Metabolism of 4-hydroxyandrostenedione and 4-hydroxytestosterone: Mass spectrometric identification of urinary metabolites

Article abstract of DOI:10.1016/j.steroids.2006.11.018

4-Hydroxyandrost-4-ene-3,17-dione is a second generation, irreversible aromatase inhibitor and commonly used as anti breast cancer medication for postmenopausal women. 4-Hydroxytestosterone is advertised as anabolic steroid and does not have any therapeutic indication. Both substances are prohibited in sports by the World Anti-Doping Agency, and, due to a considerable increase of structurally related steroids with anabolic effects offered via the internet, the metabolism of two representative candidates was investigated. Excretion studies were conducted with oral applications of 100 mg of 4-hydroxyandrostenedione or 200 mg of 4-hydroxytestosterone to healthy male volunteers. Urine samples were analyzed for metabolic products using conventional gas chromatography-mass spectrometry approaches, and the identification of urinary metabolites was based on reference substances, which were synthesized and structurally characterized by nuclear magnetic resonance spectroscopy and high resolution/high accuracy mass spectrometry. Identified phase-I as well as phase-II metabolites were identical for both substances. Regarding phase-I metabolism 4-hydroxyandrostenedione (1) and its reduction products 3β-hydroxy-5α-androstane-4,17-dione (2) and 3α-hydroxy-5β-androstane-4,17-dione (3) were detected. Further reductive conversion led to all possible isomers of 3ξ,4ξ-dihydroxy-5ξ-androstan-17-one (4, 6-11) except 3α,4α-dihydroxy-5β-androstan-17-one (5). Out of the 17β-hydroxylated analogs 4-hydroxytestosterone (18), 3β,17β-dihydroxy-5α-androstan-4-one (19), 3α,17β-dihydroxy-5β-androstan-4-one (20), 5α-androstane-3β,4β,17β-triol (21), 5α-androstane-3α,4β,17β-triol (26) and 5α-androstane-3α,4α,17β-triol (28) were identified in the post administration urine specimens. Furthermore 4-hydroxyandrosta-4,6-diene-3,17-dione (29) and 4-hydroxyandrosta-1,4-diene-3,17-dione (30) were determined as oxidation products. Conjugation was diverse and included glucuronidation and sulfatation.

Full text of DOI:10.1016/j.steroids.2006.11.018

steroids 7 2 ( 2 0 0 7 ) 278–286
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/steroids
Metabolism of 4-hydroxyandrostenedione and
4
-hydroxytestosterone: Mass spectrometric identification of
urinary metabolites
a,b,∗
a
a
a
c
Maxie Kohler
, Maria K. Parr , Georg Opfermann , Mario Thevis , Nils Schl o¨ rer ,
b
a
Franz-Josef Marner , Wilhelm Sch a¨ nzer
a
Institute of Biochemistry, German Sport University of Cologne, Carl-Diem-Weg 6, 50933 Cologne, Germany
Institute of Biochemistry, University of Cologne, Z u¨ lpicher Straße 47, 50939 Cologne, Germany
Institute of Organic Chemistry, University of Cologne, Greinstraße 4, 50939 Cologne, Germany
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
4-Hydroxyandrost-4-ene-3,17-dione is a second generation, irreversible aromatase inhibitor
and commonly used as anti breast cancer medication for postmenopausal women.
4-Hydroxytestosterone is advertised as anabolic steroid and does not have any thera-
peutic indication. Both substances are prohibited in sports by the World Anti-Doping
Agency, and, due to a considerable increase of structurally related steroids with anabolic
effects offered via the internet, the metabolism of two representative candidates was
investigated.
Received 23 October 2006
Received in revised form
20 November 2006
Accepted 22 November 2006
Published on line 17 January 2007
Keywords:
Excretion studies were conducted with oral applications of 100 mg of 4-hydroxyandro-
stenedione or 200 mg of 4-hydroxytestosterone to healthy male volunteers. Urine sam-
ples were analyzed for metabolic products using conventional gas chromatography–mass
spectrometry approaches, and the identification of urinary metabolites was based
on reference substances, which were synthesized and structurally characterized by
nuclear magnetic resonance spectroscopy and high resolution/high accuracy mass
spectrometry.
4
-Hydroxylated steroids
Formestane
-Hydroxytestosterone
4
Metabolism
Gas chromatography–mass
spectrometry
Identified phase-I as well as phase-II metabolites were identical for both substances.
Regarding phase-I metabolism 4-hydroxyandrostenedione (1) and its reduction products 3␤-
hydroxy-5␣-androstane-4,17-dione (2) and 3␣-hydroxy-5␤-androstane-4,17-dione (3) were
detected. Further reductive conversion led to all possible isomers of 3␰,4␰-dihydroxy-5␰-
androstan-17-one (4, 6–11) except 3␣,4␣-dihydroxy-5␤-androstan-17-one (5).
Out of the 17␤-hydroxylated analogs 4-hydroxytestosterone (18), 3␤,17␤-dihydroxy-5␣-
androstan-4-one (19), 3␣,17␤-dihydroxy-5␤-androstan-4-one (20), 5␣-androstane-3␤,4␤,17␤-
triol (21), 5␣-androstane-3␣,4␤,17␤-triol (26) and 5␣-androstane-3␣,4␣,17␤-triol (28) were
identified in the post administration urine specimens. Furthermore 4-hydroxyandrosta-4,6-
diene-3,17-dione (29) and 4-hydroxyandrosta-1,4-diene-3,17-dione (30) were determined as
oxidation products.
Conjugation was diverse and included glucuronidation and sulfatation.
©
2006 Elsevier Inc. All rights reserved.
∗
Corresponding author. Tel.: +49 22 149826311; fax: +49 22 14973236.
E-mail address: maxie@biochem.dshs-koeln.de (M. Kohler).
0039-128X/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2006.11.018

steroids 7 2 ( 2 0 0 7 ) 278–286
279
Further reduction leads to the 3␰,4␰-dihydroxy-5␰-an-
drostane products. Out of the eight possible isomers 3␤,4␤-
dihydroxy-5␣-androstan-17-one (4) and 3␣,4␤-dihydroxy-5␣-
androstan-17-one (9) were detected. Metabolite 4 was excreted
as glucuronide by women [6] and the sulfates of 4 and 9
were found in blood samples after intravenous application
[8]. Furthermore the oxidation product 4-hydroxyandrost-
4,6-diene-3,17-dione (29) was identified in women’s urine
specimens [8].
1
.
Introduction
4
-Hydroxyandrost-4-ene-3,17-dione (4-hydroxyandrostenedi-
one, formestane, 1) is second generation, irreversible
a
aromatase inhibitor and is structurally related to androstene-
dione, the natural substrate of the enzyme aromatase. Because
of its aromatase inhibiting activity formestane is used as anti
breast cancer medication for postmenopausal women.
4
-Hydroxyandrostenedione (1) is extensively metabolized.
The main metabolite is 4-hydroxyandrost-4-ene-3,17-dione-
-glucuronide [1,2]. In women 20% of applied formestane
4-Hydroxytestosterone
(4,17␤-dihydroxyandrost-4-en-
3-one, 18) is the 17-hydroxylated analog to formestane.
It is commercially available on the internet as anabolic
steroid for oral self-administration and does not have any
therapeutic indication. Hence, only little information is
available about its metabolism. So far, most studies dealt
with 4-hydroxytestosterone as metabolite of formestane
while one study investigated the glucuronic acid conjugates
of metabolic products of 4-hydroxytestosterone [11]. After
oral application to healthy men 4-hydroxyandrostenedione
and 4-hydroxytestosterone as well as the reduction products
2, 3 and 9 were identified. Detected oxidation products
4
is excreted as glucuronide within the first 24 h [3]. Men
transform an even higher amount of formestane (35%) into
its glucuronide [4]. Phase-I metabolism is manifold and
mainly reductive. Conjugation includes glucuronidation as
well as sulfatation. Studies were conducted with different
forms of application, male and female subjects and dif-
ferent methods of analysis of either blood or urine [5–10].
Formestane was excreted as glucuronide and sulfate after
oral administration of 500 mg by women [7,8] while it was
identified as glucuronide only after intramuscular application
to men [7]. In addition, 4-hydroxytestosterone was identified
as metabolite of formestane as glucuronide in women’s
urine [8].
were
4-hydroxyandrost-4,6-diene-3,17-dione
(29)
and
4-hydroxyandrost-1,4-diene-3,17-dione (30) [11].
Both substances are prohibited in sports by the World Anti-
Doping Agency. Formestane is categorized as anti-estrogenic
agent and 4-hydroxytestosterone is listed as anabolic steroid
[12].
In Fig. 1, an overview on the main metabolic pathways of
formestane and 4-hydroxytestosterone is illustrated. Because
the only difference of the two substances is the oxo- and
hydroxy-group at C-17, metabolism is supposedly very similar.
Identification and comparison of phase-I and phase-II
metabolites of 1 and 18 may give a better understanding of
3␣-Hydroxy-5␤-androstane-4,17-dione (3) has been found
in urine samples of men and women as glucuronide [5,7,8].
Its isomer, 3␤-hydroxy-5␣-androstane-4,17-dione (2), has not
been detected in humans but only in rats [9].
The 17-hydroxylated analogs 3␤,17␤-dihydroxy-5␣-an-
drostan-4-one (19) and 3␣,17␤-dihydroxy-5␤-androstan-4-one
(
20) were identified as glucuronidated products [5,7,8]. Sulfa-
tation was only observed for metabolite 20 [7].
Fig. 1 – Main ways of metabolism of 4-hydroxyandrostenedione and 4-hydroxytestosterone.

2
80
steroids 7 2 ( 2 0 0 7 ) 278–286
the metabolic fate of these 4-hydroxylated steroids in humans,
and for doping control purposes it may allow to identify the
applied substance by analyzing informative metabolites and
to describe a long term metabolite.
Therefore, the metabolism of 4-hydroxyandrostenedione
and 4-hydroxytestosterone was investigated. Reference
substances were synthesized for identification of urinary
metabolites. Structures and configurations were verified
by nuclear magnetic resonance spectroscopy and high res-
olution/high accuracy mass spectrometry. Identification
of metabolites was performed by gas chromatography–
mass spectrometry (GC–MS) analysis after per-trimethyl-
silylation.
2.1.
Chemicals, solvents and materials
4-Hydroxyandrostenedione was purchased from Thinker
Chemical ((purity > 99.5%) Hangzou, China), 5␤-dihydrotesto-
sterone from Sigma (St. Louis, MO, USA), [2,2,4,4- H ]-11␤-
2
4
hydroxyandrosterone was synthesized in our laboratory [14].
␤-Glucuronidase from E. coli was purchased from Roche
Molecular Diagnostics (Mannheim, Germany), and N-methyl-
N-trimethylsilyl trifluoroacetamide (MSTFA) from Chemische
Fabrik Karl Bucher (Waldstetten, Germany). Potassium car-
bonate, potassium hydrogen carbonate, potassium hydroxide,
methanol, n-hexane, hydrochloric acid, acetone, ethyl acetate
and t-butyl methyl ether (TBME) were purchased from KMF
(
St. Augustin, Germany). Ethylene glycol, pyridine and zinc
were supplied by Aldrich (Deisendorf, Germany), ammonium
iodide and ethanethiol by Fluka (Bucks, Schweitz). Other
reagents and solvents were bought from Merck (Darmstadt,
Germany). All solvents and reagents were of analytical grade,
all solutions and buffers for sample preparation were pre-
pared using deionized water (Water Lab System, Millipore,
Eschborn, Germany). Stock solutions of reference substances
were prepared in methanol. Cartridges for solid phase extrac-
tion (Chromabond C18, 6 mL, 500 mg) were purchased from
Macherey-Nagel (D u¨ ren, Germany).
2
.
Experimental
GC–MS spectra were recorded on an Agilent 6890 Series
GC System coupled to an Agilent 5973 Network Mass
Selective Detector with electron impact ionization (70 eV).
The column used was a HP5MS (15.5 m, 0.25 mm inner
diameter, 0.25 ␮m film thickness). Oven temperature was
◦
◦
◦
◦
◦
1
00–190 C by 40 C/min; 190–245 C by 5 C/min; 245–320 C
◦
by 40 C/min, 3 min hold. Injection volume was 2 ␮L in
split mode (1:10) for reference substances and in splitless
mode for samples from the administration studies. For
calculation of retention indices an almost linear oven tem-
perature was applied: 160–280 C by 5 C/min, 280–330 C
by 40 C/min, 2.5 min hold. Helium was used as carrier gas
2.2.
Synthesis of reference compounds
◦
◦
◦
2.2.1. Synthesis of 4-hydroxytestosterone (18)
◦
4-Hydroxytestosterone (18) was prepared according to a
preparation of Marsh et al. [9,15] by oxidation of 5␤-
dihydrotestosterone with oxygen in alkaline medium. Briefly,
200 mg (0.7 mmol) of 5␤-dihydrotestosterone were diluted in
(
1.5 ml/min).
total of 50 ␮g of reference compounds was per-
trimethylsilylated (per-TMS) for GC–MS measurements using
A
◦
1
00 ␮L of a mixture of MSTFA/NH4I/ethanethiol (1000:2:3,
25 mL of t-butanol (45 C). After addition of 10 mL of 962 mg
◦
v/w/v) by heating 15 min at 80 C [13].
of potassium-t-butylate in t-butanol the mixture was stirred
◦
Structure confirmation of synthesized material was
obtained by nuclear magnetic resonance (NMR) spectroscopy
and high resolution/high accuracy mass spectrometry.
NMR analyses were performed on a Bruker DRX 500
instrument, equipped with a 5 mm inverse probehead with z-
gradient coil. Samples were referenced to tetramethylsilane as
internal standard. An amount of 5 mg of each compound was
dissolved in deuterated chloroform and spectra were recorded
at room temperature. Conducted experiments verifying the
at 45 C for 6.5 h. Hydrochloric acid (7 mL, 1 M) was added
and the solution was evaporated to dryness. A volume of
20 mL of water was added and extraction was performed
three times with 20 mL of TBME each. The combined organic
layers were washed with water and evaporated to dryness. 4-
Hydroxytestosterone was purified by column chromatography
(silica gel, eluents n-hexane/ethyl acetate 40:60) and recrystal-
lized from n-hexane/ethyl acetate (6:1).
4-Hydroxytestosterone (18): 35.5 mg (0.11 mmol, 16%),
1
purity 98%; 1H NMR (500 MHz, CDCl ) ı: 2.56/2.53 (m, 2H, H-
presumed structures were H, H2 COSY, H,C HMQC, H,C HMBC
and NOESY.
3
2), 3.67 (t, 1H, H-17), 0.81 (s, 3H, H-18), 1.21 (s, 3H, H-19); 13C
High resolution/high accuracy mass spectrometry for
verification of elemental composition was accomplished
using a Thermo LTQ Orbitrap mass spectrometer in pos-
itive mode with ionization by electrospray. Analytes were
dissolved in acetonitrile/water (1:1, v/v) containing 0.1%
formic acid at concentrations of 50 ␮g/mL and introduced
into the mass spectrometer using a syringe pump. The
ionization voltage was +3.5 kV, the capillary temperature
NMR (125 MHz, CDCl ) ı: 31.6 (C-2), 193.5 (C-3), 140.4 (C-4), 140.0
(C-5), 81.5 (C-17), 10.9 (C-18), 16.9 (C-19). Elemental composi-
3
tion (protonated): C19H29O , m/z (theor.): 305.2111, m/z (exp.):
3
305.2106, error: −1.6 ppm.
2.2.2. Synthesis of 3-hydroxy-4-oxo compounds
2.2.2.1. Synthesis of 3ˇ-hydroxy-5˛-androstane-4,17-dione (2)
and 3˛-hydroxy-5ˇ-androstane-4,17-dione (3). The synthesis
was performed by hydrogenation of 4-hydroxyandrostenedi-
one with Pd/C as catalyst as described elsewhere [11]. NMR
and HRMS data of 2 are as follows.
◦
was set to 300 C. Damping gas in the linear ion trap was
helium, and gas supplied to the curved linear ion trap
was nitrogen. Spectra were acquired at a resolution of
1
00,000.
Elemental compositions were measured with a maximum
3␤-Hydroxy-5␣-androstane-4,17-dione
(2):
1H
NMR
(500 MHz, CDCl3) ı: 1.64/2.43 (m, 2H, H-2), 4.15 (dd, 1H, H-
error of 2.5 ppm. Purity of the synthesized substances was
determined from GC–MS data.
3), 2.23 (dd, 1H, H-5), 0.89 (s, 3H, H-18), 0.77 (s, 3H, H-19);
1
3C NMR (125 MHz, CDCl3) ı: 35.7 (C-2), 64.6 (C-3), 212.4 (C-4),

steroids 7 2 ( 2 0 0 7 ) 278–286
281
5
6.9 (C-5), 220.8 (C-17), 13.7 (C-18), 13.8 (C-19). Elemental
NMR analysis. Elemental composition (protonated): C19H31O3,
composition (protonated): C19H29O3, m/z (theor.): 305.2111,
m/z (exp.): 305.2111, error: −0.2 ppm.
m/z (theor.): 307.2268, m/z (exp.): 307.2263, error: −1.7 ppm.
2.2.3.2. Synthesis of 3ˇ,4˛-dihydroxy-5˛-androstan-17-one
2
(
.2.2.2. Synthesis of 3ˇ,17ˇ-dihydroxy-5˛-androstan-4-one
19) and 3˛,17ˇ-dihydroxy-5ˇ-androstan-4-one (20). For the
synthesis of the 17-hydroxylated analogs to the substances
and 3 600 mg (1.97 mmol) of testosterone were diluted in
0 mL of methanol and hydrogenated with 48 mL (2 mmol)
(10). The synthesis of 3␤,4␣-dihydroxy-5␣-androstan-17-one
(10) required the protection of the 17-oxo residue of the
starting material. Hence, 1 g (3.5 mmol) of androstenedione
was dissolved in 30 mL of benzene. After addition of 400 ␮L
of ethylene glycol and 36 mg of p-toluene sulfonic acid the
mixture was refluxed for 1 h. After cooling to ambient tem-
perature the solution was diluted with 60 mL of water and
extracted four times with 50 mL of TBME each. The combined
organic layers were washed with 60 mL of water, evaporated
to dryness and recrystallized from n-hexane/ethyl acetate
(15:2) (androstenedione-17-ethylene-ketal: 553 mg (1.8 mmol,
52%), purity 90%).
2
2
of hydrogen using Pd/C as catalyst. The solution was filtered
and the filtrate was evaporated to dryness. Products were
separated and purified by column chromatography (silica
gel, n-hexane/ethyl acetate 50:50) and dried in a vacuum
desiccator over P4O10 and potassium hydroxide.
3
␤,17␤-Dihydroxy-5␣-androstan-4-one
(19):
110 mg
1
(
0.36 mmol, 18%), purity 97%; H NMR (500 MHz, CDCl3) ı:
1
3
.64/2.42 (m, 2H, H-2), 4.14 (dd, 1H, H-3), 2.20 (dd, 1H, H-5),
.67 (t, 1H, H-17), 0.76 (s, 3H, H-18), 0.76 (s, 3H, H-19); 13C NMR
Hydroboration of the double bond and reduction of 3-
oxo residue was achieved in accordance to the preparation
by Sch a¨ nzer et al. [19]. Crystals of androstenedione-17-
ethyleneketal (1.8 mmol) were dissolved in 90 mL of diethyl
ether. Under argon as cover gas 1.98 mL (15.9 mmol) of boron
trifluoride etherate were added. A suspension of 275 mg of
LiAlH4 (7.6 mmol) in diethyl ether was added dropwise within
20 min. After stirring at room temperature for 70 min the mix-
ture was transferred into a saturated sodium sulfate solution.
Extraction was performed three times with 60 mL of diethyl
ether each, and the combined organic layers were evaporated
to dryness. Oxidation of the BH3-complex was performed by
dissolving the dried residue in 50 mL of methanol, adding
3.3 mL of sodium hydroxide (6 M) and 10.1 mL of hydrogen
peroxide and stirring at room temperature for 1 h. After dilu-
tion with 100 mL of water the preparation was extracted three
times with 150 mL of TBME each, and the combined etheral
layers were evaporated to dryness. Finally, for removal of the
protection group the dry product was dissolved in 100 mL of
acetic acid (30%) and refluxed for 30 min. After addition of
70 mL of water, it was extracted three times with 100 mL of
TBME each, and the organic layers were evaporated to dryness.
The product was purified by column chromatography (silica
gel, n-hexane/ethyl acetate 60:40) and recrystallized from ace-
tonitrile.
(
8
(
125 MHz, CDCl3) ı: 32.3 (C-2), 74.6 (C-3), 212.5 (C-4), 56.9 (C-5),
1.9 (C-17), 11.2 (C-18), 13.9 (C-19). Elemental composition
protonated): C19H31O3, m/z (theor.): 307.2268, m/z (exp.):
3
07.2260, error: −2.4 ppm.
3
␣,17␤-Dihydroxy-5␤-androstan-4-one
(20):
62 mg
1
(
0.20 mmol, 10%), purity 92%; H NMR (500 MHz, CDCl3) ı:
1
3
.59/1.99* (m, 2H, H-2), 4.04 (dd, 1H, H-3), 2.30 (dd, 1H, H-5),
.63 (t, 1H, H-17), 0.74 (s, 3H, H-18), 1.16 (s, 3H, H-19); 13C NMR
(
8
125 MHz, CDCl3) ı: 33.5* (C-2), 75.2 (C-3), 212.3 (C-4), 54.6 (C-5),
2.0 (C-17), 11.2 (C-18), 23.1 (C-19) (* was not identified unam-
biguously). Elemental composition (protonated): C19H31O3,
m/z (theor.): 307.2268, m/z (exp.): 307.2262, error: −2.0 ppm.
2
2
.2.3. Synthesis of 3,4-dihydroxy-5-androstan-17-ones
.2.3.1. Synthesis of 3ˇ,4ˇ-dihydroxy-5ˇ-androstan-17-one (8)
and 3˛,4˛-dihydroxy-5˛-androstan-17-one (11). 5␣- and 5␤-
androst-3-en-17-one were synthesized as described by Da
Silva et al. [16].
Cis-hydroxylation of the 3,4-double bond was achieved
with osmium tetroxide in analogy to literature data [17,18].
To a stirring solution of 67.5 mg (0.25 mmol) of 5␣-androst-
3
-en-17-one in 0.9 mL of pyridine 63.7 mg of OsO4 (0.25 mmol)
were added. After stirring for 2 h at room temperature
1
1
00 mg of sodium bisulfite diluted in 1.8 mL of water and
.2 mL of pyridine were added. The mixture was stirred for
3␤,4␣-Dihydroxy-5␣-androstan-17-one
(10):
15.5 mg
1
(0.05 mmol, 1.4%), purity 98%; H NMR (500 MHz, CDCl3)
ı: 1.91/1.55 (m, 2H, H-2), 3.37 (ddd, 1H, H-3), 3.31 (dd, 1H,
H-4), 1.09 (ddd, 1H, H-5), 0.89 (s, 3H, H-18), 0.88 (s, 3H, H-19);
13C NMR (125 MHz, CDCl3) ı: 28.5 (C-2), 76.9 (C-3), 75.3 (C-4),
50.8 (C-5), 216.3 (C-17), 13.7 (C-18), 13.7 (C-19). Elemental
composition (protonated): C19H31O3, m/z (theor.): 307.2268,
m/z (exp.): 307.2263, error: −1.5 ppm.
another 5 min and then extracted four times with 5 mL of
dichloromethane each. The combined dichloromethane layers
were evaporated to dryness, recrystallized from acetonitrile
and dried in a vacuum desiccator over P4O10 and potassium
hydroxide.
3
␣,4␣-Dihydroxy-5␣-androstan-17-one
(11):
(44 mg
1
(
0.14 mmol, 58%), purity > 99.5%), H NMR (500 MHz, CDCl3)
ı: 1.71/1.83 (m, 2H, H-2), 4.00 (ddd, 1H, H-3), 3.47 (dd, 1H,
H-4), 1.45 (ddd, 1H, H-5), 0.88 (s, 3H, H-18), 0.85 (s, 3H, H-19);
13C NMR (125 MHz, CDCl3) ı: 27.0 (C-2), 69.1 (C-3), 71.6 (C-4),
2.2.3.3. Synthesis of 3ˇ,4ˇ-dihydroxy-5˛-androstan-17-one (4),
3˛,4˛-dihydroxy-5ˇ-androstan-17-one (5), 3˛,4ˇ-dihydroxy-
5ˇ-androstan-17-one (6), 3ˇ,4˛-dihydroxy-5ˇ-androstan-17-
one (7) and 3˛,4ˇ-dihydroxy-5˛-androstan-17-one (9). The
synthesis of these five 3,4-dihydroxy compounds was
described elsewhere [11] and NMR and HRMS data of 4 and
9 are shown below.
4
5.6 (C-5), 221.5 (C-17), 13.8 (C-18), 12.7 (C-19). Elemental
composition (protonated): C19H31O3, m/z (theor.): 307.2268,
m/z (exp.): 307.2260, error: −2.5 ppm.
3␤,4␤-Dihydroxy-5␤-androstan-17-one (8) was prepared as
described above using 13.9 mg (0.05 mmol) of 5␤-androstenone
and 13.0 mg (0.05 mmol) of OsO4 (9.5 mg (0.03 mmol, 62%),
purity > 99.5%). The resulting amount was not adequate for
3␤,4␤-Dihydroxy-5␣-androstan-17-one
(4):
1H
NMR
(500 MHz, CDCl3) ı: 1.59/1.62 (m, 2H, H-2), 3.92 (ddd, 1H,
H-3), 3.65 (dd, 1H, H-4), 1.54 (ddd, 1H, H-5), 0.88 (s, 3H, H-18),

2
82
steroids 7 2 ( 2 0 0 7 ) 278–286
1
(
.07 (s, 3H, H-19); 13C NMR (125 MHz, CDCl3) ı: 35.0 (C-2), 70.1
C-3), 75.8 (C-4), 43.9 (C-5), 221.6 (C-17), 13.8 (C-18), 14.3 (C-19).
Elemental composition (protonated): C19H31O3, m/z (theor.):
with 5 mL of TMBE. The organic layer was separated and evap-
orated to dryness.
2
.4.3. Preparation for GC–MS
3
07.2268, m/z (exp.): 307.2263, error: −1.6 ppm.
␣,4␤-Dihydroxy-5␣-androstan-17-one (9):
1H
Residues were treated with 100 ␮L of MSTFA/ammonium
iodide/ethanethiol (1000:2:3, v/w/v, 80 C for 15 min) for per-
trimethylsilylation and analyzed as such.
3
NMR
◦
(
500 MHz, CDCl3) ı: 1.79/1.64 (m, 2H, H-2), 3.60 (ddd, 1H,
H-3), 3.79 (dd, 1H, H-4), 1.11 (ddd, 1H, H-5), 0.88 (s, 3H, H-18),
.07 (s, 3H, H-19); 13C NMR (125 MHz, CDCl3) ı: 26.1 (C-2), 72.5
C-3), 74.9 (C-4), 49.0 (C-5), 221.7 (C-17), 13.7 (C-18), 14.5 (C-19).
Elemental composition (protonated): C19H31O3, m/z (theor.):
07.2268, m/z (exp.): 307.2263, error: −1.5 ppm.
1
(
3
.
Results and discussion
3
3
.1.
Qualitative consideration of metabolism
2
.2.4. Synthesis of 5ꢀ-androstane-3ꢀ,4ꢀ,17ˇ-triols (21–28)
The transformation of 4-hydroxyandrostenedione and 4-
hydroxytestosterone by male volunteers after single oral
application resulted in the same metabolites. Phase-I
metabolism as well as conjugation of the metabolites were
identical for both administration studies.
5␰-Androstan-3␰,4␰,17␤-triols were prepared by reduction of
respective 17-oxo analogs using NaBH4. To a solution of 1 mg
of each compound in 1.1 mL of methanol/water (11:1) 2 mg of
NaBH4 were added. After 1 h at room temperature, samples
were evaporated to dryness, dissolved in 1 mL of hydrochloric
acid (0.06 M) and extracted with 5 mL of TBME.
Synthesis of 1,2- (30) and 6,7-dehydroformestane (29) was
performed as described by Marsh et al. [9].
3.1.1. Phase-I metabolism
Beside 4-hydroxyandrostenedione and 4-hydroxytestosterone
many reduction products were detected as illustrated in
Table 1. Out of the reduced metabolites with 3-hydroxy-4-oxo
residues the 3␣,5␤- and 3␤,5␣-configurations were identi-
fied. These compounds were detected as 17-hydroxylated and
17-oxo steroids each (2, 3, 19, 20). The further reduced 3,4-
dihydroxylated compounds were mainly detected as 17-oxo
steroids. Except for 3␣,4␣-dihydroxy-5␤-androstan-17-one (5)
all possible combinations of configuration were identified.
Interestingly, only three 17-hydroxylated analogs with 5␣-
configuration were determined, namely 21 (3␤,4␤), 26 (3␣,4␤)
and 28 (3␣,4␣). Furthermore the oxidation products 1,2- (30)
and 6,7-dehydroformestane (29) were identified.
2
.3.
Administration studies
Administration studies were conducted with four healthy
male volunteers who gave their written consent after eth-
ical approval by the ethical committee of the German
Sport University Cologne. Two volunteers applied 100 mg
of 4-hydroxyandrostenedione using the Promatrix product
PrimobolanTM. 4-Hydroxytestosterone was taken as 200 mg
TestobolTM by two volunteers. Analysis of the tablets was per-
formed prior to respective application [20]. Urine samples were
collected continuous for a period of 1 day and additionally 48 h
post administration of 4-hydroxytestosterone. In case of 4-
hydroxyandrostenedione urine specimens were collected over
a period of 72 h and additional morning urine aliquots were
sampled for the following 6 days.
The above mentioned 3,4,17-trihydroxylated metabolites
as well as 3␤-hydroxy-5␣-androstane-4,17-dione (2) and its 17-
hydroxy analog (19) were detected only in minor amounts and
thus excretion times demonstrated in Fig. 5 were not deter-
mined for these metabolites.
2
.4.
Preparation of urine samples
Reduction of 3-oxo-4-ene compounds like the substances
1
and 18 is common in steroid metabolism [21]. In contrast,
A volume of 2 mL of urine was spiked with 60 ␮L of the internal
standard ([2,2,4,4- H4]-11␤-hydroxyandrosterone (12 ␮g/mL)).
1,2- and 6,7-dehydrogenated products are rarely observed.
1,2-dehydrogenation is known from fluoxymesterone and
as bacterial transformation of 3-oxo-4-ene steroids such as
testosterone. 6,7-dehydrogenation, which has also been found
by Poon et al. [8] as metabolite of 4-hydroxyandrostenedione,
has further been observed in the metabolism of metandienone
[21].
2
2
.4.1. Hydrolysis of glucuronides
Samples were diluted with 1 mL of sodium phosphate buffer
(
0.8 M, pH 7). After addition of 50 ␮L (7 units) of ␤-glucuronidase
◦
from E. coli hydrolysis was performed for 1 h at 50 C. Result-
ing unconjugated metabolites were extracted after addition of
For separation and identification of all the configurations
of 3,4-dihydroxylated compounds mass spectra of the per-
trimethylsilylated derivatives were utilized.
0
.75 mL of K2CO3/KHCO3-solution (20%, 1:1, pH 9.6) with 5 mL
of TBME. The organic layer was separated and evaporated to
dryness.
As shown in Fig. 2 the 5␣-steroids produce a very abun-
dant ion at m/z 393. In contrast, the 5␤-compounds generate
intense ions at m/z 303 and 327 (Fig. 3). The origin of m/z
393 as indicated in the EI mass spectrum of 3␣,4␣-dihydroxy-
5␣-androstan-17-one (Fig. 2) was substantiated by deuterium
labelling experiments [11]. Accordingly, derivation of the ion
at m/z 303 was assigned to the same A-ring cleavage plus an
additional loss of TMSOH (−90 u).
2
.4.2. Cleavage of sulfates
The remaining aqueous layer was extracted using solid phase
cartidges. SPE-columns were dried in vacuum desiccator,
and analytes were eluted with 1 mL of methanol. Solvolysis
was perfomed after addition of 5 mL of ethyl acetate/H2SO4
◦
(
250 mL/200 ␮g) at 55 C for 1 h. 0.75 mL of potassium hydroxide
(
1 M) was added, samples were evaporated to dryness, recon-
The mass spectra of the 3,4,17-trihydroxylated steroids
showed resemblances. For the diagnostic ions at m/z 303 and
stituted in 1 mL of potassium hydroxide (1 M) and extracted

steroids 7 2 ( 2 0 0 7 ) 278–286
283
Table 1 – Identification of metabolites
Compound
Molecular weight
Retention index
Identified in administration studies
(per-TMS)
4-Hydroxy-androstenedione
4-Hydroxy-testosterone
*
*
*
1
2
3
4
5
6
7
8
9
0
1
8
9
0
1
2
3
4
5
6
7
8
9
0
518
520
520
522
522
522
522
522
522
522
522
520
520
520
524
524
524
524
524
524
524
524
516
516
2866
2846
2596
2793
2829
2740
X
X
X
X
X
X
X
X
2827
2657
2754
2581
2697
2692
2901
2757
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
*
*
*
2887
2867
2603
2811
2848
2754
2845
2659
2776
2582
2712
2707
2919
2767
2823
2835
X
X
X
X
X
X
X
X
∗
Two retention indices due to isomers resulting from per-silylation.
3
3
93 corresponding counterparts were observed at m/z 305 and
95 resulting from the additional two hydrogen atoms in posi-
tion 17. The 17-hydroxylated steroids show a characteristic
ion at m/z 215 which is assigned to be an additional loss of
90 u (TMSOH) from m/z 305.
Consideration of retention indices demonstrated that elu-
tion of the 17-oxo steroids was similar to the retention of the
corresponding 17-hydroxylated analogs. As shown in Table 1
all 17-hydroxylated compounds elute slightly later in compar-
ison to their 17-oxo analogs which results from the slightly
higher molecular weight.
Fig. 2 – Typical per-TMS mass spectrum of a
,4-dihydroxy-5␣-androstan-17-one
3␣,4␣-dihydroxy-5␣-androstan-17-one).
Comparison of retention indices of the 3,4-dihydroxylated
steroids with the known order of elution of androstanediols
[22] with different configurations at positions 3 and 5 shows
the contribution of the additional 4-hydroxy group, which first
of all results in an adjustment of retention indices to higher
values for all configurations as shown in Fig. 4. Houghton et al.
showed that androstanediols elute in the order 3␤,5␤, 3␣,5␣,
3
(
3␣,5␤ and 3␤,5␣. While the first three configurations elute
with a comparably small difference, the value for retention
of the 3␤,5␣-compound is much higher [22]. 4-Hydroxylated
analogs also show the highest retention indices for the 3␤,5␣-
configuration.
Conversion of 5␣-androstane-3␣,17␤-diol into its 3␤-isomer
results in a positive difference in retention index of 84. The
same conversion of the 5␤-steroid leads to a difference of
−
10. For the 4-hydroxylated compounds these differences
Fig. 3 – Typical per-TMS mass spectrum of a
,4-dihydroxy-5␤-androstan-17-one
3␤,4␤-dihydroxy-5␤-androstan-17-one).
have the same tendencies but higher values. Independently
from the configuration of 4-hydroxylation conversion of the
3␣-hydroxy-5␣-androstanes into their 3␤-analogs leads to
3
(

2
84
steroids 7 2 ( 2 0 0 7 ) 278–286
Table 2 – Characteristic GC–MS data of 3,4-dihydroxyandrostan-17-ones (per-TMS) (m/z 522 was set to 100%)
Compound
m/z 303 (%)
m/z 327 (%)
m/z 391 (%)
m/z 393 (%)
m/z 507 (%)
m/z 522 (%)
3␤4␤5␣
3␣4␣5␤
3␣4␤5␤
3␤4␣5␤
3␤4␤5␤
3␣4␤5␣
3␤4␣5␣
3␣4␣5␣
4
5
6
7
8
9
10
11
11.5
40.7
58.5
42.1
54.9
8.5
10.1
38.6
35.7
50.6
22.6
13.7
8.7
46.8
12.5
13.9
12.0
16.6
36.3
54.1
43.2
78.4
19.1
24.8
17.0
29.7
42.5
86.3
77.2
67.9
51.4
41.2
45.9
44.0
65.7
58.6
58.0
100
100
100
100
100
100
100
100
12.5
11.3
10.8
Fig. 5 – Conjugation and excretion times of the identified
metabolites for the 4-hydroxy-androstenedione excretion
study. Black bars show glucuonidated and white bars show
sulfated metabolites. The long term metabolite is written in
bold letters.
Fig. 4 – Demonstration of retention indices of 3,4␣- and
,4␤-Dihydroxyandrostan-17-ones as per-TMS derivatives
in comparison with retention of per-trimethylsilyated
androstanediols.
3
a positive difference of approximately 140, conversion of
the 5␤-analogs results in a negative difference of retention
indices of approximately −65. Comparison of 4␣- and 4␤-
compounds shows that for 4␣-steroids 5␤-compounds elute
earlier than their 5␣-anaolgs. In contrast, for 4␤-steroids elu-
tion of 5␤-compounds is adjusted to higher values which can
be attributed to a more bulky shape. 5␣-Compounds change to
earlier retention indices as they are more compact than their
compounds have been demonstrated earlier [11,20]. Retention
indices of these substances are listed in Table 1.
3.1.2. Excretion times of conjugated metabolites
Excretion times and periods of traceability have been inves-
tigated for the 4-hydroxyandrostenedione excretion study as
illustrated in Fig. 5. Most of the metabolites are excreted as
glucuronide and sulfate conjugates. Comparing the excre-
tion times of these metabolites (1, 18, 3, 20, 6, 7, 8, 9, 10,
11) shows that glucuronides are excreted longer than sul-
fates except for 3␤,4␤-dihydroxy-5␤-androstan-17-one (8) and
3␣,4␤-dihydroxy-5␣-androstan-17-one (9). 3␤,4␣-Dihydroxy-
5␣-androstan-17-one (10) was detected for the longest period
and found for about 90 h in urine samples as glucuronide. 4-
4␣-analogs.
Assignment of the spectra with the configurations of 3-
and 4-hydroxy groups was performed by retention indices and
intensities of characteristic ions (Tables 2 and 3).
Spectra and characteristics of 4-hydroxyandrostenedione
and 4-hydroxytestosterone as well as the 3-hydroxy-4-oxo
Table 3 – Characteristic GC–MS data of 3,4,17-androstantriols (per-TMS) (m/z 215 was set to 100%)
Compound
m/z 215 (%)
m/z 305 (%)
m/z 393 (%)
m/z 395 (%)
m/z 434 (%)
m/z 509 (%)
m/z 524 (%)
3␤4␤5␣
3␣4␣5␤
3␣4␤5␤
3␤4␣5␤
3␤4␤5␤
3␣4␤5␣
3␤4␣5␣
3␣4␣5␣
21
22
23
24
25
26
27
28
100
100
100
100
100
100
100
100
17.8
38.3
67.6
36.5
40.9
17.5
–
83.7
43.0
85.2
32.2
26.8
85.0
140.0
87.4
97.7
40.7
58.2
39.9
39.9
79.1
105.5
92.4
14.9
16.5
33.0
23.9
32.2
26.3
24.3
27.0
20.1
11.7
–
–
10.2
–
–
–
32.1
–
–
18.1
21.2
17.0
–
–
20.7

steroids 7 2 ( 2 0 0 7 ) 278–286
285
Hydroxyandrostenedione itself was found as glucuronide for
lite 6 may be glucuronidated because of its 3␣-hydroxy
group which would be similar to testosterone metabolism.
For metabolite 7 evaluation and comparison of phase-II
metabolism is difficult because a testosterone metabolite with
3␤,5␤-configuration is barely formed.
Especially the structures found in both conjugate fractions
show the exceptional conjugation of 4-hydroxylated steroids.
Metabolites 9 (3␣,4␤) and 10 (3␤,4␣) as well as 8 (3␤,4␤) and
11 (3␣,4␣) have opposite configuration of the 3- and 4-hydroxy
groups. If conjugation of the 3-hydroxy group would be simi-
lar to that of testosterone metabolites, the 4-hydroxy position
must be conjugated independently from configuration.
Because the 17-hydroxy position is known to be conjugated
in both ways, the 17-hydroxylated metabolites are not consid-
ered more closely.
6
5 h.
Considering especially the 3,4-dihydroxylated steroids
␣-compounds were detected as glucuronic acid and sul-
5
fate except for 3␤,4␤-dihydroxy-5␣-androstan-17-one (4),
which was detected as sulfate only. In contrast, the 5␤-
compounds 3␣,4␤-dihydroxy-5␤-androstan-17-one (6) and
3
␤,4␣-dihydroxy-5␤-androstan-17-one (7) have only been
found as glucuronides while 3␤,4␤-dihydroxy-5␤-androstan-
7-one (8) could be identified in both conjugate fractions.
1
The 5␣-steroids were already found in the first samples
post administration while the 5␤-steroids were not detected
until 8 h p.a. as demonstrated for substances 6, 7 and 8
in Fig. 5.
3
.1.3. Phase-II metabolism
In conclusion, qualitative consideration of phase-I
metabolism of 4-hydroxyandrostenedione and 4-hydroxytes-
tosterone demonstrates that the mainly reductively produced
urinary metabolites are identical. Phase-II metabolism includ-
ing glucuronidation and sulfatation was also comparable for
both administration studies. For inference on the applied sub-
stance a quantitative investigation and subsequent evaluation
of ratios of selected metabolic products may be an opinion.
The long term metabolite 3␤,4␣-dihydroxy-5␣-androstan-
17-one (10) was detected for about 90 h. Using selective ion
monitoring mode, a much longer detection time should be
possible and prolong the traceability of surreptitious misuse
of 4-hydroxyandrostenedione and 4-hydroxytestosterone in
sports drug testing.
The conjugation of the identified metabolites comprises glu-
curonidation and sulfatation. It cannot be deduced from
metabolic routes known from testosterone or other well inves-
tigated steroids as the 4-hydroxy group has a strong influence
on phase-II metabolism.
The metabolites 3␤-hydroxy-5␣-androstan-4,17-dione (2),
its 17-hydroxylated analog 3␤,17␤-dihydroxy-5␣-androstan-
4
-one (19) as well as 3␤,4␤-dihydroxy-5␣-androstan-17-one
(4) with its 17-hydroxylated analog 5␣-androstane-3␤,4␤,17␤-
triol (21) and 5␣-androstane-3␣,4␤,17␤-triol (26) have only
been detected in sulfate fractions, which means that they
are excreted either sulfated or as mixed conjugates. Metabo-
lites only found as glucuronides were the 3,4-dihydroxylated
compounds 3␣,4␤-dihydroxy-5␤-androstan-17-one (6) and
3
␤,4␣-dihydroxy-5␤-androstan-17-one (7) as well as the
oxidation products 29 and 30. The other metabolites,
-hydroxyandrostenedione (1), 4-hydroxytestosterone (18),
the 3-hydroxy-4-oxo compounds 3␣-hydroxy-5␤-androstane-
,17-dione (3) and its 17-hydroxylated analog 3␣,17␤-
Acknowledgement
4
The authors thank the Manfred Donike Institute for doping
analysis for supporting the presented study.
4
dihydroxy-5␤-androstan-4-one (20) as well as the 3,4-
dihydroxy-17-oxo steroids 3␤,4␤-dihydroxy-5␤-androstan-17-
one (8), 3␣,4␤-dihydroxy-5␣-androstan-17-one (9), 3␤,4␣-
dihydroxy-5␣-androstan-17-one (10) and 3␣,4␣-dihydroxy-5␣-
androstan-17-one (11) were found in both conjugate fractions.
These results are in accordance with observations from fur-
ther studies for most of the earlier detected metabolites (1, 18,
r e f e r e n c e s
[1] Dowsett M, Coombes RC. Second generation aromatase
inhibitor—4-hydroxyandrostenedione. Breast Cancer Res
Treat 1994;30(1):81–7.
3
, 20, 29) [5–8]. Metabolite 19 was found in the sulfate fraction,
[
2] Goss PE, Jarman M, Wilkinson JR, Coombes RC. Metabolism
of the aromatase inhibitor 4-hydroxyandrostenedione in
vivo. Identification of the glucuronide as a major urinary
metabolite in patients and biliary metabolite in the rat. J
Steroid Biochem 1986;24(2):619–22.
and Poon et al. [5,8] observed this metabolite as glucuronide.
Further differences were observed for the 3,4-dihydroxylated
metabolites 4 and 9. 3␤,4␤-Dihydroxy-5␣-androstan-17-one (4)
was observed as glucuronide and sulfate in earlier studies
[
3] Dowsett M, Cunningham DC, Stein RC, Evans S, Dehennin L,
Hedley A, Coombes RC. Dose-related endocrine effects and
pharmacokinetics of oral and intramuscular
[
6,8]. In the present consideration, it was only detected in
the fraction of glucuronides. Lonning et al. detected 3␣,4␤-
dihydroxy-5␣-androstan-17-one (9) as sulfate. In this study it
was additionally identified as glucuronide. In none of the men-
tioned studies 4-hydroxyandrostenedione was applied orally
to men which means that differences of conjugation may also
result from different gender or form of application.
4-hydroxyandrostenedione in postmenopausal breast
cancer patients. Cancer Res 1989;49(5):1306–12.
4] Dowsett M, Lloyd P. Comparison of the pharmacokinetics
and pharmacodynamics of unformulated and formulated
4-hydroxyandrostenedione taken orally by healthy men.
Cancer Chemother Pharmacol 1990;27(1):67–71.
5] Poon GK, Jarman M, McCague R, Davies JH, Heeremans CE,
van der Hoeven RA, Niessen WM, van der Greef J.
Identification of 4-hydroxyandrost-4-ene-3,17-dione
metabolites in prostatic cancer patients by liquid
chromatography–mass spectrometry. J Chromatogr
1992;576(2):235–44.
[
[
Considering the present results sulfatation of metabolites
and 4 can be deduced from the conjugation of testosterone
2
metabolites which are also sulfated if featuring a 3␤-hydroxy
group.
Conjugation of the oxidation products demonstrates that
-hydroxy-4-ene structures are only glucuronidated. Metabo-
4

2
86
steroids 7 2 ( 2 0 0 7 ) 278–286
[
6] Lonning PE, Geisler J, Johannessen DC, Gschwind HP,
Waldmeier F, Schneider W, Galli B, Winkler T, Blum W,
Kriemler HP, Miller WR, Faigle JW. Pharmacokinetics and
metabolism of formestane in breast cancer patients. J
Steroid Biochem Mol Biol 2001;77(1):39–47.
ketosteroids for investigation by gas-chromatography and
mass spectrometry. J Chromatogr 1980;202(3):483–6.
[14] Sch a¨ nzer W, Donike M. Synthesis of deuterated steroids for
GC/MS quantification of endogenous steorids. In: Donike M,
Geyer H, Gotzmann A, Mareck-Engelke E, editors. Recent
advances in doping analysis (2). K o¨ ln: Sport und Buch
Strauß; 1995. p. 93–112.
[
7] Poon GK, Chui YC, Jarman M, Rowlands MG, Kokkonen PS,
Niessen WM, van der Greef J. Investigation of conjugated
metabolites of 4-hydroxyandrost-4-ene-3,17-dione in
patient urine by liquid chromatography-atmospheric
pressure ionization mass spectrometry. Drug Metab Dispos
[15] Camerino B. Oxidation of 3-keto steroids in alkaline
medium. Tetrahedron Lett 1961:554–9.
[16] Da Silva EJ, Roleira FM, Sa e Melo ML, Neves AS, Paixao JA, de
Almeida MJ, Silva MR, Andrade LC. X-ray and deuterium
labeling studies on the abnormal ring cleavages of a 5
beta-epoxide precursor of formestane. Steroids
2002;67(3–4):311–9.
[17] Baran DB. Method for the cleavage of osmate esters. J Org
Chem 1960:257.
[18] Fieser LF, Fieser M. Reagents for organic synthesis. New
York: John Wiley & Sons, Inc; 1974. p. 759.
1
992;20(6):941–7.
[
[
8] Poon GK, Jarman M, Rowlands MG, Dowsett M, Firth J.
Determination of 4-hydroxyandrost-4-ene-3, 17-dione
metabolism in breast cancer patients using
high-performance liquid chromatography-mass
spectrometry. J Chromatogr 1991;565(1-2):75–88.
9] Marsh DA, Romanoff L, Williams KI, Brodie HJ, Brodie AM.
Synthesis of deuterium- and tritium-labelled
4
-hydroxyandrostene-3,17-dione, an aromatase inhibitor,
[19] Sch a¨ nzer W, Opfermann G, Donike M. Metabolism of
stanozolol: identification and synthesis of urinary
metabolites. J Steroid Biochem 1990;36(1–2):153–74.
[20] Parr MK, Opfermann G, Sch a¨ nzer W. Analytical properties of
4-hydroxysteroids and some esters. In: Sch a¨ nzer W, Geyer H,
Gotzmann A, Mareck U, editors. Recent advances in doping
analysis (12). K o¨ ln: Sport und Buch Strauß; 2004.
p. 129–38.
[21] Sch a¨ nzer W. Metabolism of anabolic androgenic steroids.
Clin Chem 1996;42(7):1001–20.
[22] Houghton E, Ginn A, Teale P, Dumasia MC, Copsey J.
Comparison of the use of mass spectrometry and methylene
unit values in the determination of the stereochemistry of
estranediol, the major urinary metabolite of
and its metabolism in vitro and in vivo in the rat. Biochem
Pharmacol 1982;31(5):701–5.
[
[
10] Foster AB, Jarman M, Mann J, Parr IB. Metabolism of
-hydroxyandrost-4-ene-3, 17-dione by rat hepatocytes. J
4
Steroid Biochem 1986;24(2):607–17.
11] Parr MK, Fusshoeller G, Opfermann G, Kohler M, Hebestreit
M, Sch a¨ nzer W. Synthesis of reference compounds for the
identification of metabolites of 4-hydroxytestosterone. In:
Sch a¨ nzer W, Geyer H, Gotzmann A, Mareck U, editors.
Recent advances in doping analysis (13). K o¨ ln: Sport und
Buch Strauß; 2005. p. 65–74.
12] The 2005 prohibitied list. World Anti-Doping Agency 2004.
www.wada-ama.org, access 27.12.06.
13] Donike M, Zimmermann J. Preparation of trimethylsilyl,
triethylsilyl and tert-butyldimethlsilyl enol ethers from
[
[
19-nortestosterone in the horse. J Chromatogr
1989;479(1):73–83.