Imitation of phase i oxidative metabolism of anabolic steroids by titanium dioxide photocatalysis

Article abstract of DOI:10.1016/j.ejps.2014.08.009

The aim of this study was to investigate the feasibility of titanium dioxide (TiO₂) photocatalysis for oxidation of anabolic steroids and for imitation of their phase I metabolism. The photocatalytic reaction products of five anabolic steroids were compared to their phase I in vitro metabolites produced by human liver microsomes (HLM). The same main reaction types - hydroxylation, dehydrogenation and combination of these two - were observed both in TiO₂ photocatalysis and in microsomal incubations. Several isomers of each product type were formed in both systems. Based on the same mass, retention time and similarity of the product ion spectra, many of the products observed in HLM reactions were also formed in TiO₂ photocatalytic reactions. However, products characteristic to only either one of the systems were also formed. In conclusion, TiO₂ photocatalysis is a rapid, simple and inexpensive method for imitation of phase I metabolism of anabolic steroids and production of metabolite standards.

Full text of DOI:10.1016/j.ejps.2014.08.009

European Journal of Pharmaceutical Sciences 65 (2014) 45–55
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences
journal homepage: www.elsevier.com/locate/ejps
Imitation of phase I oxidative metabolism of anabolic steroids
by titanium dioxide photocatalysis
Miina Ruokolainen , Minna Valkonen , Tiina Sikanen , Tapio Kotiaho a,b, Risto Kostiainen a,
a
a
a
⇑
a
Division of Pharmaceutical Chemistry and Technology, Faculty of Pharmacy, University of Helsinki, P.O. Box 56 (Viikinkaari 5E), FI-00014, Finland
Laboratory of Analytical Chemistry, Department of Chemistry, University of Helsinki, P.O. Box 55 (A.I. Virtasen aukio 1), FI-00014, Finland
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 April 2014
Received in revised form 15 August 2014
Accepted 19 August 2014
Available online 9 September 2014
The aim of this study was to investigate the feasibility of titanium dioxide (TiO ) photocatalysis for oxi-
dation of anabolic steroids and for imitation of their phase I metabolism. The photocatalytic reaction
products of five anabolic steroids were compared to their phase I in vitro metabolites produced by human
liver microsomes (HLM). The same main reaction types – hydroxylation, dehydrogenation and combina-
2
2
tion of these two – were observed both in TiO photocatalysis and in microsomal incubations. Several iso-
mers of each product type were formed in both systems. Based on the same mass, retention time and
similarity of the product ion spectra, many of the products observed in HLM reactions were also formed
Keywords:
Titanium dioxide photocatalysis
Oxidation
Steroids
Biomimetics
in TiO
also formed. In conclusion, TiO
phase I metabolism of anabolic steroids and production of metabolite standards.
Ó 2014 Elsevier B.V. All rights reserved.
2
photocatalytic reactions. However, products characteristic to only either one of the systems were
2
photocatalysis is a rapid, simple and inexpensive method for imitation of
Metabolism
Mass spectrometry
1
. Introduction
of phase I metabolism reactions (Bernadou and Meunier, 2004;
Meunier, 1992). However, yields were low for some reactions
and additional reactions, which are not observed in CYP mediated
metabolism reactions, can also occur due to lack of selectivity of
metalloporphyrins (Bernadou and Meunier, 2004). In addition,
metalloporphyrins are strong oxidizing agents and easily oxidiz-
able drugs can be overoxidized to stable products, thus bypassing
possible reactive intermediates formed in phase I metabolism reac-
tions. Therefore, the production of the desired product in sufficient
amounts requires careful selection of suitable metalloporphyrin,
oxygen donor, and reaction conditions.
Electrochemistry (EC) can mimic reactions which can be
initiated with single electron transfer step, such as N-dealkylation,
S- and P-oxidation, alcohol oxidation and dehydrogenation
(Johansson et al., 2007; Jurva et al., 2003). In contrast, reactions
initiated by hydrogen atom abstraction, such as O-dealkylation,
aliphatic hydroxylation, or hydroxylation of non-substituted aro-
matic rings, have higher oxidation potential than water, and thus,
these are less likely reactions as solvent is oxidized instead of the
organic compound (Nouri-Nigjeh et al., 2011). However, overoxi-
dation is common in hydroxylation of substituted aromatic rings,
as the hydroxylation products may be oxidized further at lower
potentials than the starting compound (Jurva et al., 2003).
The metabolism of a drug candidate in vitro is usually studied
using hepatocytes, microsomes, or recombinant enzymes. Although
drug metabolism is in general a safe and natural way to facilitate
the elimination of drugs from the body, phase I metabolism can
transform drugs into pharmacologically active or toxic compounds,
and metabolism is also an important cause for drug candidate fail-
ure. Thus, metabolism studies have shifted to earlier stages of the
drug discovery process reflecting the ‘fail early – fail cheaply’ para-
digm (Ekins et al., 2000). The increasing number of compounds to
be tested has provoked a need for faster, cheaper, or more conve-
nient alternatives to the traditional in vitro enzymatic metabolism
experiments. As the most important phase I metabolic reactions
are oxidation reactions catalyzed by cytochrome P450 (CYP) isoen-
zymes, various nonenzymatic oxidation methods, such as metallo-
porphyrins (Bernadou and Meunier, 2004), Fenton reaction (Liang
et al., 2013; Van der Steen et al., 1973; Zbaida et al., 1986), and elec-
trochemical reactions (Johansson et al., 2007; Jurva et al., 2003),
have been studied for mimicking phase I metabolism reactions.
Metalloporphyrins acting as surrogates for the active centers of
CYP enzymes have been shown to be capable of imitating all types
Fenton reaction involves hydrogen peroxide (H
ferrous salt (Fenton, 1894). Fe is oxidized to Fe , while H is
dissociated to a hydroxyl ion and a hydroxyl radical (Haber and
Weiss, 1932), which can react with various organic compounds
2
O
2
) and a
⇑
Corresponding author. Tel.: +358 2941 59134.
E-mail addresses: miina.ruokolainen@helsinki.fi (M. Ruokolainen), tiina.sikanen
2+
3+
2 2
O
@helsinki.fi (T. Sikanen), tapio.kotiaho@helsinki.fi (T. Kotiaho), risto.kostiainen@
helsinki.fi (R. Kostiainen).
http://dx.doi.org/10.1016/j.ejps.2014.08.009
0
928-0987/Ó 2014 Elsevier B.V. All rights reserved.

4
6
M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
by addition to unsaturated bonds or hydrogen abstraction from ali-
2000), however the addition of binding groups to the steroid
substrate is necessary to control the orientation of the steroid
respective to the metalloporphyrin. Jurva et al. showed that two
different aliphatic hydroxylation products of testosterone
were formed in EC-Fenton and metalloporphyrin catalyzed reac-
tions, as well as in experiments made with liver microsomes
(Johansson et al., 2007). Direct EC reactions did not, however, pro-
duce aliphatic hydroxylation reaction products with testosterone,
suggesting that direct EC is not the method of choice for mimicking
phatic carbons. Active Fe2 can be regenerated from Fe by using a
chemical reductive agent, such as ascorbic acid, or electrochemical
reduction at the working electrode (Jurva et al., 2002). Electro-
chemically assisted Fenton (EC-Fenton) reaction has been shown
to be capable of imitating hydroxylations, dealkylations and het-
eroatom oxidations (Johansson et al., 2007). However, Fenton reac-
tion is relatively non-selective (Barry et al., 2012) because of its
hydroxyl radical based mechanism (Barry et al., 2012; Mile, 2000).
+
3+
Titanium dioxide (TiO
for degradation of organic pollutants (Bhatkhande et al., 2002;
Fujishima et al., 2000). TiO can catalyze both oxidation and reduc-
tion reactions when exposed to ultraviolet (UV) light of high
enough energy to excite the electrons from the valence band to
2
) photocatalysis has been widely applied
phase
hydroxylation products and further oxidation products were
observed in TiO photocatalytic reactions of dexamethasone
(Calza et al., 2001). The hydroxyl radical based mechanism enables
aliphatic hydroxylation, which makes TiO photocatalysis
I metabolism reactions of steroids. Several aliphatic
2
2
2
a
the conduction band of TiO
2
leaving holes on valence band
potential method for the imitation of steroid metabolism. In these
biomimetic oxidation studies, comparisons with enzymatic reac-
tions, using for example human liver microsomes (HLM), are
unfortunately very limited and therefore solid conclusions on the
usefulness of the mimetic oxidation methods for predicting phase
I reactions of steroids cannot be made.
(Fig. 1). The holes and electrons can either recombine, or migrate
2
to the surface of a TiO particle, where they can react with water
or directly with organic compounds. Reactive hydroxyl radicals
Å
Åꢀ
2
(
OH) and superoxide anions (O
) are formed in aqueous solution.
photocatalysis
Few studies have described the application of TiO
2
for imitation of phase I metabolism reactions (Calza et al., 2004;
Medana et al., 2011, 2013; Nissilä et al., 2011; Raoof et al., 2013),
but on the basis of these recent reports, TiO photocatalysis seems
2
2
In this work we study the feasibility of TiO photocatalysis for
oxidizing and imitation of phase I metabolism of five anabolic ste-
roids (Fig. 2). The reaction conditions, with respect to reaction time
to be a promising alternative to imitation of drug metabolism,
since several biologically important reactions are possible. Hydrox-
ylation, dehydrogenation, N- and O-dealkylation reactions have
2
and solvent composition, are studied. The products from TiO pho-
tocatalytic oxidizing reactions are compared to those produced by
in vitro phase I metabolic reactions using human liver microsomes.
An ultra high performance liquid chromatography–high resolution
mass spectrometric (UHPLC–HRMS) method is developed for the
analysis of reaction products. The method provides high chromato-
graphic resolution for the separation and analysis of possible
2
been observed in TiO photocatalysis (Nissilä et al., 2011) and
TiO photocatalysis has been shown to produce products, which
2
are similar to the metabolites formed in vivo and in vitro (Calza
et al., 2004; Medana et al., 2011, 2013; Nissilä et al., 2011; Raoof
et al., 2013).
Steroids are lipophilic, important endogenous compounds,
which play a number of important physiological roles. Steroids
are synthesized from cholesterol via numerous enzymatic reac-
tions resulting in variety of biologically active forms. Since steroids
regulate many cellular and physiological actions, the metabolism
of steroids has been largely studied by using in vivo and in vitro
experiments. Poor water solubility of lipophilic steroids can limit
usefulness of biological in vitro metabolism assays. The solubility
of steroids can be increased by adding organic solvent to aqueous
solvents. However, even low concentrations of organic solvents
(
1
2–5%) can significantly reduce enzyme activity (Busby et al.,
999; Chauret et al., 1998; Gonzalez-Perez et al., 2012; Hickman
et al., 1998; Li, 2009).
Although biomimetic oxidation techniques can circumvent the
limitations presented by the use of organic solvents in enzymatic
metabolism assays (Bernadou and Meunier, 2004; Johansson
et al., 2007) they have seldom been used for mimicking the metab-
olism reactions of steroids and other poorly soluble compounds.
Metalloporphyrins have been shown to catalyze hydroxylation of
some steroid substrates regioselectively (Breslow et al., 1997;
Fang and Breslow, 2006; Stuk et al., 1991; Yang and Breslow,
Fig. 1. The principle of TiO
2
photocatalysis. Modified from (Bhatkhande et al.,
Fig. 2. Molecular structures, names, abbreviations, molecular formulas and mono-
isotopic masses of the test compounds used in this study. Top-left figure represents
the numbering of the different carbon atoms in the steroid structure.
2
002). See Fujishima et al. (2008) and Zhang et al. (2012) for more detailed
2
descriptions of the TiO photocatalysis mechanism.

M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
47
structural isomers. Separation of isomeric products is particularly
important in order to examine, if the oxidation site is the same
24 h. The reaction was terminated by protein precipitation with
4 M perchloric acid (10 L) and transferring the tubes into an
l
in phase I metabolism and TiO
. Materials and methods
.1. Chemicals
Methyltestosterone (MT) and nandrolone (NAN) were from
2
photocatalytic reactions.
ice bath. The precipitated HLM were removed by centrifugation
at 13,200 rpm for 10 min. The supernatants were pooled from 1,
3
, 6, 12, 18, and 24 h incubations. Blank incubations were carried
2
out similarly without the substrate. Control experiments were
also made to ensure, that perchloric acid does not oxidize the
steroids.
2
2.4. Solid phase extraction
Diosynth (Oss, the Netherlands), metandienone (MDN) and andro-
stenedione were from Steraloids (Newport, RI, US), and stanozolol
The second supernatant (450
l
L) of photocatalytic reaction and
L) from HLM incubations were
diluted each with 1 mL of 100 mM ammonium acetate buffer (pH
.0) for solid phase extraction. Waters Oasis 30 mg 1 mL cartridges
(
STZ) was from Sterling-Winthorp (New York, NY, US). Testoster-
the pooled supernatant (270
l
one (T), TiO
2
Degussa P-25, glacial acetic acid (P99.85%), ammo-
0
nium acetate, b-nicotinamide adenine dinucleotide 2 -phosphate
reduced tetrasodium salt hydrate (NADPH), sodium dihydrogen
phosphate, acetonitrile (ACN) and methanol were from Sigma–
Aldrich (Steinheim, Germany). Magnesium chloride, perchloric
acid (70–72%), disodium hydrogen phosphate and sodium hydrox-
ide were from Merck (Darmstadt, Germany). Water was purified
with a Milli-Q Plus purification system (Molsheim, France). Human
liver microsomes were purchased from BD Gentest (New Jersey,
USA). The solvents were LC–MS grade and all the reagents at least
reagent grade.
6
were conditioned with 1 mL of methanol and 1 mL of purified
water. The sample was loaded to the cartridge, which was then
washed with 1 mL of water–methanol 95:5 (v:v) and 1 mL of
water–methanol 95:5 (v:v) +2% acetic acid. Steroids and the reac-
tion products were eluted from the cartridge with 1.5 mL acetoni-
trile (testosterone, methyltestosterone, metandienone, and
nandrolone) or 1.5 mL methanol (stanozolol). The samples were
evaporated to dryness in a water bath under nitrogen. Dry residues
were reconstituted in 100 lL (HLM samples) or 50 lL (photocatal-
ysis samples) of the LC mobile phase A:B 75:25 (v:v). The eluent A
was 95:5 water–100 mM ammonium acetate (pH 6.0, adjusted
with acetic acid) and eluent B was 95:5 methanol–100 mM ammo-
nium acetate (pH 6.0). Photocatalysis samples used for studying
the effect of the organic solvent content and the duration of the
UV exposure were not purified nor enriched with solid phase
extraction.
2
.2. Photocatalytic reactions
Photocatalytic reactions were performed in duplicate, and blank
samples were prepared without a steroid. The photocatalytic
reaction mixture (500
TiO Degussa P25 particles. The effect of acetonitrile content was
ꢀ
1
lL) contained 10 lM steroid and 1 g L
2
studied with testosterone with different percentages of acetonitrile
in the reaction mixture. With water–acetonitrile 99:1, the UV
exposure times used were 0, 1, 2, 3, 4, 6, and 12 min, and with
water–acetonitrile 50:50 and 5:95, 0, 1, 5, 10, 15, 20, and 30 min.
The UV exposure times used for comparison of photocatalytic
reaction products to phase I in vitro metabolites were 2 min for
testosterone, methyltestosterone, metandienone, and nandrolone
in water–acetonitrile 99:1, and 15 min for stanozolol in water–
acetonitrile 50:50. The samples were magnetically stirred while
UV exposed from above with 5000-PC Series Dymax UV Curing
Flood Lamp (Dymax Light Curing Systems, Torrington, CT, USA).
The nominal intensity of the metal halide UV lamp was 225 mW
2.5. Liquid chromatography–mass spectrometry
Samples were analyzed with ACQUITY UPLC™ (Waters, Milford,
MA, USA) and Xevo™ Q-TOF-MS (Waters, Manchester, UK) instru-
ments. An Acquity UPLC BEH C-18 column (100 mm ꢁ 2.1 mm
i.d., 1.7 lm particle size) was used for chromatographic separation
of reaction products of each steroid. A Vanguard BEH C18
(
5 mm ꢁ 2.1 mm i.d., 1.7
l
m particle size) precolumn was used in
front of the analytical column. A fast gradient (Table A.1) was used
in the optimization of photocatalytic reaction conditions, after
which the gradients were optimized separately for the reaction
2
ꢀ1
(
cm ) . After UV exposure, the TiO
centrifugation at 13,200 rpm for 10 min. The first supernatant
470 L) was centrifuged again at 13,200 rpm for 10 min. The sec-
ond supernatant (450 L) was collected for analysis without fur-
2
particles were removed by
products of each steroid (Tables A.2–A.6). The flow rate of the
ꢀ1
mobile phase was 300
l
L min . Injection volume was 5
lL with
(
l
partial loop injection mode. The sample tray was kept at 8 °C and
the column at 40 °C. MS and MS/MS spectra were collected in the
positive ion mode using electrospray ionization (ESI). Nitrogen
l
ther pretreatment when the effect of acetonitrile content was
studied, and was pretreated as described in 2.4, when used for
comparison to metabolism samples.
ꢀ1
and argon were used as the desolvation (800 L h , 450 °C) and col-
lision gas, respectively. The source temperature was 120 °C and
capillary voltage was 3 kV. The cone voltage was 25 V and extrac-
tion cone voltage 3 V. Data were acquired with resolution of
2.3. HLM incubations
1
0,000 from 30 to 560 Da with a scan time of 0.1 s and the centroid
Phase I metabolism reactions were studied in vitro using HLM.
mass data was corrected during acquisition using an external refer-
ꢀ1
The incubations were carried out in a total volume of 100
taining 50 M substrate, 5 mM MgCl , 50 mM phosphate buffer
pH 7.4), and 0.5 mg mL total protein. Each steroid was incu-
lL con-
ence (Lockspray) leucine enkephalin (concentration 2 ng
rate 20 lL min ) at m/z 556.2771. Extracted ion chromatograms
l
L
, flow
ꢀ
1
l
2
ꢀ1
(
were generated with 20 ppm mass window. Two collision energies
were used in the MS/MS analysis of each steroid (Table A.7). There
were up to 5 simultaneous MS/MS scans in addition to the MS scan.
Metabolynx XS V4.1 software (Waters, Milford, MA, USA) was used
to search for unexpected reaction products. If a reaction product
bated separately and added into the incubation mixture in aceto-
nitrile (testosterone, methyltestosterone, metandienone, and
nandrolone) or methanol (stanozolol) solution. The final concen-
tration of organic solvent was 1%. The reaction was initiated by
addition of 5
a dry bath of 37 °C. The final concentration of NADPH in the incu-
bation mixture was 1 mM. 5 L of 20 mM NADPH was added
again after 6 h when the total incubation time was 12, 18, or
l
L NADPH (20 mM) and transferring the tubes on
with the same m/z and retention time (t
R
) as in HLM experiments
was found also in TiO photocatalytic reactions, its product ion
2
l
spectrum was compared to the product ion spectrum of the corre-
sponding metabolite.

4
8
M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
3
. Results and discussion
.1. Optimization of reaction and measurement conditions
In order to study the feasibility of TiO photocatalysis in mim-
Degradation of testosterone was rapid and the photocatalytic
reaction products formed fast in water–acetonitrile 99:1
(Fig. 4). The highest abundance of reaction products was achieved
in 2 min after which their degradation was faster than formation.
Testosterone and the oxidation reaction products were degraded
completely in 12 min. As expected, increasing acetonitrile con-
centration inhibited the reaction: The degradation of testosterone
and the formation of reaction products were slower in water–
acetonitrile 50:50 and 5:95 than in water–acetonitrile 99:1.
These results show that the addition of acetonitrile to the
3
2
icking phase I metabolism reactions of anabolic steroids, a
UHPLC–ESI/HRMS method using a q-tof mass spectrometer was
developed for the analysis of reaction products. As several isomeric
oxidation products of the selected anabolic steroids were formed in
the reactions, the chromatographic conditions were optimized
(
Tables A.2–A.6) in order to maximize the separation efficiency
2
reaction mixture provides a simple way to control the TiO phot-
for isomeric reaction products. Good chromatographic perfor-
mance was recognized as narrow and non-tailing peaks that
allowed separation of several isomeric oxidation products of the
anabolic steroids formed in the reactions (Fig. 3). The relative stan-
dard deviations of retention times were typically less than 0.25%
indicating good chromatographic repeatability that is required in
ocatalyzed oxidation reactions. In water–acetonitrile 50:50, the
abundance of reaction products increased with increasing UV
exposure time during the 30 min time frame studied, but in
water–acetonitrile 5:95, the formation of the products became
slower than their degradation by the end of the 30 min time
frame studied.
order to allow for comparison of the products formed in TiO
2
pho-
Acetonitrile was found not only to inhibit but also to direct the
photocatalytic reactions. In water–acetonitrile 99:1 and 50:50, the
same products were observed, although their relative intensities
were different (Table 1). In water–acetonitrile 50:50 and water–
acetonitrile 5:95, the formation of a dehydrogenation product
tocatalysis and HLM reactions. ESI mass spectra of the anabolic ste-
roids and their oxidation products showed an abundant protonated
molecule with few fragment ions typically formed by the loss of
one or two water molecules.
In TiO
2
photocatalysis, addition of organic solvent to water was
(Mꢀ2H, m/z 287.20, t
tion of hydroxylation (M+O, m/z 305.21, t
ation + dehydrogenation (M+Oꢀ2H, m/z 303.19,
R
3.34 min) was clearly favored over forma-
2.99 min) and hydroxyl-
2.67 min)
necessary in order to increase the solubility of anabolic steroids.
Acetonitrile was selected as organic solvent because it has weaker
radical scavenging activity than methanol (Li, 2013). Thus, it was
expected, that acetonitrile allows the photocatalytic reaction but
slows down the reaction rate by reducing hydroxyl radicals avail-
able for oxidation reactions. The effect of acetonitrile on the oxida-
tion reaction rates was studied with three different acetonitrile
concentrations: water–acetonitrile 99:1, 50:50 and 5:95 using tes-
tosterone as a model compound.
R
t
R
products, whereas these three product types were formed with
relatively similar abundances in water–acetonitrile 99:1 (Fig. 4).
The addition of acetonitrile, having electron scavenging capability,
inhibits formation of reactive oxygen species (Fig. 1) and thus
formation of hydroxylation products of the steroids. Therefore, at
higher acetonitrile concentrations steroids will rather react
2
directly with the photogenerated holes on the TiO surface via
Fig. 3. Extracted ion chromatograms of (a) m/z 305.2117 and (b) m/z 303.1960 corresponding to M+O and M+Oꢀ2H reaction products of testosterone, respectively. Peaks
observed at the same retention time and with matching MS/MS spectra in TiO photocatalysis and HLM reactions were considered identical compounds.
2

M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
49
100
200
(
a)
(b)
75
50
25
0
150
100
ACN 1%
ACN 1%
ACN 50 %
ACN 95 %
ACN 50 %
50
ACN 95 %
0
0
10
20
30
0
5
10
15
20
25
30
UV exposure time (min)
UV exposure time (min)
1
8
2
6
0
15
(
c)
(d)
1
2
9
6
3
0
1
ACN 1%
ACN 1%
ACN 50 %
ACN 95 %
ACN 50 %
ACN 95 %
0
5
10
15
20
25
30
0
5
10
15
20
25
30
UV exposure time (min)
UV exposure time (min)
Fig. 4. Effect of acetonitrile concentration on degradation of testosterone and formation of selected photocatalytic reaction products of testosterone. (a) Degradation of
testosterone (m/z 289.22, t 3.34 min), (c) formation of an M+O product (m/z 305.21, t 2.99 min), and (d)
R
3.55 min), (b) formation of an Mꢀ2H product (m/z 287.20, t
R
R
formation of an M+Oꢀ2H reaction product (m/z 303.20, t
R
2.67 min) of testosterone.
Table 1
Reaction products observed in TiO
identical compounds.
2
photocatalytic oxidation of testosterone in different acetonitrile concentrations. Peaks observed at the same retention time were considered
Reaction
m/z
t
R
(min)
ACN 1%
x
ACN 50%
x
ACN 95%
x
Dehydrogenation
Mꢀ2H
287.20
3.34
Hydroxylation
M+O
305.21
303.20
2.63
2.89
2
3
2.67
2.79
2
2
x
x
x
x
x
x
.99
.16
x
x
x
x
Hydroxylation + dehydrogenation
M+Oꢀ2H
x
x
x
x
x
x
x
x
x
x
.88
.99
x
Dihydroxylation
M+2O
321.21
319.19
2.80
3.07
3
x
x
x
.17
Dihydroxylation + dehydrogenation
2.23
x
x
M+2Oꢀ2H
Fig. 5. Examples of correlation of MS/MS spectra (CE 20 V) in TiO
2
photocatalysis and HLM reactions. (a) MS/MS spectra of M+O (m/z 305.2117, t
R
7.60 min) and (b) MS/MS
spectra of M+Oꢀ2H (m/z 303.1960, t 9.62 min) reaction product of testosterone.
R

5
0
M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
Table 2
Coverage of HLM reactions by TiO
2
photocatalysis calculated using Eq. (1).
MT (%)
Reaction/steroid
NAN (%)
100
T (%)
50
STZ (%)
0
MDN (%)
0
Dehydrogenation
Mꢀ2H
50
63
40
Hydroxylation
M+O
75
13
43
33
67
33
17
40
Hydroxylation + dehydrogenation
M+Oꢀ2H
one electron oxidation followed by deprotonation producing dehy-
drogenation products of the steroids.
2
HLM metabolism and TiO photocatalytic reactions based on the
ion chromatograms of protonated molecules recorded with
20 ppm mass window and 0.5% retention time window (Fig. 3,
Generally the formation of M+O and M+Oꢀ2H products was
less abundant and slower in water–acetonitrile 5:95 than in
water–acetonitrile 50:50, and many of these products were not
formed at all in water–acetonitrile 5:95, while some products were
only formed in water–acetonitrile 5:95 (Table 1). In most cases,
however, a greater abundance of products could be achieved in
water–acetonitrile 50:50, than in water–acetonitrile 99:1, with
longer UV exposure time. This is probably due to the slower degra-
dation of formed products when more radical scavenging acetoni-
trile is present.
Table A.3
Gradient for reaction products of nandrolone.
Time (min)
.00–10.00
10.00–20.00
20.00–22.00
B (%)
0
25 ? 40
40 ? 100
100
2
2
2.00–22.10
2.10–24.50
100 ? 25
25
Higher acetonitrile percentages with longer UV exposure times
produce greater abundance of oxidation products, which may be
useful in photocatalytic synthesis of metabolite standards. How-
ever, water–acetonitrile 99:1 and a UV exposure time of 2 min
were chosen for photocatalytic reactions for comparison to phase
I in vitro HLM metabolism reactions, because these conditions are
closer to physiological conditions. Furthermore, the reactions are
inhibited less, and are thus much faster and more convenient than
with higher acetonitrile concentrations. However, stanozolol was
not soluble enough in water–acetonitrile 99:1, and thus water–
acetonitrile 50:50 and UV exposure time of 15 min were used for
stanozolol.
Table A.4
Gradient for reaction products of testosterone.
Time (min)
.00–13.00
13.00–18.00
B (%)
0
33 ? 50
50 ? 100
100
1
2
2
8.00–20.00
0.00–20.10
0.10–22.50
100 ? 33
33
Table A.5
Gradient for reaction products of metandienone.
3.2. Comparison of reaction products from TiO
2
photocatalysis and
phase I metabolism in HLM
Time (min) B (%)
0
7
.00–7.50
.50–15.00
30 ? 40
40 ? 45
45 ? 100
100
100 ? 30
30
The main reaction types, hydroxylation and dehydrogenation
and/or hydroxylation + dehydrogenation, were mostly the same
in TiO photocatalysis and in the phase I HLM metabolism reac-
2
15.00–20.00
2
2
2
0.00–22.00
2.00–22.10
2.10–24.50
tions. Several same isomeric oxidation products were formed in
Table A.1
Gradient for reaction products of testosterone used in
optimization of photocatalytic reaction conditions. Eluent
A was 95:5 water–100 mM ammonium acetate (pH 6.0,
adjusted with acetic acid) and eluent B was 95:5 meth-
anol–100 mM ammonium acetate (pH 6.0).
Table A.6
Gradient for reaction products of stanozolol.
Time (min)
B (%)
0
4
.00–4.00
.00–11.00
30 ? 58
58 ? 64
64 ? 100
100
100 ? 30
30
Time (min)
B (%)
0.00–2.50
2.50–4.00
4.00–4.10
4.10–6.50
50 ? 100
100
100 ? 50
50
11.00–15.00
15.00–17.00
17.00–17.10
17.10–19.50
Table A.2
Gradient for reaction products of methyltestosterone.
Table A.7
Collision energies (CE) used in acquiring product ion spectra of
each steroid and its reaction products.
Time (min)
B (%)
Steroid
CE 1 (V)
CE 2 (V)
0
9
1
1
2
2
.00–9.00
30 ? 45
45 ? 50
50 ? 100
100
100 ? 30
30
.00–15.00
5.00–19.00
9.00–21.00
1.00–21.10
1.10–23.20
T
20
20
15
20
42
30
25
20
25
47
MT
MDN
NAN
STZ

M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
51
Table A.8
Reaction products of methyltestosterone found in HLM or TiO
2
photocatalysis samples classified according to their signal intensity: s = strong, m = medium, w = weak.
Reaction
[M+H]+
t
R
(min)
HLM
TiO
s
2
photocatalysis
M
303.2324
301.2168
17.85
16.60
s
w
Mꢀ2H
dehydrogenation
1
1
1
7.36
7.46
7.66
w
m
w
s
M+O
319.2273
7.71
m
hydroxylation
8
8
9
.18
.39
.46
m
w
m
w
m
m
m
w
w
s
1
1
1
1
1
1
1
1
1
0.53
1.36
2.12
2.49
2.65
3.56
3.67
3.82
4.68
s
m
w
w
m
s
w
M+Oꢀ2H
317.2117
9.32
w
m
m
hydroxylation + dehydrogenation
9.84
s
10.67
10.93
12.44
13.31
14.22
m
w
w
m
m
M+2O
dihydroxylation
335.2222
333.2066
5.44
w
7
8
.71
.36
m
m
M+2Oꢀ2H
5.50
m
dihydroxylation + dehydrogenation
Tables A.8–A.12). The identifications of matching products were
confirmed by MS/MS. Most of the MS/MS spectra correlated well
(
17b-hydroxy group, such as testosterone and nandrolone
(Schänzer, 1996). Obviously the 17 -methyl group of methyltestos-
a
Fig. 5), but five products (denoted with a in Tables A.9–A.12) with
terone, metandienone, and stanozolol inhibits their 17b-oxidation.
An intense Mꢀ2H metabolite was observed in HLM metabolism
reactions of methyltestosterone, metandienone, and stanozolol,
that is probably due to formation of a double bond in the steroid
skeleton instead of 17b-oxidation. The most obvious site of
dehydrogenation is the C6–C7 bond, since in this case the increased
matching mass and retention time were found to be different iso-
mers in HLM metabolism and TiO photocatalytic reactions.
Approximately half of the Mꢀ2H, M+O, and M+Oꢀ2H HLM reac-
tion products of all the steroids studied could be imitated by TiO
2
2
photocatalysis (Eq. (1), Table 2). However, there were also isomeric
products only observed in either of the systems (Tables A.8–A.12),
which can be attributed to the non-selectivity of the hydroxyl rad-
degree of conjugation of double bonds stabilizes the product via p-
electron delocalization. Dehydrogenation of the C6–C7 bond has
also been observed after oral administration of methyltestosterone
(Pozo et al., 2009b) and metandienone (Schänzer, 1996), but it has
been suggested to form via a labile conjugate. A few Mꢀ2H prod-
2
ical attack in TiO photocatalysis. As a whole, the main phase I HLM
metabolism reactions of methyltestosterone, nandrolone, testoster-
one, and stanozolol were imitated considerably better than those of
metandienone.
2
ucts were formed in TiO photocatalysis of methyltestosterone,
number of the same reaction products in TiO2 photocatalysis and HLM reactions
total number of metabolites produced by HLM
Coverage ¼
ꢁ 100%
ð1Þ
Mꢀ2H metabolites of structurally similar methyltestosterone,
nandrolone, and testosterone were imitated well by TiO photoca-
talysis. The main Mꢀ2H products of testosterone and nandrolone
were the same in TiO photocatalysis and HLM metabolism reac-
metandienone, and stanozolol, which were not formed in HLM
reactions.
The M+O metabolites of all the other steroids studied, except
2
metandienone, could be imitated relatively well by TiO photoca-
2
2
tions. The main Mꢀ2H product of testosterone was identified as
androstenedione by using a standard. Correspondingly, the main
Mꢀ2H product of nandrolone is most likely formed by oxidation
of 17b-hydroxyl to ketone. 17-keto metabolites are the main
excreted metabolites of anabolic steroids having a secondary
talysis. Enzymatic hydroxylation of the steroids is possible at
several sites (Bullock et al., 1995; Choi et al., 2005; Masse et al.,
1989; Parr et al., 2012; Pozo et al., 2009a; Schänzer et al., 1991).
Seven hydroxylation products were observed in HLM metabolism
of testosterone (Fig. 3). These include most likely the 1b-, 2b-,

5
2
M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
Table A.9
Reaction products of nandrolone found in HLM or TiO
2
photocatalysis samples classified according to their signal intensity: s = strong, m = medium, w = weak.
Reaction
[M+H]+
t
R
(min)
HLM
2
TiO photocatalysis
M
Mꢀ2H
275.2011
273.1855
15.00
14.00
14.51
s
s
s
s
m
m
w
dehydrogenation
1
1
4.58
4.74
s
M+O
hydroxylation
291.1960
6.78
7.79
w
w
m
w
w
m
s
s
8
9
9
.44
.39
.69
s
10.56
10.99
11.35
11.78
12.07
12.22
12.74
s
w
w
w
m
w
w
s
s
a
w
M+Oꢀ2H
289.1804
7.34
7.92
m
m
w
hydroxylation + dehydrogenation
8
8
9
9
9
9
.44
.92
.08
.20
.44
.77
m
m
s
s
m
s
10.11
10.56
11.64
s
s
m
M+2O
dihydroxylation
307.1909
3.43
3.81
w
w
w
w
w
m
w
3
4
4
6
6
.91
.64
.87
.20
.75
a
The MS/MS spectra do not correlate well enough and the products are most likely different isomers in HLM and TiO
2
photocatalytic reactions.
6b-, 11b-, 15b-, and 16b-hydroxytestosterones previously found in
4. Conclusions
HLM reactions, of which the 6b-hydroxylation product is the dom-
inant metabolite (Choi et al., 2005; Krauser et al., 2004; Yamazaki
TiO photocatalysis is a fast, simple, and inexpensive oxidation
2
and Shimada, 1997). Also minor 1
a-, 2a-, 6a
-, 7a-, 16a
-, and 18/
method. Here, we have shown that TiO photocatalysis is a suitable
2
1
2a-hydroxylation products of testosterone have been found in
method for oxidation of relatively non-polar compounds, and ace-
HLM incubations (Bullock et al., 1995). The metabolism of the
other steroids used in this study has mainly been examined using
human subjects (Masse et al., 1989; Poelmans et al., 2002; Pozo
et al., 2009a), various animal species (Biddle et al., 2009;
Blokland et al., 2005; Roig et al., 2007), or in vitro using liver prep-
arations from different animals (Bullock et al., 1995; Scarth et al.,
tonitrile can be used as solvent for hydrophobic compounds and to
slow down the degradation of formed products. Radical reaction
enables aliphatic hydroxylation in several positions, which is an
essential reaction for imitation of phase I metabolism of anabolic
steroids, and a number of important metabolism reaction products
can be produced by TiO2 photocatalysis. Therefore, TiO2 photoca-
talysis appears to be a possible method for rapid and low-cost pro-
duction of metabolite standards, if the reaction is scaled-up and
the conditions optimized for better yield.
2
010). Because of species differences in metabolism and differ-
ences between excreted metabolites and other metabolites formed
in the body, these former studies cannot be directly compared to
our in-house HLM results.
Even though the main reaction types were the same in TiO2
photocatalysis and in HLM phase I metabolism, isomeric products
can be different and products characteristic to only either one of
Several M+Oꢀ2H metabolites were formed both in HLM and
TiO
2
reactions. The M+Oꢀ2H metabolites of metandienone, meth-
yltestosterone, stanozolol, and testosterone were imitated better
by TiO photocatalysis than those of nandrolone (Table 2). Gener-
the systems are also formed. Thus, TiO photocatalysis cannot fully
2
2
imitate phase I metabolism, which is understandable when com-
ally, the M+Oꢀ2H metabolites were not imitated as well as
Mꢀ2H and M+O metabolites. The combination of two reactions
can produce more possible isomers and the probability that the
same isomers are formed in both systems is lower. In addition to
Mꢀ2H, M+O, and M+Oꢀ2H products, dihydroxylation products
paring a relatively non-selective oxidation method with inherently
highly selective enzymatic reactions. Nonetheless, in earlier stud-
ies, adequate imitation of the whole range of phase I oxidation
reactions has not been achieved with metalloporphyrins, Fenton
reaction or electrochemical methods either, and the performance
(
M+2O) were formed mainly in HLM reactions and products
of TiO photocatalysis for imitation of phase I metabolism is com-
2
formed by dihydroxylation + dehydrogenation (M+2Oꢀ2H) were
favored by TiO photocatalytic reactions. This implies that the oxi-
dation proceeds further in TiO photocatalysis. In addition, a few
parable to these methods. Most importantly, TiO2 photocatalysis
2
provides a straightforward method for imitation of phase I metab-
olism reactions under standard laboratory conditions without need
of special equipment, such as electrochemical cells. This study also
highlights the importance of optimization of chromatographic
2
other minor oxidation products were observed in TiO
ysis (Table A.12).
2
photocatal-

M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
53
Table A.10
Reaction products of testosterone found in HLM or TiO
2
photocatalysis samples classified according to their signal intensity: s = strong, m = medium, w = weak.
Reaction
[M+H]+
t
R
(min)
HLM
2
TiO photocatalysis
M
Mꢀ2H
289.2168
287.2011
16.03
15.05
s
s
s
s
dehydrogenation
1
1
5.59
5.75
w
w
m
m
M+O
305.2117
6.10
hydroxylation
7
8
9
.60
.21
.45
m
m
s
10.49
11.33
11.91
12.20
12.45
12.66
13.27
m
w
m
m
a
w
m
w
w
w
m
M+Oꢀ2H
303.1960
7.47
w
hydroxylation + dehydrogenation
7
8
8
9
9
0.11
0.56
0.72
1.31
2.79
4.29
.57
.27
.98
.16
.62
w
m
w
m
m
m
w
1
1
1
1
1
m
w
m
m
M+2Oꢀ2H
319.1909
285.1855
m
dihydroxylation + dehydrogenation
Mꢀ4H
14.48
m
2
x dehydrogenation
a
The MS/MS spectra do not correlate well enough and the products are most likely different isomers in HLM and TiO
2
photocatalytic reactions.
Table A.11
2
Reaction products of metandienone found in HLM or TiO photocatalysis samples classified according to their signal intensity: s = strong, m = medium, w = weak.
Reaction
[M+H]+
t
R
(min)
HLM
s
2
TiO photocatalysis
M
301.2168
299.2011
17.86
17.16
s
w
Mꢀ2H
dehydrogenation
1
1
7.30
7.87
w
m
s
M+O
317.2117
6.89
hydroxylation
7
7
9
9
.05
.72
.10
.82
w
m
s
m
s
1
1
1
1
1
0.64
1.92
2.15
3.55
6.58
w
m
s
w
w
m
M+Oꢀ2H
315.1960
8.25
m
hydroxylation + dehydrogenation
8
8
9
0.99
1.25
2.45
.77
.94
.89
s
w
w
w
m
w
w
a
1
1
1
m
w
M+2O
dihydroxylation
333.2066
331.1909
3.95
4
6
.64
.60
w
m
M+2Oꢀ2H
5.57
w
dihydroxylation + dehydrogenation
a
The MS/MS spectra do not correlate well enough and the products are most likely different isomers in HLM and TiO
2
photocatalytic reactions.

5
4
M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
Table A.12
Reaction products of stanozolol found in HLM or TiO
2
photocatalysis samples classified according to their signal intensity: s = strong, m = medium, w = weak.
Reaction
[M+H]+
t
R
(min)
HLM
s
2
TiO photocatalysis
M
329.2593
327.2436
13.66
12.72
s
m
Mꢀ2H
dehydrogenation
1
1
3.11
3.45
m
s
m
M+O
345.2542
5.55
hydroxylation
5
5
6
6
6
7
7
7
8
8
9
9
.67
.79
.20
.81
.89
.03
.33
.48
.45
.75
.11
.42
m
m
s
s
s
m
s
w
m
s
w
s
s
s
w
s
m
1
1
0.30
0.50
s
s
w
s
M+Oꢀ2H
343.2386
5.80
hydroxylation + dehydrogenation
6
6
6
7
7
7
7
8
8
.28
.40
.85
.03
.17
.58
.90
.55
.94
m
m
m
s
s
s
s
s
s
s
a
m
m
M+2O
361.2491
3.40
dihydroxylation
3
4
4
4
4
5
5
5
6
6
6
6
7
7
.61
.04
.65
.75
.91
.33
.46
.94
.12
.18
.32
.77
.03
.46
w
m
s
w
w
w
s
w
w
s
m
m
m
w
m
M+2Oꢀ2H
359.2335
4.17
m
dihydroxylation + dehydrogenation
4
4
4
4
5
5
5
5
5
6
7
.40
.70
.86
.90
.10
.15
.41
.59
.79
.30
.73
a
s
w
s
w
m
m
w
m
s
m
w
w
m
M+3Oꢀ2H
375.2284
3.39
trihydroxylation + dehydrogenation
3
.60
m
m
MꢀCH
4
313.2280
329.2229
11.07
demethylation
M+OꢀCH
4
5.77
m
hydroxylation + demethylation
6
6
.14
.85
m
w
a
The MS/MS spectra do not correlate well enough and the products are most likely different isomers in HLM and TiO
2
photocatalytic reactions.

M. Ruokolainen et al. / European Journal of Pharmaceutical Sciences 65 (2014) 45–55
55
conditions for separation of isomeric oxidation products in order to
get more reliable information on the feasibility of oxidative chem-
ical reactions for mimicking metabolic reactions.
Krauser, J.A., Voehler, M., Tseng, L.H., Schefer, A.B., Godejohann, M., Guengerich, F.P.,
004. Testosterone 1 beta-hydroxylation by human cytochrome P450 3A4. Eur.
2
J. Biochem. 271, 3962–3969.
Li, A.P., 2009. Evaluation of luciferin-isopropyl acetal as a CYP3A4 substrate for
human hepatocytes: effects of organic solvents, cytochrome P450 (P450)
inhibitors, and P450 inducers. Drug Metab. Dispos. 37, 1598–1603.
Li, X., 2013. Solvent effects and improvements in the deoxyribose degradation assay
for hydroxyl radical-scavenging. Food Chem. 141, 2083–2088.
Acknowledgement
Liang, S., Shiue, Y., Kuo, C., Guo, S., Liao, W., Tsai, E., 2013. Online monitoring
oxidative products and metabolites of nicotine by free radicals generation with
Fenton reaction in tandem mass spectrometry. Sci. World J. 2013, 189162-
Financial support of the work was provided by the Academy of
Finland (projects 251575 and 251629).
189162.
Masse, R., Ayotte, C., Bi, H.G., Dugal, R., 1989. Studies on anabolic-steroids. 3.
Detection and characterization of stanozolol urinary metabolites in humans by
gas-chromatography mass-spectrometry. J. Chromatogr. B. Biomed. Appl. 497,
Appendix A
1
7–37.
Medana, C., Calza, P., Giancotti, V., Dal Bello, F., Pasello, E., Montana, M., Baiocchi, C.,
011. Horse metabolism and the photocatalytic process as a tool to identify
See Tables A.1–A.12.
2
metabolic products formed from dopant substances: the case of sildenafil. Drug
Test. Anal. 3, 724–734.
Medana, C., Calza, P., Giancotti, V., Dal Bello, F., Aragno, M., Baiocchi, C., 2013. Study
of the photocatalytic transformation of synephrine: a biogenic amine relevant
in anti-doping analysis. Anal. Bioanal. Chem. 405, 1105–1113.
Meunier, B., 1992. Metalloporphyrins as versatile catalysts for oxidation reactions
and oxidative DNA cleavage. Chem. Rev. 92, 1411–1456.
Mile, B., 2000. Free radical participation in organic chemistry: Electron Spin
Resonance (ESR) studies of their structures and reactions. Curr. Org. Chem. 4,
References
Barry, S.M., Mueller-Bunz, H., Rutledge, P.J., 2012. Investigating the oxidation of
alkenes by non-heme iron enzyme mimics. Org. Biomol. Chem. 10, 7372–7381.
Bernadou, J., Meunier, B., 2004. Biomimetic chemical catalysts in the oxidative
activation of drugs. Adv. Synth. Catal. 346, 171–184.
Bhatkhande, D.S., Pangarkar, V.G., Beenackers, A.A., 2002. Photocatalytic
degradation for environmental applications?
Biotechnol. 77, 102–116.
Biddle, S.T.B., O’Donnell, A., Houghton, E., Creaser, C., 2009. Metabolism of
methyltestosterone in the greyhound. Rapid Commun. Mass Spectrom. 23,
a review. J. Chem. Technol.
55–83.
Nissilä, T., Sainiemi, L., Karikko, M., Kemell, M., Ritala, M., Franssila, S., Kostiainen, R.,
Ketola, R.A., 2011. Integrated photocatalytic micropillar nanoreactor
electrospray ionization chip for mimicking phase I metabolic reactions. Lab
Chip 11, 1470–1476.
Nouri-Nigjeh, E., Bischoff, R., Bruins, A.P., Permentier, H.P., 2011. Electrochemistry
in the mimicry of oxidative drug metabolism by cytochrome P450s. Curr. Drug
Metab. 12, 359–371.
Parr, M.K., Zöllner, A., Fußhöller, G., Opfermann, G., Schlörer, N., Zorio, M., Bureik,
M., Schänzer, W., 2012. Unexpected contribution of cytochrome P450 enzymes
CYP11B2 and CYP21, as well as CYP3A4 in xenobiotic androgen elimination –
insights from metandienone metabolism. Toxicol. Lett. 213, 381–391.
Poelmans, S., De Wasch, K., De Brabander, H.F., Van de Wiele, M., Courtheyn, D., van
Ginkel, L.A., Sterk, S.S., Delahaut, P., Dubois, M., Schilt, R., Nielen, M.,
Vercammen, J., Impens, S., Stephany, R., Hamoir, T., Pottie, G., Van Poucke, C.,
Van Peteghem, C., 2002. Analytical possibilities for the detection of stanozolol
and its metabolites. Anal. Chim. Acta 473, 39–47.
Pozo, O.J., Van Eenoo, P., Deventer, K., Lootens, L., Grimalt, S., Sancho, J.V.,
Hernandez, F., Meuleman, P., Leroux-Roels, G., Delbeke, F.T., 2009a. Detection
and structural investigation of metabolites of stanozolol in human urine by
liquid chromatography tandem mass spectrometry. Steroids 74, 837–852.
Pozo, O.J., Van Eenoo, P., Deventer, K., Lootens, L., Van Thuyne, W., Parr, M.K.,
Schänzer, W., Sancho, J.V., Hernandez, F., Meuleman, P., Leroux-Roels, G.,
Delbeke, F.T., 2009b. Detection and characterization of a new metabolite of 17
alpha-methyltestosterone. Drug Metab. Dispos. 37, 2153–2162.
7
13–721.
Blokland, M.H., van Rossum, H.J., Herbold, H.A., Sterk, S.S., Stephany, R.W., van
Ginkel, L.A., 2005. Metabolism of methyltestosterone, norethandrolone and
methylboldenone in a heifer. Anal. Chim. Acta 529, 317–323.
Breslow, R., Zhang, X., Huang, Y., 1997. Selective catalytic hydroxylation of a steroid
by an artificial cytochrome P-450 enzyme. J. Am. Chem. Soc. 119, 4535–4536.
Bullock, P., Pearce, R., Draper, A., Podval, J., Bracken, W., Veltman, J., Thomas, P.,
Parkinson, A., 1995. Induction of liver microsomal cytochrome-P450 in
cynomolgus monkeys. Drug Metab. Dispos. 23, 736–748.
Busby, W.F., Ackermann, J.M., Crespi, C.L., 1999. Effect of methanol, ethanol,
dimethyl sulfoxide, and acetonitrile on in vitro activities of cDNA-expressed
human cytochromes P-450. Drug Metab. Dispos. 27, 246–249.
Calza, P., Pelizzetti, E., Brussino, M., Baiocchi, C., 2001. Ion trap tandem mass
spectrometry study of dexamethasone transformation products on light
activated TiO
Calza, P., Pazzi, M., Medana, C., Baiocchi, C., Pelizzetti, E., 2004. The photocatalytic
process as tool to identify metabolitic products formed from dopant
substances: the case of buspirone. J. Pharm. Biomed. Anal. 35, 9–19.
Chauret, N., Gauthier, A., Nicoll-Griffith, D.A., 1998. Effect of common organic
solvents on in vitro cytochrome P450-mediated metabolic activities in human
liver microsomes. Drug Metab. Dispos. 26, 1–4.
Choi, M.H., Skipper, P.L., Wishnok, J.S., Tannenbaum, S.R., 2005. Characterization of
testosterone 11 beta-hydroxylation catalyzed by human liver microsomal
cytochromes P450. Drug Metab. Dispos. 33, 714–718.
Ekins, S., Ring, B.J., Grace, J., McRobie-Belle, D.J., Wrighton, S.A., 2000. Present and
future in vitro approaches for drug metabolism. J. Pharmacol. Toxicol. Methods
2
surface. J. Am. Soc. Mass Spectrom. 12, 1286–1295.
a
Raoof, H., Mielczarek, P., Michalow, K.A., Rekas, M., Silberring, J., 2013. Synthesis of
metabolites of paracetamol and cocaine via photooxidation on TiO
by UV light. J. Photochem. Photobiol. B 118, 49–57.
2
catalyzed
Roig, M., Segura, J., Ventura, R., 2007. Quantitation of 17 beta-nandrolone
metabolites in boar and horse urine by gas chromatography–mass
spectrometry. Anal. Chim. Acta 586, 184–195.
Scarth, J.P., Spencer, H.A., Hudson, S.C., Teale, P., Gray, B.P., Hillyer, L.L., 2010. The
application of in vitro technologies to study the metabolism of the androgenic/
anabolic steroid stanozolol in the equine. Steroids 75, 57–69.
4
4, 313–324.
Fang, Z., Breslow, R., 2006. Metal coordination-directed hydroxylation of steroids
with a novel artificial P-450 catalyst. Org. Lett. 8, 251–254.
Fenton, H.J.H., 1894. LXXIII.-Oxidation of tartaric acid in presence of iron. J. Chem.
Soc., Trans. 65, 899–910.
Fujishima, A., Rao, T.N., Tryk, D.A., 2000. Titanium dioxide photocatalysis. J.
Photochem. Photobiol. C:Photochem. Rev. 1, 1–21.
Fujishima, A., Zhang, X., Tryk, D.A., 2008. TiO
phenomena. Surf. Sci. Rep. 63, 515–582.
Gonzalez-Perez, V., Connolly, E.A., Bridges, A.S., Wienkers, L.C., Paine, M.F., 2012.
Impact of organic solvents on cytochrome P450 probe reactions: filling the gap
with (S)-Warfarin and Midazolam hydroxylation. Drug Metab. Dispos. 40,
Schänzer, W., 1996. Metabolism of anabolic androgenic steroids. Clin. Chem. 42,
1001–1020.
Schänzer, W., Geyer, H., Donike, M., 1991. Metabolism of metandienone in man –
identification and synthesis of conjugated excreted urinary metabolites,
determination of excretion rates and gas-chromatographic mass-
spectrometric identification of bis-hydroxylated metabolites. J. Steroid
Biochem. Mol. Biol. 38, 441–464.
2
photocatalysis and related surface
Stuk, T.L., Grieco, P.A., Marsh, M.M., 1991. Site-selective hydroxylation of steroids
2
136–2142.
via oxometalloporphinates covalently linked to ring-D
– introduction of
Haber, V.F., Weiss, J., 1932. Über die Katalyse des Hydroxyperoxydes.
Naturwissenschaften 20, 948–950.
Hickman, D., Wang, J.P., Wang, Y., Unadkat, J.D., 1998. Evaluation of the selectivity
of in vitro probes and suitability of organic solvents for the measurement of
human cytochrome P450 monooxygenase activities. Drug Metab. Dispos. 26,
hydroxyl-groups into the C(9) and C(12) position of 5-alpha-androstanes. J.
Org. Chem. 56, 2957–2959.
Van der Steen, J., Timmer, E.C., Westra, J.G., Benckhuysen, C., 1973. 4-
Hydroperoxidation in fenton oxidation of antitumor agent cyclophosphamide.
J. Am. Chem. Soc. 95, 7535–7536.
Yamazaki, H., Shimada, T., 1997. Progesterone and testosterone hydroxylation by
cytochromes P450 2C19, 2C9, and 3A4 in human liver microsomes. Arch.
Biochem. Biophys. 346, 161–169.
Yang, J., Breslow, R., 2000. Selective hydroxylation of a steroid at C-9 by an artificial
cytochrome P-450. Angew. Chem. Int. Ed. Engl. 39, 2692–2694.
Zbaida, S., Kariv, R., Fischer, P., Silman-Greenspan, J., Tashma, Z., 1986. Reaction of
cimetidine with Fenton reagent. Eur. J. Biochem. 154, 603–605.
Zhang, L., Mohamed, H.H., Dillert, R., Bahnemann, D., 2012. Kinetics and
2
07–215.
Johansson, T., Weidolf, L., Jurva, U., 2007. Mimicry of phase I drug metabolism –
novel methods for metabolite characterization and synthesis. Rapid Commun.
Mass Spectrom. 21, 2323–2331.
Jurva, U., Wikstrom, H.V., Bruins, A.P., 2002. Electrochemically assisted Fenton
reaction: reaction of hydroxyl radicals with xenobiotics followed by on-line
analysis with high-performance liquid chromatography/tandem mass
spectrometry. Rapid Commun. Mass Spectrom. 16, 1934–1940.
Jurva, U., Wikstrom, H.V., Weidolf, L., Bruins, A.P., 2003. Comparison between
electrochemistry/mass spectrometry and cytochrome P450 catalyzed oxidation
reactions. Rapid Commun. Mass Spectrom. 17, 800–810.
mechanisms of charge transfer processes: in photocatalytic systems:
review. J. Photochem. Photobiol. C: Photochem. Rev. 13, 263–276.
a