Efficient Synthesis of 12-Oxochenodeoxycholic Acid Using a 12α-Hydroxysteroid Dehydrogenase from Rhodococcus ruber

Article abstract of DOI:10.1002/adsc.201900849

12α-Hydroxysteroid dehydrogenase (12α-HSDH) has the potential to convert cheap and readily available cholic acid (CA) to 12-oxochenodeoxycholic acid (12-oxo-CDCA), a key precursor for chemoenzymatic synthesis of the therapeutic bile acid ursodeoxycholic acid (UDCA). In this work, a native nicotinamide adenine dinucleotide (NAD⁺)-dependent 12α-hydroxysteroid dehydrogenase (Rr12α-HSDH) from Rhodococcus ruber was identified using a structure-guided genome mining (SSGM) approach, which is based on the structure of cofactor pocket and the conserved nicotinamide cofactor binding motif alignment. Rr12α-HSDH was heterologously overexpressed in Escherichia coli BL21 (DE3), purified and characterized. The purified Rr12α-HSDH showed a high oxidative activity of 290 U mg^?1_protein toward CA, with a catalytic efficiency (k_cat/K_M) of 5.10×10³ mM^?1 s^?1. In a preparative biotransformation (100 mL), CA (200 mM, 80 g L^?1) was efficiently converted to 12-oxo-CDCA in 1 h, with a 85% isolated yield and a space-time yield (STY) of up to 1632 g L^?1 d^?1. Furthermore, Rr12α-HSDH was shown to be able to catalyze the oxidation of other 12α-hydroxysteroids at high substrate loads (up to 200 mM), giving the corresponding 12-oxo-hydroxysteroids in 71%–85% yields, indicating the great potential of Rr12α-HSDH as a promising biocatalyst for the synthesis of various therapeutic bile acids. (Figure presented.).

Full text of DOI:10.1002/adsc.201900849

Title: Efficient Synthesis of 12-Oxochenodeoxycholic Acid Using a 12α-
Hydroxysteroid Dehydrogenase from Rhodococcus ruber
Authors: Shou-Cheng Shi, Zhi-Neng You, Ke Zhou, Qi Chen, Jiang
Pan, Xiao-Long Qian, Jian-He Xu, and Chun-Xiu Li
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201900849
Link to VoR: http://dx.doi.org/10.1002/adsc.201900849

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
VERY IMPORTANT PUBLICATION
COMMUNICATION
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Efficient Synthesis of 12-Oxochenodeoxycholic Acid Using a
12α-Hydroxysteroid Dehydrogenase from Rhodococcus ruber
Shou-Cheng Shi,a, # Zhi-Neng You,a, # Ke Zhou,a Qi Chen,a, b Jiang Pan,a, b Xiao-Long
Qian,c Jian-He Xu,a, b, * and Chun-Xiu Li a, b,
*
a
Laboratory of Biocatalysis and Synthetic Biotechnology, State Key Laboratory of Bioreactor Engineering, East China
University of Science and Technology, 130 Meilong Road, Shanghai 200237, P. R. China
Fax: (+86)-21-6425 0840; e-mail: jianhexu@ecust.edu.cn or chunxiuli@ecust.edu.cn
Shanghai Collaborative Innovation Centre for Biomanufacturing, School of Biotechnology, East China University of
Science and Technology, 130 Meilong Road, Shanghai 200237, P. R. China
Suzhou Bioforany EnzyTech Co. Ltd., No. 8 Yanjiuyuan Road, Economic Development Zone, Changshu, Jiangsu
215512, P. R. China
b
c
#
These authors contributed equally to this work.
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Abstract: 12α-Hydroxysteroid dehydrogenase (12α- which is a traditional Chinese medicine used for
thousands of years.[1] Clinically, 1e is an effective
HSDH) has the potential to convert cheap and readily
available cholic acid (CA) to 12-oxochenodeoxycholic acid
drug for the treatment of cholesterol gallstones and
cholestatic diseases.[2] Traditionally, 1e is either
obtained from bear bile powder or produced
chemically from cholic acid (CA, 1a, Scheme 1),
(12-oxo-CDCA), a key precursor for chemoenzymatic
synthesis of the therapeutic bile acid ursodeoxycholic acid
(UDCA). In this work, a native nicotinamide adenine
dinucleotide
(NAD+)-dependent
12α-hydroxysteroid
which is the most abundant and cheapest bile acid.[1,3]
However, on the one hand, it is inhuman and
unsustainable to extract 1e from the bile of live bears.
On the other hand, chemical synthesis requires
multiple protection and deprotection steps, leading to
a low/moderate overall yield (about 30%–53%), and
the inevitable use of toxic and hazardous reagents
(e.g., pyridine and CrO3) generates copious amounts
of waste.[4] Therefore, it is much more sustainable
and economical enzymatic synthesis of 1e from 1a,
given the availability of 1a as a cheap raw material
and the eco-friendly nature of the enzymatic reaction
conditions.[3]
dehydrogenase (Rr12α-HSDH) from Rhodococcus ruber
was identified using a structure-guided genome mining
(SSGM) approach, which is based on the structure of
cofactor pocket and the conserved nicotinamide cofactor
binding motif alignment. Rr12α-HSDH was heterologously
overexpressed in Escherichia coli BL21 (DE3), purified
and characterized. The purified Rr12α-HSDH showed a
high oxidative activity of 290 U mg−1protein toward CA, with
a catalytic efficiency (kcat/KM) of 5.10 × 103 mM–1 s–1. In a
preparative biotransformation (100 mL), CA (200 mM, 80
g L−1) was efficiently converted to 12-oxo-CDCA in 1 h,
with a 85% isolated yield and a space-time yield (STY) of
up to 1632 g L−1 d−1. Furthermore, Rr12α-HSDH was
shown to be able to catalyze the oxidation of other 12α-
hydroxysteroids at high substrate loads (up to 200 mM),
giving the corresponding 12-oxo-hydroxysteroids in
71%−85% yields, indicating the great potential of Rr12α-
HSDH as a promising biocatalyst for the synthesis of
various therapeutic bile acids.
The enzymatic synthesis of 1e requires the use of a
very important class of enzymes, hydroxysteroid
dehydrogenases (HSDHs), which can perform
oxidoreduction of the hydroxyl/carbonyl groups on
the core structure of hydroxysteroids.[5] These
enzymes show very high regioselectivity toward the
hydroxyl groups at different positions (e.g., C3, C7,
and C12) of the steroid nucleus (Scheme 1).[3]
Keywords: biocatalysis; 12α-hydroxysteroid
dehydrogenase; NAD+-dependence; structure-guided
genome mining; 12-oxochenodeoxycholic acid
Furthermore,
HSDHs
display
excellent
stereoselectivity for each of these positions by
oxidizing hydroxyl groups oriented either below (α-
OH) or above (β-OH) the plane of the steroid
molecule.[6] To date, many HSDHs with known
protein sequences have been reported. These include
Ursodeoxycholic acid (UDCA, 1e, Scheme 1) is an
active pharmaceutical ingredient of bear bile powder,
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
3α-HSDHs from rat liver, Comamonas testosteroni,
and Eggerthella lenta;[7] 7α-HSDHs from Escherichia
coli HB101, Eubacterium sp. VPI 12708, Clostridium
sordelli, Bacteroides fragilis, Clostridium absonum
(formerly Clostridium sardiniense), Clostridium
(NADP+)-dependent 12α-HSDH with published
sequence was identified from Clostridium sp. strain
ATCC 29733/VPI C48-50 and showed low catalytic
activity.[7c,12] Ridlon et al. discovered several new
12α-HSDHs from Eggerthella sp. CAG:298,
Clostridium hylemonae, Clostridium scindens, and
Clostridium hiranonis, respectively.[7c,10]
difficile, Stenotrophomonas
maltophilia, and
Brevundimonas sp.;[8] and 7β-HSDHs from
Collinsella aerofaciens,[9a] Clostridium absonum,[8e]
Ruminococcus gnavus,[9b] Ruminococcus torques,[9c]
Lactobacillus spicheri,[8g] and Clostridium sp.
Marseille.[8h] However, 12α-HSDHs with known gene
sequences are rarely reported.[10]
Nevertheless, all of these documented 12α-
HSDHs
are
stringently
NADP+-dependent
oxidoreductases (Table 1, entries 3−7). When
compared with NADP+, nicotinamide adenine
dinucleotide (NAD+) is much less costly and more
stable under industrial conditions, making it the
preferred cofactor for practical application of
dehydrogenases.[13] Most recently, a new NAD+-
dependent 12α-HSDH from Eggerthella lenta
(El12α-HSDH) was reported by the group of Ulf
12α-Hydroxysteroid dehydrogenase (12α-HSDH;
EC 1.1.1.176) is a key enzyme for chemoenzymatic
synthesis of 1e from 1a.[6] 12α-HSDHs catalyze the
oxidation of the 12α-hydroxyl group of 1a, forming
12-oxochenodeoxycholic acid (12-oxo-CDCA, 1b),
which can be subjected to a Wolff-Kishner-Huang
reduction to remove the C12 carbonyl group. The 7α-
hydroxylated chenodeoxycholic acid (CDCA, 1c) is
then epimerized to 7β-hydroxylated 1e through the
stable intermediate 7-oxo-lithocholic acid (7-oxo-
LCA, 1d) using two stereo-complementary enzymes,
7α-HSDH and 7β-HSDH (Scheme 1).[6,11] Up to now,
the nicotinamide adenine dinucleotide phosphate
Hanefeld. However, El12α-HSDH showed
a
productivity of only 4 g L−1 d−1 as a biocatalyst for
the synthetic application.[14] Therefore, the
identification of novel NAD+-dependent 12α-HSDHs
with high enzymatic activity for efficient and
economical synthesis of 1e precursor will promote
sustainable and cost-efficient production of
therapeutic bile acids.
Scheme 1. 12α-HSDH mediated oxidation of 1a as the first step of chemoenzymatic synthesis of 1e.
Table 1. Specific activity of 12α-HSDHs toward 1a.[a]
Entry
Enzyme
NCBI accession no.
Sequence identity
[%]
Cofactor dependence
Specific activity
[U mg−1]
1
2
3
4
Rr12α-HSDH
El12α-HSDH
Cl12α-HSDH
Eg12α-HSDH
WP_026137560.1
WP_114518444.1
ERJ00208.1
100
NAD+
NAD+
289 ± 13
76 ± 2.1
40 ± 0.8
59 ± 0.3
43
42
NADP+
NADP+
CDD59475.1
39
5
6
7
rCHYL
rCSCI
rCHIR
EEG75500.1
EDS06338.1
EEA85268.1
42
41
38
NADP+
NADP+
NADP+
103 ± 3
84 ± 2
197 ± 4
[a]
Specific activity of 12α-HSDHs was measured in 100 mM potassium phosphate, pH 8.0, containing 1.0 mM NAD+ or
NADP+, and 2 mM substrate 1a at 30°C using purified enzyme.
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
(A)
(B)
(C)
Leu39
Val37
Gly50
Arg51
Asn52
NADH
NADPH
NADH
Asp38
Asp36
Phe40
Gly26
Ala14 Asp38
Glu53
Gly16
Figure 1. Structural models of (A) Cl12α-HSDH with NADPH, (B) El12α-HSDH with NADH, and (C) Rr12α-HSDH
with NADH.
Herein, we discovered a new NAD+-dependent
Rr12α-HSDH (NCBI accession no.
WP_114518444.1) as the template, and carried out an
alignment of the conserved cofactor-binding motifs.
Twenty eight enzyme genes with negatively charged
residues (Asp) on the N-terminal of hydroxyl-binding
loop were selected from 1000 sequences (Figure S1),
cloned into pET28a vector, and heterologously
overexpressed in Escherichia coli BL21 (DE3)
(Table S1). The activity of each enzyme (using crude
enzyme extracts as the catalysts) toward 1a was
tested using NAD+ as a cofactor. One of the tested
WP_026137560.1), which was identified from
Rhodococcus ruber using a structure-guided genome
mining (SGGM) approach based on the 3D structure
of the enzyme’s cofactor pocket and the alignment of
a conserved nicotinamide cofactor-binding motif.
Rr12α-HSDH possesses the highest specific activity
toward 1a when compared with other 12α-HSDHs
(Table 1). Up to 200 mM (80 g L−1) of 1a was
efficiently converted by Rr12α-HSDH to 1b with a
space-time yield (STY) of as high as 1632 g L−1 d−1
(Table 5, entry 7). Moreover, Rr12α-HSDH can bio-
oxidize other 12α-bile acids and their glyco and tauro
conjugates with 99% conversion, forming the
corresponding 12-oxo-hydroxysteroids (Table 4),
which are important precursors for the synthesis of
pharmaceutically relevant bile acids.
enzymes,
Rr12α-HSDH
originating
from
Rhodococcus ruber, showed a significant activity on
1a and was thus chosen for further studies. No
activity was detected when using NADP+ as a
cofactor, indicating that Rr12α-HSDH is a native
NAD+-dependent enzyme. As shown in Figure 1C, a
DVD motif was present on the hydroxyl-binding loop
of the NAD+-dependent Rr12α-HSDH. The
negatively charged Asp36 residue may form
hydrogen bonds with the 2′-hydroxyl group of the
adenosine ribose in NADH, which facilitates the tight
binding of NADH with the enzyme. However, this
position in all the NADP+-dependent 12α-HSDHs
was replaced by a phosphate-binding motif (GRN)
(Figure S6). The results demonstrated that SGGM
strategy could enable effective and target-focused
identification of the unknown but naturally occurring
NAD+-dependent dehydrogenases.
Rr12α-HSDH is shown to be a new 12α-HSDH
because it shares only 38%–43% of its identity with
the reported El12α-HSDH (Eggerthella lenta),
Cl12α-HSDH (Clostridium sp. strain ATCC
29733/VPI C48-50), Eg12α-HSDH (Eggerthella sp.
CAG: 298), rCHYL (Clostridium hylemonae DSM
15053), rCSCI (Clostridium scindens ATCC 35704),
and rCHIR (Clostridium hiranonis DSM 13275)
(Table 1). To characterize the enzymatic properties of
To discover novel and native NAD+-dependent
12α-HSDHs, a structure-guided genome mining
(SGGM) strategy was proposed. Firstly, we
structurally analyzed the phosphate- and hydroxyl-
binding pockets of Cl12α-HSDH and El12α-HSDH,
respectively. The phosphate-binding motif (GRN) in
NADP+-dependent Cl12α-HSDH (Figure 1A) was
found to be replaced by a hydroxyl-binding motif
(DLF) in the NAD+-dependent El12α-HSDH (Figure
1B). This change in the amino acid residues located
on the phosphate/hydroxyl-binding loops would be
responsible
for
NAD(H)/NADP(H)
cofactor
recognition.[8g,14,15] In general, NADP+-dependent
enzymes have larger phosphate-binding pockets with
positively charged residues (such as Arg51 in Cl12α-
HSDH, Figure 1A) that interact with the negatively
charged 2′-phosphate group of NADPH (phospho-)
adenosine ribose, while NAD+-dependent enzymes
prefer negatively charged residues (such as Asp38 in
El12α-HSDH, Figure 1B).[13c,15,16] Therefore, NAD+-
dependent enzymes could be rationally discovered by
discriminating the residues on the cofactor pocket.
Subsequently, we performed a pBLAST search of
the NCBI nonredundant database using the protein
sequence of El12α-HSDH (NCBI accession no.
Rr12α-HSDH,
the
N-terminal
His-tagged
recombinant enzyme was purified by nickel affinity
chromatography and analyzed by sodium dodecyl
sulfate polyacrylamide gel electrophoresis (Figure
S2). As shown in Table 1, the oxidation activity of
purified Rr12α-HSDH was 290 U mg−1
(under
protein
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
standard conditions), which is the highest specific
activity toward 1a for 12α-HSDHs reported to date.
Rr12α-HSDH showed higher catalytic activity at a
temperature range of 35–45°C, and exhibited the
highest activity at 40°C, retaining 70% activity at
30°C (Figure S3). The pH properties of the purified
Rr12α-HSDH in both the reductive and oxidative
reactions are shown in Figure S4. Rr12α-HSDH
displayed a maximum activity at pH 10.0 for 1a
dehydrogenation with NAD+ as a cofactor and an
optimum at pH 6.0 for 1b reduction with NADH as a
cofactor. The thermostability of Rr12α-HSDH was
also examined, giving half-lives (t1/2) of 72, 20, and
0.25 h, respectively, at 30, 40, and 50°C (Figure S5),
implying that the enzyme was quite stable at 30°C
but deactivated at higher temperatures. Hence, the
enzymatic synthesis reactions were performed at
30°C so as to avoid the negative influences of the
poor thermostability.
The turnover frequency (kcat) of Rr12α-HSDH for 1a
was 128 s−1, affording the catalytic efficiency (kcat/Km)
of 5.10 × 103 mM−1 s−1, which is 88-fold greater than
that of El12α-HSDH. In addition, compared with
El12α-HSDH, Rr12α-HSDH showed a much better
affinity to cofactor NAD+ with a Km of 0.063 mM
(Table 2, entry 3), which is 15 times lower than
El12α-HSDH (0.92 mM) (Table 2, entry 4),
indicating that Rr12α-HSDH has a greater application
prospect in enzymatic synthesis. The kinetic
parameters of Rr12α-HSDH toward other 12α-
hydroxysteroids were also measured, as summarized
in Table 2 (entries 5−11). Rr12α-HSDH exhibited
relatively high kcat values for all the 12α-
hydroxysteroids tested (78.3–141 s−1) and the highest
kcat was observed when using glycocholic acid (GCA,
6a) as substrate. The Km values of Rr12α-HSDH for
majority of the 12α-hydroxysteroids ranged from
0.025 to 0.046 mM, indicating that the enzyme has
excellent affinity to these substrates except for
deoxycholic acid (DCA, 4a, Km = 0.138 mM). The
results demonstrated that Rr12α-HSDH can be used
as a potential biocatalyst for the production of various
health-promoting bile acids.
The kinetic parameters of 12α-HSDHs toward 1a
and cofactor were determined. As shown in Table 2
(entries 1−2), the Michaelis−Menten constant (Km) of
Rr12α-HSDH for 1a was 0.025 mM, displaying a 25-
fold higher affinity than El12α-HSDH (0.64 mM).
Table 2. Kinetic parameters of recombinant Rr12α-HSDH and El12α-HSDH.[a]
Entry
Substrate
Vmax
Km
[mM]
kcat
kcat/Km
(R1 =, R2 =)
[s−1]
[mM−1 s−1]
[μmolmin1mg1]
1
1a (α-OH, H)
1a (α-OH, H)[b]
NAD+[c]
290 ± 13
78 ± 2.1
292 ± 5
65 ± 6
0.025 ± 0.005
0.640 ± 0.060
0.063 ± 0.005
0.920 ± 0.160
0.041 ± 0.003
0.046 ± 0.008
0.138 ± 0.003
0.035 ± 0.007
0.042 ± 0.008
0.038 ± 0.004
0.042 ± 0.002
128 ± 6
36.5 ± 1.0
129 ± 2
5.10 × 103
2
57.0
3
2.03 × 103
33.3
4
NAD+[d]
30.6 ± 2.6
78.3 ± 1.8
106 ± 6
5
2a (Ketone, H)
3a (β-OH, H)
4a (H, H)
178 ± 4
240 ± 13
276 ± 21
319 ± 15
320 ± 15
295 ± 8
293 ± 4
1.91 × 103
2.30 × 103
8.80 × 102
4.01 × 103
3.35 × 103
3.40 × 103
3.10 × 103
6
7
121 ± 9
8
5a (α-OH, Taurine)
6a (α-OH, Glycine)
7a (H, Taurine)
8a (H, Glycine)
140 ± 7
9
141 ± 7
10
130 ± 4
11
129 ± 2
[a]
Kinetic parameters were determined at different substrate concentrations (0–10 mM) in the presence of 1 mM NAD+
using purified Rr12α-HSDH, and [b] El12α-HSDH, respectively. [c] Kinetic parameters were determined at different NAD+
concentrations (0–1.0 mM) in the presence of 2 mM 1a using purified Rr12α-HSDH, and El12α-HSDH, respectively.
[d]
The kcat values were calculated considering a subunit size of 28.0 and 28.3 kDa for Rr12α-HSDH and El12α-HSDH,
respectively.
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
Table 3. Optimization of reaction conditions for the oxidation of 1a catalyzed by Rr12α-HSDH.[a]
Entry
1
Rr12α-HSDH
NAD+
[mM]
Time
[h]
Conv.
[%][b]
1a
[mM]
[g L−1]
50
1
0.2
0.5
>99
2
3
4
5
6
100
100
200
200
200
1
1
2
2
5
0.2
0.4
0.4
0.6
0.5
24
1
82
>99
50
12
1.5
0.5
>99
>99
[a]
Reaction conditions: 10 mL reaction mixture, containing 50–200 mM substrate 1a, 1.5-fold equivalent pyruvate, 1–5 g
L−1 each of lyophilized E. coli cells harboring Rr12α-HSDH and LdLDH genes respectively, 0.2–0.6 mM NAD+, and
[b]
potassium phosphate buffer (100 mM, pH 8.0), was incubated at 30°C and 220 rpm.
reversed-phase high-performance liquid chromatography (RP-HPLC).
Conversion was determined by
To evaluate the feasibility of utilizing Rr12α-
HSDH for the biocatalytic process, the pyruvate/LDH
system was chosen for the regeneration of cofactor
NAD+. Compared with other NAD+ regeneration
a load of 0.6 mM cofactor (Table 3, entry 5). For cost
effectiveness, taking into account the internal
cofactor content of lyophilized E. coli cells,[18] the
reaction was carried out using 5 g L−1 biocatalyst and
0.5 mM of external NAD+. Accordingly, 200 mM
substrate was transformed completely with 99%
conversion within 0.5 h (Table 3, entry 6), indicating
the great potential of Rr12α-HSDH as a biocatalyst
for practical applications. In the 10-mL scale
biotransformation, 50 mM CA could be rapidly
converted to 12-oxo-CDCA. But when the substrate
loading was gradually increased to 200 mM, both the
concentrations of biocatalyst and cofactor had to be
increased simultaneously to achieve complete
oxidation of CA, indicating that the enzymes and
cofactor might have been partially inactivated during
the reaction. Further work will be required to
elucidate the mechanisms that lead to the loss of
enzyme activity and/or degradation of the cofactor.
Finally, in order to demonstrate the practical
capability of Rr12α-HSDH for synthetic reactions,
the oxidation of various 12α-hydroxysteroids were
performed on a preparative scale (100 mL) under the
optimized conditions. Consequently, 200 mM 1a was
completely oxidized (>99% conversion) into 1b by
Rr12α-HSDH within 1 h. After extraction and
purification, 1b was isolated in 85% yield (Table 4,
entry 1), giving a STY of up to 1632 g L−1 d−1, which
represents the highest reported productivity to the
best of our knowledge (Table 5). For substrates 7-
systems
such
as
O2/NOX
and
α-
ketoglutarate/glutamate dehydrogenase system, the
lactate dehydrogenase (LdLDH) identified from
Lactobacillus delbrueckii subsp. bulgaricus LdLDH
possesses higher specific activity (331
U
mg−1protein).[11a,17a] In practical industrial applications,
the supply of O2 could be very problematic,[17b] and
α-ketoglutarate is very expensive since it is used for
medical treatments and as a chemical building
block.[17c] The pyruvate/LDH system utilized to
regenerate cofactor NAD+ is more efficient and
economical. Initial experiments were performed in a
10 mL reaction system with 50 mM 1a, 0.2 mM
NAD+, 75 mM pyruvate, and 1 g L−1 each of
lyophilized E. coli cells harboring Rr12α-HSDH and
LdLDH genes, respectively. As a result, 1a was
completely converted to 1b within 0.5 h (Table 3,
entry 1). An increase of substrate loading to 100 mM
without changing other conditions led to 82%
conversion in 24 h (Table 3, entry 2). When 0.4 mM
NAD+ was added to the reaction mixture, Rr12α-
HSDH completely oxidized 100 mM 1a within 1 h
(Table 3, entry 3). The substrate loading could be
further raised to 200 mM, but the conversion was
only 50% after 12 h (Table 3, entry 4). It was found
that 99% conversion could be achieved in 1.5 h with
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
oxo-deoxycholic acid (7-oxo-DCA, 2a), ursocholic
acid (UCA, 3a), and 4a, the corresponding products
7,12-dioxo-lithocholic acid (7,12-dioxo-LCA, 2b),
12-oxoursodeoxycholic acid (12-oxo-UDCA, 3b),
and 12-oxo-lithocholic acid (12-oxo-LCA, 4b) were
isolated after 3.5−4.5 h of reaction in 75%, 81%, and
82% yields, respectively, albeit at a slightly lower
substrate loading (150 mM) (Table 4, entries 2−4).
Their respective STYs, 240, 333, and 295 g L−1 d−1,
were obviously lower than that of 1b. This might be
attributed to the slightly lower kcat of Rr12α-HSDH
toward 2a and 3a and higher Km toward 4a than 1a
(Table 2, entries 5−7). Following the same protocol,
200 mM substrates of taurocholic acid (TCA, 5a),
glycocholic acid (GCA, 6a), taurodeoxycholic acid
(TDCA, 7a), and glycodeoxycholic acid (GDCA, 8a)
were totally converted within 1−2 h, affording the
corresponding 12-oxo-hydroxysteroids with 99%
conversion (Table 4, entries 5−8). These results
demonstrated that Rr12α-HSDH is an efficient
biocatalyst for the technically competitive and
economically viable bioproduction of 12-oxo-
hydroxysteroids, all acting as precursors for
manufactory of therapeutic bile acids.
Table 4. Oxidation of hydroxysteroids catalyzed by Rr12α-HSDH.[a]
[a] Reaction conditions: 100 mL reaction mixture, containing 150−200 mM substrate, 1.5-fold equivalent pyruvate, 5 g L−1
lyophilized E. coli cell harboring Rr12α-HSDH gene, 5 g L−1 lyophilized E. coli cell harboring LdLDH gene, 0.5 mM
[b]
NAD+, and potassium phosphate buffer (100 mM, pH 8.0), was incubated at 30°C and 220 rpm.
determined by RP-HPLC. [c] Isolated yields.
Conversion was
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
Table 5. Comparison of Rr12α-HSDH with other reported catalysts in synthesis of 1b from 1a.
Entry
Catalyst
Time
[h]
Conv.
/Yield [%]
E factor[a]
Atom
efficiency
[%][b]
STY
Ref.
1a
[mM]
[g L−1 d−1]
1
Chemical
catalysts[c]
312.5
21
47.5
55.4
48.5
68
Hofmann[4]
2
3
4
12α-HSDH[d]
12ɑ-HSDH[e]
12ɑ-HSDH[f]
El12α-HSDH[g]
100
100
12
4
>99/88
92
1.8
6.1
3.3
2.5
64.3
87.1
81.9
95.8
240
37
9
Carrea et al.[19]
Fossati et al.[20]
24
12
24
98
Giovannini
et al.[21]
Tonin et al.[14]
5
6
10
>99
4
El12α-HSDH[h]
Rr12ɑ-HSDH[i]
10
4
1
>99
4.4
75.2
81.9
24
Tonin et al.[14]
7
200
>99/85
1.08
1632
This work
[a]
The calculations did not take into account the waste generated for the preparation of the catalysts and the downstream
processes. The detailed calculations of E factor and [b] atom efficiency are listed in Supporting Information. [c] The 400 mL
[d]
reaction mixture contained 50 g 1a (312.5 mM). 12α-HSDH was extracted from Clostridium group P. 70 mL reaction
mixture contained 0.37 U/mL 12α-HSDH and 0.06 mM NADP+.
Purified 12ɑ-HSDH from Clostridium sp. 10 mL
[e]
reaction mixture contained 20 U/mL 12α-HSDH and 0.1 mM NADP+. [f] Partially purified 12ɑ-HSDH from calcoaceticus
lwoffii. 100 mL reaction mixture contained 0.028 U/mL 12α-HSDH and 0.2 mM NAD+.
Purified El12α-HSDH
[g]
(recombinant E. coli/pET28a-El12ɑ-HSDH) combined with NOX (75 mL reaction mixture contained 0.1 U/mL El12α-
HSDH and 0.5 mM NAD+), and [h] MDH (50 mL reaction mixture contained 0.1 U/mL El12α-HSDH and 0.5 mM NAD+),
[i]
respectively. Lyophilized cells of recombinant E. coli/pET28a-Rr12ɑ-HSDH. 100 mL reaction mixture contained 10
U/mL Rr12ɑ-HSDH and 0.5 mM NAD+.
The enzymatic synthesis of 1b from 1a mediated
by the combination of Rr12ɑ-HSDH and LDH
showed an atom efficiency of 81.9% (Table 5),
slightly lower than that obtained by the El12α-HSDH
combined with NOX, but still acceptable compared
with the chemical route. More importantly, as shown
in Table 5, the E factor (environmental factor) for the
production of 1b catalyzed by Rr12ɑ-HSDH was
estimated to be 1.08, in contrast to those for other
reported biocatalysts and 55.4 for the chemical route,
indicating the significant environmental benefit of
biocatalysts, especially Rr12ɑ-HSDH.
biocatalyst for highly efficient and environmentally
benign production of important scaffolds of
pharmaceutical bile acids. Further reaction
engineering for pilot-scale and large-scale
applications is ongoing in our laboratory.
Experimental Section
Structure-Guided Genome Mining of New 12α-HSDHs
To discover novel and native NAD+-dependent 12α-
HSDHs, a structure-guided genome mining (SGGM)
strategy was adopted. We performed a pBLAST search of
In summary, a new and native NAD+-dependent
Rr12α-HSDH with high catalytic efficiency was
discovered from Rhodococcus ruber by the proposed
structure-guided genome mining approach (SGGM).
Rr12α-HSDH showed higher specific activity toward
1a than other reported 12α-HSDHs. By stepwise
optimization of reaction conditions, as much as 80 g
L−1 (200 mM) of 1a could be completely oxidized
within 1 h, and efficiently converted to 1b, with an
isolated yield of 85% and a STY of up to 1632 g L−1
d−1. The biocatalytic process gave an atom efficiency
of 81.9% and an E factor of 1.08. Furthermore,
various 12α-hydroxysteroids were bio-oxidized to the
corresponding 12-oxo-hydroxysteroids by Rr12α-
HSDH at high substrate loads (up to 200 mM) and in
good yields (up to 85%), indicating that E.
coli/pET28a-Rr12α-HSDH can serve as an attractive
the
NCBI
nonredundant
database
(http://www.ncbi.nlm.nih.gov) using the protein sequence
of El12α-HSDH (NCBI accession no. WP_114518444.1)
as the template, and carried out an alignment of the
conserved cofactor-binding motifs. To ensure the diversity
of sequences, 1000 candidate sequences with 20−60%
similarities to the template sequence were selected. Based
on the fact that the residues located on the
phosphate/hydroxyl-binding loops would be responsible
for NAD(H)/NADP(H) cofactor recognition, 28 putative
12α-HSDH sequences encoding a negatively charged
amino acid residue (Asp) on the N-terminal of hydroxyl-
binding loop were chosen among the 1000 candidate genes
and used for cloning, expression, and activity detection
(Supporting Information).
Activity Assays and Kinetic Parameters
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
Enzyme activity was determined by spectrophotometric
assay at 30°C measuring the oxidation of NAD(P)H or
reduction of NAD(P)+ at 340 nm (ɛ = 6.22 mM−1 cm−1).
The reaction mixture (1 mL) was composed of 970 μL
potassium phosphate buffer (KPi, 100 mM, pH 8.0), 10 μL
of substrate stock solution (200 mM), 10 μL NAD(H) or
NADP(H) solution (100 mM), and 10 μL purified enzyme
with an appropriate concentration. One unit of the enzyme
activity is defined as the amount of enzyme that catalyzes
the reduction (or oxidation) of 1 μmol NAD(P)(H) per
minute under the assay conditions. Kinetic parameters
were determined at different substrate concentrations (0–
10 mM) and cofactor (NAD+ or NADP+) concentrations
(0–1.0 mM). The resultant data were adjusted to a
Michaelis-Menten model using Origin 8.6.
Wang, Acta Crystallogr., Sect. F: Struct. Biol. Cryst.
Commun. 2017, 73, 246–252.
[6] D. Monti, E. E. Ferrandi, I. Zanellato, L. Hua, F.
Polentini, G. Carrea, S. Riva, Adv. Synth. Catal. 2009,
351, 1303–1311.
[7] a) J. E. Pawlowski, M. Huizinga, T. M. Penning, J.
Biol. Chem. 1991, 266, 8820–8825; b) C. Grimm, E.
Maser, E. Möbus, G. Klebe, K. Reuter, R. Ficner, J.
Biol. Chem. 2000, 275, 41333–41339; c) S. M. Mythen,
S. Devendran, C. Méndez-García, I. Cann, J. M. Ridlon,
Appl. Environ. Microbiol. 2018, 84, e02475-17.
[8] a) T. Yoshimoto, H. Higashi, A. Kanatani, X. S. Lin, H.
Nagai, H. Oyama, K. Kurazono, D. Tsuru, J. Bacteriol.
1991, 173, 2173–2179; b) S. F. Baron, C. V. Franklund,
P. B. Hylemon, J. Bacteriol. 1991, 173, 4558–4569; c)
J. P. Coleman, L. L. Hudson, M. J. Adams, J. Bacteriol.
1994, 176, 4865–4874; d) M. J. Bennett, S. L.
McKnight, J. P. Coleman, Curr. Microbiol. 2003, 47,
475–484; e) E. E. Ferrandi, G. M. Bertolesi, F.
Polentini, A. Negri, S. Riva, D. Monti, Appl. Microbiol.
Biotechnol. 2012, 95, 1221–1233; f) D. Bakonyi, W.
Hummel, Enzyme Microb. Technol. 2017, 99, 16–24; g)
F. Tonin, L. G. Otten, I. W. C. E. Arends,
ChemSusChem 2019, 12, 3192–3203; h) X. Chen, Y.
Cui, J. Feng, Y. Wang, X. Liu, Q. Wu, D. Zhu, Y. Ma,
Adv. Synth. Catal. 2019, 361, 2497–2504.
Preparative Synthesis of 12-Oxo-hydroxysteroids
The biotransformation was conducted at 30°C and 220 rpm
for 24 h in a 100 mL solution containing 150–200 mM
12α-hydroxysteroids, 1.5-fold equivalent pyruvate, 5 g L-1
each of lyophilized E. coli cells harboring Rr12α-HSDH
and LdLDH genes respectively, 0.5 mM NAD+, and
potassium phosphate buffer (KPi, 100 mM, pH 8.0). The
conversion of substrates was determined by RP-HPLC
(Supporting Information). After complete conversion, the
solution was acidified with 1 M HCl till pH 3.0, and the
precipitated product was filtered and washed with 50−100
mL of ethanol. Subsequently, the ethanol is evaporated,
and the powder was dried, and purified by silica gel
column chromatography eluting with a mixture of
dichloromethane and methanol (20/1–10/1) to yield pure
12-oxo-hydroxysteroids. The product was characterized by
NMR spectroscopy and high-resolution mass spectrometry
(HRMS) (Supporting Information).
[9] a) L. Liu, A. Aigner, R. D. Schmid, Appl. Microbiol.
Biotechnol. 2011, 90, 127–135; b) J. Y. Lee, H. Arai, Y.
Nakamura, S. Fukiya, M. Wada, A. Yokota, J. Lipid
Res. 2013, 54, 3062–3069; c) M. M. Zheng, R. F.
Wang, C. X. Li, J. H. Xu, Process Biochem. 2015, 50,
598–604.
[10] H. Doden, L. A. Sallarn, S. Devendran, L. Ly, G.
Doden, S. L. Daniel, J. M. P. Alves, J. M. Ridlon, Appl.
Environ. Microbiol. 2018, 84, e00235-18.
Acknowledgements
[11] a) M. M. Zheng, K. C. Chen, R. F. Wang, H. Li, C. X.
Li, J. H. Xu, J. Agric. Food. Chem. 2017, 65, 1178–
1185; b) M. Braun, B. Sun, B. Anselment, D. Weuster-
Botz, Appl. Microbiol. Biotechnol. 2012, 95, 1457–
1468; c) D. Bakonyi, A. Wirtz, W. Hummel, Z.
Naturforsch. 2012, 67, 1037–1044; d) L. Liu, M. Braun,
G. Gebhardt, D. Weuster-Botz, R. Gross, R. D. Schmid,
Appl. Microbiol. Biotechnol. 2013, 97, 633–639; e) B.
Sun, C. Kantzow, S. Bresch, K. Castiglione, D.
Weuster‐ Botz, Biotechnol. Bioeng. 2013, 110, 68–77;
f) B. Sun, F. Hartl, K. Castiglione, D. Weuster‐ Botz,
Biotechnol. Progr. 2015, 31, 375–386.
This work was financially supported by The National Key
Research and Development Program of China (No.
2018YFC1706200), The National Natural Science Foundation of
China (Nos. 21536004, 21871085, and 31500592), and the
Fundamental Research Funds for the Central Universities (No.
22221818014). We are grateful to Dr. Yu-Cong Zheng and Dr.
Chao Shou, both from East China University of Science and
Technology, for their experimental assistance.
References
[1] Y. Feng, K. Siu, N. Wang, K. M. Ng, S. W. Tsao, T.
Nagamatsu, Y. Tong, J. Ethnobiol. Ethnomed. 2009, 5,
2.
[12] M. Braun, H. Link, L. Liu, R. D. Schmid,
Weuster‐ Botz, D. Biotechnol. Bioeng. 2011, 108,
1307–1317.
[2] T. Ikegami, Y. Matsuzaki, Hepatol. Res. 2008, 38,
123–131.
[3] a) F. Tonin, I. W. C. E. Arends, Beilstein J. Org. Chem.
2018, 14, 470–483; b) T. Eggert, D. Bakonyi, W.
Hummel, J. Biotechnol. 2014, 191, 11–21.
[4] a) A. F. Hofmann, Acta Chem. Scand. 1963, 17, 173–
186; b) P. S. Dangate, C. L. Salunke, K. G. Akamanchi,
Steroids 2011, 76, 1397–1399.
[5] a) S. Savino, E. E. Ferrandi, F. Forneris, S. Rovida, S.
Riva, D. Monti, A. Mattevi, Proteins Struct. Funct.
Bioinf. 2016, 84, 859–865. b) R. Wang, J. Wu, D. K.
Jin, Y. Chen, Z. Lv, Q. Chen, Q. Miao, X. Huo, F.
[13] a) C. You, R. Huang, X. L. Wei, Z. G. Zhu, Y. H. P.
Zhang, Synth. Syst. Biotechnol. 2017, 2, 208–218; b) P.
M. Thompson, N. J. Turner, ChemCatChem 2017, 9,
3833–3836; c) J. K. B. Cahn, C. A. Werlang, A.
Baumschlager, S. Brinkmann-Chen, S. L. Mayo, F. H.
Arnold, ACS Synth. Biol. 2017, 6, 326–333; d) A. M.
Chánique, L. P. Parra, Front. Microbiol. 2018, 9, 194.
[14] F. Tonin, N. Alvarenga, J. Z. Ye, I. W. C. E. Arends,
U. Hanefeld, Adv. Synth. Catal. 2019, 361, 2448–2455.
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
[15] Z. N. You, Q. Chen, S. C. Shi, M. M. Zheng, J. Pan,
X. L. Qian, C. X. Li, J. H. Xu, ACS Catal. 2019, 9,
466–473.
Rahman, C. C. Xu, W. Qin, Bioresour. Bioprocess.
2018, 5, 3. c) K. Y. San, G. N. Bennett, S. J. Berrıos-
́
Rivera, Vadali, R. V.; Y. T. Yang, E. Horton, F. B.
Rudolph, B. Sariyar, K. Blackwood, Metab. Eng. 2002,
4, 182–192.
[16] a) N. S. Scrutton, A. Berry, R. N. Perham, Nature
1990, 343, 38–43; b) A. Lerchner, A. Jarasch, W.
Meining, A. Schiefner, A. Skerra, Biotechnol. Bioeng.
2013, 110, 2803–2814.
[19] G. Carrea, R. Bovara, R. Longhi, S. Riva, Enzyme
Microb. Technol. 1985, 7, 597–600.
[17] a) S. Kochhar, N. Chuard, H. Hottinger, J. Biol.
Chem. 1992, 267, 20298–20301. b) F. Sha, Y. Zheng, J.
Chen, K. Chen, F. Cao, M. Yan, P. Ouyang, Green
Chem. 2018, 20, 2382–2391. c) B. Beer, A. Pick, V.
Sieber, Metab. Eng. 2017, 40, 5–13.
[20] E. Fossati, F. Polentini, G. Carrea, S. Riva,
Biotechnol. Bioeng. 2006, 93, 1216–1220.
[21] P. P. Giovannini, A. Grandini, D. Perrone, P. Pedrini,
G. Fantin, M. Fogagnolo, steroids 2008, 73, 1385–1390.
[18] a) Q. Li, B. Huang, Q. He, J. Lu, X. Li, Z. Li, H. Wu,
Q. Ye, Bioresour. Bioprocess. 2018, 5, 41. b) M. S.
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900849
Advanced Synthesis & Catalysis
COMMUNICATION
Efficient Synthesis of 12-Oxochenodeoxycholic
Acid Using a 12α-Hydroxysteroid
Dehydrogenase from Rhodococcus ruber
Adv. Synth. Catal. Year, Volume, Page – Page
Shou-Cheng Shi,+a Zhi-Neng You,+a Ke Zhou,a Qi
Chen,a, b Jiang Pan,a, b Xiao-Long Qian,c Jian-He
Xu,a, b,* and Chun-Xiu Lia, b,
*
10
This article is protected by copyright. All rights reserved.