6
92
Z.-X. Jiang et al. / Steroids 70 (2005) 690–693
2
.7. 3-Oxo-4-aza-5α-androst-1-ene-17β-(N-tert-
and the precipitate was stirred until it hydrolyzed to jelly,
butylcarboxamide) (1)
the yield was 74.3%. Subsequently, this product was reacted
with tert-BuOH in concentrated sulfuric and acetic acids, and
6 was converted to 3-oxo-4-azaandrosta-5,16-diene-17-(N-
tert-butylcarboxamide) 7 in 79.9% yield. The introduction
of 17-(N-tert-butyl-carboxamide) was convenient and gave
a higher yield.
A 50 mL three neck round bottom flask equipped with
a nitrogen inlet, reflux condenser, magnetic stirrer, and a
septum inlet was charged with 8 (1.0 g, 2.67 mmol), DDQ
(1.0 g) and dioxane (7 mL). BSTFA (3 mL) was added to
this suspension with a syringe while stirring. Heating the
mixture at reflux for 10 min yielded a clear solution. The
solution was refluxed for 18 h at the end of which com-
plete disappearance of starting material was observed by
TLC. The suspension was cooled to room temperature and
added to a mixture of 30 mL CH2Cl2 and 6 mL of 1% aque-
oussodiumbisulfitesolution, precipitatingthehydroquinone,
which was separated by filtration. The CH2Cl2 layer of the
filtrate was separated, extracted twice with 25 mL of 5%
NaHSO3, concentrated to thick oil, and triturated with di-
ethyl ether (50 mL). The solid was isolated and dried, yielding
Subsequently, the reduction of 7 was carried out over
Pd/C in C2H OH in the presence of hydrogen to give 3-
5
oxo-4-aza-5␣-androstane-17-(N-tert-butylcarboxamide) 8.
Under these conditions, with the steric hindrance of the 19-
CH3 group, the 16-ene-17-(N-tert-butylcarboxamide) was
nearly reduced to 17-(N-tert-butylcarboxamide), and the
reduction of the 5-ene had high selectivity for the 5␣-H
(5-H: 5␣-H < 2%). Finally, dehydrogenation of 8 was
carried out by treatment with a suitable dehydrogenat-
ing agent, such as chloranil, benzeneseleninic anhydride,
or dichlorodicyanobenzoquinone (DDQ), in an anhydrous
solvent, such as chlorobenzene, dioxane, xylene, toluene,
or benzene, and optionally, in the presence of BTSFA
[bis(trimethylsilyl)trifluoro acetamide]. To optimize the
yield, we chose DDQ and BSTFA in dioxane to dehydro-
genate 8 in order to give finasteride 1 in 70.5% yield [21].
Using this sequence, finasteride was synthesized from 4-
androstene-3,17-dione in 18.6% overall yield.
◦
white analytical sample 1 (0.70 g, 70.5%): mp = 253–255 C,
◦
−1
Ref. [1] mp = 252–254 C. IR (KBr) (cm ): 3429, 3240,
3
1
2
1
1
1
114, 2967, 2936, 1687, 1668, 1599, 1450, 1383, 1364, 1255,
219, 1127, 814. H MNR (CDCl3) δ (ppm): 6.78 (d, 2H,
-H), 5.80 (dd, 2H, 1-H), 5.37 (brs, 1H, 4-NH), 5.07 (brs,
H, 17-CONH), 3.32 (m, 1H, 5␣-H), 2.14 (dd, 1H, H-17),
.35 (s, 9H, C(CH3)3), 0.97 (s, 3H, 19-CH3), 0.69 (s, 3H,
8-CH3).
1
References
3
. Results and discussion
[
1] Rasmusson GH, Reynolds GF, Utne T, Jobson RB, Primka RL,
Berman C, et al. Azasteroids as inhibitors of rat prostatic 5␣-
reductase. J Med Chem 1984;27(12):1690–701.
4
-Androstene-3,17-dione 2 is an important intermediate
for the synthesis of steroid since it is commercially avail-
able in high purity. Oxidation of 4-androsten-3,17-dione has
been reported to yield 5,17-dioxo-A-nor-3,5-secoandrostan-
[
2] Rasmusson GH, Reynolds GF, Steinberg NG, Walton E, Patel GF,
Liang T, et al. Azasteroids: structure–activity relationships for in-
hibition of 5␣-reductase and of androgen receptor binding. J Med
Chem 1986;29:2298–315.
3
-oic acid 3 via two methods: (1) oxidation with Von Rudloff
solution in 81.5% yield [15] and (2) ozonolysis with ozone
in 93.0% yield [16]. For large-scale manufacture, we chose
method (1) to oxidize 2 for the synthesis of 3. The ring closure
of 3 was carried out with ammonium acetate in acetic acid
at high temperature to give 4-azaandrost-5-ene-3,17-dione
[3] Rasmusson GH, Reynolds GF. 17beta-substituted-4-aza-5alpha-
androstenones and their use as 5alpha-reductase inhibitors. Merck
&
Co. Inc., USA. Europe Patent 85-155096, 18 September 1985.
[
[
[
[
[
[
4] Heinzl S. Androgenetic alopecia: finasteride treated hair loss. Med
Monatsschr Pharm 1999;22(4):124–7.
5] Tosti A, Piraccini BM. Finasteride and the hair cycle. J Am Acad
Dermatol 2000;42(5 Pt 1):848–9.
6] Coltman CA, Thompson IM, Feigl P. Prostate cancer prevention trial
4
[
, which was simpler than the previously reported methods
1,12].
In order to introduce 17-(N-tert-butylcarboxamide), we
(PCPT) update. Eur Urol 1999;35(5–6):544–7.
7] Li XJ, Chen LG, Fang F, Wang XJ. Graphical synthetic routes of
finasteride. Chin J Pharm 2001;32(5):245–7.
8] Zheng JH, Xu F, Liao QJ. New synthesis of finasteride. Chin J Pharm
2003;34(3):105–7.
9] Zheng JH, Xu F, Liao QJ. Synthesis of finasteride, a new drug
for the treatment of begin prostatic hyperplasia. Chin J Med Chem
chose the Ritter reaction by using ␣,-unsaturated carboni-
trile to react with tert-BuOH in concentrated sulfuric and
aceticacids[10,11]. SinceGascandNedelecreportedthesuc-
cessful introduction of the 17-cyano-17␣-hydroxyl group
in 3,3-ethylene-dioxy-19-norandtosta-5(10),9(11)-dien-17-
one [17], the cyanohydrin reaction has been thoroughly inves-
tigated [18,19]. In our approach, 4 was selectively converted
to 4-aza-5-androstene-17␣-hydroxy-17-cyano-3-one 5 in
1996;6(3):203–6.
[
10] Ritter JJ, Minieri PP. A new reaction of nitriles. I. Amides from
alkenes and mono-nitriles. J Am Chem Soc 1948;70:4045–8.
[11] Plaut H, Ritter JJ. A new reaction of nitriles. VI. Unsaturated amides.
J Am Chem Soc 1951;73:4076–7.
[12] Solomons WZ, Doorenbos NJ. Synthesis and antimicrobial proper-
ties of 17--amino-4-aza-5␣-androstane and derivatives. J Pharm Sci
9
1
1.4% yield. Dehydration of 5 with POCl3 in pyridine at
20 C gave 4-aza-androst-5,16-diene-17-cyano-3-one 6, a
◦
small difference as described in Ref. [20]. The warm mixture
1974;63(1):19–23.
◦
was cooled to about 60–70 C, then dropped with vigorous
[
13] Zoltan T, Judit H, Janos S, Laszlo K, Gabor B. Novel process
stirring into a mixture of ice-water and 37% aqueous HCl,
for preparing 17-substituted-4-azaandrostane derivatives. Richter