuki coupling
reaction [14]. Here we wish to describe a preliminary report
to utilize commercially available heteroaryl boronic acids via
the palladium-mediated Suzuki cross-coupling reaction, to
synthesize the 17-heteroaryl androstenes. The reaction was
depicted in the following Scheme 1.
5␣-R, could be effective in the treatment of prostatic cancer by
achieving total androgen blockade. This new type of inhibitors
would be a promising candidate in prostatic cancer therapy [6].
Recently, 17-heteroaryl substituted androstenes (3 and 4)
have been found to have good inhibition activity against
ꢀ
P45017␣ [6]. We and others have reported that 17-(2 -
ꢀ
ꢀ
imidazolyl), 4 -imidazolyl, 3 -pyrazolyl [1], as well as 1H-
imidazolyl and 1H-1,2,3-triazolyl-androsta-5,16-dien-3␤-ol [7]
ꢀ
were very potent inhibitors for P45017␣. 3 -Pyridyl-androsta-
5
2.
Experimental
ꢀ
,16-dien-3␤-ol is already in clinical trials [6] and 17-(5 -
isoxazolyl)-androsta-4,16-dien-3-one (L-39) which was both
active against P45017␣and 5␣-R will soon enter phase I clini-
cal trials [7]. From our experience, a 16,17-double bond would
enhance inhibition activity, whereas the basic steroidal skele-
ton of 4-en-3-one, such as L-39, would also have inhibition
activity against 5␣-R [6].
Pd(PPh3)4 and all nine heteroaryl boronic acids (mainly 2-
or 3-furanyl, benzofuranyl, thienyl and benzothienyl boronic
acid, etc. as shown in Table 1) were purchased from Aldrich
without further purification. 17-Iodoandrosta-5,16-dien-3␤-ol
(1) was prepared by Potter’s procedure as reported [11] with
some modifications by us. Oppenauer oxidation of (1) gave
17-iodoandrosta-4,16-dien-3-one (2) in good yield. 1H NMR
spectra were recorded with a JOEL-AL300 or Bruker 500 instru-
ment. FAB-MS were recorded on ZAB-HF-3FMS spectrometer,
and ESI-MS were recorded on MDS SCIEX QSTAR spectrometer,
Elemental analyses were performed on a Varo EL III elemen-
tal analyser. Melting points were determined on TX4 melting
point apparatus fitted with a microscope and are uncorrected.
One of the approaches to synthesize 17-heteroaryl
androstene was cyclo-condensation of the appropriate con-
structed steroidal 17-side chain with some reagents. For
ꢀ
examples, 4 -imidazole was synthesized via 21-acetoxy preg-
nenolone with formaldehyde and ammonia (Weigenhagen
ꢀ
method), and 2 -imidazole was formed via 17-carboxaldehyde
ꢀ
ꢀ
with glyoxal and NH4OH [1]; Gabriel condensation gave 5 -
substituted thiazole and Hantzch procedure yielded the 4 -
substituted thiazole [8], while 21-bromo pregnenolone with
thiourea afforded 2 -aminothiazole steroids [9]. But these
2.1.
Representative procedure
ꢀ
cyclo-condensation approaches were not convenient and
sometimes were rather tedious and gave the low yield, espe-
cially when there was 16,17-double bond. For example, 2 -
The synthesis of 17-iodoandrosta-5,16-dien-3␤-ol (1) and 17-
iodoandrosta-4,16-dien-3-one (2).
ꢀ
imidazole was only obtained in less than 5% yield, although it
was a very potent inhibitor [1].
17-Iodoandrosta-5,16-dien-3ˇ-ol (1)
Into a 100 ml round-bottomed flask, fitted with a mag-
netic stirrer bar, was placed dehydroepiandrosterone (DHEA)
(1.783 g, 6.19 mmol) and ethanol (40 ml), and then hydrazine
hydrate (2.0 ml, 40 mmol) was added, followed by a solu-
tion of hydrazine sulfate (6 mg, 0.038 mmol) in water (2 ml).
After stirring at room temperature for 4 days (TLC showed
the reaction completed), the mixture was poured into cold
water (80 ml) with stirring. The precipitate was dried at
room temperature, dissolved in dioxane (38 ml), and triethy-
lamine (7 ml, 51.8 mmol) added with stirring. Iodine (3.20 g,
25.2 mmol) was added in portions over 30 min and stirring
continued for a further 1 h. The mixture was poured into
9% Na SO (30 ml), and precipitate collected by filtration,
As part of our ongoing medicinal chemistry program to
search for more active P45017␣ and 5␣-R inhibitors, we are
anxious to explore a convenient synthesis method which
would introduce 17-heteroaryl groups into the androstene
skeleton directly. In particular, we required a versatile and
simple method of varying the 17-heteroaryl moiety in order
to prepare scaffolds for chemical libraries. We noticed that
the recent work described by Haidar et al. [10], who used
lithium trimethoxy (5-pyrimidyl) boronate in the presence of
bis-(triphenylphospphine) palladium(II) chloride as catalyst,
ꢀ
to introduce the 5 -pyrimidyl group into the 17-position of
androstene, and the approach was in a similar fashion as
described by Potter et al. [11] who used lithium trimethoxy(4-
2
3
washed with cold water (3 × 20 ml), and recrystallized from
Scheme 1

steroids 7 1 ( 2 0 0 6 ) 585–590
587
ethanol to give white needle crystals (1) (1.516 g, 62.0%) mp
◦
◦
1
74–176 C (lit. [11] 175–176 C).
17-Iodoandrosta-4,16-dien-3-one (2)
From a solution of (1) (200 mg, 0.503 mmol) in dry toluene
25 ml) and cyclohexanone (4 ml, 38.6 mmol) was distilled off
(
part of the solvent (3–5 ml) to eliminate moisture. After allow-
◦
ing to cool to 90 C, Al(O-i-Pr)3 (220 mg, 1.12 mmol) was added
and the mixture heated under reflux for 8 h, and then allowed
to cool, poured into water (25 ml) with stirring. The organic
solution was separated out and the aqueous solution was
extracted with ether (3 × 25 ml). The combined organic solu-
tion was washed with water (3 × 20 ml) and brine (20 ml),
and then dried over anhydrous Na2SO4, and concentrated.
Chromatography, on elution with EtOAc/Petroleum ether (PE)
(
1/4, v/v), afforded 2 (198 mg). Recrystallized from acetone-
PE to give 2, crystals (150 mg, 75.4%). mp 165–168 C. 1H
◦
NMR(CDCl3): ı (ppm) = 6.12 (m, 1H, 16-H), 5.72 (s, 1H, 4-H), 1.19
(
s, 3H, 19-CH3), 0.76 (s, 3H, 18-CH3). Anal. Calcd for (C19H25OI):
C, 57.58; H, 6.36. Found: C, 57.58; H, 6.49.
2
1
.1.1. General procedure for Suzuki coupling with
7-iodoandrost-5,16-dien-3ˇ-ol (1)
A 25 ml flask equipped with a magnetic stirring bar and fitted
with a condenser was charged with 17-iodoandrost-5,16-dien-
3␤-ol (1) (199 mg, 0.50 mmol), Pd(PPh3)4 (58 mg, 0.05 mmol),
heteroaryl boronic acid (0.50 mmol), benzene (10 ml), MeOH
(
2 ml) and 2 M Na2CO3 (0.5 ml), and flushed with nitrogen. The
◦
reaction was monitored by TLC. After being stirred at 80 C for
the period shown in Table 1, the reaction mixture was treated
with water (20 ml) at room temperature. The product was
extracted with Et2O (3 × 20 ml). The combined ether extracts
were dried over MgSO4. Flash column chromatography (FCC),
eluted with EtOAc-PE (1/5, v/v) was employed to separate the
product. The following compounds were prepared unless oth-
erwise noted.
ꢀ
2
1
(
1
.1.1.1. 17-(3 -Furanyl)androsta-5,16-dien-3ˇ-ol (Entry 1). mp
◦
◦
1
87–189 C (lit. [9] 186–189 C). H NMR (500 MHz, CDCl3) ı
ꢀ
ꢀ
ppm): 7.47 (s, 1H, 2 -H), 7.36 (t, 1H, J = 1.8 Hz, 5 -H), 6.48 (dd,
H, J = 1.8 Hz, 4 -H), 5.83 (m, 1H, ꢀ -H), 5.38 (m, 1H, ꢀ -H), 3.53
ꢀ
16
6
(
m, 1H, 3␣-H), 1.08 (s, 3H, 19-CH3), 0.96 (s, 3H, 18-CH3).
ꢀ
2
2
(
1
.1.1.2. 17-(3 -Thienyl)androsta-5,16-dien-3ˇ-ol (Entry 3). mp
◦
◦
1
16–219 C (lit. [9] 215–219 C). H NMR (500 MHz, CDCl3) ı
ꢀ
ꢀ
ppm): 7.24 (s, 1H, 2 -H), 7.23 (d, 1H, J = 0.6 Hz, 5 -H), 7.15 (dd,
H, J = 0.6 Hz, 4 -H), 5.91 (m, 1H, ꢀ -H), 5.37 (m, 1H, ꢀ -H), 3.52
ꢀ
16
6
(
m, 1H, 3␣-H), 1.04 (s, 3H, 19-CH3), 1.00 (s, 3H, 18-CH3).
ꢀ
2
1
(
.1.1.3. 17-(2 -Thienyl)androsta-5,16-dien-3ˇ-ol (Entry 4). mp
◦
◦
1
92–195 C (lit. [9] 190–195 C). H NMR (300 MHz, CDCl3) ı
ꢀ
ꢀ
ppm): 7.14 (d, J = 0.3 Hz, 1H, 5 -H), 7.04 (d, 1H, J = 0.2 Hz, 3 -H),
.97 (dd, 1H, J = 0.3, 0.2 Hz, 4 -H), 5.98 (m, 1H, ꢀ16-H), 5.39 (m,
H, ꢀ -H), 3.55 (m, 1H, 3␣-H), 1.07 (s, 3H, 19-CH3), 1.00 (s, 3H,
ꢀ
6
1
1
6
8-CH3).
ꢀ
2
6
.1.1.4. 17-(2 -Benzofuranyl)androsta-5,16-dien-3ˇ-ol (Entry
◦
1
). mp 133–135 C. H NMR (500 MHz, CDCl3) ı: 7.49 (d,
J = 0.6 Hz, 1H, 5 -H), 7.46 (d, 1H, J = 0.3 Hz, 8 -H), 7.30 (dd, 1H,
J = 0.3, 0.2 Hz, 7 -H), 7.20 (dd, 1H, 0.6,0.2 Hz, 6 -H), 6.72 (s, 1H,
-H), 6.14 (m, 1H, ꢀ16-H), 5.40 (m, 1H, ꢀ -H), 3.53 (m, 1H,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
5
3

5
88
steroids 7 1 ( 2 0 0 6 ) 585–590
ꢀ
16
4
3
(
␣-H), 1.09 (s, 3H, 19-CH3), 1.02 (s, 3H, 18-CH3). FAB-MS: 389
1H, 3 -H), 6.38 (m, 1H, ꢀ -H), 5.76 (m, 1H, ꢀ -H), 1.28 (s, 3H,
M + H+).
+
19-CH ), 1.03 (s, 3H, 18-CH ). ESI-MS: 387 (M + H ).
3
3
ꢀ
ꢀ
2.1.2.4. 17-(2 -Benzothienyl)androsta-4,16-dien-3-one (Entry
2
.1.1.5. 17-(2 -Benzothienyl)androsta-5,16-dien-3ˇ-ol (Entry 7).
◦
1
16). mp 203–205 C. 1H NMR (300 MHz, CDCl3) ı: 7.76 (d, 1H,
◦
mp 210–212 C. H NMR (300 MHz, CDCl3) ı: 7.75 (d, J = 1.8 Hz,
H, 8 -H), 7.70 (t, 1H, J = 1.1 Hz, 5 -H), 7.32 (dd, 1H, J = 1.8, 1.1 Hz,
-H), 7.27 (s, 1H, 3 -H), 7.23 (dd, 1H, J = 1.8, 1.1 Hz, 7 -H), 6.12
m, 1H, ꢀ16-H), 5.40 (m, 1H, ꢀ -H), 3.55 (m, 1H, 3␣-H), 1.10 (s,
H, 19-CH3), 1.05 (s, 3H, 18-CH3). Anal. Calcd. for (C27H32OS) C
0.15%, H 7.97%; Found: C 79.93%, H 7.84%.
ꢀ
ꢀ
ꢀ
ꢀ
1
6
J = 1.8 Hz, 8 -H), 7.67 (t, 1H, J = 1.1 Hz, 5 -H), 7.33 (dd, 1H, J = 1.8,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
1.1 Hz, 6 -H), 7.27 (s, 1H, 3 -H), 7.22 (dd, 1H, J = 1.8, 1.1 Hz, 7 -H),
6.12 (m, 1H, ꢀ16-H), 5.76 (m, 1H, ꢀ -H), 1.25 (s, 3H, 19-CH3),
1.10 (s, 3H, 18-CH3). Anal. Calcd. for (C27H30OS): C 80.55, H
7.51; Found: C 80.63, H 8.00.
5
4
(
3
8
ꢀ
ꢀ
2
.1.1.6. 17-(3 -Benzothienyl)androsta-5,16-dien-3ˇ-ol (Entry 8).
2.1.2.5. 17-(3 -Benzothienyl)androsta-4,16-dien-3-one
◦
1
(entry17). mp 230–232 C. 1H NMR (500 MHz, CDCl3) ı: 7.90 (d,
◦
1H, J = 1.2 Hz, 8 -H), 7.85 (t, 1H, J = 1.0 Hz, 5 -H), 7.39 (s, 1H, 2 -H),
mp 209–211 C. H NMR (500 MHz, CDCl3) ␦ (ppm): 7.91 (d,
J = 1.2 Hz, 1H, 8 -H), 7.85 (d, 1H, J = 1.0 Hz, 5 -H), 7.38 (s, 1H, 2 -H),
.34 (dd, 1H, J = 1.0, 0.04 Hz, 6 -H), 7.32 (dd, 1H, J = 1.2, 0.04 Hz,
-H), 6.02 (m, 1H, ꢀ16-H), 5.42 (m, 1H, ꢀ -H), 3.56 (m, 1H, 3␣-
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
7
7
7.37 (d, 1H, J = 1.0 Hz, 6 -H), 7.32 (dd, 1H, J = 1.2, 0.1 Hz, 7 -H),
6.01 (m, 1H, ꢀ16-H), 5.77 (m, 1H, ꢀ -H), 1.22 (s, 3H, 19-CH3),
1.05 (s, 3H, 18-CH3). FAB-MS: 403 (M + H ).
ꢀ
5
4
+
H), 1.09 (s, 3H, 19-CH3), 1.04 (s, 3H, 18-CH3). Anal. Calcd. for
(
C27H32OS): C 80.15, H 7.97; Found: C 79.93, H 7.84.
ꢀ ꢀ
.1.2.6. 17-[1 -(Phenylsulfonyl)-3 -indolyl]androsta-4,16-dien-
2
◦
1
ꢀ
ꢀ
3-one (Entry 18). mp 155–158 C. H NMR (500 MHz, CDCl ) ı:
2
3
.1.1.7. 17-[1 -(Phenylsulfonyl)-3 -indolyl]androsta-5,16-dien-
3
ꢀ ꢀꢀ
8.01 (d, J = 8.5 Hz, 1H, 2 -H), 7.88 (d, J = 7.5 Hz, 2H, 2 -H), 7.71 (d,
◦
1
ˇ-ol (Entry 9). mp 249–251 C. H NMR (300 MHz, CDCl3)
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀꢀ
1H, J = 8.0 Hz, 4 -H), 7.55–7.52 (m, 2H, 3 -H), 7.46–7.43 (m, 2H,
ı: 8.02 (d, J = 8.1 Hz, 1H, 2 -H), 7.88–7.85 (m, 2H, 2 -H), 7.71
(
2
1
1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
4 ,7 -H), 7.35–7.32 (m, 1H, 6 -H), 7.27–7.24 (m,1H, 5 -H), 6.15 (s,
d, 1H, J = 6.9 Hz, 4 -H), 7.55–7.50 (m, 2H,3 -H), 7.46–7.41 (m,
1H, ꢀ16-H), 5.77 (s, 1H, ꢀ -H), 1.26 (s, 3H, 19-CH ), 1.04 (s, 3H,
18-CH ). FAB-MS: 526 (M + H ).
4
ꢀ
ꢀ
ꢀ
ꢀ
H, 4 ,7 -H), 7.35–7.22 (m, 2H, 5 , 6 -H), 6.15 (dd, 1H, J = 3.0,
3
+
.8 Hz, ꢀ16-H), 5.40 (d, 1H, J = 5.4 Hz, ꢀ5-H), 3.53 (m, 1H, 3␣-H),
3
.09 (s, 3H, 19-CH3), 1.02 (s, 3H, 18-CH3). Anal. Calcd. for
(
2
C33H37O3SN): C 75.14, H 7.02, N 2.66; Found: C 75.23, H 6.87,
.61.
3.
Results and discussion
The 17-iodoandrostene (1) was prepared by the method of
Potter et al. [11] reported in 1997 with the following improve-
ments. Dehydroepiandorstone was treated with hydrazine to
give the 17-hydrazone which on treatment with iodine in the
presence of 1,1,3,3-tetramethylquanidine. However, this two
step procedure was tedious and 1,1,3,3-tetramethylquanidine
expensive. We then discovered that the 17-hydrazone could be
decomposed directly with I2 and triethylamine to give similar
yields and the procedure was both simple and economical.
The Suzuki cross-coupling of heteroaryls, although a pro-
cess of great synthetic potential [15,16], has not been used to
its full extent due to the unavailability of the corresponding
heteroaryl boronic acids. Hence, our experiments only start on
those heteroaryl boronic acids which are commercially avail-
able. They are mainly furan and thiophene derivative as shown
in Table 1.
2
1
.1.2. General procedure for Suzuki coupling with
7-iodoandrost-4,16-dien-3-one (2)
A 25 ml flask equipped with a magnetic stirring bar and a
condenser, was charged with 17-iodoandrost-4,16-dien-3-one
(
100 mg, 0.25 mmol), Pd(PPh3)4 (30 mg, 0.025 mmol), and het-
eroaryl boronic acid (0.25 mmol), benzene (6 ml), MeOH (2 ml)
and 2 M Na2CO3 (0.5 ml), and flushed with nitrogen. the reac-
tion was monitored by TLC. After being stirred at 80 C for
the period shown in the Table 1, the reaction mixture was
treated with water (10 ml) at room temperature. The product
was extracted with Et2O (3 × 10 ml). The combined Et2O solu-
tion was dried over MgSO4. Flash column chromatography and
eluted with EtOAc-PE (1/5, v/v) was employed to separate the
products.
◦
ꢀ
2
.1.2.1. 17-(3 -Furanyl)androsta-4,16-dien-3-one (Entry 10).
Palladium-mediated reaction of 17-iodoandrostenes (1)
and (2) with the boronic acid derivatives under the Suzuki
arylation condition (cat. Pd (PPh3)2, Na2CO3/MeOH) mostly led
to the desired cross-coupled product in moderate yield (see
◦
1
mp 110–111 C. H NMR (500 MHz, CDCl3) ı(ppm): 7.47 (s, 1H,
ꢀ
ꢀ
ꢀ
2
5
-H), 7.37 (t, 1H, J = 1.8 Hz, 5 -H), 6.48 (dd, 1H, J = 1.8 Hz, 4 -H),
.83 (m, 1H, ꢀ16-H), 5.75 (m, 1H, ꢀ -H), 1.08 (s, 3H, 19-CH3), 0.96
4
+
ꢀ
ꢀ
ꢀ
(
s, 3H, 18-CH3). ESI-MS: 337 (M + H ).
Table 1). 17-(3 -Furanyl)-, (3 or 2 -thienyl)-androsta-5,16-dien-
␤-ol (Entries 1, 3 and 4) were identical with that reported by
3
ꢀ
2
.1.2.2. 17-(2 -thienyl) androsta-4,16-dien-3-one (Entry 13).
Burkhart et al. [9], which were obtained by reacting with a
17-keto steroid with lithiofuran or lithiothiophene and dehy-
dration of the resulting tertiary alcohol in two steps. Although
the total yields were similar, Suzuki coupling was direct and
convenient. Generally, the 5-ene-3␤-ol (1) gave better yields
than the 4-ene-3-one (2) (compare Entries 1–10; 7–16; 8–17
and 9–18). Among these coupling reactions 3-furan boronic
acids afforded the highest yield (60%, Entry 1), whereas 2-furan
boronic acids failed to give any coupling product (Entries 2
and 11), however, 2-benzofuran boronic acid did give the cou-
◦
1
mp 139–141 C. H NMR (500 MHz, CDCl3) ı: 7.15 (d, J = 5.1,
ꢀ
ꢀ
3
3
1
.0 Hz, 1H, 5 -H), 7.04 (d, 1H, J = 3.0 Hz, 3 -H), 6.97 (dd, 1H, J = 5.1,
.0 Hz, 4 -H), 5.97 (m, 1H, ꢀ16-H), 5.75 (s, 1H, ꢀ -H), 1.22 (s, 3H,
ꢀ
4
+
9-CH3), 1.05 (s, 3H, 18-CH3). ESI-MS: 353 (M + H ).
ꢀ
2
1
.1.2.3. 17-(2 -Benzofuranyl)androsta-4,16-dien-3-one (Entry
◦
1
5). mp 214–218 C. H NMR (500 MHz, CDCl3) ı: 7.52 (d,
ꢀ
ꢀ
J = 0.6 Hz, 1H, 5 -H), 7.48 (d, 1H, J = 0.3 Hz, 8 -H), 7.26 (dd, 1H,
J = 0.3, 0.2 Hz, 7 -H), 7.16 (dd, 1H, J = 0.6, 0.2 Hz, 6 -H), 6.83 (s,
ꢀ
ꢀ

steroids 7 1 ( 2 0 0 6 ) 585–590
589
Scheme 2
pling product albeit in poor yield (ca.10%, Entries 6 and 15).
The 2-and 3-thienyl and benzothienyl boronic acids, as well
as 3-indolyl boronic acid yielded the Suzuki product in about
position to heteroatom, This is the preliminary report and the
further work on Suzuki coupling of steroids with imidazole or
pyrimidine boronic acids is on going.
3
0% yield. Surprisingly, the 4-ene-3-one (2) did not yield any
of the desired product with 3-thienyl boronic acid (Entry 12) or
-chloro-2-thienyl boronic acid (Entry 14). These results can
be explained as indicated below:
5
Acknowledgement
This work was Supported by the National Science Foundation
of China, (No. 3017110, China).
3
.1.
The deiodination of 17-iodoandrostenes (1) and
(
2)
r e f e r e n c e s
Recently, dehalogenation has been reported in the Suzuki
reaction [17,18] and we observed that the deiodination reac-
tion side reaction was rather large in our case because the iodo
compound is not stable. Quite a lot of the 17-deiodo byprod-
uct was recovered in almost every reaction. For example, up to
[1] Ling YZ, Li J, Liu Y, Kato K, Clus GT, Brodie AMH.
17-Imidazolyl, pyrazolyl, and isoxazolyl androstene
derivatives. Novel steroidal inhibitors of human
cytochrome C17,20-lyase(P45017␣). J Med Chem
2
5% and 18% yield of the 17-deiodo compound was recovered
in Entries 6 and 13, respectively, A possible mechanism of its
formation is shown in Scheme 2.
1997;40(20):3297–304.
[
2] Trachtenberg J. Ketoconazole therapy in advanced
prostatic cancer. J Urol 1984;132(1):61–3.
3
.2.
The deboronation of heteroaryl boronic acid
[3] Williams G, Kerle DJ, Ware H, Doble A, Dunlop H, Smith C,
et al. Objective responses to ketoconazole therapy in
patients with relapsed progressive prostate cancer. Br J
Urol 1986;58:45–51.
The protodeboronation was recognized as a problem for het-
eroaryl boronic acids when the boronic acid was in the ortho
position to heteroatom. 2-Pyridine boronic acid is unstable
especially in protic solvents, while its 3- and 4-pyridyl species
are stable [19]. This would explain the coupling reaction with
furan boronic acid. Furan-3-boronic acid gave a highest yield
[
4] Stoner E. The clinical development of a 5␣-reductase
inhibitor, finasteride. J Steroid Biochem Mol Biol
1990;37(3):375–8.
[5] Geller J. Effect of finasteride, a 5␣-reductase inhibitor of
prostate tissue androgen and prostate specific antigen. J
Clin Endocrinol Metab 1990;71:1552–5.
(
60%) in our case (Entry 1), while its 2-boronic acid failed
[
6] Njar VCO, Brodie AMH. Inhibitors of 17␣-hydroxylase
C17,20-Lyase (CYP17): potential agents for the treatment of
prostate cancer. Curr Pharm Des 1999;5(3):163–80.
to yield any product as the boronic acid in 2-furanboronic
acid is adjacent to oxygen. But benzofurano-2-boronic acid,
in which the benzene ring might disperse the electronega-
tive effect of oxygen, did give 10% yield. Because the electron
withdrawing effect of sulfur is less then oxygen, 2-thienyl
boronic acid did yield a coupling product in moderate yield
[
7] Handratta VD, Jelovac D, Long BJ, Kataria R, Nnane IP, Njar
VCO, et al. Potent CYC17 inhibitors: improved syntheses.
Pharmacokinetics and anti-tumor activity in the LNCaP
human prostate cancer model. J Steroid Biochem Mol Biol
2
004;92:155–65.
(
30%, Entries 4 and 13) while a strong electron withdrawing
[
[
8] Zhu N, Ling YZ, Lei XP, Handratta V, Brodie AMH. Novel
chlorine at the 5-position (Entries 5 and 14) failed to yield
any coupled product except the deiodo steroid (7) after 22 h
reflux. If the reaction proceeded at room temperature for 45 h
only trace amounts of coupled product could be detected
on TLC.
ꢀ
P45017␣inhibitor: 17-(2 -oxazolyl)androstene derivatives.
Steroids 2003;68:603–11.
9] Burkhart JP, Gates CA, Laughlin ME, Resvick RJ, Peer NP.
Inhibitor of steroid C17(20) lyase with C-17-heteroaryl
steroid. Bioorg Med Chem 1996;4(9):1411–20.
In summary, we have utilized herteroaryl boronic acids
via palladium-mediated Suzuki cross-coupling reaction with
[10] Haidar S, Ehmer PB, Hartmann RW. Novel steroidal
pyrimidyl inhibitors of P450 17(17␣-hydroxylase
C17,20-lyase). Pharm Pharm Med Chem 2001;334:373–4.
1
7-iodoandrostene and found that it would be a simple and
[
11] Potter GA, Hardcastle IR, Jarm M. A convenient, large-scale
synthesis of abiraterone
convenient alternative for introduce heteroaryl group into 17-
position of steroid. The yields were moderate (from 10% to
acetate[3␤-acetoxy-17-(3-pyridyl)androsta-5, 16-diene]. A
potential new drug for the treatment of prostate cancer.
Org Prep Proc Int 1997;29(1):123–34.
6
0%) although it has not been optimized. The yield might be
poor to nil, especially when the boronic acid is at the ortho

5
90
steroids 7 1 ( 2 0 0 6 ) 585–590
[
[
12] Cai WL, David H, Brown R. A practical synthesis of
pyridylboranes via magnesium-halogen exchange. Synlett
[17] Ghosez L, Franc C, Denonne F, Cuisinier C, Touillaux R.
Studies of palladium-catalyzed coupling reaction for
preparation of hindered 3-arylpyrroles relevant to
(−)-rhazinilam and its analogues. Can J Chem
2001;79:1827–39.
2
002;2:273–4.
13] Potter GA, Barrie SE, Jarman M, Rowlands MG. Novel
steroidal inhibitors of human cytochrome
P45017␣(17␣-hydroxylase C17,20-lyase): potential agents for
the treatment of prostatic cancer. J Med Chem
[18] Handy ST, Bregman H, Lewis J, Zhang XL, Zhang Y. An
unusual dehalogenation in the suzuki coupling of
4-bromopyrrole-2-carboxylates. Tetrahedron Lett
2003;44:427–30.
[19] Ishiyama T, Ishida K, Miyaura N. Synthesis of pinacol
arylboronates via cross-coupling reaction of
bis(pinacolato)diboron with chloroarene catalyzed by
palladium(0)–tricyclohexylphosphine complexes.
Tetrahedron 2001;57:9813–6.
1
995;38(13):2463–71.
[
[
14] Tyrrel E, Brookes P. The synthesis and application of
heterocyclic boronic acids. Synthesis 2003;4:469–83.
15] Kalinin VN. Carbon-cabon bond formation in heterocycles
using Ni- and Pd-catalyzed reactions. Synthesis
1
992:413–32.
[
16] Undheim K, Benneche T. Metallation and metal-assisted
bond formation in ␲-electron deficient hetereocycles. Acta
Chem Scand 1993;47:102–21.