Hydroxyl groups at C-3 and at C-17 of the unnatural enantiomer, ent-androsta-5,9(11)-diene-3β,17β-diol are oxidised by cholesterol oxidase from Rhodococcus erythropolis

Article abstract of DOI:10.1016/S0040-4039(00)02005-0

The ent-androsta-4,9(11)-diene-3β,17β-diol 1b and ent-androsta-5,9(11)-diene-3β,17β-diol 2b prepared from chiral dione 3, were oxidised by cholesterol oxidase with kinetic parameters close to those of the natural steroids 1a and 2a. In the preparative oxi

Full text of DOI:10.1016/S0040-4039(00)02005-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 505–507
Hydroxyl groups at C-3 and at C-17 of the unnatural
enantiomer, ent-androsta-5,9(11)-diene-3b,17b-diol are oxidised
by cholesterol oxidase from Rhodococcus erythropolis
b
a
a
a
a,
Dai Kitamoto, Serge Dieth, Alain Burger, Denis Tritsch and Jean-Fran c¸ ois Biellmann *
a
Laboratoire de Chimie Organique Biologique, UMR CNRS 7509, Facult e´ de Chimie, Universit e´ Louis Pasteur,
1
rue Blaise Pascal, Fr-67008 Strasbourg, France
b
National Institute of Materials and Chemical Research, MITI, Tsukuba, Ibaraki 305-8565, Japan
Received 15 September 2000; revised 17 October 2000; accepted 8 November 2000
Abstract—The ent-androsta-4,9(11)-diene-3b,17b-diol 1b and ent-androsta-5,9(11)-diene-3b,17b-diol 2b prepared from chiral
dione 3, were oxidised by cholesterol oxidase with kinetic parameters close to those of the natural steroids 1a and 2a. In the
preparative oxidation the final product was ent-androsta-4,9(11)-diene-3,17-dione 5. So the enzyme catalysed the oxidation of the
hydroxyl groups at C-3 and C-17, whereas the natural enantiomers were only oxidised at C-3. © 2001 Elsevier Science Ltd. All
rights reserved.
Cholesterol oxidase (EC 1.1.3.6) catalyses the oxidation
of cholesterol into cholest-5-en-3-one and the isomerisa-
tion of cholest-5-en-3-one to cholest-4-en-3-one. Previ-
differ by the position of rings B and D as shown in Fig.
1. Due to this partial symmetry, cholesterol oxidase
may not be enantiospecific and could oxidise ent-
steroids.
1
ously, we found that the enantioselectivity of the
oxidation of bi- and tricyclic alcohols with cholesterol
2
–4
oxidase from Rhodococcus erythropolis,
was lower
The ‘natural’ enantiomers 1a and 2a were prepared
5
6
with the tricyclic than with the bicyclic alcohols. We
then realised that the enantiomeric steroids have more
similarities than may be seen on first examination. If
the structure of the natural steroid turned by 180°
around an axis joining C-3 to C-13 (‘upside-down’)
is compared to the structure of the ent-steroid, the
hydroxyl group at C-3, rings A and C and the
methyl groups at position C-18 and C-19 of both
enantiomers are superimposed. The enantiomers
from cortisol acetate. We also prepared ent-androsta-
4,9(11)-diene-3b,17b-diol 1b and ent-androsta-5,9(11)-
7
–10
diene-3b,17b-diol 2b from chiral dione 3 (Scheme 1).
To improve yields we modified the reaction conditions
of the published synthesis. The kinetic parameters (kcat,
K ) for cholesterol oxidase were determined by a mod-
m
2
,11
ification of the method of Smith and Brooks.
An-
drosta-4-ene-3b,17b-diol and androsta-5-ene-3b,17b-
diol were used as references.
1
8
H
H
18
18
R
R
R
19
19
19
1
3
H
H
H
H
H
3
H
H
3
H
3
HO
HO
HO
H
3β-hydroxy-sterol
ent-3β-hydroxy-sterol
Figure 1. Structures of the enantiomers of 3b-hydroxy-sterol.
*
0
Corresponding author. E-mail: jfb@chimie.u-strasbg.fr
040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)02005-0

5
06
D. Kitamoto et al. / Tetrahedron Letters 42 (2001) 505–507
OH
OH
OH
17
17
H
H
H
H
H
3
H
3
HO
HO
HO
HO
2a
2b
1
a
OH
OH
O
17
H
H
H
3
H
O
HOOC
8 steps
O
3
4
1b
Scheme 1.
The first order rate constant for this process (k=
corresponding 3-keto-4-ene. The mechanism of the iso-
merisation of the enantiomers remains to be studied.
−
1
0.00684 min ; t1/2=101 min) as determined from the
The unnatural enantiomers 1b and 2b were oxidised
with rate constants close to those of the corresponding
natural enantiomers 1a and 2a (Table 1). The enzyme
was found to catalyse not only the oxidation at the
hydroxyl group at C-3 of products 1b and 2b, but also
the isomerisation of the double bond of compound 2b
as shown by the formation of the conjugated ketone,
which is used to measure the enzymatic rate.
The unnatural enantiomers 4, 1b and 2b may bind
again to the active site of cholesterol oxidase in the
‘
upside down’ and ‘backward’ mode to account for the
oxidation of the hydroxyl group at C-17. In contrast,
the natural enantiomers 1a and 2a were thus not bound
to the active site in the ‘backward’ mode. The resolu-
tion of the racemic steroidal diol with cholesterol oxi-
dase could be performed owing to the enantiospecificity
of the oxidation at C-17 but not at C-3. Other struc-
tures related to natural steroids might be substrates of
cholesterol oxidase. The enantiospecificity should be
determined for other enzymes of the steroid
metabolism: such as 5-ene-3-keto steroid isomerase,
hydroxysteroid dehydrogenase and hydroxysteroid sulfo-
transferase. These enzymes are neither regio- nor
stereospecific due to the ‘wrong-way’ binding mode, i.e.
5
In the preparative oxidation in biphasic system, the
final reaction product from 1b and 2b was not the
expected ent-dehydrotestosterone 4, but ent-androsta-
1
2
4
,9(11)-diene-3,17-dione 5 (Scheme 2). By chromato-
graphy we detected the ent-dehydrotestosterone 4 as an
intermediate.
The oxidation rate of the hydroxyl group at C-17 must
be significant, but slower than the oxidation at C-3 and
the double bond isomerisation.
16,17
the ‘backward’ and/or ‘upside-down’ mode.
ent-
Steroids may act as inhibitors for some of these en-
zymes.
Another interesting feature is the isomerisation of the
b,g-enone to the a,b-enone on unnatural enantiomer
2
b. In cholesterol oxidase from Brevibacterium
13,14 361
Acknowledgements
sterolicum,
the terminal carboxylate of Glu , the
base involved in the isomerisation, is positioned over
15
the b-face of the bound sterol and is quite mobile. In
the cholesterol oxidase used here, the base involved in
the isomerisation, likely positioned over the b-face of
the bound substrate, is also quite mobile, since both
enantiomers of 3-keto-5-ene were isomerised to the
We thank Roussel-Uclaf (France) for the supply of the
chiral dione 3, Roche Diagnostics (Penzberg, Germany)
for the cholesterol oxidase and Dr. R. Graf (Universite´
Louis Pasteur, Strasbourg) for the NMR experiments.
Table 1. Comparison of kinetic parameters of the natural and unnatural enantiomers of androstenediols for cholesterol
oxidase from Rhodococcus erythropolis
−
1
mg−1
kcat/Km (×103 L min−1 mg−1
)
Compounds (range of concentrations used, mM)
Km (mM)
kcat (mmol min
)
Androsta-4-ene-3b,17b-diol (18–190)
66
383
525
6.5
36
0.37
0.32
0.20
0.05
0.04
0.03
5.6
0.8
0.4
7.5
1.1
0.6
Androsta-4,9(11)-diene-3b,17b-diol (1a) (49–396)
ent-Androsta-4,9(11)-diene-3b,17b-diol (1b) (136–820)
Androsta-5-ene-3b,17b-diol (3.7–158)
Androsta-5,9(11)-diene-3b,17b-diol (2a) (11–125)
ent-Androsta-5,9(11)-diene-3b,17b-diol (2b) (21–116)
52

D. Kitamoto et al. / Tetrahedron Letters 42 (2001) 505–507
507
O
OH
OH
17
17
H
H
H
3
H
H
3
H
HO
HO
O
2b
5
1b
Scheme 2.
References
11. Cholesterol oxidase as a 3 M NaCl aqueous solution
from Roche Diagnostics dialysed against 0.1 M phos-
phate buffer (pH 7.4) had a specific activity of 66 U/mg
at 25°C towards cholesterol as specified by supplier. A
stock solution of steroid in 2-propanol was diluted to the
proper concentration with 2-propanol so that a constant
volume of solution (10 mL) was added to the assay (980
mL of buffer at 30°C). The reaction was initiated by the
addition of cholesterol oxidase (100–110 mg of protein) in
the buffer (10 mL), and was followed by measuring the
UV absorbance at 8–10 different substrate concentra-
tions. The slope of the first 10% of the reaction was
determined by linear regression and converted into mmol/
1
. For a review, see: Maclachlan, J.; Wotherspoon, A. T. L.;
Ansell, R. O.; Brooks, C. J. W. J. Steroid Biochem. Mol.
Biol. 2000, 72, 169–195.
2
. Smith, A. G.; Brooks, C. J. W. Biochem. J. 1977, 167,
1
21–129.
3
. Sojo, M.; Bru, D.; Lopez-Molina, D.; Garcia-Carmona,
F.; Arg u¨ elles, J.-C. Appl. Microbiol. Biotechnol. 1997, 47,
5
83–589.
4
. Elalami, A.; Kreit, J.; Filali-Malfout, A.; Boudrant, J.;
Germain, P. R. P. World J. Microbiol. Biotechnol. 1999,
1
5, 579–585.
. Dieth, S.; Tritsch, D.; Biellmann, J.-F. Tetrahedron Lett.
995, 36, 2243–2246, and unpublished results.
. Audouin, M.; Levisalles, J. Bull. Soc. Chim. Fr. 1985,
280–1284.
. Nomin e´ , G.; Amiard, G.; Torelli, V. Bull. Soc. Chim. Fr.
968, 3664–3673.
. Belle, C.; Guarna, A.; Paoli, P.; Dapporto, P. J. Chem.
Res. (M) 1992, 0901–0909, J. Chem. Res. (S) 1992,
22–123.
. Rychnovsky, S.; Mickus, D. E. J. Org. Chem. 1992, 57,
732–2736.
5
6
7
8
−
1
−1
min/mg using o248=15000 M cm for dehydrotestos-
1
−
1
−1
terone and o250=14200 M cm for testosterone.
2. The oxidation of the hydroxyl group at C-17 of the
unnatural enantiomers was also observed with cholesterol
oxidase from Pseudomonas fluorescens, but not with the
enzymes from Brevibacterium sp. and Streptomyces sp.
3. Li, J.; Vrielink, A.; Brick, P.; Blow, D. M. Biochemistry
1
1
1
1
1
1
1
993, 32, 11507–11515.
4. Vrielink, A.; Lloyd, L. F.; Blow, D. M. J. Mol. Biol.
991, 219, 533–554.
1
9
1
2
5. Kass, I. J.; Sampson, N. S. Biochemistry 1998, 37,
1
0. ent-Dehydrotestosterone 4 was shown to be a single
1
enantiomer by chiral HPLC and by H NMR of the
17990–18000.
(
R)-Mosher’s ester of 2b. See: Dale, J. A.; Dull, D. L.;
16. Schwab, J. M.; Henderson, B. S. Chem. Rev. 1990, 90,
1203–1245.
Mosher, H. S. J. J. Org. Chem. 1969, 34, 2543–2549 [The
modifications will be published in a full paper].
17. Biellmann, J. F. Acc. Chem. Res. 1986, 19, 321–328.
.
.