Regio- and stereoselective reduction of 17-oxosteroids to 17β-hydroxysteroids by a yeast strain Zygowilliopsis sp. WY7905

Article abstract of DOI:10.1016/j.steroids.2016.11.002

The reduction of 17-oxosteroids to 17β-hydroxysteroids is one of the important transformations for the preparation of many steroidal drugs and intermediates. The strain Zygowilliopsis sp. WY7905 was found to catalyze the reduction of C-17 carbonyl group of androst-4-ene-3,17-dione (AD) to give testosterone (TS) as the sole product by the constitutive 17β-hydroxysteroid dehydrogenase (17β-HSD). The optimal conditions for the reduction were pH 8.0 and 30 °C with supplementing 10 g/l glucose and 1% Tween 80 (w/v). Under the optimized transformation conditions, 0.75 g/l AD was reduced to a single product TS with >90% yield and >99% diastereomeric excess (de) within 24 h. This strain also reduced other 17-oxosteroids such as estrone, 3β-hydroxyandrost-5-en-17-one and norandrostenedione, to give the corresponding 17β-hydroxysteroids, while the C-3 and C-20 carbonyl groups were intact. The absence of by-products in this microbial 17β-reduction would facilitate the product purification. As such, the strain might serve as a useful biocatalyst for this important transformation.

Full text of DOI:10.1016/j.steroids.2016.11.002

Accepted Manuscript
Regio- and stereoselective reduction of 17-oxosteroids to 17β-hydroxysteroids
by a yeast strain Zygowilliopsis sp. WY7905
Yuanyuan Liu, Yu Wang, Xi Chen, Qiaqing Wu, Min Wang, Dunming Zhu,
Yanhe Ma
PII:
S0039-128X(16)30165-9
http://dx.doi.org/10.1016/j.steroids.2016.11.002
STE 8051
DOI:
Reference:
To appear in:
Steroids
Received Date:
Revised Date:
Accepted Date:
19 July 2016
29 September 2016
6 November 2016
Please cite this article as: Liu, Y., Wang, Y., Chen, X., Wu, Q., Wang, M., Zhu, D., Ma, Y., Regio- and stereoselective
reduction of 17-oxosteroids to 17β-hydroxysteroids by a yeast strain Zygowilliopsis sp. WY7905, Steroids (2016),
doi: http://dx.doi.org/10.1016/j.steroids.2016.11.002
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Regio- and stereoselective reduction of 17-oxosteroids to 17β-hydroxysteroids by
a yeast strain Zygowilliopsis sp. WY7905
1
Yuanyuan Liu , Yu Wang , Xi Chen , Qiaqing Wu , Min Wang , Dunming
,2,⊥
2,⊥
2
2,*
1
2
Zhu and Yanhe Ma
,*
2
1
Key Laboratory of Industrial Fermentation Microbiology, Ministry of Education,
College of Biotechnology, Tianjin University of Science & Technology, Tianjin
300457, China
2
National Engineering Laboratory for Industrial Enzymes and Tianjin Engineering
Research Center of Biocatalytic Technology, Tianjin Institute of Industrial
Biotechnology, Chinese Academy of Sciences, Tianjin 300308, China
⊥
These authors contributed equally.
Corresponding Author:
Qiaqing Wu, Dunming Zhu, Tianjin Institute of Industrial Biotechnology, Chinese
Academy of Sciences, 32 Xi Qi Dao, Tianjin Airport Economic Area, Tianjin 300308,
China
Email: wu_qq@tib.cas.cn, zhu_dm@tib.cas.cn
1

Abstract
The reduction of 17-oxosteroids to 17β-hydroxysteroids is one of the important
transformations for the preparation of many steroidal drugs and intermediates. The
strain Zygowilliopsis sp. WY7905 was found to catalyze the reduction of C-17
carbonyl group of androst-4-ene-3,17-dione (AD) to give testosterone (TS) as the sole
product by the constitutive 17β-hydroxysteroid dehydrogenase (17β-HSD). The
optimal conditions for the reduction were pH 8.0 and 30°C with supplementing 10 g/l
glucose and 1% Tween 80 (w/v). Under the optimized transformation conditions, 0.75
g/l AD was reduced to a single product TS with > 90% yield and > 99%
diastereomeric excess (de) within 24 h. This strain also reduced other 17-oxosteroids
such as estrone, 3β-hydroxyandrost-5-en-17-one and norandrostenedione, to give the
corresponding 17β-hydroxysteroids, while the C-3 and C-20 carbonyl groups were
intact. The absence of by-products in this microbial 17β-reduction would facilitate the
product purification. As such, the strain might serve as a useful biocatalyst for this
important transformation.
Keywords: Zygowilliopsis sp; 17β-Reduction; Steroid biotransformation; Androst-4-
ene-3,17-dione; Testosterone
2

1. Introduction
1
7β-Hydroxysteroids
such
as
testosterone
(TS),
estradiol,
1,2-
dehydrotestosterone are of pharmaceutical importance and one of the straightforward
methods for their synthesis is the stereoselective reduction of 17-oxosteroids [1-3].
The ability of microorganisms to reduce 17-oxo- to 17β-hydroxysteroids was firstly
reported nearly a century ago and the 17β-reduction reaction was widely observed and
played important roles in microorganisms [3]. A variety of 17-oxosteroids were
reported to be reduced to 17β-hydroxysteroids by microorganisms of different
taxonomy: bacteria, fungi, yeasts, algae or protozoa [3-6]. As such, microbial 17β-
reduction of 17-oxosteroids was envisaged as a promisingly convenient and effective
method for the production of 17β-hydroxysteroids [3].
However, microbial 17β-reduction of 17-oxosteroids were usually companied
with other reactions such as hydroxylation, reduction of double bond and
dehydrogenation [3]. For example, 17-carbonyl reduction and hydroxylation at C-6β,
C-14α, C-15α positions occurred in the biotransformation of androst-1,4-dien-3,17-
dione (ADD) by filamentous fungus Mucor racemosus, resulting in two
monohydroxylated and three dihydroxylated products as well as 17β-hydroxyandrost-
1,4-dien-3-one [7]. A double reduction of both 17-oxo group and 1,2-C=C bond of
ADD by Mycobacterium sp. NRRL B-3683 was observed [8]. Therefore, we initiated
an effort to search for new microbial strains for effectively converting 17-oxosteroids
to 17β-hydroxysteroids without concurrence of other reactions. In this study, a new
Zygowilliopsis yeast in our laboratory was identified and found to effectively
transform androst-4-ene-3,17-dione (AD) to TS. In addition, the microbial reduction
3

conditions were optimized and the ability of this strain to reduce other 17- oxosteroids
was described.
2. Experimental
2.1. Materials and analytical methods
The 17-oxosteroids (≥97% of purity), standard samples of 17β-hydroxysteroids
(
≥98% of purity) and other chemical reagents and solvents were purchased from
chemical companies. 17α-Hydroxy-androst-4-en-3-one was prepared by the reduction
of AD with a carbonyl reductase available in our laboratory according to the
precedure described in the supplementary materials (section 3). Silica gel (Qingdao
Haiyang ChemicalCo., 200-300 mesh) was used for the column chromatography. Thin
layer chromatography (TLC) was developed with ethyl acetate and petroleum ether
(
6:4, v/v) and visualized by UV light (254 nm) and then by phosphomolybdic acid
colored. High performance liquid chromatography (HPLC) was performed with the
mixture of water and methanol (35:65, v/v) as eluent at a flow rate of 0.5 ml/min and
detection wavelength of 254 nm, 210 nm, 230 nm. The retention times for AD and TS
were 12.3 and 14.4 min, respectively. HPLC analysis was performed with the column
CHIRALPAK OD-H [Daicel Chemical Industries Ltd, 4.6 mm × 250 mm, n-
hexane/isopropanol (7:3, v/v), flow rate: 0.5 ml/min, UV254 nm]. The retention times
were 6.4 and 7.2 min for 17β-hydroxy-androst-4-en-3-one (TS) and 17α-hydroxy-
androst-4-en-3-one.
2.2. Microorganisms, conditions of cultivation and preparation of resting cells
4

Strains were isolated from soil samples in our previous studies [9,10] and stored
in our laboratory at -80°C with 20 % glycerol (v/v). The strains (5%, v/v) were
inoculated into 20 ml of liquid growth culture medium (15 g/l glucose, 5 g/l yeast
.
extract, 5 g/l peptone, 1 g/l K
2
HPO
4
, 1 g/l KH
2
PO
4
, 1 g/l NaCl, 0.5 g/l MgSO
4
7H
2
O,
pH 7.0) and grown at 30°C on a rotatory shaker at 200 rpm. After 12 h growth, 1 ml
of pre-culture liquid was transferred into 50 ml of fresh liquid growth medium with
same composition as described above and continued to grow at 30°C on a rotatory
shaker at 200 rpm for 24 h. The cells were harvested by centrifugation at 7875
0 min (4°C) and washed twice with phosphate buffer (100 mM, pH 7.0). The
resulting resting cells were used for the steroid transformation.
×
g, for
1
2.3. Screening for TS-producing strains
The resting cells (100 g/l, 0.2 g wet weight) obtained as described above were
resuspended in phosphate buffer (100 mM, pH 7.0) and added into 10 mL erlenmeyer
flask, containing glucose (10 g/l, 30 mg) and AD (1.5 g/l, 3 mg) dissolved in N,N-
dimethyl-formamide (DMF) (1%, v/v of DMF/buffer). The final reaction volume was
2
ml and the reactions were carried out at 30 °C, 200 rpm for 24 h under aerobic
conditions in the shaker. The whole reaction mixtures were extracted with an equal
volume of ethyl acetate, centrifuged, and the 1ml of supernatant was dried and
subjected to HPLC analysis. The yield (%) of product, the conversion (%) of substrate
and the de value were determined and calculated based on the areas of related peaks.
For the yield (%) of product, the conversion (%) of substrate, the amounts of product
and substrate in the reaction mixture were calibrated by the standard curve obtained
with known concentration of standard samples of product and substrate.
2.4. Phenotypic, biochemical characterization and 26S rDNA sequence analysis
5

Morphocytology of strain WY7905 was observed by light microscope (Leica
DM5000B, Germany). Morphological, biochemical, and physiological tests of strain
WY7905 were carried out as described in the literature [11]. Assimilation and
fermentation tests were conducted as described in the reference [12]. Total genomic
DNA was extracted using TIANamp Yeast DNA kit (Tiangen Biotech, Beijing,
China). The 26S rDNA was amplified by PCR (95°C for 3 min, one cycle; 95°C for
30 s/65°C for 30 s/72°C for 1 min 25 cycles; 72°C for 10 min ,one cycle) using
-
1
dNTPs
mix
(0.2
mmol·l ),
universal
primers
NL4:
NL1:
5'-
5'-
GCATATCAATAAGCGGAGGAAAAG-3',
-1
GGTCCGTGTTTCAAGACGG-3'; 0.5 µmol·l each), genomic DNA template (100
ng) and polymerase (2.5 U) in a final volume of 50 µl. PCR products were purified
by using Cycle-pure kits (Tiangen Biotech, Beijing, China) and sequenced. Highly
homologous strains were obtained by comparing this sequence with the GenBank
database using nucleotide BLAST. A phylogentic tree was constructed by using
ClustalX version 2.0, MEGA version 4.0 based on the homologous 26S rDNA
sequences.
2.5. The inductive effect of steroidal compounds on the 17β-HSD activity
Strain WY7905 was inoculated in 20 ml of culture medium at 30°C on a rotatory
shaker at 200 rpm. After growing for 12 h, 1 ml of the culture was transferred into 50
ml of growth medium containing 100 µg/ml of different steroidal compounds (estrone,
hydrocortisone, cortodoxone, 3β-hydroxyandrost-5-en-17-one (DHEA), AD, TS,
progesterone) and grown at 30°C for 24 h. The cells were harvested as described in
section 2.2. The resting cells (0.2 g, wet weight) were resuspended in phosphate
buffer (100 mM, pH 7.0) and added into 10 mL erlenmeyer flask, containing glucose
6

(
15 g/l, 30 mg) and AD (1.5 g/l, 3 mg) dissolved in DMF (1%, v/v of DMF/buffer).
The final reaction volume was 2 ml and the reactions were carried out at 30°C, 200
rpm for 24 h under aerobic conditions in the shaker. The whole reaction mixtures
were extracted with an equal volume of ethyl acetate. Then the extract was dried and
analyzed by HPLC to determine the conversion and yield.
2.6. Time course of 17β-HSD production of Zygowilliopsis sp. WY7905
Zygowilliopsis sp.WY7905 was cultivated as described in section 2.2. The OD600
of the cell culture was measured by ultraviolet spectrophotometer at various time
intervals. The cells were collected at these time intervals as described in section 2.2
and used to determine the 17β-HSD activity by measuring the yield of TS as
described in section 2.5.
2.7. Optimization of reaction conditions for the 17β-reduction of AD by the resting
cells
The resting cells (0.2 g, wet weight) resuspended in different pH buffers (100
mM, sodium acetate buffer for pH 4.0-6.0, phosphate buffer for pH 6.0-8.0, Tris-HCl
buffer for pH 8.0-9.0, Na CO /NaHCO buffer for pH 9.0-10.0) were added into 10 ml
2
3
3
erlenmeyer flask containing glucose (15 g/l, 30 mg) and AD (1.5 g/l, 3 mg) dissolved
in DMF (1 %, v/v of DMF/buffer). The final reaction volume was 2 ml and the
reactions were carried out at 30°C, 200 rpm for 24 h under aerobic conditions. The
reaction samples were worked-up and analyzed to evaluate the pH effect on the
transformation as described in section 2.5. To optimize the reaction temperature, the
reaction was carried out in phosphate buffer (100 mM, pH 8.0) at 20, 25, 30, 35, 37 or
40°C for 24 h, respectively. The reaction mixture was worked-up and analyzed as
described in section 2.5.
7

Similarly, the reaction was performed with different final concentrations of
glucose (0, 5, 10, 15, 20 or 25 g/l) in phosphate buffer (100 mM, pH 8.0). The
reaction mixture was shaken at 30°C for 24 h, worked-up and analyzed as described
in section 2.5.
AD (1.5 g/l, 3 mg) dissolved in different kinds of co-solvents (1%, v/v) or Tween
80 (0.5%, w/v) was added into 10 ml erlenmeyer flask containing glucose (10 g/l, 20
mg) and resting cells (0.2 g, wet weight) resuspended in phosphate buffer (100 mM,
pH 8.0). The reaction mixture (2 ml) was carried out at 30°C, 200 rpm for 24 h.
Similarly, AD (1.5 g/l, 3 mg) dissolved in different concentrations of Tween 80 was
added to the reaction mixture, and the reaction was carried out at 30°C for 24 h. The
reaction mixture was worked-up and analyzed as described in section 2.5.
AD (0.25-1.5 g/l) dissolved in Tween 80 (1%, w/v) were added into 10 ml
erlenmeyer flask containing glucose (10 g/l, 20 mg) and resting cells (0.2g, wet
weight) resuspended in phosphate buffer (100 mM, pH 8.0), and the resulting
mixtures (2 ml) were shaken at 30°C and 200 rpm for 24 h. Then the reaction
mixtures were worked-up and analyzed as described in section 2.5. Under the
optimized substrate concentration, the time course of bioreduction of AD by
Zygowilliopsis sp. WY7905 was investigated. Samples were drawn at the different
intervals, worked-up and analyzed as described in section 2.5.
2.8. Biotransformation of AD by growing cells of Zygowilliopsis sp. WY7905
Zygowilliopsis sp. WY7905 was inoculated into 20 ml growth medium and
incubated at 30°C for 12 h. Then 1 ml of pre-culture liquid was added into 20 ml
fermentation medium (pH 7.0) consisted of 30 g/l of glucose, 5 g/l of yeast extract, 5
2 4 2 4
g/l of peptone, 10.23 g/l of K HPO , 9.05 g/l of KH PO , 1 g/l of NaCl and 0.5 g/l of
8

4 2
MgSO .7H O. The resulting medium was cultivated at 30°C for 12 h. AD (0.5-2 g/l)
dissolved in the Tween 80 (1%, w/v) was added into the culture. 5 g/l of glucose was
added into the culture at intervals of 12 h within 72 h and the transformation was
implemented at 30°C under aerobic conditions. Samples were drawn at the time
intervals of 12 h, worked-up and analyzed as described in section 2.5.
2.9. Transformation of C-18, C-19 and C-21 steroids by the resting cells
C-18, C-19 and C-21 steroids (0.75 g/l, 1.5 mg), such as estrone (3), 9-
dehydroandrostenedione (7), norandrostenedione (9), estra-4,9-diene-3,17-dione (11),
9-hydroxyandrost-4-ene-3,17-dione (13), etc separately dissolved in Tween 80 (1%
1
w/v) were added into 2 ml of reaction mixtures containing glucose (10 g/l, 10 mg) and
resting cells (0.2 g wet weight) resuspended in phosphate buffer (100 mM, pH
8.0).The mixture was shaken at 30°C for 24 h, then worked-up and analyzed as
described in section 2.5.
For the preparative scale reactions, different substrates (0.75 g/l, 150 mg) in
Tween 80 (1% w/v) were added into the reaction mixtures (200 ml) containing
glucose (10 g/l, 2 g) and resting cells (20 g wet weight) resuspended in phosphate
buffer (100 mM, pH 8.0). The reactions were carried out at 30°C for 24 h in shaker.
The whole reaction mixtures were extracted three times with ethyl acetate (200 ml),
and the organic extracts were dried over anhydrous sodium sulfate. The solvents were
evaporated under vacuum to give crude products, which were purified by silica gel
chromatography with ethyl acetate / petroleum ether as eluent to give pure products.
In order to determine the configuration of 17-hydroxyl, NMR analyses of the products
1
were performed (see below and supplementary materials section 4). H NMR data of
isolated products testosterone (2) (137.2 mg, 93% yield), 17β-estradiol (4) (15.2 mg,
9

11% yield) and 19-nortestosterone (10) (75.6mg, 53% yield) were compared with
1
those of authentic standard samples, H NMR data of androst-5-ene-3β,17β-diol (6)
1 13
64.2 mg, 47% yield) were compared to the literature data [13]. H, C and 2D NMR
(
(
1
1
1
1
1
13
1
13
H- H COSY, H- H NOESY, H- C HMBC, H- C HMQC) of products ∆-9(11)-
testosterone (8) (76.5 mg, 52% yield), 17β-hydroxy-estra-4,9-diene-3-one (12) (90.45
mg, 61% yield) and 19-hydroxytestosterone (14) (123.7 mg, 84% yield) were
conducted, and the results were summarized in Table 1, which showed that 17β-
hydroxyl products were obtained in all the cases.
2.9.1 Testosterone (2)
1
H NMR (400 MHz, CDCl ) δ 5.72 (s, 1H), 3.65 (t, 1H J = 4.0 Hz ), 2.25-2.46 (m,
3
4
1
3
H), 1.99-2.13 (m, 2H), 1.82-1.89 (m, 2H), 1.66-1.73 (m, 1H), 1.54-1.66 (m, 3H),
.39-1.52 (m, 2H), 1.27-1.34 (m, 1H), 1.19 (s, 3H), 1.05-1.13 (m, 1H), 0.89-1.05 (m,
1
H), 0.79 (s, 3H). The H NMR spectrum was indentical to the standard sample.
2.9.2 17β-Estradiol (4)
1
H NMR (400 MHz, CD OD) δ 7.00 (d, J = 8.0 Hz, 1H), 6.46 (dd, J = 5.6, 8.0 Hz,
3
1
H), 6.41 (d, J = 2.4 Hz, 1H), 3.58 (t, J = 8.8 Hz, 1H ), 2.61-2.79 (m, 2H), 2.14-2.28
(
m, 1H), 1.74-2.20 (m, 4H), 1.55-1.68 (m, 1H), 1.02-1.51 (m, 7H), 0.69 (s, 3H). The
H NMR spectrum was indentical to the standard sample.
1
2.9.3 Androst-5-ene-3β,17β-diol (6)
1
H NMR (400 MHz, CDCl ) δ 5.25 (d, J = 5.2 Hz, 1H), 3.42 (m, 1H ), 3.24 (m,
3
1
0
H ), 2.01-2.20 (m, 2H), 1.62-1.97 (m, 5H), 1.28-1.58 (m, 7H), 1.01-1.06 (m, 1H),
1
.95-1.05 (m, 2H), 0.94 (s, 3H), 0.80-0.90 (m, 2H), 0.64 (s, 3H). The H NMR
spectrum was indentical to that in the literature [13].
1
0

2.9.4 19-Nortestosterone (10)
1
H NMR (400 MHz, CDCl ) δ 5.82 (s, 1H), 3.66 (t, J = 8.0 Hz, 1H ), 2.35-2.53
3
(
m, 2H), 2.18-2.37 (m, 3H), 2.01-2.51 (m, 2H), 1.76-1.92 (m, 3H), 1.20-1.69 (m, 6H),
1
.95-1.18 (m, 3H), 0.82-0.80 (m, 1H), 0.80 (s, 3H). The H NMR spectrum was
0
indentical to the standard sample.
3. Results and discussion
3.1 Identification of TS-producing strains
A total of eighty strains from our laboratory were tested for their ability to
produce TS from AD. As shown in Table 2, eight strains were able to reduce AD to
TS with > 99% de and one strain (WY97-3) reduced AD less stereoselectively. The
strain WY7905 showed higher conversion for 17-ketone reduction. On the basis of
these results, the strain WY7905 was selected for further studies. Colonies of the
strain WY7905 on nutrient agar were milky, wet and smooth. Cells of the strain were
oval, non-motile, non-spore. The physiological characteristic of WY7905 was
investigated (supplementary materials Table S1 and Fig. S1). Comparative database
analysis of 26S rDNA sequences suggested that strain WY7905 was very similar with
strain in the yeast of Zygowilliopsis sp. The sequence similarity between WY7905 and
some Zygowilliopsis strains was ≥98%, revealing that strain WY7905 belongs to the
yeast Zygowilliopsis. Until now there was been no report about 17β-reduction of 17-
oxosteroids by Zygowilliopsis sp. The strain WY7905 was deposited at CGMCC
under accession number 2.5599.
3.2 The induction effect of steroidal compounds on the 17β-HSD activity
1
1

The inducible character of 17β-HSD from Comamonas testosteroni [14],
Streptomyces hydrogenans [15] and other bacteria had been reported, and only small
amounts of constitutive enzymes (1-5%) were found in cells grown in the absence of
steroidal compounds [3]. The possible inducibility of 17β-HSD from Zygowilliopsis
sp. WY7905 was examined by adding different steroidal compounds into the culture
medium. The strain did not grow on androstenedione or testosterone as a sole source
of carbon. As shown in Fig. 1, addition of steroidal compounds into the growth
medium did not enhance the 17β-HSD activity, indicating the 17β-HSD of WY7905
might be constitutive enzyme. For the fungal, Cochliobolus lunatus [16], besides
having inducible 11β-HSD and 20β-HSD activities, this strain was also found to have
a constitutive 17β-HSD activity by the reduction reaction with intact cells. A 17β-
HSD from Schizosaccharomyces pombe was also reported as a constitutive enzyme
[
17].
3.3 Time course of 17β-HSD production of Zygowilliopsis sp.WY7905
The cell growth and 17β-HSD production of Zygowilliopsis sp. WY7905 were
monitored by measuring the OD600 and yield of TS at different cultivation times. As
shown in Fig. 2, the 17β-HSD activity increased in parallel with cell growth during
the period of 24 h. The 17β-HSD activity of Zygowilliopsis sp. WY7905 reached the
highest at 24 h, and then decreased slightly. This is different from Williopsis saturnus
var. mrakii AJ-5620 [18], in which reductase activity decreased obviously after about
2
0 h. At all stages of WY7905 cell growth, the stereoselectivity of product was
unchanged (de >99%).
.4 Reaction condition optimization of the 17β-reduction of AD by resting cells
3
1
2

The reaction conditions were optimized by using resting cells of Zygowilliopsis
sp. WY7905 grown until optimal culture stage. The effects of buffer pH and
temperature on the reduction of AD were studied, and the results are presented in Figs
3
and 4. The yield was < 10% when pH was below 5.0. The conversion of AD to TS
improved with the increase of buffer pH from 5.0 to 8.0. The strain showed high 17β-
reduction activity at a broad pH range with the optimal pH being 8.0. The 17β-
reduction of AD by Zygowilliopsis sp. WY7905 reached highest conversion when the
biotransformation was carried out at 30°C.
Cofactor recycling is one of the most crucial issues encountered in bioreductions
and the cofactor regeneration with glucose as a co-substrate is preferred because of its
availability and low cost [19]. In the reduction reaction of AD by Zygowilliopsis sp.
WY7905, as shown in Fig. 5, the addition of glucose was able to accelerate the 17β-
reduction of AD and the substrate conversion reached maximum with 10 g/l glucose
in the reaction system. Higher glucose concentration had no effect on reduction,
suggesting that 10 g/l glucose was enough for cofactor regeneration.
Steroids have very low aqueous solubility which is in the range of 1–100 µM [20].
The low steroid aqueous solubility results in poor availability of the substrate to
whole-cell biocatalysts. In order to improve the solubility of steroids and further
increase the biocatalytic efficiency, the effects of various co-solvents besides Tween
80 on the 17β-reduction of AD were evaluated. As shown in Fig. 6, compared to the
organic co-solvents, higher yield of TS was obtained when Tween 80 was used to
improve the availability of the substrate. The effect of different concentrations of
Tween 80 on the bioconversion of AD by Zygowilliopsis sp. WY7905 was studied
1
3

(
Fig. 7). The yield of TS reached the highest when substrate was dissolved in 1% (w/v)
Tween 80.
The 17β-reduction of AD by Zygowilliopsis sp. WY7905 at different substrate
concentration was studied, and the results are presented in Fig. 8. When the
concentration of AD was 0.5 g/l and 0.75 g/l, the yields of TS achieved 96% and 93%
within 24 h, respectively. As the substrate concentration was increased up to 1.0 g/l,
the yield of TS decreased to below 80% at 24 h. The time course of 17β-reduction of
AD (0.75 g/l) catalyzed by Zygowilliopsis sp. WY7905 under the optimized reaction
conditions is presented in Fig. 9. Within 24 h, the bioreduction of AD was almost
completed and only TS was detected. With longer reaction time, the yield and
configuration of the product were unchanged (de >99%) .
3.5. 17β-reduction of AD by growing cells of Zygowilliopsis sp.WY7905
1
7β-reduction of AD with growing cells of Zygowilliopsis sp. WY7905 was also
investigated. As shown in Fig. 10, at 0.5 g/l, 1 g/l and 2 g/l substrate concentration,
the yield of TS achieved above 90% within 120 h and TS was the only product. At the
substrate concentration of 2.5 g/l, the yield of TS was above 80% within 120 h.
Currently, some other microbial strains have been reported to reduce 17-
oxosteroids to 17β-hydroxysteroids by growing cells. For example, Aspergillus
brasiliensis converted ADD to 17β-hydroxyandrost-1,4-dien-3-one at the final
substrate concentration of 1 g/l in 5 days. During this conversion, two hydroxylated
metabolites were detected [21]. Mucor piriformis could transform DHEA to androst-
5
-ene-3β,17β-diol with other by-products [22]. Similarly, Mucor racemocus
transformed AD to TS and produced other metabolite including C-6β, C-14α, C-15α
hydroxylated products [7]. Nostoc ellipsosporum and Nostoc muscorum transformed
1
4

AD and ADD to the corresponding 17-hydroxysteroids at substrate concentration of
.5 g/l in 7 days [23, 24]. Some yeasts could also produce TS from AD. For instance,
0
Saccharomyces cerevisiae could reduce AD to TS in the absence and presence of
cyclodextrin at substrate concentration of 2 g/l in 5 days with 27% and 90% yield of
product [25]. For Zygowilliopsis sp. WY7905, the growing cells of strain could
transform AD to TS in the absence and presence of Tween 80 at subtrate
concentration of 2 g/l in 5 days with 35% and 90% yield of product. Candida albicans
and Schizosaccharomyces pombe could reduce 17-ketosteroids to 17β-
hydroxysteroids, but the conversion efficiency was at low level [26, 27]. The
generation of by-products commonly observed in the microbial reduction of 17-
oxosteroids to 17β-hydroxysteroids resulted in the difficulty in product purification,
thus limiting their synthetic applications. In this study, no by-product was detected in
the 17β-reduction of AD by the resting cells and growing cells of Zygowilliopsis sp.
WY7905, resulting in TS as the sole product in greater than 90% yields at the
substrate concentration of 0.75 g/l and 2 g/l within 24 h and 120 h, respectively.
3.6. Transformation of other oxo-steroids by the resting cells
Biotransformation of other C18-, C19- and C21-steroids were also studied under
the optimized reaction conditions by using the resting cells of Zygowilliopsis sp.
WY7905. The results are summarized in Table 3. It can be seen that steroids
containing C-17 carbonyl groups were reduced to the corresponding 17β-
hydroxysteroids. 17β-reduction reactions of some steroids by microorganism were
reported for the first time, such as 9-dehydroandrostenedione, 19-hydroxytestosterone,
estra-4,9-diene-3,17-dione, the products of which are significant drug intermediates.
1
5

As shown in Table 3, TS and ethisterone, only containing C-3 carbonyl group,
were not reduced by Zygowilliopsis sp. WY7905, indicating that the strain did not
exhibit 3α/3β-HSD activity. In addition, the TS was not oxidized to AD in the
presence of glucose, indicating that the oxidative activity was much weaker than
reductive activity. C-20 carbonyl groups of C21-steroids were also barely reduced by
this strain. These results indicated that the strain WY7905 had highly selective 17β-
HSD activity under the conditions studied.
4. Conclusion
In this study, a 17β-HSD producing strain was identified as Zygowilliopsis sp.
WY7905, based on physiological biochemical characteristics, and 26S rDNA gene
sequence. Zygowilliopsis sp. WY7905 reduced AD to TS without the formation of by-
products. The reaction conditions (pH, temperature, amount of glucose, co-solvent)
were optimized for the bio-reduction with the resting cells of Zygowilliopsis sp.
WY7905. Under the optimized conditions, 0.75 g/l AD was transformed to TS with
93% yield within 24 h. The strain also reduced other 17-oxosteroids giving the
corresponding 17β-hydroxysteroids with no by-product being detected, but showed
highest activity toward AD. Almost no reduction activity was observed for steroids
with only C-3 or C-20 carbonyl group. The 17β-HSDs in Zygowilliopsis sp. WY7905
existed as constitutive enzymes. Cloning of the new 17β-HSD genes from the strain,
and characterization and synthetic applications of the recombinant enzymes are in
progress in our laboratory. These results will be reported in due course.
Abbreviations
TS
Testosterone
1
6

AD
4-Androstene-3,17-dione
Hydroxysteroid dehydrogenase
Androst-1,4-dien-3,17-dione
3β-Hydroxyandrost-5-en-17-one
N,N-Dimethylformamide
Dimethyl sulfoxide
HSD
ADD
DHEA
DMF
DMSO
EA
Ethyl acetate
MTBE
Methyl tertiary butyl ether
Acknowledgments
This work was financially supported by National High-Tech Research &
Development Program of China (863 Program, No. 2011AA02A211), STS Program
of Chinese Academy of Sciences (KFJ-SW-STS-164) and Tianjin Municipal Science
＆Technology Commission (11 ZCZDSY08900).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at
http://dx.doi.org/…
References
[
1] W.K. Sumanasekera, L. Zhao, M. Ivanova, D.D. Morgan, E.L. Noisin, R.S.
Keynton, C.M. Klinge, Effect of estradiol and dihydrotestosterone on
1
7

hypergravity-induced MAPK signaling and occludin expression in human
umbilical vein endothelial cells, Cell Tissue Res. 324 (2006) 243-253.
[
[
[
[
[
[
2] R. Glaser, C. Dimitrakakis, Testosterone and breast cancer prevention, Maturitas
82 (2015) 291-295.
3] M.V. Donova, O.V. Egorova, V.M. Nikolayeva, Steroid 17β-reduction by
microorganisms- a review, Process Biochem. 40 (2005) 2253-2262.
4] K. Kristan, T. Lanišnik Rižner, Steroid-transforming enzymes in fungi, J. Steroid
Biochem. 129 (2012) 79-91.
5] M.V. Donova, O.V. Egorova, Microbial steroid transformations: current state and
prospects, Appl. Microbiol. Biotechnol. 94 (2012) 1423-1447.
6] M.V. Donova, Transformation of steroids by actinobacteria: A review, Appl.
Biochem. Micro. 43 (2007) 1-14.
7] M.A. Faramarzi, M. Badiee, N. Zolfaghary, Y.M. Tabatabaei, M. Amini,
Microbial conversion of androst-1,4-dien-3,17-dione by Mucor racemosus to
hydroxysteroid-1,4-dien-3-one derivatives, J. Mol. Catal. B: Enzym. 84 (2009)
1021-1025.
[
[
8] W.H. Liu, C.W. Kuo, K.L. Wu, C.Y. Lee, W.Y. Hsu, Transformation of
cholesterol to testosterone by Mycobacterium sp, J. Ind. Micriobiol. Biot. 13
(
1994) 167-171.
9] Y. Wang, J.J. Li, Q.Q. Wu, D.M. Zhu, Microbial stereospecific reduction of 3-
quinuclidinone with newly isolated Nocardia sp. and Rhodococcus erythropolis,
J. Mol. Catal. B: Enzym. 88 (2013) 14-19.
1
8

[
10] H.L. Zhang, X.Z. Gao, J. Ren, J.H. Feng, T.C. Zhang, Q.Q. Wu, D.M. Zhu,
Enzymatic hydrogenation of diverse activated alkenes. Identification of two
Bacillus old yellow enzymes with broad substrate profiles, J. Mol. Catal. B:
Enzym. 105 (2014) 118-125.
[
[
[
11] J.A. Barnett, R.W. Payne, D. Yarrow, Yeasts: characteristics and identification.
Cambridge University Press, Cambridge, UK (2000).
12] C.P. Kurtzman, J.W. Fell, The yeasts, a taxonomic study. Elsevier, New York
(
1998).
13] M. Choudhary, S. Zafar, N.T. Khan, A. Rahman, Biotransformation of
dehydroepiandrosterone with Macrophomina phaseolina and β-glucuronidase
inhibitory activity of transformed products, J. Enzym. Inhib. Med. Chem. 27
(
2012) 348-355.
[
[
14] J.E Cabrera, J.L.P. Paz, S. Genti-Raimondi, Steroid-inducible transcription of the
β/17β-hydroxysteroid dehydrogenase gene (3β/17β-HSD) in Comamonas
3
testosteroni, J. Steriod Biochem. Mol. Biol. 73 (2000) 147-152.
15] B. Bauer, L. Trager, Enzyme induction in Streptomyces hydrogenans.
Comparison of the effects of different steroids to increase the activity of 3α, 20β-
hydroxysteroid dehydrogenase and of 3β, 17β-hydroxysteroid dehydrogenase, Z.
Allg. Microbiol. 22 (1982) 287-92.
[
16] A. Plemenitas, M. Zakeljmavric, R. Komel, Hydroxysteroid dehydrogenase of
Cochliobolus Lunatus, J. Steroid Biochem. Mol. Biol. 29 (1988) 371-372.
1
9

[
17] J. Dlugonꢀ
ꢀ
and cell morphology of Schizosaccharomyces pombe, Antonie van.
Leeuwenhoek 73 (1998) 189-194.
[
[
18] I. Kira, T. Suzuki, N. Onishi, K. Watanabe, E. Nakanishi, Microbial production
of (S)-1-phenyl-1,3-propanediol by stereospecific reduction of 3-hydroxy-1-
phenylpropane-1-one, J. Mol. Catal. B: Enzym. 56 (2009) 283-287.
19] Y. Nie, Y. Xu, X.Q. Mu, Highly enantioselective conversion of racemic 1-
phenyl-1,2-ethanediol by stereoinversion involving a novel cofactor-dependent
oxidoreduction system of Candida parapsilosis CCTCC M203011, Org. Process
Res. Dev. 8 (2004) 246-251.
[
[
[
20] R.Goetschel, R Bar, Formation of mixed crystals in microbial conversion of
sterols and steroids, Enzyme Microb. Technol. 14 (2001) 462-469.
21] D.R. Brannon, F.W. Parrish, B.J. Wiley, L. Long, Microbial transformation of a
series of androgens with Aspergillus tamarii, J. Org. Chem. 32 (1967) 1521-1527.
22] D. Madyastha, T. Josegh, Transformation of dehydroepiandrosterone and
pregnenolone by Mucor piriformis, Appl. Microbiol. Biotechnol. 44 (1995) 339-
343.
[
23] M.A. Faramarzi, M.T. Yazdi, H. Ghotinroudi, M. Amini, Nostoc muscorum: a
regioselective biocatalyst for 17-carbonyl reduction of androst-4-en-3,17-dione
and androst-1,4-dien-3,17-dione, Ann. Microbiol. 56 (2006) 253-256.
2
0

[
[
[
[
24] Z. Moradpour, M. Torshabi, M.A. Faramarzi, Microalgal transformation of
androst-4-en-3, 17-dione by Nostoc ellipsosporum, Res. J. Microbiol. 3 (2006)
289-293.
25] Y. Singer, H. Shity, R. Bar, Microbial transformations in a cyclodextrin medium.
Part 2. Reduction of androstenedione to testosterone by Saccharomyces
cerevisiae, Appl. Microbiol. Biotechnol. 35 (1991) 731-737.
26] R. Ghraf, E. Rodney, S. Oza, H. Schriefers, Transformation of C18-steroids,
C19-steroids and C21-steroids by cultures of Candida Albicans, J. Steroid
Biochem. Mol. Biol. 6 (1975) 1531-1536.
27] T. Pajic, M. Vitas, D. Zigon, A. Pavko, S.L. Kelly, R. Komel, Biotransformation
of steroids by the fission yeast Schizosaccharomyces pombe, Yeast 15 (1999)
639-645.
2
1

1
13
Table 1. H-NMR and C-NMR data of compounds (δ in ppm).
8
12
14
Carbon
1
3
1
13
1
13
1
C
H
C
H
2.49
C
H
1
2
3
4
5
6
7
8
9
31.7
32.8
199.4
123.9
169.9
34.2
33.8
37.5
144.7
41.0
118.6
38.4
41.3
47.6
24.1
30.6
81.6
10.5
26.1
2.01, 1.06
2.55, 2.34
37.2
25.9
199.9
122.2
157.4
30.9
27.2
39.5
146.5
125.5
25.7
36.6
43.0
51.2
23.5
30.7
81.4
10.6
34.6
36.0
203.4
126.8
172.0
34.7
33.3
37.6
52.3
45.5
24.3
38.2
44.1
55.8
22.3
30.7
82.3
11.8
66.1
2.41, 1.77
2.77, 2.29
2.90, 2.50
5.74
5.69
5.87
2.47, 2.48
2.10,
2.21
2.42, 2.35
1.90, 1.27
2.25
2.52, 2.38
1.93, 1.05
1.73
0.95
1
1
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
7
8
9
5.53
2.12, 1.94
2.80, 2.14
1.95, 1.25
1.63, 1.32
1.86, 1.03
1.25
1.78, 1.38
2.11, 1.53
3.75
1.15
1.68, 1.39
2.12, 1.52
3.70
1.06
1.72, 1.54
1.98, 1.48
3.58
0.83
4.04, 3.83
0.76
1.35
0.94
Table 2. Transformation of AD to TS by resting cells of different strains.
Strain number 8101
97-3
52.2
97
8002
17.77
99
0902
18.01
99
0802
11.82
99
0801
18.23
99
2401
17.04 30.59 10.21
99 99 99
7905
93-3
Yield (%)
16.68
99
a
de (%)
a
The diastereomeric excess values were measured by HPLC analysis.
Table 3. Transformation of oxo-steroids by the resting cells of Zygowilliopsis sp.
WY7905.
a
b
Entry
Substrate
Product
Conversion (%)
De (%)
2
2

O
OH
1
2
3
96
11.8
48
99
99
99
O
O
Androst-4-ene -3,17-dione
1)
O
Testosterone (2)
(
OH
HO
HO
Estrone (3)
17β-estradiol (O4H)
O
HO
HO
3
β-hydroxyandrost-5-en-
androst-5-ene-3β,17β-diol
(6)
1
7-one (5)
O
OH
4
O
52.5
99
9
-
O
Dehydroandrostenedione
7)
O
∆-9(11)-testosterone(8)
(
OH
5
6
53
99
99
O
O
Norandrostenedione(9)
O
19-nortestosterone (10)
OH
61.5
O
O
Estra-4,9-diene-3,17-
dione(11)
O
17β-hydroxy-estra-4,9-
diene-3 one(12)
OH
HO
HO
7
8
9
85.3
99
O
O
1
9-hydroxyandrost-4-ene-
3
1
9-hydroxytestosterone (14)
,17-dione(13)
OH
c
-
-
O
Testosterone (2)
OH
c
-
-
O
Ethisterone (15)
HO
O
O
OH
1
0
-
-
Trace
Trace
O
Cortisone (16)
HO
O
HO
OH
1
1
O
2
3

Hydrocortisone (17)
a
The conversions were determined by HPLC analysis.
b
c
The diastereomeric excess values were measured by NMR.
No conversion was observed.
2
4

Fig. 1. The effect of different steroids on the 17β-reduction activity of Zygowilliopsis
sp. WY7905 (□) Conversion of AD; (■) yield of TS.
Fig. 2. Time-course of cell growth and enzyme production by Zygowilliopsis sp.
WY7905(●) cell growth;(□) yield of TS.
Fig. 3. Effect of pH on the 17β-reduction of AD by Zygowilliopsis sp. WY7905 (□)
Conversion of AD; (●) yield of TS.
Fig. 4. Effect of reaction temperature on the 17β-reduction of AD by Zygowivlliopsis
sp. WY7905 (□) Conversion of AD; (●) yield of TS.
Fig. 5. Effect of glucose addition on the 17β-reduction of AD by Zygowivlliopsis sp.
WY7905 (□) Conversion of AD; (■) yield of TS.
Fig. 6. Effect of different co-solvents and on Tween 80 the 17β-reduction of AD by
Zygowivlliopsis sp. WY7905 (□) Conversion of AD; (■) yield of TS.
Fig. 7. Effect of various Tween 80 concentrations on the 17β-reduction of AD by
Zygowivlliopsis sp. WY7905 (□) Conversion of AD; (■) yield of TS.
Fig. 8. Effect of substrate concentration on the 17β-reduction of AD by
Zygowivlliopsis sp. WY7905 (□) Conversion of AD; (●) yield of TS.
Fig. 9. Time course of the 17β-reduction of AD by Zygowivlliopsis sp. WY7905 (□)
Conversion of AD; (●) yield of TS.
Fig. 10. Biotransformation of AD by the growth cells of Zygowivlliopsis sp. WY7905
2
5

Fig. 1
Fig. 2
2
6

Fig. 3
Fig. 4.
2
7

Fig. 5.
Fig. 6
2
8

Fig. 7.
Fig. 8.
2
9