Transformations of 4- and 17α-substituted testosterone analogues by Fusarium culmorum

Article abstract of DOI:10.1016/j.steroids.2005.05.005

A series of 4- and/or 17α-substituted testosterone analogues has been incubated with the hydroxylating fungus Fusarium culmorum AM282. It was found that 19-norandrostenedione, 19-nortestosterone, 4-methoxytestosterone, 4-methyltestosterone, and 4-chloro-17α-methyltestosterone were hydroxylated exclusively or mainly at the 6β-position. The mixtures of 6β-, 15α-, and 12β- or 11α-monohydroxy derivatives were obtained from 17α-methyltestosterone and 17α-ethyl-19- nortestosterone - the substrates with alkyl group at C-17α. 4-Chlorotestosterone was predominantly hydroxylated at 15α-position, but the reaction was accompanied by the reduction of 4-en-3-one system, which proceeded in the sequence: reduction of ketone to 3β-alcohol and then reduction of the double 4,5 bond. The results obtained indicate an influence of stereoelectronic and steric effects of substitutes on regioselectivity of the hydroxylation of 4-en-3-one steroids by F. culmorum.

Full text of DOI:10.1016/j.steroids.2005.05.005

Steroids 70 (2005) 817–824
Transformations of 4- and 17␣-substituted testosterone
analogues by Fusarium culmorum
∗
´
Alina Swizdor , Teresa Kołek
Department of Chemistry, Agricultural University, Norwida 25, 50-375 Wrocław, Poland
Received 18 February 2005; received in revised form 26 April 2005; accepted 9 May 2005
Available online 14 July 2005
Abstract
A series of 4- and/or 17␣-substituted testosterone analogues has been incubated with the hydroxylating fungus Fusarium culmorum
AM282. It was found that 19-norandrostenedione, 19-nortestosterone, 4-methoxytestosterone, 4-methyltestosterone, and 4-chloro-17␣-
methyltestosterone were hydroxylated exclusively or mainly at the 6␤-position. The mixtures of 6␤-, 15␣-, and 12␤- or 11␣-monohydroxy
derivatives were obtained from 17␣-methyltestosterone and 17␣-ethyl-19-nortestosterone—the substrates with alkyl group at C-17␣. 4-
Chlorotestosterone was predominantly hydroxylated at 15␣-position, but the reaction was accompanied by the reduction of 4-en-3-one
system, which proceeded in the sequence: reduction of ketone to 3␤-alcohol and then reduction of the double 4,5 bond.
The results obtained indicate an influence of stereoelectronic and steric effects of substitutes on regioselectivity of the hydroxylation of
4-en-3-one steroids by F. culmorum.
© 2005 Elsevier Inc. All rights reserved.
Keywords: Microbial hydroxylation; Anabolic; Fusarium culmorum
1. Introduction
[10,11]. Although the hydroxylations proceeded with signif-
icant regio- and stereospecifities, we observed formation of
both mono- and dihydroxy products, with OH groups in dif-
ferent positions of the steroid skeleton. We have noticed that
the substrates with 4-ene-3-one group were hydroxylated in
one of the following positions: 6␤ or 15␣ or 12␤, or both
12␤ and 15␣ simultaneously. What is more, hydroxylation
at 6␤ took place only in the cause of substrates with an oxy-
gen function at C-17. These results “fit” well to the Brannon
and Jones models [12]; therefore, in our opinion F. culmo-
rum seems to be a suitable model for studies on correlation
between the course of the enzymatic hydroxylation and a
substrate structure.
A common feature of filamentous fungi is their ability
to transform steroids, often by regio- and stereoselective
hydroxylation. Although the site of hydroxylation is often
a characteristic feature of a fungal genus or species [1,2],
some of the collected data indicate that the structure of a sub-
strate can also influence the position of the newly introduced
hydroxyl group.
There are reports that fungi belonging to the genus Fusar-
ium have been used in steroid transformations [1–9]; how-
ever, very little systematic studies on the correlation between
the structure of a substrate and the course of hydroxylation
have been carried out.
In this paper, we have focused on biotransformations
by means of F. culmorum of the following substrates: 19-
norandrostenedione, 19-nortestosterone, and its 17␣-ethyl
analogue, and the testosterone derivatives with additional
substituents at C-4 or/and C-17.
In our previous work, we investigated transformations
of the 4-en-3-one and 3␤-hydroxy-5-ene steroid hormones
and their derivatives by means of Fusarium culmorum
∗
It is commonly known that structural modifications
of steroid hormones influence their biological activity.
Corresponding author. Fax: +48 71 3284124.
E-mail address: alina@ozi.ar.wroc.pl (A. Swizdor).
´
0039-128X/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2005.05.005

´
A. Swizdor, T. Kołek / Steroids 70 (2005) 817–824
818
For example, removal of 19-methyl group in testosterone
enhances the anabolic to androgenic ratio. Also, high
anabolic activity is attributed to the 4-substituted testos-
terone analogues (especially 4-halogenated compounds).
Addition of 17␣-methyl group to testosterone halves both the
anabolic and androgenic activity but leaving the activity ratio
unchanged. 17␣-AlkylsubstituentprotectstheC-17hydroxyl
function from oral inactivation. Studies on metabolism of
these compounds in human and in other mammals are well
documented [13–28], but the structures of the hydroxylated
metabolites have not been fully elucidated [22–28], due to
the lack of the specific reference standards.
250 ◦C, 2 ◦C/min–300 ◦C/5 min), and for 8 (220 ◦C/min,
10 ◦C/min–270 ◦C, 8 ◦C/min–300 ◦C/9 min) or equipped
with a HP-1 column (cross-linked methyl siloxane,
25 m × 0.32 mm × 0.52 ␮m film thickness) for 3–5
(200 ◦C/min, 6 ◦C/min–270 ◦C, 10 ◦C/min–300 ◦C/2 min).
Column chromatography was performed using silica gel and
hexane/acetone mixtures with gradually increasing acetone
content as eluent. Structures of the biotransformation
products were determined by means of 1H NMR (the spectra
were recorded on DRX 300 MHz Bruker spectrometer and
measured in CDCl3 with TMS as an internal standard).
3. Results
2. Experimental
3.1. Biotransformations of steroids with F. culmorum
2.1. Microorganism
The following substrates were examined: 19-nortesto-
sterone (1), 19-norandrostenedione (2), 4-methoxytesto-
sterone (3), 4-methyltestosterone (4), 4-chlorotestosterone
(5), 17␣-methyltestosterone (6), 17␣-ethyl-19-nortesto-
sterone (7), and 4-chloro-17␣-methyltestosterone (8). The
results are given in Table 1. Only the metabolites the yield of
which exceeded 10% of all the products (determined by GC)
are listed. The structures of substrates and their metabolites
are showed in Fig. 1.
The microorganism F. culmorum AM282 used in the
present study was obtained from the collection of the Institute
of Biology and Botany of the Medical University of Wrocław.
2.2. Conditions of cultivation and transformation
The strain of F. culmorum was maintained on a Sabouraud
4% dextrose agar slope and was freshly subcultured before
using in the biotransformation experiment.
Three hundred milliliters Erlenmeyer flasks, each contain-
ing 100 ml of the medium consisting of 3% glucose and 1%
peptone, were inoculated with a suspension of F. culmorum
and then incubated for 3 days at 20 ◦C in a rotary shaker. After
full growth of the microorganisms, 20–25 mg of a substrate
in 1 ml of acetone was added to it, and the transformation
was continued for a few days under the same conditions.
Microbial conversions of substrates were monitored by TLC
and GC.
3.2. Identification of metabolites
All the products obtained proved to be secondary alco-
hols. Position and configuration of the introduced hydroxyl
groups were determined by the chemical shifts and the shape
of CHOH signals in the 1H NMR spectra. Assumed structures
were confirmed by comparison of chemical shifts of selected,
diagnostically significant signals in the NMR spectra of the
substrates and the products and the literature data [25,29–31].
1H NMR data is shown in Table 2.
2.3. Isolation and identification of the biotransformation
products
3.2.1. 6β-Hydroxy derivatives
1
In H NMR spectra of the metabolites with a hydroxyl
All the fermentation media were extracted three times
with 15–20 ml of chloroform. Then the solvent was evapo-
rated off and the transformation products in the residues were
analyzed by TLC, GC and separated by silica gel column
chromatography. TLC analysis was carried out using Merck
Kieselgel 60 F254 plates with hexane/acetone (2:1 or 3:2,
v/v) as eluent. Visualization of the steroids was performed by
spraying the plates with a mixture of methanol/concentrated
sulphuric acid (1:1, v/v) and heating at 100 ◦C until the
colors developed. GC analysis was performed using a
Hewlett-Packard 5890A Series II GC instrument (FID,
carrier gas H2 at flow rate of 2 ml/min), equipped with
group at 6␤-position a narrow triplet of 6␣-proton, and the
downfield shift of 19-H3 signal (about 0.2 ppm with respect to
the substrate) confirmed that a 6␤-hydroxyl group had been
introduced. Chemical shifts of the 6␣-proton in the spec-
tra of metabolites without any additional substituent at C-4:
δ = 4.35–4.38 ppm for 6␤, 17␤-dihydroxy compound (1a, 6a,
and 7a), and δ = 4.42 ppm for 6␤-hydroxy-17-oxo compound
2a were in accordance with those reported by Kirk et al.
for 6␤-hydroxytestosterone and 6␤-hydroxyandrostenedione
[31].
6␣-H signal of the 4-substituted 6␤-hydroxy products
(3a, 4a–4d, 5a, and 8a) was significantly shifted downfield
due to the influence of a substituent at C-4. In the case
of chlorine as a 4-substituent, the difference of the shifts
reached as much as 0.96–0.98 ppm. The analogous change
of the chemical shifts was observed for 6␤-hydroxy-4-
a
HP-5 column (cross-linked 5% Ph–Me–siloxane,
25 m × 0.32 mm × 0.52 ␮m film thickness) for 1,2
(temperature conditions: 220 ◦C/min, 8 ◦C/min–270 ◦C,
2 ◦C/min–300 ◦C/2 min), for 6,7 (220 ◦C/min, 6 ◦C/min–

´
A. Swizdor, T. Kołek / Steroids 70 (2005) 817–824
819
Table 1
Biotransformation of steroids by Fusarium culmorum
Substrate
Incubation time (days)
Metabolite
% (by GC)
Rt (min)
19-Nortestosterone (1)
3
6␤-Hydroxy-19-nortestosterone (1a)
6␤-Hydroxy-19-norandrostenedione (2a)
32
48
13.92
13.66
19-Norandrostenedione (2)
3
6␤-Hydroxy-19-nortestosterone (1a)
6␤-Hydroxy-19-norandrostenedione (2a)
15
66
13.92
13.66
4-Methoxytestosterone (3)
4-Methyltestosterone (4)
4
2
6␤-Hydroxy-4-methoxyandrostenedione (3a)
80
11.98
6␤-Hydroxy-4-methylandrostenedione (4b)
6␤-Hydroxy-4-methyltestosterone (4a)
76
10
9.80
10.09
10
6␤,15␣-Dihydroxy-4-methylandrostenedione (4c)
6␤,11␣-Dihydroxy-4-methylandrostenedione (4d)
32
17
12.11
11.74
4-Chlorotestosterone (5)
1.5
6␤-Hydroxy-4-chloroandrostenedione (5a)
15␣-Hydroxy-4-chloroandrostenedione (5b)
3␤,15␣-Dihydroxy-4-chloro-4-androstene-17-one (5c)
3␤,15␣-Dihydroxy-4␣-chloro-5␣-androstan-17-one (5d)
10
11
22
39
12.53
13.28
12.45
10.70
17␣-Methyltestosterone (6)
2
1
1
6␤-Hydroxy-17␣-methyltestosterone (6a)
15␣-Hydroxy-17␣-methyltestosterone (6b)
12␤-Hydroxy-17␣-methyltestosterone (6c)
50
22
22
22.22
24.13
23.63
17␣-Ethyl-19-nortestosterone (7)
4-Chloro-17␣-methyltestosterone (8)
6␤-Hydroxy-17␣-ethyl-19-nortestosterone (7a)
15␣-Hydroxy-17␣-ethyl-19-nortestosterone (7b)
11␣-Hydroxy-17␣-ethyl-19-nortestosterone (7c)
43
22
12
20.55
22.65
22.16
6␤-Hydroxy-4-chloro-17␣-methyltestosterone (8a)
15␣-Hydroxy-4-chloro-17␣-methyltestosterone (8b)
73
27
15.71
18.15
chloro-1,2-dehydro-17␣-methyltestosterone in comparison
with 6␤-hydroxy-1,2-dehydro-17␣-methyltestosterone [25].
The presence of 4-methyl group resulted in the change of
6␣-H signal position by 0.64 ppm. The same influence of
a methyl group was observed by Azerad and co-workers
in the case of hydroxylation of 1,4a-dimethyl-4,4a,5,6,7,8-
hexahydro-2(3H)-naphthalenone [32]. In the spectrum of 3a
the chemical shift of 6␣-H signal was shifted downfield by
0.78 ppm, as a result of deshielding effect of C-4-metoxyl
group. The observed effect of the substituents at C-4 on
chemical shifts of 6␣-H decreased in the following order:
Cl > OCH3 > CH3, and was in accordance with that reported
in [33].
tional hydroxyl group at C-6, the signal of 15␤-proton was
moved by 0.09 ppm.
3.2.3. 6β,11α-Dihydroxy-4-methylandrostenedione (4d)
11␣-Hydroxylation was characterized by the broad 11␤-
H signal at δ = 4.1 ppm, and the exceptionally low-field 1␤-H
signal at δ = 2.7 ppm, which was shifted due to the spa-
tial proximity of 11␣-OH. Also, the above-mentioned fea-
tures indicating 6␤-hydroxylation were present. Changes in
the chemical shift of 19-H3 with respect to 6␤-hydroxy-4-
methylandrostenedione (4a) (+0.14 ppm) confirmed the pres-
ence of 11␣-hydroxyl group.
Additionally, 6␤-hydroxyl group effects the chemical
shift of 4-CH3 or 4-OCH3 protons, moving it downfield by
0.1 ppm.
3.2.4. 3β,15α-Dihydroxy-4-chloro-4-androsten-17-one
(5c)
The 1H NMR spectrum of 5c showed two broad
CHOH signals, which are typical for axial protons. 15␣-
Hydroxylation was indicated by spectral similarities to other
15␣-alcohols. Reduction of the 3-keto group was recognized
by up-field shift of 19-H3 signal with respect to the 15␣-
hydroxy-4-chloroandrostenedione (5b) (−0.2 ppm, which is
in accordance with the literature [34]).
3.2.2. 15α-Hydroxy derivatives
The presence of 15␣-hydroxyl group was indicated by the
signal of 15␤-H of the shape of a triplet split into doublet
(Wh > 20 Hz). Its chemical shift in the spectra of 15␣,17␤-
dihydroxy metabolites (6b, 7b, and 8b) and the 15␣-
hydroxy,17-oxo metabolites (5b–5c) (δ = 4.10–4.13 ppm and
4.41–4.45 ppm, respectively) was in accordance with those
reported in literature for 15␣-hydroxytestosterone and 15␣-
hydroxyandrostenedione [31]. In the all of 15␣-hydroxy-17-
oxo products, the signal of 16␤-H was observed at δ ≈ 3 ppm
(dd, J16␣,16␤ = 19.3 Hz, J15␤,16␤ = 8 Hz). In the spectrum of
6␤,15␣-dihydroxy product 4c, in the presence of the addi-
3.2.5. 3β,15α-Dihydroxy-4α-chloro-5α-androstan-
17-one (5d)
In the 1H NMR spectrum of 5d all the characteristic
features associated with 15␣-hydroxylation of 17-keto com-
pounds (see above) were present. Significant up-field shift of
19-H3 singlet indicated that the 4,5 double bond was reduced.

´
820
A. Swizdor, T. Kołek / Steroids 70 (2005) 817–824
Fig. 1. Structures of substrates and products of their transformations.
A special feature of the spectrum were the two broad signals
of axial protons: a triplet at δ = 3.76 ppm (4␤-H) and a charac-
teristic multiplet of 3␣-H at δ = 3.55 ppm. These signals could
be identified as 4␣-H and 3␤-H in 5␤-series, or 4␤-H and
3␣-H in 5␣-series, respectively. The assignment to the 5␣-H
configurationwassupportedbythelowerchemicalshiftofthe
H-4 triplet compared to the one observed for 5␤-series [35].
3.2.7. 11α-Hydroxy-17α-ethyl-19-nortestosterone (7c)
The shape of the multiplet at δ = 3.91 ppm (triplet split into
doublet) is similar to that of 11␤-H in 1H NMR spectrum of
6␤,11␣-dihydroxy-4-methylandrostenedione(4d). Chemical
shift of 18-H3 group is in accordance with the one calculated
from increments given by Kirk [31].
3.2.6. 12β-Hydroxy-17α-methyltestosterone (6c)
The location of a hydroxyl group at the 12␤-position was
proved by the shape of the CHOH signal (the axial proton is
coupled with two other protons only), and a downfield shift
of 17␣-methyl protons. The chemical shifts of the significant
signals in 1H NMR of this metabolite are consistent with the
literature data [29].
4. Discussion
Our previous research showed that the strain F. culmorum
AM282 is capable of effective regioselective hydroxylation
of 4-en-3-one steroid hormones (testosterone, androstene-
dione, progesterone) [10]. We also showed that the hydrox-
ylations of these hormones at 6␤-, 12␤-, and 15␣-positions

Table 2
Chemical shifts of the substrates and their metabolites in 1H NMR spectra (δ ppm, J Hz; in CDCl3)
Compound
4-H
17␣-H
18-H3
19-H3
CHOH
Other significant signals
Remarks
[29]
19-Nortestosterone (1)
19-Norandrostenedione (2)
6␤-Hydroxy-19-nortestosterone (1a)
5.82s
5.82s
5.88s [5.88]
3.66t (J = 8.4)
–
3.66t
0.81s
0.95s
0.82 [0.81]
–
–
–
–
–
–
–
–
(J = 8.5)
4.38t (J = 2.5) (6␣-H) [4.37t, J = 2.5]
[3.65t, J = 8.1]
6␤-Hydroxy-19-norandrostenedione (2a)
4-Methoxytestosterone (3)
6␤-Hydroxy-4-methoxyandrostenedione (3a)
4-Methyltestosterone (4)
6␤-Hydroxy-4-methylandrostenedione (4b)
6␤-Hydroxy-4-methyltestosterone (4a)
6␤,15␣-Dihydroxy-4-methyl-androstenedione (4c)
5.89s
–
0.94s
0.77s
0.94s
0.79s
0.95s
0.83s
0.98s
–
4.42t (J = 2.5) (6␣-H)
–
5.19t (J = 2.7) (6␣-H)
–
–
–
–
–
–
–
–
3.57t (J = 9.0)
–
3.65t (J = 9.0)
–
3.65t (J = 9.0)
–
1.18s
1.40s
1.18s
1.40s
1.38s
1.45s
3.57brs (4-OCH3)
3.66brs (4-OCH3)
1.78s (4-CH3)
1.89s (4-CH3)
1.87s (4-CH3)
1.85s (4-CH3); 3.00dd (J = 19.3; J = 8.0)
(16␤-H)
1.90s (4-CH3); 2.70dt (J = 14.5; J = 4.5)
(1␤-H)
–
–
2.99dd (J = 19.3; J = 8.0) (16␤-H)
2.97dd (J = 19.3; J = 8.0) (16␤-H)
5.05t (J = 2.9) (6␣-H)
5.00t (J = 2.9) (6␣-H)
5.05t (J = 3.1) (6␣-H); 4.50m
(Wh = 23.0 Hz) (15␤-H)
5.10t (J = 3.0) (6␣-H); 4.10dt
(J = 11.0; J = 4.5) (11␤-H)
–
5.36t (J = 3.0) (6␣-H)
4.41m (Wh = 23.5 Hz) (15␤-H)
4.40m (Wh = 16.0 Hz) (3␣-H);
4.45m (Wh = 24 Hz) (15␤-H)
3.55m (Wh = 23.0 Hz) (3␣-H);
4.41m (Wh = 23 Hz) (15␤-H)
–
4.35t (J = 2.9) (6␣-H) [4.36t, J = 2.8]
4.13m (Wh = 20 Hz) (15␤-H)
3.76dd (J = 10.8; J = 4.6) (12␣-H)
[3.75dd, J = 11; J = 4.8]
–
4.37t (J = 2.8) (6␣-H)
4.12m (Wh = 22.2 Hz) (15␤-H)
3.91dt (J = 10.0; J = 4.5) (11␤-H)
–
6␤,11␣-Dihydroxy-4-methyl-androstenedione (4d)
–
–
0.98s
1.55s
4-Chlorotestosterone (5)
–
–
–
–
3.66t (J = 8.0)
0.80s
0.98s
0.96s
0.93s
1.24s
1.45s
1.27
6␤-Hydroxy-4-chloro-androstenedione (5a)
15␣-Hydroxy-4-chloro-androstenedione (5b)
3␤,15␣-Dihydroxy-4-chloro-4-androstene-17-one (5c)
–
–
–
1.07s
3␤,15␣-Dihydroxy-4␣-chloro-5␣-androstan-17-one (5d)
3.76t (J = 10.0)
–
0.90s
0.89s
2.99dd (J = 19.3; J = 8.0) (16␤-H)
17␣-Methyltestosterone (6)
5.72s
5.82s [5.82]
5.73s
–
–
–
–
0.91s
1.21s
1.21s (17␣-CH3)
6␤-Hydroxy–17␣-methyltestosterone (6a)
15␣-Hydroxy–17␣-methyltestosterone (6b)
12␤-Hydroxy-17␣-methyltestosterone (6c)
0.93s [0.94]
0.92s [0.91]
0.97s [0.96]
1.40s [1.40]
1.21s [1.21]
1.21s [1.20]
1.22s (17␣-CH3) [1.22]
1.35s (17␣-CH3) [1.34]
1.35s (17␣-CH3) [1.34]
[25]
[30]
[29]
5.73s [5.72]
17␣-Ethyl-19-nortestosterone (7)
5.81s
5.88s
5.82
5.81s
–
–
–
–
–
–
–
–
0.93s
0.96s
0.95s
0.95s
0.91s
0.93s
0.92s [0.94]
-
-
-
-
0.97t (J = 7.2) (17␣-CH2CH3)
1.07t (J = 7.2) (17␣-CH2CH3)
1.01t (J = 7.2) (17␣-CH2CH3)
0.98t (J = 7.2) (17␣-CH2CH3)
1.21s (17␣-CH3)
6␤-Hydroxy-17␣-ethyl-19-nortestosterone (7a)
15␣-Hydroxy-17␣-ethyl-19-nortestosterone (7b)
11␣-Hydroxy-17␣-ethyl-19-nortestosterone (7c)
4-Chloro-17␣-methyltestosterone (8)
6␤-Hydroxy-4-chloro-17␣-methyltestosterone (8a)
15␣-Hydroxy-4-chloro-17␣-methyltestosterone (8b)
1.24s
1.42s
1.26s [1.27]
–
–
5.33t (J = 3.0) (6␣-H)
4.10m (Wh = 21.5 Hz) (15␤-H)
1.21s (17␣-CH3)
1.35s (17␣-CH3) [1.37]
[30]

´
A. Swizdor, T. Kołek / Steroids 70 (2005) 817–824
822
Fig. 2. Mechanism of hydroxylation by means of P450 hydroperoxo-iron
species involving insertion of OH+.
were carried out by the same enzyme complex, and that a
nature of a substituent at C-17 had a significant effect on the
regioselectivity of the reaction.
Fig. 3. Enzyme–substrate complexes with different orientation of steroid
molecules.
Continuing our research on relationship between a
substrate structure and a hydroxylation profile we car-
ried out the transformation of the following substrates:
19-nortestosterone, 19-norandrostenedione and C-4 and/or
C-17 substituted testosterone analogues. Some of the
substrates (19-nortestosterone (1), 19-norandrostendione
(2), 4-metoxytestosterone (3), and 4-methyltestosterone
(4)) underwent selective 6␤-hydroxylation, whereas 4-
chlorotestosterone (5) gave mainly 15␣-hydroxy derivatives
(in above 70% yield) (Table 1). Time course experiments
evidently indicated (data not presented) that the first stage of
the transformation was an oxidation of the hydroxyl group at
C-17, which means that the isolated products were formed
from 17-oxo substrates. The results obtained showed that
both steric and stereoelectronic factors have an impact on
the regioselectivity of hydroxylations by means of F. cul-
morum. They are consistent with results obtained for other
microorganisms [5,36].
The 19-nor-substrates (1,2) underwent regioselective 6␤-
hydroxylation, whereas the transformation of androstene-
dione, carried out under the same conditions, gave 6␤- and
15␣-hydroxy derivatives (ratio 1:2, respectively) [10]. Lack
of a methyl group in the 1-3 cis position promoted stereo-
electronically favoured 6␤-hydroxylation.
It might be expected that the presence of a substituent at
C-4 would prevent the hydroxylation at the relatively close
6␤-position, due to the steric hindrance. It was then highly
surprising to find that the 6␤-hydroxy derivatives were the
only products of the transformations of 4-methyltestosterone
(4) and 4-metoxytestosterone (3). Also, a 6␤-hydroxy com-
pound was the major product of the biotransformation of one
ofthe17␣-alkylsubstrates:4-chloro-17␣-methyltestosterone
(8).
a result, the hydroxylation occurs in the allylic, axial 6␤-
position.
The biotransformations of the 17␣-alkyl derivatives: 17␣-
methyltestosterone (6) and 17␣-ethyl-19-nortestosterone (7)
gave mixtures of 6␤-, 15␣- and, respectively, 12␤- or
11␣-alcohols. The mixtures contained less 6␤-allyl alcohol
than in the case of the transformation of 17␣-unsubstituted
testosterone. Monitoring of the progress of the testosterone
biotransformation (by means of GC) showed that testos-
terone initially undergoes 6␤-hydroxylation [10]—no other
hydroxy derivatives were detected in the reaction after 12 h of
incubation. The lower contents of 6␤-hydroxy products in the
transformations of the 17␣-alkyl compounds may arise from
steric factors associated with the 17␣-substituent. These may
effect formation of the enzyme–substrate complex with par-
ticipation of the 17␤-hydroxyl group (III, Fig. 3). At the same
time, the 17␣-substituent is a drawback for 1-3 cis hydrox-
ylation at 15␣-position, therefore the hydroxylation at the
equivalent 12␤-position occurs (II, Fig. 3).
Similar effect of substituents on the regioselectivity of the
hydroxylation was observed in transformations of proges-
terone and 17␣-acetoxy-progesterone by the strain Calonec-
tria decora [38]. Progesteron was hydroxylated at 12␤,15␣-
positions, whereas 17␣-acetoxy-progesterone only at 12␤-
position.
The 11␣-hydroxy derivative of 17␣-ethyl-19-
nortestosterone (7) is probably formed via the complex
which usually leads to 12␤-alcohol (II, Fig. 3), but in the
presence of the bulkier ethyl group (compared to the methyl
one) the more distant 11␣-position is hydroxylated. The
different stereochemistry of the newly introduced hydroxyl
group (11␣- instead of 12␤-) may arise from the general
tendency of the strain to hydroxylation of equatorial C H
bonds.
The results presented above indicate that the influence of
a substituent at C-4 on the reaction of enzymatic hydrox-
ylation is similar to the one observed in electrophilic sub-
stitution in aromatic systems. This support the mechanism
proposed by Newcomb et al. [37], which assumes a transfer
of OH+ cation from hydroperoxo-iron species P450 to the
substrate molecule (Fig. 2). An electrodonating or/and pos-
sessing free valence electrons substituent at C-4 promotes
(by means of ␲-electrons) attack of OH+ and decreases elec-
trodeficient properties of the intermediate oxo-cation. As
4-Chlorotestosterone (5) was transformed in a different
way: apart from the hydroxylation and the oxidation of C-
17 hydroxyl group, the reduction of 4-en-3-one system in
the A-ring was observed. The first step of this reduction
was formation of the 3␤-allyl alcohol (Fig. 4). Microbial
reduction of an ␣,␤-unsaturated ketone to an allyl alco-

´
A. Swizdor, T. Kołek / Steroids 70 (2005) 817–824
823
Fig. 4. Proposed metabolic pathway of 4-chlorotestosterone by F. culmorum. (i) and (ii) were isolated only after 12 or 24 h of transformation.
hol is a rarely observed process [39,40]. Much more often,
there are ␣-chlorosubstituted ␣,␤-unsaturated ketones that
undergo this type of transformation. This is because the
presence of the chlorine atom decreases unsaturated charac-
ter of a conjugate ketone [41]. The fact that 4-chloro-17␣-
methyltestosterone does not undergo the reduction in the
A ring indicates that the presence of the 17-oxo group has
an impact on this reaction. The products of the transforma-
tion of 4-chlorotestosterone were mainly the 15␣-hydroxy
derivatives. 4-Chloroandrostendione (i) and 3␤-hydroxy-4-
chloro-4-androsten-17-one (ii) (Fig. 4) were identified in the
reaction mixture after 12 and 24 h of the transformation, but
theywereabsentinthefinalmixture(datanotpresented). This
suggest that they also underwent the C-15 hydroxylation.
The course of transformations of the tested steroids by F.
culmorum indicated that the 6␤-hydroxy metabolism played
minor role when the metabolism of the A-ring was rapid, as
it was in the case of 4-chlorotestosterone (5). Similar depen-
dence was observed in the metabolism of anabolic steroids
in humans [25].
References
[1] Charney W, Herzog HL. Microbial transformations of steroids. New
York/London: Academic Press; 1967.
[2] Akhrem AA, Titow YA. Steroids and mikroorganisms. Moscow:
Nauka Press; 1970.
ˇ
[3] Capek A, Hancˇ O. Microbiological transformation of steroids XII.
Transformation of steroids by different species and strains of Fusaria.
Folia Microbiol 1960;5:251–6.
[4] Gluber A, Tamm Ch, Juchasz G, Weiss-Berg E, Zu¨cher W. Uber die
¨
Substratspezifita¨t der Hydroxylierung von Androstan- und Pregnan
derivaten durch Fusarium lini. Helv Chim Acta 1963;46:889–906.
[5] Defaye G, Luche MJ, Chambaz EM. Microbiological 7- and 15-
hydroxylations of C-19 steroids. J Steroid Biochem 1978;9:331–6.
[6] Cotillon AC, Doostzadech J, Morfin R. The inducible and
cytochrome P450-containing dehydroepiandrosterone 7␣-hydroxy-
lating enzyme system of Fusarium moniliforme. J Steroid Biochem
Mol Biol 1997;62:467–75.
[7] Cotillon AC, Morfin R. Transformation of 3-hydroxy-steroids by
Fusarium moniliforme. J Steroid Biochem Mol Biol 1999;68:229–37.
[8] Wilson MR, Gallimore WA, Reese PB. Steroid transformation
with Fusarium oxysporum var. cubense and Colletotrichum musae.
Steroids 1999;64:834–43.
´
[9] Jakkel A, Ilko˝y E, Horva´th G, Pallagi I, Su¨to˝ J, Ambrus G. Micro-
In general, the biotransformations of the tested sub-
strates led mainly to monohydroxy products. Only 4-
methyltestosterone (4), when subjected to the reaction for
longer time (10 days), gave 6␤,11␣- and 6␤,15␣-dihydroxy
derivatives. In our earlier works with F. culmorum we also
observed formation of dihydroxy products in the case of other
relatively lipofilic substrates [10,11].
Twelve of the metabolites obtained are the new com-
pounds. The hydroxy derivatives isolated from the biotrans-
formation of 4-chlorotestosterone may be helpful in tracing
the fate of this anabolic steroid in mammals.
bial hydroxylation of 13␤-ethyl-4-gonene-3,17-dione. J Mol Catal B:
Enzymatic 1998;5:385–7.
´
[10] Kołek T, Swizdor A. Biotransformation XLV. Transformations of 4-
ene-3-oxo steroids in Fusarium culmorum culture. J Steroid Biochem
Mol Biol 1998;67:63–9.
[11] Kołek T. Biotransformation XLVII: transformations of 5-ene steroids
in Fusarium culmorum culture.
J Steroid Biochem Mol Biol
1999;71:83–90.
[12] Brannon DR, Parrish FW, Wiley BJ, Long Jr L. Microbial trans-
formation of a series of androgens with Aspergillus tamarii. J Org
Chem 1967;32:1521–7.
[13] Mosbach EH, Shefer S, Abell LL. Identification of the fecal metabo-
lites of 17␣-methyltestosterone in the dog. J Lipid Res 1968;9:93–7.

´
A. Swizdor, T. Kołek / Steroids 70 (2005) 817–824
824
[14] Sta`rka L, Siekmann L, Hoppen HO, Breuer H. Stoffwechsel von
4-Chlortestosteron in der Leber des Menschen. Arzneim-Forsch
1969;19:2022–5.
[15] Castegnaro E, Sala G. Absorpion and metabolism of 4-
chlorotestosterone acetate by oral route. Steroids Lipids Res
1973;4:184–92.
[28] Williams TM, Kind AJ, Hyde WG, Hill DW. Characterization of uri-
nary metabolites of testosterone, methyltestosterone, mibolerone and
boldenone in greyhound dogs. J Vet Pharmacol Ther 2000;23:121–9.
[29] Huszcza E, Dmochowska-Gładysz J. Transformations of testosterone
and related steroids in Absidia glauca culture. J Basic Microbiol
2003;43:113–20.
[16] Templeton JF, Jackson CC. Metabolism of 17␣-methyltestosterone
in the rabbit: C-6 and C-16 hydroxylated metabolites. Steroids
1983;42:115–21.
[17] Scha¨nzer 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;275:23–48.
[18] Thieme D, Anielski P, Grosse J, Sachs H, Mueller RK. Identyfication
of anabolic steroids in serum, urine, sweat and hair comparison if
metabolic patterns. Anal Chim Acta 2003;483:299–306.
[19] Ayotte C, Goudreault D, Charlebois A. Testing for natural and
[30] Von Schubert K, Schlegel J, Groh H, Rose G, Ho¨rhold VIII Cla¨re.
Struktur-Stoffwechel-Beziehungen bei der mikrobiellen Hydrierung
unterschiedlich substituierter Testosteronderivate. Endokrinologie
1972;59:99–114.
[31] Kirk DN, Toms HC, Douglas C, White KA, Smith KE, Latif S, et al.
A survey of the high-field 1H NMR spectra of the steroid hormones,
their hydroxylated derivatives, and related compounds. J Chem Soc
Perkin Trans 1990;2:1567–94.
[32] Hammoumi A, Girault JP, Azerad R, Revial G, d’Angelo J. Microbial
hydroxylation and functionaliztion of synthetic policyclic enones.
Tetrahedron Asymmetry 1993;4:1295–306.
[33] Bhacca NS, Williams DH. Applications of NMR Spectroscopy in
organic chemistry. San Francisco/London/Amsterdam: Holden-Day,
Inc.; 1964. p. 189–190.
[34] Zu¨rcher RF. Protonenresonanzspektroskopie und Steroidstruktur II.
Die Lage der C-18- und C-19-Methylsignale in Abha¨ngigkeit
von den Substituenten am Steroidgeru¨st. Helv Chim Acta
1963;46:2054–88.
[35] Kaufman G, Schumann G, Horhold C. Influence of 1-double bond
and 11␤-hydroxy on stereospecific microbial reductions of 4-en-3-
oxo steroids. J Steroid Biochem 1986;25:561–6.
[36] Holland HL, Brown FM, Chenchaiah PC, Chernishenko MJ, Khan
SH, Appa Rao J. Microbial hydroxylation of steroids. Part 12.
Hydroxylation of testosterone and related steroids by Gnomonia fruc-
ticola ATCC 11430. Can J Chem 1989;67:268–74.
[37] Newcomb M, Shen R, Choi S-Y, Toy PH, Hollenberg PF, Vaz
ADN, et al. Cytochrome P450-catalyzed hydroxylation of mecha-
nistic probes that distinguish between radicals and cations. evidence
for cationic but not for radical intermediates. J Am Chem Soc
2000;122:2677–86.
synthetic anabolic agents in human urine.
1996;687:3–25.
J Chromatogr B
[20] Walshe M, O’Keeffe M, Le Bizec B. Studies on the determination
of chlorotestosterone and its metabolites in bovine urine. Analyst
1998;123:2687–91.
[21] Rendic S, Nolteernsting E, Scha¨nzer W. Metabolism of anabolic
steroids by recombinant human cytochrome P450 enzymes. Gas
chromatographic-mass spectrometric determination of metabolites. J
Chromatogr B 1999;735:73–83.
[22] Schoene C, Nedderman ANR, Haughton E. Preliminary study
of the metobolism of 17␣-methyltestosterone in horses utiliz-
ing gas chromatography-mass spectrometric techniques. Analyst
1994;119:2537–42.
[23] Andre´ F, Le Bizec B, Montrade MP, Maume D, Monteau F,
Marchand P. Developments in residue assays and metabolism
study of growth-promoters by mass spectrometric analysis. Analyst
1994;119:2529–35.
[24] Leyssens L, Royackers E, Gielen B, Missotten M, Schoofs J,
Czech J, et al. Metabolites of 4-chlorotestosterone acetate in cat-
tle urine as diagnostic markers for its illegal use. J Chromatogr B
1994;654:43–54.
[25] Scha¨nzer W, Horning S, Donike M. Metabolism of anabolic
steroids in humans: synthesis of 6␤-hydroxy metabolites of 4-chloro-
1,2-dehydro-17␣-methyltestosterone, fluoxymesterone, and metan-
dienone. Steroids 1995;60:353–66.
[38] Kieslich K. New examples of microbial transformations in pharma-
ceutical chemistry. Bull Soc Chim France 1980;1–2:9–17.
[39] Dmochowska-Gładysz J, Kołek T, Siewin´ski A. Reduction of testos-
terone, androstenedione and derivatives by Aphanocladium album. J
Basic Microbiol 1989;29:3–12.
[26] Le Bizec B, Momtrade MP, Monteau F, Gaudin I, Andre F. 4-
Chlorotestosterone acetate metabolites in cattle after intramuscular
and oral administrations. Clin Chem 1998;44:973–84.
[27] Williams TM, Kind AJ, Hill DW. In vitro biotransformation of
anabolic steroids in canines. J Vet Pharmacol Ther 2000;23:57–
66.
[40] Porter RBR, Gallimore WA, Reese PB. Steroid transformations with
Exophiala jeanselmei var. lecanii-corni and Ceratocystis paradoxa.
Steroids 1999;64:770–9.
[41] Ringold HJ, Ramachandran S, Forchielli E. The effect of electron-
withdrawing and electron-releasing substituents on the enzymic
reduction of steroids. Biochim Biophys Acta 1964;82:143–57.