Derivatives of 16α-hydroxy-dehydroepiandrosterone with an additional 7-oxo or 7-hydroxy substituent: Synthesis and gas chromatography/mass spectrometry analysis

Article abstract of DOI:10.1016/j.steroids.2005.04.002

Derivatives of 16α-hydroxy-dehydroepiandrosterone, which have an additional oxygen substituent at position 7 (oxo or hydroxy group), were synthesized. Firstly, 17,17-dimethoxyandrost-5-ene-3β,16α-diyl diacetate was prepared and then oxidized with a complex of chromium(VI) oxide and 2,5-dimethylpyrazole to the respective 7-oxo derivative. This key intermediate was both deprotected or reduced by l-Selectride or sodium borohydride in the presence of cerium(III) chloride and then deprotected to give 7-oxo, 7α-hydroxy and 7β-hydroxy derivatives of 16α-hydroxy-dehydroepiandrosterone. The target compounds were characterized by ¹H and ¹³C NMR spectra and in the form of O-methyloxime-trimethylsilyl derivatives, by gas chromatography/mass spectrometry methods.

Full text of DOI:10.1016/j.steroids.2005.04.002

Steroids 70 (2005) 739–749
Derivatives of 16␣-hydroxy-dehydroepiandrosterone with an
additional 7-oxo or 7-hydroxy substituent: Synthesis and gas
ଝ
chromatography/mass spectrometry analysis
a
a,∗
b
b
b
ˇ
Vladim ´ı r Pouzar , Ivan Cern y´ , Martin Hill , Marie Bi cˇ ´ı kov a´ , Richard Hampl
a
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Prague, Czech Republic
b
Institute of Endocrinology, Prague, Czech Republic
Received 13 December 2004; received in revised form 6 April 2005; accepted 6 April 2005
Available online 23 May 2005
Abstract
Derivativesof16␣-hydroxy-dehydroepiandrosterone, whichhaveanadditionaloxygensubstituentatposition7(oxoorhydroxygroup), were
synthesized. Firstly, 17,17-dimethoxyandrost-5-ene-3␤,16␣-diyl diacetate was prepared and then oxidized with a complex of chromium(VI)
oxide and 2,5-dimethylpyrazole to the respective 7-oxo derivative. This key intermediate was both deprotected or reduced by l-Selectride®
or sodium borohydride in the presence of cerium(III) chloride and then deprotected to give 7-oxo, 7␣-hydroxy and 7␤-hydroxy derivatives
1
13
of 16␣-hydroxy-dehydroepiandrosterone. The target compounds were characterized by H and C NMR spectra and in the form of O-
methyloxime-trimethylsilyl derivatives, by gas chromatography/mass spectrometry methods.
©
2005 Elsevier Inc. All rights reserved.
Keywords: Steroid; Synthesis; Dehydroepiandrosterone; NMR spectra; GC–MS
1
. Introduction
DHEA metabolite 3␤-hydroxyandrost-5-ene-7,17-dione (7-
oxo-DHEA). The latter steroid was shown to act as a potent
thermogenic agent by influencing oxidative metabolism in
mitochondria in a similar way as thyroid hormones; see e.g.,
[3] and other references of these authors. 7-oxo-DHEA is also
an intermediate in the epimerization of 7␣-hydroxy-DHEA
to the 7␤-isomer and vice versa, which proceeds in mam-
malian liver and also in a non-enzymatic way in a mild acidic
milieu [4].
On the other hand, 16␣-hydroxy-DHEA represents a prod-
uct of a concurrent enzymatic reaction. 16␣-Hydroxylation
of DHEA and other C19- and C18 steroids occurs not only in
liver, but in many other tissues. 16␣-Hydroxy-DHEA is the
precursor of fetal estrogens and has been detected in the um-
bilical blood of newborns [5]. Since the development of its
first immunoassay, 16␣-hydroxy-DHEA has been considered
a normal constituent of human serum where it is present in a
lownanomolarrange [6]. Increased levelsof16-hydroxylated
estrogens have been associated with an increased risk of
cancer of the female reproductive system and, more recently,
with some systemic autoimmune diseases. Because 16␣-
Metabolites of dehydroepiandrosterone (DHEA), hydrox-
ylated in position 7 such as 7␣-hydroxy-DHEA, its 7␤-
epimer, and the respective androst-5-ene-3␤,7␰,17␤-triols,
are now believed to possess immunomodulatory and im-
munoprotective effects, as demonstrated by a number of re-
ports published in the last two decades [1]. These compounds
are formed in various mammalian tissues by the action of
two 7-hydroxylating enzymes of the cytochrome P450 series,
differing in their characteristics and tissue- and species local-
ization; see ref. [1] and other references therein. Of particu-
lar importance are the actions of these steroids in the brain
[2]. All of these steroids are present in nanomolar concen-
trations in human blood, together with another 7-oxygenated
ଝ
Part of this work was presented at the 39th Conference “Advances in
Organic, Bioorganic and Pharmaceutical Chemistry”, Nymburk, Czech Re-
public, November 26–28, 2004.
∗
Corresponding author.
ˇ
E-mail address: cerny@uochb.cas.cz (I. Cern y´ ).
0
039-128X/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2005.04.002

7
40
V. Pouzar et al. / Steroids 70 (2005) 739–749
1
3
hydroxy-DHEA is a precursor of the latter steroids, we hy-
pothesized that increased formation of 16␣-hydroxy-DHEA
could be a risk factor in the development of these diseases;
in other words, it may counteract the beneficial effects of
instruments ( C at 125.7 MHz) under the above conditions;
secondary referencing was performed using the solvent sig-
nal at position δ(CDCl3) = 77.0 and δ(CD3OD) = 49.0.
Thin-layer chromatography (TLC) was performed on sil-
ica gel G (ICN Biochemicals). Detection was accomplished
by spraying with concentrated sulfuric acid followed by heat-
ing. PreparativeTLCwasdoneonplates(200 mm × 200 mm)
with 0.4 mm layer thickness. For silica gel G, spraying with
7-hydroxylated DHEA metabolites; for review see [5].
Since both 7␰- and 16␣-hydroxylations proceed as con-
current reactions in the same tissues, one may expect that
metabolites containing both 7- and 16␣-hydroxy groups are
formed and are also present in blood. Their ratio in blood
and eventually in various tissues (liver, brain, sex organs),
may differ in healthy subjects and patients with autoimmune
diseases or disorders of brain function.
0
.1% morine in methanol and UV light (360 nm) visualiza-
tion were used for detection. Conjugate ketones 8 and 9 were
purified on TLC-Silica gel 60 GF254 (Merck, detection by
54 nm UV light). For column chromatography, neutral sil-
2
The isolation of the derivatives of 16␣-hydroxy-DHEA,
which have an additional oxygen substituent at position 7
ica gel Kieselgel 60 (Merck) was used. Prior to evaporation
on a rotary evaporator in vacuo (0.25 kPa, bath temperature
◦
(oxo or hydroxy group), was reported from urine of the pa-
4
0 C), solutions in organic solvents were dried over anhy-
tient with adrenal carcinoma [7]. In this work, however, only
the respective 7-oxo derivative was unambiguously synthe-
sized; for the 7␣-hydroxy derivative, the chemical synthesis
was unsuccessful. To our knowledge, the corresponding 7␤-
hydroxy derivative was nor isolated, nor synthesized.
drous sodium sulfate.
6␣-Bromoandrost-5-en-3␤-ol (2) was prepared accord-
ing to ref. [9].
1
2.2. Derivatives of 16α-hydroxy-DHEA
Published synthetic pathways for the preparation of 16␣-
hydroxy-7-oxo-DHEA differed in the order of introduction of
oxygen substituents into positions 7 and 16 [3,7]. During the
synthesis of 7,16-dihydroxy derivatives, the situation is more
complicated, notonlyduetothesmoothmutualisomerization
of the 7-hydroxy derivatives in acid medium [4], but also due
to the possible isomerization of 16␣-hydroxy-17-oxo deriva-
tives to 17␤-hydroxy-16-oxo derivatives in alkaline medium
2.2.1. 17,17-Dimethoxyandrost-5-ene-3β,16α-diol (3)
(a) Modification of the known procedure [10]. A hot solu-
tion of bromo ketone 2 (2.20 g, 6.0 mmol) in methanol
(65 ml) was added to a refluxed solution of sodium (4.4 g,
191 mmol) in methanol (130 ml), and the refluxing was
continued for 10 min. The reaction mixture was then
poured on ice (ca. 600 g). Methanol and part of the water
were evaporated in vacuo, and the residue was allowed
to stand in the refrigerator overnight. A separated crude
product was collected by suction and washed with water
until the filtrate was neutral. Crystallization from a small
volume (35 ml) of hot acetone afforded 1.43 g (68%) of
[8,9]. This had to be respected in the process of selecting
protecting groups for the synthesis.
As follows from the above analysis, the chemical synthe-
sis of derivatives of 16␣-hydroxy-DHEA with an additional
oxygen substituent at position 7 is a challenging task. Conse-
quently, the main aim of the present work is to develop their
reliable synthesis and unambiguously prove their structure.
The importance of this type of compounds consists in their
use as standards in the broader study of steroid metabolism
disorders.
◦
hydroxy acetal 3, m.p. 178–181 C, [␣]D −67 (c 0.59,
◦
acetone). Literature [10] gives m.p. 177–179 C (ace-
tone). H NMR (400 MHz, CDCl3 + CD3COOD): 5.342
1
ꢀ
(
1 H, bd, J ≈ 5, H-6); 4.285 (1 H, dd, J = 8.9, J = 2.0,
H-16␤); 3.528 (1 H, m, W = 32, H-3␣); 3.459 (3 H, s,
OCH3); 3.367 (3 H, s, OCH3); 0.993 (3 H, s, 3 × H-19);
0.794 (3 H, s, 3 × H-18).
2
. Experimental
(
b) Aqueous KOH (0.4 M, 0.5 ml) was added to a solution
of acetate 4 (44 mg, 0.1 mmol) in tetrahydrofuran (1 ml)
and methanol (0.5 ml) mixture. After stirring at room
temperature for 3 h, the reaction mixture was poured into
ethyl acetate (70 ml) and washed with saturated aqueous
NaCl (2 × 25 ml). The solvent was evaporated in vacuo,
and the residue was crystallized from acetone. The yield
of diol 3, identical with the above described sample, was
27 mg (76%).
2
.1. General
Melting points were determined on a Boetius micro melt-
ing point apparatus (Germany). Optical rotations were mea-
◦
sured at 25 C on an AUTOPOL IV polarimeter (Rudolph
Research Analytical, USA), and [␣]D values are given in
−
1
2
−1
−1
1
0
deg cm g . Infrared spectra (wavenumbers in cm )
1
were recorded on a Bruker IFS 88 spectrometer. H NMR
spectra were taken on Brucker AVANCE-400 and AVANCE-
◦
5
00 instruments (400 and 500 MHz, FT mode) at 23 C in
2.2.2. 17,17-Dimethoxyandrost-5-ene-3β,16α-diyl
diacetate (4)
Diol 3 (701 mg, 2.0 mmol) was dissolved in pyridine
(6.5 ml), and the solution was cooled in an ice bath. Acetic
anhydride (0.97 ml, 10 mmol) was added, and the mix-
CDCl3 (unless stated otherwise) with tetramethylsilane as the
internal standard. Chemical shifts are given in ppm (δ-scale);
coupling constants (J) and widths of multiplets (W) are given
in Hz. 13C NMR spectra were taken on Bruker AVANCE-500

V. Pouzar et al. / Steroids 70 (2005) 739–749
741
ture was left at room temperature overnight. The reaction
mixture was then poured onto ice (100 g), and the sep-
arated product was filtered off and dissolved in benzene
s, CH3COO); 1.050 (3 H, s, 3 × H-19); 0.989 (3 H, s, 3 × H-
1
18). The product was identical (TLC, IR, H NMR) with an
authentic sample [11].
(100 ml). This solution was washed successively withice cold
5
% aqueous citric acid, water, saturated aqueous KHCO3,
2
2
.3. Derivatives of 16α-hydroxy-7-oxo-DHEA
and water. The solvent was evaporated in vacuo, and the
residue was chromatographed on ten preparative silica gel
plates in a mixture of benzene/ether (90:10). The yield of
.3.1. 7,17-Dioxoandrost-5-ene-3β,16α-diyl diacetate
(7)
◦
diacetate 4 was 738 mg (85%), m.p. 177–179 C (ether),
A suspension of chromium(VI) oxide (1.50 g, 15.0 mmol)
[
(
␣]D −110 (c 0.51, CHCl3). IR spectrum (CCl4): 1736
◦
in dichloromethane (12 ml) was stirred at −25 C under argon
with3,5-dimethylpyrazole(1.50 g, 15.6 mmol). After15 min,
a solution of olefin 6 (389 mg, 1.0 mmol) in dichloromethane
C O, acetate); 1670 (C C); 1246, 1033 (C O, acetate);
1
1
167, 1133, 1088, 1069, 1046 (C O C O C). H NMR
(
400 MHz, CDCl3): 5.355 (1 H, bd, J ≈ 5, H-6); 5.218 (1
(2 ml) was added dropwise, and stirring was continued at
ꢀ
H, dd, J = 9.7, J = 3.1, H-16␤); 4.602 (1 H, m, W = 32, H-
␣); 3.488 (3 H, s, OCH3); 3.284 (3 H, s, OCH3); 2.092
3 H, s, CH3COO); 2.031 (3 H, s, CH3COO); 1.011 (3 H, s,
× H-19); 0.908 (3 H, s, 3 × H-18). Analysis calculated for
C H38O (434.6): C, 69.10; H, 8.81. Found: C, 69.39; H,
◦
−
20 C for 4 h. The reaction mixture was then diluted with
3
(
3
a mixture of benzene/ethyl acetate (7:3, 30 ml) and filtered
through a short column of silica gel (10 g) layered with celite.
The column was washed with the same solvent mixture,
and the solvents were evaporated in vacuo. Chromatogra-
phy on a column of silica gel (30 g) in a mixture of ben-
zene/ethyl acetate (95:5 to 90:10) afforded 283 mg (70%)
2
5
6
8
.93.
2
.2.3. 3β,16α-Dihydroxyandrost-5-en-17-one (5)
◦
of ketone 7, m.p. 199–202 C (methanol), [␣]D −73 (c 0.57,
A solution of 4-toluenesulfonic acid monohydrate
◦
CHCl3). The literature [7] gives m.p. 198–200 C (methanol),
(3.9 mg, 0.021 mmol) in water (0.3 ml) was added to a solu-
[
1
1
␣]D −65.3 (CHCl3). IR spectrum (CHCl3): 1733 (C O,
tion of acetal 3 (35 mg, 0.1 mmol) in acetone (2.0 ml), and
the reaction mixture was stirred at room temperature for
3
the residue was dissolved in ethyl acetate (5 ml) and wa-
ter (1 ml). The aqueous phase was extracted with ethyl ac-
etate (3 ml), and the combined organic phases were washed
with water (2 × 2 ml). The solvent was evaporated in vacuo,
and the residue was chromatographed on a preparative sil-
ica gel plate in a benzene/acetone (80:20) mixture. The yield
7-ketone and acetate); 1671 (C O, 7-ketone); 1631 (C C);
1
246, 1036(C O, acetate);1018(C O). HNMR(400 MHz,
0 min. The solvents were then evaporated in vacuo, and
CDCl3): 5.760 (1 H, bd, J = 1.8, H-6); 5.434 (1 H, dd, J = 9.1,
ꢀ
J = 1.3, H-16␤); 4.728 (1 H, m, W = 32, H-3␣); 2.115 (3 H,
s, CH3COO); 2.056 (3 H, s, CH3COO); 1.242 (3 H, s, 3 × H-
19); 1.001 (3 H, s, 3 × H-18).
2.3.2. 3β,16α-Dihydroxyandrost-5-ene-7,17-dione (8)
◦
1
of 5 was 19 mg (63%), m.p. 186–188 C (methanol/water).
Literature [11] gives 185–188 C. H NMR (400 MHz,
(a) Sulfuric acid (3 M, 2.8 ml, 8.4 mmol) was added to
a suspension of diacetate 7 (201 mg, 0.5 mmol) in
methanol (28 ml) and stirred at room temperature for
48 h. Methanol was then evaporated in vacuo, and
the residue was treated with ethyl acetate (50 ml)
and water (10 ml). The aqueous phase was extracted
with ethyl acetate (30 ml), and the combined organic
phases were washed with water (2 × 20 ml). The sol-
vent was evaporated in vacuo, and the residue was chro-
matographed on two preparative silica gel plates in a
mixture of chloroform/acetone (70:30). The yield of hy-
◦
CDCl3 + CD3COOD): 5.377 (1 H, bd, J ≈ 5, H-6); 4.377
(1 H, bd, J = 8.2, H-16␤); 3.527 (1 H, m, W = 32, H-3␣); 1.036
(
3 H, s, 3 × H-19); 0.981 (3 H, s, 3 × H-18). The product
1
was identical (TLC, IR, H NMR) with the authentic sample
[11].
2
.2.4. 17-Oxoandrost-5-ene-3β,16α-diyl diacetate (6)
A solution of 4-toluenesulfonic acid monohydrate
(3.9 mg, 0.021 mmol) in water (0.3 ml) was added to a solu-
◦
tion of acetal 4 (43 mg, 0.1 mmol) in acetone (2.0 ml), and
droxy ketone 8 was 124 mg (78%), m.p. 238–243 C
the reaction mixture was stirred at room temperature for
(decomposition, methanol), [␣]D −67 (c 0.40, ethanol).
◦
1
20 h. The solvents were then evaporated in vacuo, and the
Literature [7] gives m.p. 236–243 C (methanol), [␣]D
◦
residue was dissolved in ether (5 ml) and water (1 ml). The
aqueous phase was extracted with ether (3 ml), and the com-
bined organic phases were washed with water (2 × 2 ml). The
solvent was evaporated in vacuo, and the residue was chro-
matographed on a preparative silica gel plate in a mixture of
−92 (ethanol); literature [3] gives m.p. 235–239 C.
1
H NMR (400 MHz, CDCl3 + CD3COOD): 5.749 (1 H,
bd, J = 1.8, H-6); 4.402 (1 H, bd, J = 8.5, H-16␤);
3.699 (1 H, m, W = 32, H-3␣); 1.254 (3 H, s, 3 × H-
13
19); 0.996 (3 H, s, 3 × H-18). For C NMR, see
benzene/ether (90:10). The yield of 6 was 22 mg (57%), m.p.
Table 1.
◦
1
1
67–169 C (ethyl acetate/hexane). The literature [11] gives
(b) A solution of 4-toluenesulfonic acid monohydrate
(3.9 mg, 0.021 mmol) in water (0.3 ml) was added to a
suspension of acetal 10 (36 mg, 0.1 mmol) in acetone
(2.0 ml), and the reaction mixture was stirred at room
◦
1
64–167 C. H NMR (400 MHz, CDCl3): 5.398 (1 H, bd,
ꢀ
J ≈ 5, H-6); 5.425 (1 H, dd, J = 9.1, J = 1.1, H-16␤); 4.608
(
1 H, m, W = 32, H-3␣); 2.121 (3 H, s, CH3COO); 2.038 (3 H,

7
42
V. Pouzar et al. / Steroids 70 (2005) 739–749
Table 1
(3 H, s, 3 × H-19); 0.914 (3 H, s, 3 × H-18). Analysis cal-
culated for C H O7 (448.6): C, 66.94; H, 8.09. Found: C,
66.71; H, 7.87.
1
3C NMR chemical shifts of 16␣-OH-DHEA derivatives
25
36
Carbon
8a
13a
16b
1
2
3
4
5
6
7
8
9
36.16
36.57
37.95
30.46c
69.78
31.01
71.10f
36.76
143.29
125.60
72.26
48.10i
48.11i
41.45
19.92
32.94
47.67
39.78
31.24
70.98f
219.98
13.88
d
32.08e
2.3.4. 3β,16α-Dihydroxy-17,17-dimethoxyandrost-5-en-
g
71.91
7-one (10)
38.31
42.87
146.70
124.75
64.93
43.71j
43.83
k
Aqueous KOH (0.4 M, 1 ml) was added to a solution of
166.95
125.57
200.79
43.07
44.04
41.70
20.14
33.50
47.67
49.98
30.73c
70.87
219.13
14.07
17.37
acetate 9 (90 mg, 0.2 mmol) in tetrahydrofuran (2 ml) and
stirred at room temperature for 3 h. The reaction mixture was
then poured into ethyl acetate (100 ml) and washed with satu-
rated aqueous NaCl (2 × 30 ml). The solvent was evaporated,
and the residue was crystallized from an acetone/hexane mix-
f
h
h
j
10
11
12
13
14
15
16
17
18
19
20.83
◦
e,l
ture. The yield of diol 10 was 58 mg (79%), m.p. 210–212 C,
32.39
48.21
38.57
32.39
72.29
221.07
14.24
18.65
[␣]D −205 (c 0.30, CHCl3). IR spectrum (CHCl3): 3611
(O H, free); 3518 broad (O H, bonded); 1671 (C O); 1631
d
e,l
g
1
(
C C); 1169, 1054 (C O C O C). H NMR (400 MHz,
CDCl3 + CD3COOD): 5.696 (1 H, bd, J = 1.7, H-6); 4.284
ꢀ
(
1 H, dd, J = 9.5, J = 2.2, H-16␤); 3.669 (1 H, m, W = 32,
19.09
H-3␣); 1.186 (3 H, s, 3 × H-19); 0.795 (3 H, s, 3 × H-18).
Analysis calculated for C21H32O (364.5): C, 69.20; H, 8.85.
a
In CDCl3; b in CD3OD;
be interchanged; undeterminable value; l signal of two carbons.
c–j
5
tentative assignment, values in columns may
k
Found: C, 69.17; H, 9.04.
2.4. Derivatives of 7,16α-dihydroxy-DHEA
temperature for 30 min. The solvents were evaporated
in vacuo, and the residue was dissolved in ethyl acetate
2
3
.4.1. 17,17-Dimethoxy-7α-hydroxyandrost-5-ene-
β,16α-diyl diacetate (11)
(5 ml) and water (1 ml). The aqueous phase was ex-
®
l-Selectride (1 M solution in tetrahydrofuran, 0.6 ml)
tracted with ethyl acetate (3 ml), and the combined or-
ganic phases were washed with water (2 × 2 ml). The
solvent was evaporated in vacuo, and the residue was
chromatographed on a preparative silica gel plate in a
mixture of chloroform/acetone (70:30). The yield of hy-
droxy ketone 8, identical with the above described sam-
ple, was 21 mg (67%).
◦
was added under argon at −78 C to a solution of ketone
9
(224 mg, 0.5 mmol) in tetrahydrofuran (6 ml), and the mix-
◦
ture was stirred at −78 C for 5 h. Water (0.35 ml) was then
◦
added, and the mixture was warmed to 0 C. An aqueous
6
M NaOH (0.65 ml, 3.9 mmol) was added, followed by 30%
hydrogen peroxide (0.65 ml), and stirring was continued at
room temperature for 30 min. The mixture was diluted with
ether (200 ml), and the organic layer was washed successively
with ice cold 5% aqueous citric acid, water, saturated aque-
ous KHCO3, and water. The solvent was evaporated, and the
residue was chromatographed on four preparative silica gel
plates in a mixture of benzene/ethyl acetate (60:40). The yield
of the hydroxy derivative 11 was 138 mg (61%), amorphous
foam, [␣]D −143 (c 0.41, CHCl3). IR spectrum (CHCl3):
3603 (O H, free); 3529, 3444 (O H, bonded); 1730 (C O);
1665 (C C); 1252 (C O, acetate); 1192, 1134, 1070, 1037
2
.3.3. 17,17-Dimethoxy-7-oxoandrost-5-ene-3β,16α-
diyl diacetate (9)
A suspension of chromium(VI) oxide (1.50 g, 15.0 mmol)
◦
in dichloromethane (12 ml) was stirred at −25 C under argon
with3,5-dimethylpyrazole(1.50 g, 15.6 mmol). After15 min,
a solution of olefin 4 (435 mg, 1.0 mmol) in dichloromethane
(
−
3 ml) was added dropwise, and stirring was continued at
◦
20 C for 4 h. The reaction mixture was then diluted with
1
a mixture of benzene/ethyl acetate (30 ml, 7:3) and filtered
through a short column of silica gel (10 g) layered with
celite. The column was washed with the same solvent mix-
ture, and the solvents were evaporated in vacuo. The residue
was chromatographed on eight preparative silica gel plates
in a mixture of benzene/ether (80:20). The yield of ketone 9
(C O C O C). HNMR(400 MHz, CDCl3 + CD3COOD):
ꢀ
5.606 (1 H, dd, J = 5.1, J = 1.5, H-6); 5.233 (1 H, dd, J = 9.9,
ꢀ
J = 3.2, H-16␤); 4.637 (1 H, m, W = 32, H-3␣); 3.796 (1 H, bt,
J ≈ 4.5, H-7␤); 3.490 (3 H, s, CH3O); 3.297 (3 H, s, CH3O);
2.079 (3 H, s, CH3COO); 2.031 (3 H, s, CH3COO); 1.000
(3 H, s, 3 × H-19); 0.911 (3 H, s, 3 × H-18). Analysis cal-
culated for C H38O7 (450.6): C, 66.64; H, 8.50. Found: C,
◦
was 308 mg (69%), m.p. 214–215 C (ether), [␣]D −167 (c
25
0
1
.25, CHCl3). IR spectrum (CHCl3): 1732 (C O, acetate);
65.51; H, 8.68.
671 (C O, 7-ketone); 1632 (C C); 1249, 1039 (C O, ac-
1
etate). H NMR (400 MHz, CDCl3): 5.701 (1 H, bd, J = 1.6,
2.4.2. 17,17-Dimethoxyandrost-5-ene-3β,7α,16α-triol
(12)
ꢀ
H-6); 5.260 (1 H, dd, J = 9.4, J = 3.2, H-16␤); 4.716 (1 H, m,
W = 32, H-3␣); 3.478 (3 H, s, CH3O); 3.285 (3 H, s, CH3O);
Aqueous 0.4 M KOH (1.25 ml, 0.5 mmol) was added to
a solution of acetate 11 (113 mg, 0.25 mmol) in tetrahydro-
2
.092 (3 H, s, CH3COO); 2.053 (3 H, s, CH3COO); 1.202

V. Pouzar et al. / Steroids 70 (2005) 739–749
743
furan (3 ml) and stirred at room temperature for 48 h. The
reaction mixture was then poured into ethyl acetate (100 ml),
washed with water (3 × 50 ml), and the solvent was evap-
orated in vacuo. The yield of chromatographically (TLC)
pure product 12 was 79 mg (86%). The analytical sample
(400 MHz, CDCl3 + CD3COOD): 5.289 (1 H, bs, H-6); 5.258
ꢀ
(1 H, dd, J = 9.9, J = 2.9, H-16␤); 4.619 (1 H, m, W = 32, H-
ꢀ
3␣); 3.867 (1 H, dt, J = 8.2, J = 2.0, H-7␣); 3.487 (3 H, s,
CH3O); 3.289 (3 H, s, CH3O); 2.094 (3 H, s, CH3COO);
2.041 (3 H, s, CH3COO); 1.053 (3 H, s, 3 × H-19); 0.917
(3 H, s, 3 × H-18). Analysis calculated for C H38O7 (450.6):
◦
was obtained by crystallization from ether, m.p. 225–228 C,
25
decomposition, [␣]D −153 (c 0.33, CHCl3). IR spectrum
C, 66.64; H, 8.50. Found: C, 66.72; H, 8.46.
(
KBr): 3640, 3415 shoulder, 3310 shoulder (O H); 1668
C C); 1054 (C O); 1192, 1132, 1037 (C O C O C).
(
2.4.5. 17,17-Dimethoxyandrost-5-ene-3β,7β,16α-triol
(15)
1
H NMR (400 MHz, CDCl3 + CD3COOD): 5.567 (1 H, dd,
ꢀ
ꢀ
J = 5.4, J = 1.6, H-6); 4.319 (1 H, dd, J = 9.1, J = 2.2, H-16␤);
Aqueous 0.4 M KOH (1.25 ml, 0.5 mmol) was added to
a solution of acetate 14 (113 mg, 0.25 mmol) in tetrahydro-
furan (3 ml) and stirred at room temperature for 48 h. The
reaction mixture was then poured into saturated aqueous
NaCl (25 ml); the separated product was collected on a filter
3
3
3
.816 (1 H, bt, J ≈ 4.5, H-7␤); 3.569 (1 H, m, W = 32, H-3␣);
.468 (3 H, s, CH3O); 3.365 (3 H, s, CH3O); 0.973 (3 H, s,
× H-19); 0.799 (3 H, s, 3 × H-18). Analysis calculated for
C21H34O (366.5): C, 68.82; H, 9.35. Found: C, 68.97; H,
5
9
.36.
and dried in vacuo. Crystallization from methanol afforded
◦
58 mg (63%) of hydroxy derivative 15, m.p. 253–255 C,
2
.4.3. 3β,7α,16α-Trihydroxyandrost-5-en-17-one (13)
A solution of 4-toluenesulfonic acid monohydrate
[␣]D −22 (c 0.31, DMSO). IR spectrum (KBr): 3401, 3340
shoulder, 3287 shoulder (O H); 1056, 1019 (C O); 1190,
1
(3.9 mg, 0.021 mmol) in water (0.3 ml) was added to a so-
1126, 1083 (C O C O C). H NMR (400 MHz, DMSO-
lution of acetal 12 (37 mg, 0.1 mmol) in acetone (2.0 ml) and
stirred at room temperature for 30 min. The reaction mixture
was then poured into ethyl acetate (150 ml), washed with
water (2 × 30 ml), and the solvent was evaporated in vacuo.
The residue was chromatographed on a preparative silica
gel plate in a mixture of chloroform/acetone (50:50). The
d ): 5.123 (1 H, bs, H-6); 3.549 (1 H, bt, J ≈ 8, H-16␤); 4.084
6
(1 H, m, W = 20, H-7␣); 3.382 (3 H, s, CH3O); 3.225 (3 H, s,
CH3O); 0.941 (3 H, s, 3 × H-19); 0.755 (3 H, s, 3 × H-18).
Analysis calculated for C21H34O (366.5): C, 68.82; H, 9.35.
5
Found: C, 68.76; H, 9.27.
◦
yield of 13 was 23 mg (71%), m.p. 211–213 C (methanol),
2.4.6. 3β,7β,16α-Trihydroxyandrost-5-en-17-one (16)
A solution of 4-toluenesulfonic acid monohydrate
(3.9 mg, 0.021 mmol) in water (0.3 ml) was added to a so-
lution of acetal 15 (37 mg, 0.1 mmol) in acetone (2.0 ml) and
stirred at room temperature for 30 min. The reaction mixture
was then poured into ethyl acetate (150 ml), washed with
water (2 × 30 ml), and the solvent was evaporated in vacuo.
The residue was chromatographed on a preparative silica gel
plate in a mixture of chloroform/acetone (50:50). The yield
[
3
1
␣]D −60 (c 0.31, dimethyl sulfoxide). IR spectrum (KBr):
430 broad (O H); 1747 (C O); 1664 (C C); 1055, 1034,
1
010 (C O). H NMR (500 MHz, CD3OD): 5.579 (1 H, dd,
ꢀ
J = 5.3, J = 1.7, H-6); 4.355 (1 H, d, J = 7.9, H-16␤); 3.870
(
1 H, m, W = 11, H-7␣); 1.038 (3 H, s, 3 × H-19); 0.961 (3 H,
13
s, 3 × H-18). For C NMR see Table 1. Analysis calculated
for C19H28O4 (320.4): C, 71.22; H, 8.81. Found: C, 71.03;
H, 8.66.
◦
of 16 was 25 mg (77%), m.p. 214–217 C, decomposition
2
3
.4.4. 17,17-Dimethoxy-7β-hydroxyandrost-5-ene-
β,16α-diyl diacetate (14)
A 0.4 M solution of cerium(III) chloride heptahydrate in
(ether-acetone), [␣]D +118 (c 0.12, CHCl3). IR spectrum
(KBr): 3417 broad (O H); 1745 (C O); 1054, 1045, 1028,
1
1013 (C O). H NMR (500 MHz, CDCl3 + CD3COOD):
methanol (1.25 ml, 0.5 mmol) was added to a stirred solution
of ketone 9 (224 mg, 0.5 mmol) in tetrahydrofuran (3 ml).
Sodium borohydride (18 mg, 0.48 mmol) was added to this
mixture in small portions over 3 min, and stirring was con-
tinued for 10 min. The mixture was then poured into ether
5.297 (1 H, bt, J ≈ 2, H-6); 4.351 (1 H, bd, J = 8.2, H-
ꢀ
16␤); 3.944 (1 H, dt, J = 8.1, J = 2.2, H-7␣); 3.540 (1 H,
m, W = 32, H-3␣); 1.078 (3 H, s, 3 × H-19); 0.984 (3 H, s,
13
3 × H-18). For C NMR see Table 1. Analysis calculated
for C19H28O4 (320.4): C, 71.22; H, 8.81. Found: C, 71.36; H,
8.57.
(60 ml) and washed with ice cold 5% aqueous citric acid
60 ml). The aqueous phase was extracted with ether (60 ml),
(
and the collected organic layers were washed successively
with ice cold, 5% aqueous citric acid, water, saturated aque-
ous KHCO3, and water. The solvent was evaporated, and
the residue was chromatographed on four preparative sil-
ica gel plates in a mixture of benzene/ethyl acetate (70:30).
The yield of the hydroxy derivative 14 was 163 mg (72%),
2.5. Derivatives 7-hydroxy-DHEA
2.5.1. 7α-Hydroxy-17-oxoandrost-5-en-3β-yl
acetate (17)
A sample was prepared from DHEA according to the
literature [12] in the 41% yield: 1H NMR (400 MHz,
◦
ꢀ
m.p. 177–179 C (ether-hexane), [␣]D −36 (c 0.27, CHCl3).
CDCl3 + CD3COOD): 5.676 (1 H, dd, J = 5.3, J = 1.7, H-6);
IR spectrum (CHCl3): 3626 shoulder (O H, free); 3597
4.649 (1 H, m, W = 32, H-3␣); 3.975 (1 H, bt, J ≈ 4, H-7␤);
2.045 (3 H, s, CH3COO); 1.037 (3 H, s, 3 × H-19); 0.887
(3 H, s, 3 × H-18).
(
O H, bonded); 1726 (C O); 1675 (C C); 1253 (C O, ac-
1
etate); 1188, 1128, 1084, 1069 (C O C O C). H NMR

7
44
V. Pouzar et al. / Steroids 70 (2005) 739–749
2
.5.2. 7β-Hydroxy-17-oxoandrost-5-en-3β-yl
mass spectrometer equipped with a quadrupole electron im-
acetate (18)
pact detector and a fixed electron voltage of 70 eV.
A sample was prepared from DHEA according to the
literature [12] in the 43% yield: 1H NMR (400 MHz,
CDCl3 + CD3COOD): 5.344 (1 H, bt, J ≈ 2, H-6); 4.628 (1 H,
2.6.3. Derivatization
In the first step, 10 ␮g of the steroids under investiga-
tion were derivatized with 50 ␮l of a 2% solution of O-
ꢀ
m, W = 32, H-3␣); 3.973 (1 H, dt, J = 8.2, J = 2.1, H-7␣);
◦
2
.044 (3 H, s, CH3COO); 1.096 (3 H, s, 3 × H-19); 0.907
methylhydroxylamine hydrochloride in pyridine at 60 C for
(3 H, s, 3 × H-18).
1 h to deactivate the oxo groups. Then, the mixtures were
dried under a stream of nitrogen. To deactivate the hydroxy
groups, the dried residues were derivatized with 50 ␮l Sylon
2
.5.3. 3β,7α-Dihydroxyandrost-5-en-17-one (19)
◦
BFT (99% BSTFA + 1% TMCS) at 90 C for 45 min. Sub-
Hot 5% aqueous Na2CO3 (2.0 ml, 1.0 mmol) was added
sequently, the derivatization agent was evaporated under a
stream of nitrogen and the steroid derivatives were dissolved
in 1 ml of isooctane. Then, 4 ␮l portions of the mixtures cor-
responding to 40 ng of the steroids were injected into the
GC–MS system.
to a refluxed solution of acetate 17 (173 mg, 0.5 mmol) in
methanol (10 ml), and the refluxing was continued for 75 min.
Methanol and part ofthe waterwere thenevaporated in vacuo,
and the mixture was diluted with water and extracted with
dichloromethane (3 × 20 ml). The collected extracts were
washed with water (2 × 50 ml), dried, and the solvent was
evaporated in vacuo. Crystallization of the residue from a
mixture of acetone-hexane afforded 135 mg (89%) of diol
2.6.4. GC–MS analysis
GC separation was carried out using a Zebron ZB-50 cap-
◦
illary column (15 m × 0.25 mm) with 0.15 ␮m film thick-
1
9, m.p. 178–180 C, [␣]D −80 (c 0.14, CHCl3). Litera-
◦
ness, Cat. no. 7EG-G004-05, (Phenomenex, St. Torrance,
ture [13] gives m.p. 183–184 C, [␣]D −73 (c 0.5, CHCl3).
◦
1
CA, USA). The temperature of the injection port was 300 C.
H NMR (400 MHz, CDCl3 + CD3COOD): 5.653 (1 H, dd,
ꢀ
The following protocol was used: Splitless high-pressure
J = 5.3, J = 1.7, H-6);3.986(1 H, bt, J ≈ 4, H-7␤);3.582(1 H,
◦
injection for 1 min at 100 kPa, 1 min delay at 120 C and
m, W = 32, H-3␣); 1.025 (3 H, s, 3 × H-19); 0.890 (3 H, s,
◦
3
0 kPa, then, a steep gradient 40 C/min and 8.5 kPa/min
up to 220 C and 51 kPa, a gentle gradient 2.9 C/min and
.5 kPa/min up to 240 C and 54.5 kPa, and a steep gradient
0 C/min and 9 kPa/min up to 310 C and 70 kPa followed
3
× H-18).
◦
◦
◦
0
4
2
.5.4. 3β,7β-Dihydroxyandrost-5-en-17-one (20)
◦
◦
Hot 5% aqueous Na2CO3 (2.0 ml, 1.0 mmol) was added
by a 2 min plateau. The duration of the analysis was 14.2 min.
The detector voltage was at 1.65 kV, and the sampling rate
to a refluxed solution of acetate 18 (173 mg, 0.5 mmol) in
methanol (10 ml), and the refluxing was continued for 75 min.
Methanol and part ofthe waterwere thenevaporated in vacuo,
and the mixture was diluted with water and extracted with
dichloromethane (3 × 20 ml). The collected extracts were
washed with water (2 × 50 ml), dried, and the solvent was
evaporated in vacuo. Crystallization of the residue from a
mixture of acetone-hexane afforded 127 mg (83%) of diol 20,
◦
was 0.25 s. The temperature of the interface was 320 C. The
response was recorded in total ion current mode recording the
mass spectrum from the effective mass 70 amu up to 700 amu
at the scan speed 4000 amu/s.
3. Results and discussion
◦
m.p. 213–215 C, [␣]D +63 (c 1.3, CHCl3). Literature [13]
◦
1
gives m.p. 214–215 C, [␣]D +71 (c 0.4, CHCl3). H NMR
400 MHz, CDCl3 + CD3COOD): 5.321 (1 H, bt, J ≈ 2, H-6);
.976 (1 H, dt, J = 8.1, J = 2.0, H-7␣); 3.586 (1 H, m, W = 32,
H-3␣); 1.081 (3 H, s, 3 × H-19); 0.905 (3 H, s, 3 × H-18).
3.1. Synthesis
(
3
On the basis of the model experiments, the approach of the
“
16-substituent first” was selected (Scheme 1). As a starting
compound, the fully protected derivative of 16␣-hydroxy-
DHEA was necessary with hydroxy groups in positions 3
and 16 protected against oxidation and the 17-oxo group
protected from reduction. These requirements were fulfilled
with the derivative 4, which had hydroxy groups protected
as acetates and a 17-oxo group protected as dimethyl acetal.
Moreover, the protected ␣-ketol moiety in the D ring was re-
sistant against unnecessary rearrangement into the isomeric
2
2
.6. GC–MS analysis
.6.1. Chemicals
The analytical grade solvents were from Merck (Darm-
stadt, Germany). The derivatization agents Sylon BFT
and O-methylhydroxylamine hydrochloride were from Su-
pelco (Bellefonte, PA, USA) and Sigma (St. Louis, MO,
USA).
17␤-hydroxy-16-oxo derivative.
For syntheses of all three target compounds 8, 13, and 16,
2
.6.2. Instruments
The GC–MS system by Shimazdu (Kyoto, Japan), con-
the same starting compound, 17,17-dimethoxy-16␣-hydroxy
derivative 3, was used (Scheme 2). This compound was easily
available from the 16␣-bromo-17-oxo derivative 2 by mod-
ification of the procedure described in the literature [10].
sisted of GC17A gas chromatograph equipped with an auto-
matic flow control, an AOC-20 autosampler and a QP5050A

V. Pouzar et al. / Steroids 70 (2005) 739–749
745
Scheme 1. Retrosynthetic analysis.
Its acetylation with a mixture of acetic anhydride and pyri-
dine gave a fully protected derivative 4. Using this deriva-
tive, the possibility of selective removal of the protecting
groups for the hydroxy group and oxo group was checked.
Acid deprotection under controlled conditions afforded the
ketone-diacetate 6, whereas alkaline hydrolysis gave the diol-
ketal 3.
For introduction of a 7-oxo group into a molecule of
fully protected derivative 4, oxidation with a complex of
chromium(VI) oxide and 2,5-dimethylpyrazole [14] was
used, and ketone 9 was prepared. The removal of acetyl
groups from compound 9 in alkaline medium yielded diol
10, which subsequently gave the known 16␣-hydroxy-7-oxo-
DHEA 8 [3,7] by acidic splitting of the acetal moiety. An
Scheme 2. (i) CuBr2, MeOH; (ii) MeONa, MeOH; (iii) Ac2O, Py; (iv) KOH, H2O/THF; (v) TsOH, Me2CO/H2O; (vi) 3,5-diMepyrazole-CrO3, CH2Cl2,
◦
−
20 C; (vii) 3 M H2SO4, MeOH.

7
46
V. Pouzar et al. / Steroids 70 (2005) 739–749
®
◦
Scheme 3. (i) NaBH4, CeCl3, MeOH/THF; (ii) 1. l-Selectride , THF, −78 C, 2. NaOH, H2O2; (iii) KOH, H2O/THF; (iv) TsOH, Me2CO/H2O.
alternative synthesis started from the known [10,11] 3␤,16␣-
diacetoxy-17-oxo derivative 6, which was transformed by
oxidation with a complex of chromium(VI) oxide – 2,5-
dimethylpyrazole [14] to 7,17-dione 7. After removal of the
acetyl protecting groups, the known target compound 16␣-
hydroxy-7-oxo-DHEA 8 [11] was prepared by an indepen-
dent way.
merization was observed. The structure of both isomers was
assigned mainly on the basis of H NMR spectroscopy (see
below).
1
3.2. NMR spectra of 7- and 16-hydroxy derivatives of
DHEA
Synthesis of epimeric 7,16␣-dihydroxy derivatives DHEA
The observed shape (apparent multiplicity) of the proton
signals in positions substituted by the hydroxy group de-
pends on the measurement conditions for both the 7-hydroxy
[15,16] and 16␣-hydroxy [11] derivatives of DHEA. This is
the reason why there are differences among published values
given in the literature, especially for the various 7-hydroxy
derivatives of DHEA. In the present study, the spectra of the
model derivatives of DHEA were acquired under the same
conditions, including those of the 7-oxo compounds 5 and 6
and the 7-hydroxy derivatives with acetylated (17 and 18),
and free (19 and 20) 3␤-hydroxy groups. The spectra in
deuteriochloroform were measured with additional perdeu-
terioacetic acid (1%) in order to achieve exchange of protons
in the hydroxy groups and consequently to simplify the spec-
tra. Acetic acid at this concentration and during a time period
of tens of minutes did not cause marked cleavage of the acetal
1
3 and 16 started from the 7-oxo derivative 9 mentioned
above (Scheme 3). In the synthesis of the first isomer, the
␣-hydroxy derivative 11 was prepared from 9 by controlled
7
®
stereoselective reduction with l-Selectride [15]. Alkaline
hydrolysis removed the acetyl protecting groups to give triol
2. Finally, a ketal moiety was split off by 4-toluenesulfonic
1
acid, and 7␣,16␣-dihydroxy-DHEA 13 was obtained.
The second isomer 7␤,16␣-dihydroxy derivative 16 was
prepared by reduction of 9 with sodium borohydride in the
presence of cerium(III) chloride [12] and successive depro-
tection of the obtained 7␤-hydroxy derivative 14 via inter-
mediate 15 analogously as above. Both isomeric derivatives
3 and 16 differed on TLC, so any mutual isomerization at
position 7 during the synthesis could be monitored. Both syn-
theses were sufficiently stereoselective, and no apparent iso-
1

V. Pouzar et al. / Steroids 70 (2005) 739–749
747
protecting group in position 17. The 5% aqueous solution of
citric acid, used in the work-up of reaction mixtures, also did
not cleave this protecting group. For its removal, a stronger
acid such as 4-toluenesulfonic acid was necessary.
ical shift value of H-6, which differed significantly for both
isomers [12]. For the 7␣-isomer, the value of H-6 was about
5.6 ppm, and for the 7␤-isomer, it was about 5.3 ppm. The
value of J(6,7) was also different: for the 7␣-hydroxy deriva-
tive, it amounted to about 5 Hz, and for the 7␤-hydroxy
derivative, it was about 2 Hz. The shape of the signal was fur-
The spectra of all 17-oxo derivatives with 16␣-hydroxy
and 16␣-acetoxy groups contained a characteristic doublet
of 16␤-H with a coupling constant J(15␤,16␤) of 8–9 Hz.
The coupling of 16␤-H with 15␣-H was close to zero, and
it was observable only for some compounds in the series.
Thecorresponding17,17-dimethoxyderivativeshaddifferent
geometry at ring D and the coupling constant J(15␣,16␤) was
slightly higher beeing about 2–3 Hz. The isomeric 17-oxo
derivatives with a 16␤-acetoxy group differed significantly,
giving the 16␣-H as a characteristic triplet with a coupling
constant of about 8.5 Hz [3].
4
ther influenced by the allylic interaction J(4␤,6) amounting
to about 2 Hz. Signal H-7 was further split by the interaction
with H-8, and again, the value was different for both isomers;
it was about 4 Hz for the 7␣- and 8 Hz for the 7␤-isomer. For
5
the 7␤-hydroxy derivative, an additional coupling J(4␤,7␣)
of about 2 Hz was discernible. The final shape of the sig-
nals depended on particular coupling constants and could be
different for any combination of substituents.
The configurations of the hydroxy groups at position
1
For the 7-hydroxy derivatives, it was possible to assign
the configuration of a hydroxy group on the basis of a chem-
7, assigned on the basis of the H NMR spectra, were in
Fig. 3. Chromatogram of the MO-TMS derivatives of 3␤,7␤,16␣-
trihydroxyandrost-5-en-17-one (16) as recorded on the effective masses 117
and 129.
Fig. 1. Chromatogram of the MO-TMS derivative of 3␤,16␣-dihydroxy-
androst-5-ene-7,17-dione (8) as recorded on the effective masses 117 and
1
29.
Fig. 2. Chromatogram of the MO-TMS derivatives of 3␤,7␣,16␣-
trihydroxyandrost-5-en-17-one (13) as recorded on the effective masses 117
and 129.
Fig. 4. Mass spectra of the MO-TMS derivatives of 3␤,16␣-dihydroxy-
androst-5-ene-7,17-dione (8); for better clarity, only the effective masses
greater than 300 amu are shown.

7
48
V. Pouzar et al. / Steroids 70 (2005) 739–749
typical for a number of steroids with the 3␤-OTMS-5-ene
structure, including the compounds evaluated. All measured
compounds gave two peaks due to the presence of both E and
Z isomers of 17-(O-methyloxime) derivatives. Both peaks
displayed nearly the same mass spectra, thus, in the Figs. 4–6,
only the mass spectra of the stronger peaks with shorter re-
tention times together with their interpretation were included.
The expected fragmentation patterns are shown in Figs. 4–6
and were in accordance with the proposed structures. The ac-
quired retention times and the fragmentation gave a base for
the use of GC–MS in the screening of these 16␣-hydroxy-
DHEA derivatives in biological samples.
4. Conclusions
Fig. 5. Mass spectra of the MO-TMS derivatives of 3␤,7␣,16␣-
trihydroxyandrost-5-en-17-one (13); for better clarity, only the effective
masses greater than 300 amu are shown.
The synthesis of derivatives of 16␣-hydroxy-DHEA with
an additional oxygen substituent at position 7, i.e., 7-oxo
derivative 8 and both epimeric 7-hydroxy derivatives 13 and
1
6, was developed and their structures were proved by NMR
accordance with the configurations predicted by the known
reaction course of the reduction of a 7-oxo group with the
employed reduction reagents [12,15,16].
spectroscopy. These compounds were designed for further
use as standards in gas chromatography/mass spectrometry
assays. Monitoring of their occurrence in biological tissues
can lead to better knowledge of related steroid metabolism
pathways, for example, those associated with brain disorders
or autoimmune diseases.
The 13C NMR spectra for target compounds 8, 13 and
1
6 are summarized in Table 1. Chemical shifts of particular
carbons were in accordance with the corresponding struc-
tures. On the basis of the available spectral data, isomeric
1
7␤-hydroxy-16-oxo derivatives could be excluded due to
the absence of a characteristic chemical shift of C-17 at about
Acknowledgements
8
5 ppm [17].
We thank Mrs. D. Hyb sˇ ov a´ for skillful technical assis-
tance. We are indebted to Dr. S. Va sˇ ´ı cˇ kov a´ for taking the IR
spectraandDr. P. Fiedlerforinterpretingthem. Ourthanksare
due to Mrs. M. Snopkov a´ for the measurements of 400 MHz
3
.3. GC–MS analysis
The chromatograms of the MO-TMS derivatives of com-
1
ˇ
H NMR spectra and to Dr. D. Saman for the 500 MHz H
1
pounds 8, 13, and 16 are shown in Figs. 1–3. The responses
NMR and 13C spectra. Elemental analyses were carried out
in the Analytical Laboratory (Dr. L. V a´ lkov a´ , Head). This
work was carried out as a part of the CAS research project
Z4 055 0506 and supported by the Grant Agency of the Czech
Republic No. 203/03/0142.
were recorded on the effective masses 117 and 129 that are
References
[1] Morfin R. Involvement of steroids and cytochromes P450 species
in the triggering of immune defenses. J Steroid Biochem Mol Biol
2002;80:273–90.
[
2] Morfin R, St a´ rka L. Neurosteroid 7-hydroxylation products in the
brain. Int Rev Neurobiol 2001;46:79–95.
[
3] Reich IL, Reich HJ, Kneer N, Lardy H. Ergosteroids V: preparation
and biological activity of various D-ring derivatives in the 7-oxo-
dehydroepiandrosterone series. Steroids 2002;67:221–33.
[
4] Hampl R, St a´ rka L. Epimerisation of naturally occurring C-19
steroid allylic alcohols by rat liver preparations. J Steroid Biochem
1969;1:47–56.
Fig. 6. Mass spectra of the MO-TMS derivatives of 3␤,7␤,16␣-
trihydroxyandrost-5-en-17-one (16); for better clarity, only the effective
masses greater than 300 amu are shown.
[5] Hampl R, St a´ rka L. Minireview: 16␣-hydroxylated metabolites of de-
hydroepiandrosterone and their biological significance. Endocr Regul
2000;34:161–3.

V. Pouzar et al. / Steroids 70 (2005) 739–749
749
[
6] Furuya K, Yoshida T, Takagi S, Kanbegawa A, Yamashita
H, Kurosawa Y, et al. Radioimmunoassay of 16␣-hydroxy-
dehydroepiandrosterone and its sulfate. Steroids 1976;27:797–812.
7] Okada M, Fukushima DK, Gallagher TF. Isolation and characteriza-
the synthesis of 3␤-acetoxy-7␣- and 7␤-fluoroandrost-5-en-17-one.
Steroids 1996;61:453–60.
[13] Defaye G, Luche MJ, Chambaz EM. Microbiological 7- and 15-
hydroxylations of C-19 steroids. J Ster Biochem 1978;9:331–6.
[14] Salmond WG, Barta MA, Havens JL. Allylic oxidation with
[
5
tion of 3␤-hydroxy-ꢀ -steroids in andrenal carcinoma. J Biol Chem
5
1
959;234:1688–92.
3,5-dimethylpyrazole chromium trioxide complex. Steroidal ꢀ -7-
[8] Fishman J. Rearrangement of steroidal ring D ketols. J Am Chem
Soc 1960;82:6143–7.
ketones. J Org Chem 1978;43:2057–9.
[15] Pouzar V, Slav ´ı kov a´ T, Cern y´ I. Synthesis of (19E)-3␤,7␣-dihydroxy-
ˇ
[
9] Numazawa M, Nagaoka M, Osawa Y. Stereospecific synthesis of
17-oxoandrost-5-en-19-al 19-(O-carboxymethyl)oxime, a new hapten
for 7␣-hydroxydehydroepiandrosterone (3␤,7␣-dihydroxyandrost-5-
en-17-one). Steroids 1998;63:454–8.
1
6␣-hydroxy-17-oxo steroids by controlled alkaline hydrolysis of
corresponding 16-bromo 17-ketones and its reaction mechanism. J
Org Chem 1982;47:4024–9.
ˇ
[16] Pouzar V, Slav ´ı kov a´ T, Cern y´ I. Synthesis of (19E)-3␤,7␤-dihydroxy-
[
10] Hassner A, Catsoulacos P. Some reactions of steroidal ␣-bromo ke-
17-oxoandrost-5-en-19-al 19-(O-carboxymethyl)oxime, a new hapten
for 7␤-hydroxydehydroepiandrosterone (3␤,7␤-dihydroxyandrost-
5-en-17-one). Collect Czech Chem Commun 1997;62:109–
23.
tones. J Org Chem 1966;31:3149–52.
ˇ
[
11] Pouzar V, Cern y´ I, Lap cˇ ´ı k O, Hill M, Hampl R. Synthesis
of two haptens of 16␣-hydroxydehydroepiandrosterone (3␤,16␣-
dihydroxyandrost-5-en-17-one). Steroids 2003;68:149–58.
[17] Doller D, Gros EG. Carbon-13 nuclear magnetic resonance spec-
tral data of steroidal vicinal ketols and related compounds. Steroids
1991;56:168–72.
[12] Marwah P, Thoden JB, Powell DR, Lardy HA. Steroidal allylic fluo-
rination using diethylaminosulfur trifluoride: a convenient method for