Identification of drostanolone and 17-methyldrostanolone metabolites produced by cryopreserved human hepatocytes

Article abstract of DOI:10.1016/j.steroids.2008.11.002

Methyldrostanolone (2α,17α-dimethyl-17β-hydroxy-5α-androstan-3-one) was synthesized from drostanolone (17β-hydroxy-2α-methyl-5α-androstan-3-one) and identified in commercial products. Cultures of cryopreserved human hepatocytes were used to study the biot

Full text of DOI:10.1016/j.steroids.2008.11.002

Steroids 74 (2009) 306–314
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Identification of drostanolone and 17-methyldrostanolone metabolites
produced by cryopreserved human hepatocytes
Julie Gauthiera, Danielle Goudreaulta, Donald Poirierb, Christiane Ayottea,∗
a INRS-Institut Armand-Frappier, 531, boul. des Prairies, Laval, Québec H7V 1B7, Canada
b CHUQ-CHUL Research Center and Université Laval, Québec, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 13 November 2008
Methyldrostanolone (2␣,17␣-dimethyl-17␤-hydroxy-5␣-androstan-3-one) was synthesized from
drostanolone (17␤-hydroxy-2␣-methyl-5␣-androstan-3-one) and identified in commercial products.
Cultures of cryopreserved human hepatocytes were used to study the biotransformation of drostanolone
and its 17-methylated derivative. For both steroids, the common 3␣- (major) and 3␤-reduced metabolites
were identified by GC–MS analysis of the extracted culture medium and the stereochemistry confirmed
by incubation with 3␣-hydroxysteroid dehydrogenase. Structures corresponding to hydroxylated
metabolites in C-12 (minor) and C-16 were proposed for other metabolites based upon the evaluation
of the mass spectra of the pertrimethylsilyl (TMS-d0 and TMS-d9) derivatives. Finally, on the basis of
the GC–MS and 1H NMR data and through chemical synthesis of the 17-methylated model compounds,
structures could be proposed for metabolites hydroxylated in C-2. All the metabolites extracted from
hepatocyte culture medium were present although in different relative amounts in urines collected
following the administration to a human volunteer, therefore confirming the suitability of the cry-
opreserved hepatocytes to generate characteristic metabolites and study biotransformation of new
Keywords:
Methyldrostanolone
Drostanolone
Metabolites
Hepatocytes
GC–MS
steroids.
© 2008 Elsevier Inc. All rights reserved.
1. Introduction
(nortestosterone). Others, such as tetrahydrogestrinone (13-ethyl-
17-hydroxy-18,19-dinor-17-pregna-4,9,11-trien-3-one) [7], DMT (a
The list of anabolic agents including anabolic androgenic
steroids, prohibited in sports is published by the World Anti-Doping
Agency [1]. In order to detect their utilization, urine samples col-
lected from athletes are tested utilizing comprehensive GC–MS
[2–4] or LC–MS [5,6] methods targeting characteristic, long-lasting
metabolites. Those metabolites were generally identified from
specimens collected following the administration of pharmaceu-
tical preparations to healthy volunteers.
In nearly 10 years, as a perverse consequence of the Dietary
Supplement Health and Education Act adopted in the mid-nineties
in the USA, new substances openly appeared on the market in
products labelled with a rather confusing nomenclature, using
new terminology such as “pro-hormone” or “designer supplement”.
Steroids such as androstenedione (androst-4-en-3,17-dione), dehy-
droepiandrosterone (DHEA) were referred to as testosterone
pro-hormones, while others such as boldione (androsta-1,4-
dien-3,17-dione) and norandrostenedione (estr-4-en-3,17-dione
or the delta-5 isomer) were related respectively to bolde-
none (17␤-hydroxyandrosta-1,4-dien-3-one) and to nandrolone
mixture of 17␤-hydroxy-17␣-methyl-5␣-androst-2-en and the
3-en isomer) [8], norbolethone (13-ethyl-17-hydroxy-18,19-dinor-
17-pregn-4-en-3-one) [9] and methyldrostanolone (17␤-hydroxy-
2␣,17␣-dimethyl-5␣-androstan-3-one) [10] more clandestinely
shared, were characterised by anti-doping laboratories following
seizures, denunciations or through the analysis of athletes’ sam-
ples. No clinical data are available for such molecules; hormonal
properties and health problems [11–14] are discovered a poste-
riori. As the adverse effects of these compounds are not known,
it is difficult to justify on ethical grounds the administration of
these steroids to humans for the purpose of investigating their
metabolism.
Freshly isolated human hepatocytes were successfully utilized
to produce metabolites of androstenedione, norandrostenedione,
tetrahydrogestrinone and gestrinone [15–17], correlating positively
with the results of human excretion studies. Microsomal and S9
fractions of human liver homogenates have been utilized for the for-
mation of phases I and II metabolites of 17-alkylated steroids such
as methyltestosterone (17␤-hydroxy-17␣-methylandrost-4-en-3-
one) and methandienone (17␤-hydroxy-17␣-methylandrosta-1,4-
dien-3-one) [18]. Even so, as microsomes are artefacts, these being
vesicles formed by the endoplasmic reticulum after rupture of the
cell, some enzymes and cofactors can be lost [19]; in this respect,
∗
Corresponding author. Tel.: +1 514 630 8806; fax: +1 450 686 5614.
E-mail address: christiane.ayotte@iaf.inrs.ca (C. Ayotte).
0039-128X/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2008.11.002

J. Gauthier et al. / Steroids 74 (2009) 306–314
307
the use of human hepatocytes for biotransformation offer a distinct
advantage.
tive detector equipped with an 5973 auto-sampler with instrument
control and data analysis by the Enhance ChemStation software
(Agilent Technologies, New Castle, USA) with EI ionization at 70 eV.
A DB-5MS column (SGE) (30 m × 0.22 mm i.d. × 0.25 ␮m film thick-
ness) (Agilent, New Castle, USA) was utilized with the following
temperature program of the oven: 100 ◦C (0.5 min) to 240 ◦C at
20 ◦C/min to 278 ◦C at 6 ◦C/min and 320 ◦C (10 ◦C/min, held 4 min).
The temperature of the injector and the transfer line were respec-
tively270and310 ◦C. Thecarrier gas washelium BIP (BOC, Montréal,
Canada).
Human hepatocytes have been successfully cryopreserved,
thus maintaining their viability and showing comparable drug-
metabolizing enzyme activities to freshly isolated ones; they are
reported to afford under the right conditions, reliable results
for short-term metabolic studies and evaluation of cytotoxicity
[20–23]. The identification of metabolites of new steroid molecules
could be greatly simplified in the incubation medium of cryop-
reserved hepatocytes, given the absence of interfering substances
present in urine matrices.
Having worked on the identification of unknown metabolites in
an athlete’s sample already containing the metabolites of several
anabolic agents, we postulated that a 17-methylated derivative of
drostanolone could now be available. In fact, a few months later,
we identified the presence of 17-methyldrostanolone in hormonal
supplements named Superdrol and Methasterone. Already, a case of
cholestatic jaundice and IgA nephropathy in a bodybuilder has been
associated with that substance [24].
The metabolites of drostanolone have been partially identified
[25,26], while only major reduced metabolites of methyldrostano-
lone were described from human excretion studies [27,28].
The goals of this study were to characterize 17-methyldros-
tanolone and investigate its biotransformation (phase I metabo-
lites) and that of drostanolone in vitro with human cryopreserved
hepatocytes.
2.3. NMR
1H spectra were recorded at 400 MHz using a Brucker AVANCE
400 spectrometer (Billerica, MA, USA). The chemical shifts (ı) are
expressed in ppm and referenced to chloroform (7.26 ppm).
2.4. Synthesis of
2˛,17˛-dimethyl-17ˇ-hydroxy-5˛-androstan-3-one
2.4.1. 3,3-Ethylenedioxy-2˛-methyl-5˛-androstan-17ˇ-ol (2)
To a solution of drostanolone (1) (200 mg, 0.657 mmol) in anhy-
drous dichloromethane (3 mL) at room temperature under argon
were added anhydrous ethylene glycol (240 ␮L, 4.30 mmol), tri-
ethyl orthoformate (370 ␮L, 2.22 mmol) and p-toluene sulfonic
acid monohydrate (13 mg, 0.07 mmol). The solution was stirred
overnight at room temperature. The reaction mixture was then
poured in water and extracted three times with dichloromethane.
The organic layer was dried with MgSO4, filtered and evaporated
under reduce pressure. The resulting crude product was purified by
flash chromatography (silica gel) using ethyl acetate:hexane (3:7,
v:v) as eluent to give 2 (179 mg, 78%). 1H NMR (CDCl3): 0.72 (s,
3H, CH3-19), 0.83 (s, 3H, CH3-18), 0.83 (d, J = 6.6 Hz, 3H, CH3-2␣),
0.70–2.10 (residual CH and CH2), 3.62 (t, J = 8.5 Hz, 1H, CH-17␣), 3.93
(m, 4H, 2 × CH2O).
2. Experimental
2.1. Chemicals and reagents
The following reagents were purchased from Sigma–Aldrich
(Oakville, Canada): tryptan blue, Dulbecco’s Modified Eagle’s
Medium (DMEM) (without red phenol), nicotinamide ade-
nine dinucleotide phosphate NADPH, trimethylsilyliodide
(TMIS), N-methyl-N-(trimethylsilyl)-trifluoracetamide (MSTFA),
pyridine, trimethylsilylchloride (TMCS) and triethylamine.
N,O-bis(trimethyl-d9-silyl) acetamide (BSA-d9) and trimethyl-
d9-silylchloride (TMCS-d9) were obtained from C/D/N Isotopes
(Pointe-Claire, Canada). Distilled water was purified on a Milli-Q
Gradient system from Millipore (Fisher Scientific, Whithy, Canada).
Ethanethiol was obtained from Fisher Scientific (Whithy, Canada),
acetonitrile, diethyl ether, methanol, hexane, and dichloromethane
from Caledon (Georgetown, Canada). Sodium phosphate monoba-
sic (NaH2PO4), sodium bicarbonate (NaHCO3) ACS grade and
sodium carbonate (Na2CO3) were obtained from VWR (Batavia,
USA). The Krebs–Henseleit buffer and the cryopreserved cells
were purchased from Celsis-In Vitro Technologies (Baltimore,
USA). The 3␣-hydroxysteroid dehydrogenase (EC-1.1.1.50) was
purchased from Sigma–Aldrich (Oakville, Canada). The authentic
standards of estr-4-en-17␤-ol-3-one (19-nortestosterone) and
of 17␤-hydroxy-2␣-methyl-5␣-androstan-3-one (drostanolone)
were obtained from Steraloids Inc. (Newport, USA).
2.4.2. 3,3-Ethylenedioxy-2˛-methyl-5˛-androstan-17-one (3)
To a solution of alcohol 2 (68 mg, 0.195 mmol) in anhydrous
dichloromethane (20 mL) at 0 ◦C was added Dess–Martin perio-
dinane (91 mg, 0.215 mmol). The solution was stirred for 40 min
at room temperature. The reaction mixture was then poured in
water and extracted three times with dichloromethane. The organic
layer was dried with MgSO4, filtered and evaporated under reduce
pressure. The resulting crude product was purified by flash chro-
matography (silica gel) using ethyl acetate:hexane (15:85, v:v) as
eluent to give 3 (56 mg, 82%). 1H NMR (CDCl3): 0.83 (d, J = 6.7 Hz,
3H, CH3-2␣), 0.85 (s, 6H, CH3-19 and CH3-18), 0.70–2.10 (residual
CH and CH2), 2.43 (dd, J1 = 8.6 Hz and J2 = 19.4 Hz, 1H of CH2-16),
3.93 (m, 4H, 2 × CH2O).
2.4.3. 3,3-Ethylenedioxy-2˛,17˛-dimethyl-5˛-androstan-17ˇ-ol
(4)
A
solution of anhydrous cerium (III) chloride (260 mg,
Anhydrous reactions were performed in oven-dried glass-
ware under positive argon pressure using commercially available
anhydrous solvents (Gibbstown, USA), except THF distilled from
sodium/benzophenone ketyl under argon. Chemical reagents for
synthesis were purchased from Aldrich Chemical Co. (Milwaukee,
USA). Flash chromatography was performed on Silicycle 60 230-
400-mesh silica gel (Québec, Canada).
1.05 mmol) in anhydrous THF (4.5 mL) was stirred overnight under
argon at room temperature before the addition of a 1.6 M solution of
methyl lithium in diethyl ether (650 ␮L, 1.05 mmol) at −78 ◦C. The
resulting solution was stirred for 30 min and the ketone 3 (38 mg,
1.09 mmol) was added in THF at −78 ◦C and the mixture stirred
for an additional 50 min. The resulting solution was then poured
in water and extracted three times with ethyl acetate. The organic
layer was dried with MgSO4, filtered and evaporated under reduce
pressure. The resulting crude product was purified by flash chro-
matography (silica gel) using ethyl acetate:hexane (15:85, v:v) as
eluent to give 4 (20 mg, 51%). 1H NMR (CDCl3): 0.83 (d, J = 6.6 Hz,
3H, CH3-2␣), 0.84 and 0.85 (2s, 3H, CH3-19 and CH3-18), 1.20
2.2. GC–MS
Analyses in full scan mode (50–750 amu) were performed on
Agilent 6890 gas chromatograph coupled to a 5973 mass selec-

308
J. Gauthier et al. / Steroids 74 (2009) 306–314
(s, 3H, CH3-17␣), 0.70–1.95 (residual CH and CH2), 3.93 (m, 4H,
2 × CH2O).
just before the incubations to 100 ␮M. Typically, a volume of 250 ␮L
of the cell suspension was sampled in 24-well plates to which
250 ␮L of the solution of steroid were added (final concentration of
steroid: 50 ␮M). Incubations were carried out for 4 h or 5 h at 37 ◦C
under an atmosphere of O2:CO2 (95:5). The enzymatic reactions
were stopped by adding 500 ␮L of acetonitrile in each well. Once
transferred in Eppendorf tubes, the samples were vortexed and cen-
trifuged for 10 min at 14,000 rpm. The metabolites were isolated by
solid phase extraction on Sep Pak C18 cartridges (Waters, Montréal,
Canada) as described previously [4]. The steroids were converted
to trimethylsilyl-d0 or -d9 ethers or trimethylsilyl enol-ethers as
described previously, which were analyzed as such by GC–MS.
For scale up experiments, 16 vials containing approximately
8 × 106 cells were placed in 6-well plates (the content of one vial
per well) and incubated with proportional amount of drostanolone
(1) in the same conditions. The metabolites extracted by solid
phase extraction from the incubation media were pooled, solvent
was dried under nitrogen and the residue dissolved in 25 ␮L of
methanol. Metabolite1C was isolated by reverse phase HPLC on Agi-
lent model 1100 equipped with a diode array detector and utilizing
Agilent Zorbax SB-C18 column (5 ␮m, 9.4 mm × 250 mm) (Agilent
Technologies, New Castle, USA); the following gradient of acetoni-
trile and water was used: 50:50 (held 10 min) increased to 65%
acetonitrile in 15 min (held 10 min) to 100% acetonitrile in 10 min
(held 25 min).
2.4.4. 2˛,17˛-Dimethyl-17ˇ-hydroxy-5˛-androstan-3-one
(methyldrostanolone) (5)
To a solution of compound 4 (14 mg, 0.04 mmol) in acetone
(700 ␮L) was added a 10% (v:v) aqueous solution of HCl (280 ␮L).
The solution was stirred for 1 h at room temperature and sodium
bicarbonate was then added up to neutralisation. The reaction mix-
ture was extracted three times with dichloromethane, dried with
MgSO4, filtered and evaporated under reduce pressure. The result-
ing crude product was purified by flash chromatography (silica gel)
using ethyl acetate:hexane (10:90, v:v) as eluent to give 5 (11 mg,
90%). 1H NMR (CDCl3): 0.87 (s, 3H, CH3-18), 1.00 (d, J = 6.5 Hz, 3H,
CH3-2␣), 1.08 (s, 3H, CH3-19), 1.21 (s, 3H, 17␣-CH3), 0.70–2.20
(residual CH and CH2), 2.32 (t, J = 14.0 Hz, 1H), 2.47 (m, 1H).
2.5. Synthesis of
2ˇ,17˛-dimethyl-5˛-androstane-2˛,3ˇ,17ˇ-triol (11)
2.5.1. 2-Methylene-17˛-methyl-5˛-androstane-3ˇ,17ˇ-diol (9)
This compound was prepared from commercially available
oxymetholone (6) as reported by Counsell and Klimstra [29] except
that the acetylation step was omitted and a solution of LiAlH4 in
THF (1.0 M) was used instead of LiAlH4 powder in dioxane.
2.7. Enzymatic reduction with 3˛-hydroxysteroid dehydrogenase
2.5.2. 17ˇ-Hydroxy-17˛-methyl-2-oxirane-5˛-androstane-3-one
(10)
Stock solutions of steroid substrates in methanol (0.01 M) were
diluted in 0.1 M phosphate buffer (pH 6.8) to final concentration of
10 ␮M. NADPH (10 mM) and the enzyme (1 U/mL) were added to
each sample and incubations were performed for 12 h at 37 ◦C in a
water bath. The steroids were extracted with 5 mL of diethyl ether
and treated as described previously.
To
a solution of alkene 9 (45 mg, 0.143 mmol) in chlo-
roform (3 mL) and benzene (3 mL) was added a solution of
m-chloroperbenzoic acid (35 mg, 0.172 mmol) in chloroform (1 mL)
and the mixture stirred overnight. The resulting solution was
diluted with chloroform (20 mL) and then successively washed with
a solution of sodium bisulfite (10%), sodium bicarbonate (10%) and
brine. The organic phase was dried using a phase separator (Biotage)
and evaporated under reduce pressure. The crude product was puri-
fied by flash chromatography (silica gel) using ethyl acetate/hexane
(3:7, v:v) as eluent to give 10 (23 mg, 50%). 1H NMR (CDCl3): 0.84 (s,
3H, CH3-18), 0.88 (s, 3H, CH3-19), 1.21 (s, 3H, CH3-17␣), 0.70–1.85
(residual CH and CH2), 2.56 and 3.17 (2d, J = 5.1 Hz, 2H, CH2O of
oxirane), 3.85 (dd, 1H, CH-3␣, J1 = 5.3 Hz, J2 = 11.3 Hz).
3. Results
3.1. Synthesis of
2˛,17˛-dimethyl-17ˇ-hydroxy-5˛-androstan-3-one
(methyldrostanolone)
As schematized in Fig. 1, the four-step synthesis of methyl-
drostanolone (5) from drostanolone (1) involved a sequential
oxidation and methylation in C-17 after protection of the ketone
in C-3. The presence of a singlet signal at 1.21 ppm in the 1H NMR
spectra is consistent with the stereochemistry of the 17␣-methyl
group, while the presence of the 2␣-methyl group lead to a doublet
signal at 1.00 ppm (2␣-CH3) for both steroids 1 and 5. The mass
spectrum of methyldrostanolone was characterised by the pres-
ence of the molecular ion at m/z 318 and of ions at m/z 303, 300
and 285 (–CH3, –H2O). The major characteristic ions of the TMS-
ether, TMS-enol derivative were attributed to the molecular ion at
m/z 462, to the presence of the 17-Me, 17-OTMS groups (m/z 143
and 130) and to the fragmentation in the A-ring with ions at m/z 141
(C2(Me)–C3(OTMS)), 157, also present in drostanolone (1) [3,10].
2.5.3. 2ˇ,17˛-Dimethyl-5˛-androstane-2˛,3ˇ,17ˇ-triol (11)
Oxirane 10 (25 mg, 0.075 mmol) was suspended in a solution
of anhydrous diethylether, tetrahydrofurane and dioxane (3:2:2,
v:v:v) (7 mL) and LiAlH4 (315 ␮L, 1.0 M THF) was added under atmo-
sphere of argon at room temperature. The solution was stirred at
room temperature for 24 h. The organic layer was diluted with ethyl
acetate (20 mL), washed with a Rochelle salt solution (10%), brine,
dried with MgSO4, filtered and evaporated under reduce pressure.
The crude product was purified by flash chromatography (silica gel)
using ethyl acetate/hexane (1:1, v:v) as eluent to give 11 (8 mg, 32%).
1H NMR (CDCl3): 0.84 (s, 3H, CH3-18), 0.88 (s, 3H, CH3-19), 1.21 (s,
3H, CH3-17␣), 1.27 (s, 3H, CH3-2␤), 0.60–1.85 (residual CH and CH2),
1.91 (d, J = 13.1 Hz, 1H), 3.56 (dd, J1 = 4.6 Hz and J2 = 11.5 Hz, CH-3␣).
2.6. Production of metabolites by incubation of cryopreserved
hepatocytes
3.2. Incubations with cryopreserved human hepatocytes
The phase
I urinary metabolites of 19-nortestosterone in
According to the instructions of the supplier, the cells were
thawed in a thermostatic water bath at 37 ◦C, washed and sus-
pended in 2.5 mL of Krebs–Henseleit buffer (KHB). Viability was
assessed by tryptan blue uptake and the vials contained at least
5 × 106 cells [30].
humans are well known and are formed by the action of 3- and 17-
hydroxysteroid dehydrogenases and 5-reductases [31–34]. Having
utilized nortestosterone as a control substrate, the supernatant
fractions were found to contain after incubations with cryopre-
served hepatocytes (4 h), 19-norandrosterone (3␣-hydroxy-5␣-
Stock methanolic solutions (0.01 M in solvent) of nortestos-
terone, methyldrostanolone and drostanolone were diluted in KHB
androstan-17-one),
19-noretiocholanolone
(3␣-hydroxy-5␤-
androstan-17-one) (minor) and 19-norepiandrosterone (3␤-

J. Gauthier et al. / Steroids 74 (2009) 306–314
309
Fig. 1. Synthesis of methyldrostanolone (5) from drostanolone (1). Reagents and conditions: (a) ethylene glycol, triethyl orthoformate, p-TSA, CH2Cl2, rt; (b) Dess–Martin
periodinane, CH2Cl2, rt; (c) MeLi, CeCl3, THF, −78 ◦C; and (d) HCl 10%/acetone (4:6, v:v), rt.
hydroxy-5␣-androstan-17-one) (major), along with several
hydroxylated metabolites as reported previously from incubations
of fresh hepatocytes [15].
HOTMS and CH3 producing ions at m/z 374, 359, 284 and 269. The
integrity of the D-ring is shown by the presence of the fragment
at m/z 143 while the absence of ions at m/z 141 and 157 further
supports the reduction of the ketone in C-3. The corresponding
metabolites 1A and 1B showing molecular ions at m/z 450 and char-
acteristic ions at m/z 360, 270, 255 and 129 (D-ring fragmentation
[36]) were identified in the incubation mixture of drostanolone (1).
These metabolites are known to be excreted in humans [26]. The
structure and stereochemistry of the 3␣-isomers were confirmed
by the incubation of parent compounds 1 and 5 in the presence of
3␣-hydroxysteroid dehydrogenase [37,38], which lead to the obten-
tion of single isomers, respectively 1A and 5A (results not shown).
The addition of a hydroxyl group in C-16 modifies the frag-
mentation of 17-methylsteroids, shifting D-ring characteristic ions
at m/z 143 and 130 to m/z 231 and 218, the intensity of the lat-
ter being increased [39]. Those two intense ions were present in
the mass spectrum of two minor metabolites 5E and 5G possess-
ing also intense molecular ions at m/z 552 and 550 respectively,
as shown in Fig. 3. Without attempting to further determine
3.3. Incubations with drostanolone and methyldrostanolone
Similar metabolites were formed following incubations of
human hepatocytes with drostanolone (1) and methyldrostanolone
(5) as substrates. The results are therefore presented simulta-
neously. A typical chromatogram from the GC–MS analysis of
the pertrimethylsilyl mixture of metabolites 5A–5H of methyl-
drostanolone (5) is shown in Fig. 2.
The first two metabolites 5A and 5B of methyldrostanolone (5)
were identified as respectively the 3␣- and 3␤-isomers of 2␣,17␣-
dimethyl-17␤-hydroxy-5␣-androstan-3-ol, previously reported as
the major metabolites excreted following administration of a cap-
sule of the commercial product to a human volunteer [28,35]. The
molecular ions of the pertrimethylsilyl derivatives are present at
m/z 464 along with ions corresponding to the sequential losses of
Fig. 2. Total ion chromatogram of the GC–MS analysis (full scan mode) of metabolites (TMS-derivatives) isolated from incubations of cryopreserved human hepatocytes with
methyldrostanolone (5).

310
J. Gauthier et al. / Steroids 74 (2009) 306–314
Fig. 3. Mass spectrum and proposed structures of the pertrimethylsilyl derivative of methyldrostanolone 16-hydroxylated metabolites 5E (upper panel) and 5G (lower panel)
produced from incubations with cryopreserved human hepatocytes.
Fig. 4. Mass spectrum of the pertrimethylsilyl derivative of the major isomer of methyldrostanolone metabolite 5C (upper panel) and drostanolone metabolite 1C (lower
panel) produced from incubations with cryopreserved human hepatocytes.

J. Gauthier et al. / Steroids 74 (2009) 306–314
311
to the sequential losses of three HOTMS groups. Informative ions
suggesting the presence of the third hydroxyl group in A-ring, prob-
ably in C-2, are present at m/z 420, 421 (TMS-d9, 438, 439), 332
(TMS-d9, 341) containing respectively two and one –OTMS groups.
These ions are also found in the mass spectra of the corresponding
drostanolone metabolite at m/z 407, 408 (TMS-d9, 425, 426) and
318 (TMS-d9, 327), while an ion at m/z 157 (TMS-d9, 166) is present
in both metabolites.
Having postulated that the metabolite could bear either
a
2-hydroxy or a 2-hydroxylmethyl group, readily available
17␤-hydroxy-2-(hydroxymethylene)-17␣-methyl-5␣-androstan-
3-one, oxymetholone (6), was utilized as the starting material
to the synthesis of representative molecules (Fig. 5). Firstly, 2␣-
hydroxymethyl-17␤-hydroxy-17␣-methyl-5␣-androstan-3-one
(7) and the corresponding 3␣-OH and 3␤-OH derivatives (8A
and 8B) were obtained by the reduction of oxymetholone (6)
with sodium borohydride in methanol and purification by flash
chromatography [41]. The mass spectra of the pertrimethylsilyl
derivative of isomers 8A and 8B differ significantly from those of
the metabolites 5C and 5F, being characterised by the presence
of low intensity molecular ions at m/z 552 and ions at m/z 462,
447, 372, 357, 262 and 247 attributed to the sequential losses of
HOTMS groups (−90 amu) as shown in Fig. 6. On that basis, the
2-hydroxymethyl structure was excluded.
Four isomers can be produced from the reduction of the ketone
in C-3 (3␣-OH and 3␤-OH) and the hydroxylation in C-2 (2␣-
OH, 2␤-CH3 or 2␤-OH, 2␣-CH3). By a sequence of three steps
presented in Fig. 5, the second model compound 11, the isomer
possessing the 3␤-OH, 2␣-OH, 2␤-CH3 stereochemistry was pre-
pared from oxymetholone (6). The reduction of the hydroxyenone
system of 6 with LiAlH4, followed by the epoxidation of 9 with m-
chloroperbenzoic acid to provide the oxirane 10 as the major isomer
and a regioselective opening of 10 with LiAlH4 gave the tertiary
alcohol 11. The 1H NMR spectrum of 11 shows clearly four singlet
signals at 0.84, 0.88, 1.21 and 1.27 ppm corresponding to methyl
groups in C-13 (CH3-18), C-10 (CH3-19), C-17␣ and C-2, respec-
tively, and a doublet of doublet signal at 3.56 ppm for the CH-3.
From the analysis of the NOESY spectrum of 11, it was also possi-
ble to confirm the CH3-2␤ stereochemistry (a NOE effect between
CH3-19 and CH3-2␤) and the 3␤-OH stereochemistry (no NOE effect
between CH3-2␤ and CH-3␣). The large coupling constants of the
signal at 3.56 ppm (dd) are consistent with a hydroxyl group in the
3␤- instead of the 3␣-configuration.
Fig. 5. Synthesis of 2␣-hydroxymethyl-17␤-hydroxy-17␣-methyl-5␣-androstan-
3-one (7), the corresponding 3␣-OH and 3␤-OH derivatives (8A and 8B)
and 2␤,17␣-dimethyl-5␣-androstane-2␣,3␤,17␤-triol (11) from oxymetholone (6).
Reagents and conditions: (a) NaBH4, MeOH, rt; (b) LiAlH4, dioxane/THF (1:1, v:v),
reflux; (c) m-CPBA, CHCl3/benzene (1:1, v:v), rt; and (d) LiAlH4 (1.0 M in THF),
diethylether/THF/dioxane (3:2:2, v:v:v), rt.
the stereochemistry at C-3 and C-16, the peaks were attributed
to the 16-hydroxylated metabolites of methyldrostanolone (5G)
and its reduced 3-ol metabolite (5E). Equally typical and easily
recognizable fragmentation of pertrimethylsilyl 17-methylsteroids
hydroxylated in C-12 was observed in the mass spectrum of another
very minor metabolite 5D showing the characteristic intense ion at
m/z 170 (presumptive molecular ion at m/z 550) [40].
The mass spectra of the pertrimethylsilyl (TMS-d0) major
metabolite 5C and of the corresponding drostanolone metabolite
1C are shown in Fig. 4. With a mass spectrum almost identical,
metabolite 5F appears to be a minor isomer of 5C. Molecular ion
at m/z 552 (538 for drostanolone metabolite) suggests reduction
in C-3 and hydroxylation, while the presence of ions at m/z 143
and 130 excludes hydroxylation in D-ring. Other ions at m/z 462
(TMS-d9, 480), 372 (TMS-d9, 381) and 282 (TMS-d9, 282) are due
The mass spectrum of the TMS-derivative of synthe-
sized 2␤,17␣-dimethyl-5␣-androstane-2␣,3␤,17␤-triol (11) is
Fig. 6. Structure and mass spectrum of the pertrimethylsilyl derivative of 2␣-hydroxymethyl-17␣-methyl-5␣-androstan-3␣,17␤-diol (8A) obtained by chemical synthesis
from oxymetholone (6).

312
J. Gauthier et al. / Steroids 74 (2009) 306–314
Fig. 7. Mass spectrum and proposed fragmentation pathways of the pertrimethylsilyl derivative of 2␤,17␣-dimethyl-5␣-androstane-2␣,3␤,17␤-triol (11) produced by chemical
synthesis from oxymetholone (6).
Fig. 8. Proposed structures and fragmentation pathways for the pertrimethylsilyl derivatives of drostanolone and methyldrostanolone 2-hydroxylated metabolites 1C and 5C
(major isomers) extracted from incubations with cryopreserved human hepatocytes.
reproduced in Fig. 7. The fragmentation pathways proposed are
presented as well. The mass spectrum is almost identical to the
one obtained for methyldrostanolone metabolite 5C. The retention
times differ therefore suggesting a different stereochemistry at C-2
and C-3. The major reduced metabolites 1A and 5A possess 3␣-OH
groups and that stereochemistry should prevail in all other major
metabolites reduced in C-3.
Drostanolone (1) being available in larger amounts, large scale
incubations with cryopreserved human hepatocytes were carried
out and metabolite 1C could be isolated from the mixture by HPLC
in a quantity sufficient to allow the acquisition of its 1H NMR spec-
trum. The protons of the three intact methyl groups in C-2, C-10
(CH3-19) and C-13 (CH3-18) appear as singlets at respectively 1.24,
1.00 and 0.74 ppm, the CH-3 as a sharp multiplet at 3.55 ppm, the
CH-17 as a doublet of triplets at 3.63 and the OH produces a weak
doublet signal at 3.50 ppm.
The comparison of the key NMR signals of CH3-2 and CH-3 of
synthesized compound 11 with those of isolated compound 1C,
confirms the presence of a tertiary methyl group at C-2 and of a
–CHOH at C-3 but suggests different stereochemistry. This is sup-
ported by the very similar chemical shifts of CH3-2 at 1.27 and
1.24 ppm and CH-3 at 3.56 and 3.55 ppm in compounds 11 and
1C respectively. The shape of the CH-3 signals differs, indicating
a probable 3␣-OH and 2␤-OH, 2␣-CH3 structure for the major
isomers of drostanolone and methyldrostanolone metabolites 1C
and 5C (Fig. 8). While it is not necessary within the scope of this
study to fully confirm the stereochemistry of the metabolites at
this point, further work is currently under progress to confirm the
formation of 2-hydroxylated metabolites from other steroids as
well.
trimethylsilyl derivative of metabolite 5H therefore suggesting a
2,3,16,17-tetrol structure. The mass spectrum shows the presence
of the molecular ion at m/z 640, of ions at m/z 550, 460 and 370
(sequential losses of HOTMS), 231 and 218 (due to C16 -OTMS, C17
-
OTMS-Me) and small ions at m/z 508, 420 and 157 corresponding
to the fragmentation induced by the vicinal 2,3-OTMS groups in the
A-ring as observed with metabolite 5C and synthesized compound
11.
4. Discussion
Eight metabolites, for which structures were proposed, com-
pounds 5A to 5H, were generated by the incubation of
cryopreserved human hepatocytes with methyldrostanolone (5) as
summarized in Fig. 9 (corresponding drostanolone (1) metabolites
(17␣-H) not shown). Whether the hepatocytes were collected from
a single donor or a pool of five, the same metabolites were formed
in slightly varying relative abundance.
Under the conditions employed, cryopreserved hepatocytes
when incubated during 4 or 5 h in the presence of drostanolone
or methyldrostanolone, produced phase I metabolites that are
expected to be found in human urines; the glucuro- and sulfo-
conjugation phase II processes were however not observed. Only
the 3␣- (major) and the 3␤-reduced metabolites have been
reported following the administration of methyldrostanolone (5)
to human volunteers [27,28]. However, once identified from incu-
bation mixtures, a careful re-examination of the urine samples
circulated amongst laboratories as part of an educational test
revealed the presence of the same metabolites hydroxylated in C-2,
C-12 and C-16, and mixed diols (proposed structures). The forma-
tion of metabolites from the incubation of hepatocytes is a static
process, the substrate and metabolites formed remaining in the
Finally, the same fragmentation plus that of 16-hydroxylated
17-methyl steroids was observed in the mass spectrum of the per-

J. Gauthier et al. / Steroids 74 (2009) 306–314
313
Fig. 9. Proposed pathways and structures of metabolites 5A to 5H for the biotransformation of methyldrostanolone (5) in cryopreserved human hepatocytes.
medium for further reaction. In vitro dynamic processes of distribu-
tion, metabolism, conjugation and excretion result in urine samples
containing different metabolites in varying amounts depending
amongst other factors, upon the mode and time of administra-
tion.
analysis of drostanolone metabolite 1C. This was most useful in
proposing a structure. In our view, the relatively high cost of human
hepatocytes is counterbalanced by a more efficient structural anal-
ysis by-passing ethically difficult administration of impure steroids
to human volunteers.
The corresponding metabolites were extracted from cultures
of cryopreserved hepatocytes with drostanolone (1). In humans,
the conjugated urinary metabolites were reported as being mainly
3␣-hydroxy-2␣-methyl-5␣-androstan-17-one, and minor 3␣,17␤-
dihydroxy (1A) and 3␤,17␤-dihydroxysteroids (1B), all produced
by the 3␣-,␤- and 17␤-hydroxysteroid dehydrogenase enzymes
[26]. These metabolites were detected in the incubation mixture
along with other hydroxylated metabolites, not yet identified in
human urine. Studies reported the presence of unexpected hydrox-
ylated metabolites, but no structure assignment was attempted
[26,27].
There are clear advantages to studying biotransformation path-
ways in culture of cryopreserved hepatocytes, one of which being
the simplified structural analysis of the metabolites. The urine
matrix is much more complex and substances are often co-
extracted with the metabolite, interfering with the GC–MS analysis.
For example, the pertrimethylsilyl derivative of the 2,3,16,17-tetrol
5H elutes in a region of the chromatogram where several metabo-
lites of corticosteroids are present. The mass spectrum recorded for
the peak corresponding to the urinary metabolite 5H extracted first
by solid phase (SPE), and liquid–liquid extraction following enzy-
matic hydrolysis of the glucuroconjugates, is compared in Fig. 10 to
the one extracted from culture medium. While the latter can readily
be analyzed, the characteristic ions are lost in the mixed spectrum
of the urinary metabolite.
Fig. 10. Mass spectrum of the pertrimethylsilyl 2,3,16,17-tetrol metabolite 5H of
methyldrostanolone (5) obtained from the incubation with cryopreserved human
hepatocytes (upper panel) and excreted in human urine (lower panel), the latter
being interfered by coeluting substances (corticosteroids metabolites).
It is also possible to purify enough material from large-scale
incubations of cryopreserved hepatocytes to allow the 1H NMR

314
J. Gauthier et al. / Steroids 74 (2009) 306–314
[20] Hengstler JG, Utesch D, Steinberg P, Platt KL, Diener B, Ringel M, et al. Cryop-
Acknowledgements
reserved primary hepatocytes as a constantly available in vitro model for the
evaluation of human and animal drug metabolism and enzyme induction. Drug
Metab Rev 2000;32(1):81–118.
This project has been carried out with the support of the
World Anti-Doping Agency and of NSERC (Grant to Julie Gauthier).
Thanks to Christine Cadot and René Maltais for the chemical syn-
thesis of methyldrostanolone, 17␤-hydroxy-2␣-hydroxymethyl-
17␣-methyl-5␣-androstan-3-one and 2␤,17␣-dimethyl-5␣-andro-
stane-2␣,3␤,17␤-triol. The guidance of Mary Gregory for incuba-
tions of hepatocytes is gratefully acknowledged.
[21] Li AP, Gorycki PD, Hengstler JG, Kedderis GL, Koebe HG, Rahmani R, et al. Present
status of the application of cryopreserved hepatocytes in the evaluation of
xenobiotics: consensus of an international expert panel. Chem Biol Interact
1999;121(1):117–23.
[22] Skett P, Roberts P, Khan S. Maintenance of steroid metabolism and hormone
responsiveness in cryopreserved dog, monkey and human hepatocytes. Chem
Biol Interact 1999;121(1):65–76.
[23] Li AP, Lu C, Brent JA, Pham C, Fackett A, Ruegg CE, et al. Cryopreserved
human hepatocytes: characterization of drug-metabolizing enzyme activities
and applications in higher throughput screening assays for hepatotoxicity,
metabolic stability, and drug–drug interaction potential. Chem Biol Interact
1999;121(1):17–35.
[24] Jasiurkowski B, Raj J, Wisinger D, Carlson R, Zou L, Nadir A. Cholestatic jaundice
and IgA nephropathy induced by OTC muscle building agent superdrol. Am J
Gastroenterol 2006;101(11):2659–62.
References
[1] WADA. 2006. Available from: http://www.wada-ama.org/rtecontent/
document/Explanatory Note 2006.pdf. [updated 2006; cited].
[2] Goudreault D, Masse R. Studies on anabolic steroids. 4. Identification of
new urinary metabolites of methenolone acetate (Primobolan) in human by
gas chromatography/mass spectrometry. J Steroid Biochem Mol Biol 1990;
37(1):137–54.
[3] Goudreault D, Masse R. Studies on anabolic steroids. 6. Identification of urinary
metabolites of stenbolone acetate (17␤-acetoxy-2-methyl-5 alpha-androst-
1-en-3-one) in human by gas chromatography/mass spectrometry. J Steroid
Biochem Mol Biol 1991;38(5):639–55.
[4] Ayotte C, Goudreault D, Charlebois A. Testing for natural and synthetic anabolic
agents in human urine. J Chromatogr B: Biomed Appl 1996;687(1):3–25.
[5] Deventer K, Eenoo PV, Delbeke FT. Screening for anabolic steroids in doping
analysis by liquid chromatography/electrospray ion trap mass spectrometry.
Biomed Chromatogr 2006;20(5):429–33.
[25] Schanzer W. Metabolism of anabolic androgenic steroids. Clin Chem
1996;42(7):1001–20.
[26] de Boer D, de Jong EG, Maes RA, van Rossum JM. The methyl-5 alpha-
dihydrotestosterones mesterolone and drostanolone; gas chromatographic/
mass spectrometric characterization of the urinary metabolites.
Biochem Mol Biol 1992;42(3–4):411–9.
J Steroid
[27] Van Eenoo P, Delbeke FT. Metabolism and excretion of anabolic steroids in
doping control—new steroids and new insights. J Steroid Biochem Mol Biol
2006;101(4–5):161–78.
[28] Parr MK, Opfermann G, Schanzer W, editors. Detection of new 17-alkylated
anabolic steroids on WADA 2006 list. Manfred Donike workshop on dope anal-
ysis. 2006.
[6] Deventer K, Van Eenoo P, Mikulcikova P, Van Thuyne W, Delbeke FT. Quanti-
tative analysis of androst-4-ene-3,6,17-trione and metabolites in human urine
after the administration of a food supplement by liquid chromatography/ion
[29] Counsell RE, Klimstra PD. Anabolic agents. 2-Methylene-5alpha-androstane
derivatives. J Med Chem 1963;6(November):736–8.
[30] In Vitro T. Products>Cryopreserved Hepatocytes. Available from: <http://www.
invitrotech.com/cryo hepatocytes.cfm> [cited].
trap-mass spectrometry.
2005;828(1-2):21–6.
J Chromatogr B: Analyt Technol Biomed Life Sci
[31] Ayotte C. Significance of 19-norandrosterone in athletes’ urine samples. Br J
Sports Med 2006;40(Suppl. 1 (July)):i25–9.
[7] Catlin DH, Sekera MH, Ahrens BD, Starcevic B, Chang YC, Hatton CK. Tetrahydro-
gestrinone: discovery, synthesis, and detection in urine. Rapid Commun Mass
Spectrom 2004;18(12):1245–9.
[8] Sekera MH, Ahrens BD, Chang YC, Starcevic B, Georgakopoulos C, Catlin DH.
Another designer steroid: discovery, synthesis, and detection of ‘madol’ in urine.
Rapid Commun Mass Spectrom 2005;19(6):781–4.
[9] Catlin DH, Ahrens BD, Kucherova Y. Detection of norbolethone, an anabolic
steroid never marketed, in athletes’ urine. Rapid Commun Mass Spectrom
2002;16(13):1273–5.
[10] Ayotte C, Goudreault D, Cyr D, Gauthier J, Ayotte P, Larochelle C, Poirier D,
editors. Characterisation of chemical and pharmacological properties of new
steroids related to doping of athletes. Manfred Donike workshop, 24th cologne
workshop on dope analysis. 2006.
[11] HealthCanada. Health Canada advises consumers not to use unauthorized prod-
ucts containing anabolic steroids. Ottawa: Health Canada; 2006.
[12] Hartgens F, Kuipers H. Effects of androgenic-anabolic steroids in athletes. Sports
Med 2004;34(8):513–54.
[13] Bahrke MS, Yesalis CE. Abuse of anabolic androgenic steroids and related sub-
stances in sport and exercise. Curr Opin Pharmacol 2004;4(6):614–20.
[14] Kutscher EC, Lund BC, Perry PJ. Anabolic steroids: a review for the clinician.
Sports Med 2002;32(5):285–96.
[15] Levesque JF, Gaudreault M, Aubin Y, Chauret N. Discovery, biosynthesis, and
structure elucidation of new metabolites of norandrostenedione using in vitro
systems. Steroids 2005;70(4):305–17.
[16] Levesque JF, Gaudreault M, Houle R, Chauret N. Evaluation of human hepatocyte
incubation as a new tool for metabolism study of androstenedione and noran-
drostenedione in a doping control perspective. J Chromatogr B: Analyt Technol
Biomed Life Sci 2002;780(1):145–53.
[32] Schanzer W, Breidbach A, Geyer H, Van Kuk C, Nolteernsting E, Thevis M, editors.
Metabolism of nortestosterone, norandrostenedione and norandrostenediol;
identification of 3 alpha-hydroxyestr-4-en-17-one glucuronide and 3alpha,
16alpha-dihydroxy-5alpha-estran-17-one glucuronide and sulphate. Manfred
Donike workshop; 18th cologne workshop on dope analysis. 2000.
[33] Masse R, Laliberte C, Tremblay L, Dugal R. Gas chromatographic/mass spectro-
metric analysis of 19-nortestosterone urinary metabolites in man. Biomed Mass
Spectrom 1985;12(3):115–21.
[34] Engel LL, Alexander J, Wheeler M. Urinary metabolites of administered 19-
nortestosterone. J Biol Chem 1958;231(1):159–64.
[35] Rodchenkov G, Sobolevsky T, Sizoi V, editors. New designer anabolic steroids
from internet. Recent advances in doping analysis. Proceedings of the Manfred
Donike workshop on dope analysis. 2006.
[36] Diekman J, Djerassi C. Mass spectrometry in structural and stereochemical
problems. CXXV. Mass spectrometry of some steroid trimethylsilyl ethers. J
Organic Chem 1967;32(4):1005–12.
[37] Steckelbroeck S, Jin Y, Gopishetty S, Oyesanmi B, Penning TM. Human
cytosolic 3alpha-hydroxysteroid dehydrogenases of the aldo-keto reductase
superfamily display significant 3beta-hydroxysteroid dehydrogenase activ-
ity: implications for steroid hormone metabolism and action. J Biol Chem
2004;279(11):10784–95.
[38] Penning TM, Jin Y, Steckelbroeck S, Lanisnik Rizner T, Lewis M.
Structure–function of human
3 alpha-hydroxysteroid dehydrogenases:
genes and proteins. Mol Cell Endocrinol 2004;215(1–2):63–72.
[39] Schanzer W, Donike M. Metabolism of anabolic-steroids in man-synthesis and
use of reference substances for identification of anabolic-steroid metabolites.
Anal Chim Acta 1993;275(April 1 (1–2)):23–48.
[40] Schanzer W, Geyer H, Donike M. Metabolism of metandienone in man: identifi-
cation and synthesis of conjugated excreted urinary metabolites, determination
of excretion rates and gas chromatographic–mass spectrometric identifica-
tion of bis-hydroxylated metabolites. J Steroid Biochem Mol Biol 1991;38(4):
441–64.
[17] Levesque JF, Templeton E, Trimble L, Berthelette C, Chauret N. Discovery, biosyn-
thesis, and structure elucidation of metabolites of a doping agent and a direct
analogue, tetrahydrogestrinone and gestrinone, using human hepatocytes. Anal
Chem 2005;77(10):3164–72.
[18] Kuuranne T, Leinonen A, Thevis M, Schanzer W, Pystynen KH, Kostiainen
R, editors. Metabolism of “new” anabolic steroids: development of in vitro
methodology in metabolite production and analytical techniques. Manfred
Donike workshop 24th cologne workshop on dope analysis. 2006.
[19] Voet DVJ. Biochemistry. Paris: De Boeck Université; 1998.
[41] Oda T, Sato R, Sato Y. The hydrogenation of a-hydroxymethylene-ketone
derivatives to a-hydroxymethyl-ketone derivatives with a cell-free system of
Streptomyces cinereocrocatus. Chem Pharm Bull 1989;37:502–5.