Flavin Oxidoreductase-Mediated Regeneration of Nicotinamide Adenine Dinucleotide with Dioxygen and Catalytic Amount of Flavin Mononucleotide for One-Pot Multi-Enzymatic Preparation of Ursodeoxycholic Acid

Article abstract of DOI:10.1002/adsc.201900111

Ursodeoxycholic acid (UDCA), a pharmaceutical ingredient widely used in clinics, can be prepared from chenodeoxycholic acid (CDCA) by the epimerization of the 7α-OH group. In this study, a nicotinamide adenine dinucleotide (NAD⁺) regeneration system was developed by using flavin oxidoreductase (FR) and flavin mononucleotide (FMN). Only catalytic amount of FMN is required for the effective NAD⁺ recycling. FR/FMN system was then applied in the oxidation of CDCA to 7-ketolithocholic acid (7-keto-LCA) by NAD⁺-dependent 7α-hydroxysteroid dehydrogenase (Bs-7α-HSDH) from Brevundimonas sp., which showed extremely high enzyme activity toward CDCA (k_cat/K_m=8050 s^?1 ? mM^?1). When Escherichia coli whole cells coexpressing Bs-7α-HSDH and FR genes were used as biocatalyst, CDCA (50 mM) was completely converted to 7-keto-LCA with the turnover number of FMN being 227 and 58.8 g ? L^?1 ? d^?1 space-time yield of 7-keto-LCA. For the reduction of 7-keto-LCA, nicotinamide adenine dinucleotide phosphate (NADPH)-dependent 7-β-hydroxysteroid dehydrogenase (Cm-7β-HSDH) from Clostridium sp. Marseille was employed with alcohol dehydrogenase from Thermoanaerobacter brockii (TbADH) and iso-propanol as co-factor regeneration system. When E. coli whole cells coexpressing Cm-7β-HSDH and TbADH genes were used as biocatalyst, 40 mM 7-keto-LCA was reduced to UDCA with 26.8 g ? L^?1 ? d^?1 space-time yield. The oxidation and reduction were then carried in a one-pot concurrent mode, 12.5 mM CDCA was completely converted to UDCA. The epimerization of CDCA to UDCA proceeded to completion at the substrate concentration of 30 mM in the one-pot sequential process. Therefore, the complete conversion of CDCA to UDCA in one-pot has been realized by employing 7α-HSDH and 7β-HSDH of different co-factor specificities with independent co-factor recycling systems. The cholic acids, especially UDCA, exert inhibitive effect on the activities of these enzymes, preventing the complete epimerization of 7α-OH at higher substrate loading. This inhibition issue should be solvable by engineering the involved enzymes, that is currently pursued in our laboratory. (Figure presented.).

Full text of DOI:10.1002/adsc.201900111

Title: Flavin Oxidoreductase-Mediated Regeneration of Nicotinamide
Adenine Dinucleotide with Dioxygen and Catalytic Amount of
Flavin Mononucleotide for One-Pot Multi-Enzymatic Preparation
of Ursodeoxycholic Acid
Authors: Xi Chen, Yunfeng Cui, Jinhui Feng, Yu Wang, Xiangtao Liu,
Qiaqing Wu, Dunming Zhu, and yanhe ma
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.201900111
Link to VoR: http://dx.doi.org/10.1002/adsc.201900111

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Flavin Oxidoreductase-Mediated Regeneration of Nicotinamide
Adenine Dinucleotide with Dioxygen and Catalytic Amount of
Flavin Mononucleotide for One-Pot Multi-Enzymatic
Preparation of Ursodeoxycholic Acid
†
†
Xi Chen,a Yunfeng Cui,a Jinhui Feng,a Yu Wang,a Xiangtao Liu, a Qiaqing Wu, a
Dunming Zhua* and Yanhe Maa
a
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
E-mail: zhu_dm@tib.cas.cn.
† The 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. Ursodeoxycholic acid (UDCA), a pharmaceutical When E. coli whole cells coexpressing Cm-7-HSDH and
ingredient widely used in clinics, can be prepared from TbADH genes were used as biocatalyst, 40 mM 7-keto-
chenodeoxycholic acid (CDCA) by the epimerization of the LCA was reduced to UDCA with 26.8 g·L-1·d-1 space-time
7-OH group. In this study, a nicotinamide adenine yield. The oxidation and reduction were then carried in a
dinucleotide (NAD+)regeneration system was developed by one-pot concurrent mode, 12.5 mM CDCA was completely
using flavin oxidoreductase (FR) and flavin mononucleotide converted to UDCA. The epimerization of CDCA to UDCA
(FMN). Only catalytic amount of FMN is required for the proceeded to completion at the substrate concentration of
effective NAD+ recycling. FR/FMN system was then applied 30 mM in the one-pot sequential process. Therefore, the
in the oxidation of CDCA to 7-ketolithocholic acid (7-keto- complete conversion of CDCA to UDCA in one-pot has
LCA)
by
NAD+-dependent
7α-hydroxysteroid been realized by employing 7α-HSDH and 7-HSDH of
dehydrogenase (Bs-7α-HSDH) from Brevundimonas sp., different co-factor specificities with independent co-factor
which showed extremely high enzyme activity toward recycling systems. The cholic acids, especially UDCA,
CDCA (kcat/ Km = 8050 s-1mM-1). When Escherichia coli exert inhibitive effect on the activities of these enzymes,
whole cells coexpressing Bs-7α-HSDH and FR genes were preventing the complete epimerization of 7-OH at higher
used as biocatalyst, CDCA (50 mM) was completely substrate loading. This inhibition issue should be solvable
converted to 7-keto-LCA with the turnover number of FMN by engineering the involved enzymes, that is currently
being 227 and 58.8 g·L-1·d-1 space-time yield of 7-keto- pursued in our laboratory.
LCA. For the reduction of 7-keto-LCA, nicotinamide
adenine dinucleotide phosphate (NADPH)-dependent 7--
Keywords: NAD+ regeneration; chenodeoxycholic acid
(CDCA); one-pot reaction; ursodeoxycholic acid (UDCA);
epimerization
hydroxysteroid dehydrogenase (Cm-7-HSDH) from
Clostridium sp. Marseille was employed with alcohol
dehydrogenase from Thermoanaerobacter brockii (TbADH)
and iso-propanol as co-factor regeneration system.
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
group of CA derivatives. 7α-HSDHs have been found
Introduction
[13]
in Escherichia coli [12], Clostridium absonum
,
[14]
Pseudomonas sp. B-083
,
Acinetobacter
Ursodeoxycholic acid (UDCA) is a pharmaceutical
ingredient widely used in clinics for the treatment of
gallstone disease and liver disease associated with
cystic fibrosis [1]. Latest development in the synthesis
of UDCA has been summarized [2]. Cholic acid (CA)
from bovine bile is the most available starting
material, but the most direct synthesis of UDCA is
using chenodeoxycholic acid (CDCA) as the starting
material. CDCA is widely distributed in the bile acids
of poultry or can be prepared from cholic acids.
Conversion of CDCA to UDCA involves two
transformations: the oxidation of 7α-OH group to
carbonyl group, and the reduction of 7-carbonyl group
to 7-OH group (Scheme 1) [3]. Both chemical and
biocatalytic approaches have been explored for the 7-
OH epimerization [2]. For the biocatalytic process, 7α-
hydroxysteroid dehydrogenases (7α-HSDH) and 7-
hydroxysteroid dehydrogenases (7-HSDH) have
been used for the oxidative and reductive steps,
respectively. The epimerization of CDCA to UDCA
have been achieved by using a redox-neutral cascade
reaction with two nicotinamide adenine dinucleotide
phosphate (NADP+)-dependent dehydrogenases.
However, the conversion was limited by the intrinsic
reaction equilibrium and a mixture of CDCA, UDCA
and 7-ketolithocholic acid (7-keto-LCA) was obtained
[3d]. Therefore, sequential two-step process has been
adopted to enforce the complete epimerization of
CDCA to UDCA, in which the enzymes for the
calcoaceticus lwoffi [15], Eubacterium sp. strain VPI
[16]
12708
and Bacteroides fragilis [17]. Some of them
have been used in the preparation of 7-oxo bile acid
[3d, 18]
and proved to be a powerful biocatalyst
.
Glutamate dehydrogenase/α-ketoglutarate [17a], lactate
[3d,
18b]
dehydrogenase/pyruvate
dehydrogenase/acetone or acetaldehyde
,
alcohol
have been
[19]
developed for the in situ regeneration of cofactor
NAD(P)+ in the transformation of CDCA to 7-keto-
LCA employing 7α-HSDHs. No over-oxidation of the
product was reported because of the excellent
selectivity of 7α-HSDHs. In these systems, usually 2-
3 equivalents of α-ketoglutarate or pyruvate, or large
excess of acetone or acetaldehyde was required as the
co-substrate to drive the reaction to completion,
raising the operation cost. Laccase/Meldola’s blue
system showed its potential for the oxidation of CA,
with 0.005 eq of Meldola’s blue being added into the
reaction, but the acidic pH should be selected at the
expense of enzyme activity of 7α-HSDH because
laccase showed low enzyme activity at neutral to
alkaline pH [20]. Nicotinamide adenine dinucleotide
(NADH) oxidases (NOX) have been employed to
regenerate NAD+ with dioxygen (O2) being reduced
[21]
[22]
to hydrogen peroxide (H2O2)
or water (H2O)
,
but NOX have the same problem with
laccase/Meldola’s blue system, the relative low
enzyme activity of NOX made it becoming the rate
limiting step. Light-emitting diode (LED) light
sources promote the hydride transfer from NAD(P)H
to FMN [23], but the transmittance of light would
become a problem when the reaction is scaled up to
tons production. Flavin oxidoreductase (FR) has been
reported with high activity at a wide range of pH
toward oxidation of NADH [24] and/or NADPH [25] by
flavin mononucleotide (FMN). However, until now
FR/FMN system has not been used to regenerate the
cofactor NAD+ or NADP+. Flavin oxidoreductase
(MgFR) from Mycobacterium goodii X7B [26] showed
678 U/mg (pH 7.0) specific activity toward the
hydride transfer from NADH to FMN [24], yielding
NAD+ and the reduced FMN (FMNH2) [27]. The fast
autoxidation of FMNH2 [28] suggests that it may be not
necessary to use the stoichiometric amount of FMN.
Herein, we report an effective FR-mediated
regeneration of NAD+ by using O2 and catalytic
amount of FMN as the oxidizing agent, which is
coupled with a highly active NAD+-dependent 7α-
HSDH from Brevundimonas sp. to achieve the
efficient conversion of CDCA to 7-keto-LCA,
followed by the reduction using 7-HSDH and
[4]
oxidation step were inactivated by heat-treatment
[5]
or removed as “tea-bag” before the reduction was
initiated. In this context, we adopted an alternative
[5,6]
approach
to addressing this equilibrium issue, in
which cofactor-specific 7α-HSDH and 7-HSDH of
different cofactors were employed with suitable co-
factor recycling systems.
Scheme 1 Preparation of UDCA from CDCA through 7-
keto-LCA
For the regioselective oxidation of 7α-hydroxyl group
of CDCA, some chemical oxidizing reagents, such as
N-bromosuccinimide (NBS) [7], o-iodoxybenzoic acid
(IBX) [8], pyridinium chlorochromate (PCC)
,
[9]
NaClO [10], ceric ammonium nitrate (CAN) in
[11]
conjunction with sodium bromate (NaBrO3)
have
alcohol dehydrogenase from Thermoanaerobacter
[29]
been reported. Although chemical oxidation has been
widely studied, but there are at least one of the
problems such as relatively higher temperature, use of
organic solvent, over oxidation of product or toxic
heavy metal-based oxidizing agents. On the other
hand, 7α-HSDH catalyzes the regio- and
stereoselective oxidative reaction of 7α-hydroxyl
brockii (TbADH)
in one-pot fashion for the
preparation of UDCA (Scheme 2).
Scheme 2 One-pot multi-enzymatic preparation of UDCA
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
Results and Discussion
The influence of substrate addition mode was
investigated. Feeding the substrate in batches
shortened the reaction time from 6 to 4 h at substrate
concentration of 25 mM (Table 1, entries 7 and 8).
Without addition of external NAD+, the reaction
conversion was only 94%, and the complete
conversion of CDCA required longer reaction time of
30 h (Table 1, entry 9). When substrate loading was
increased to 50 mM and no extra NAD+ was added,
only 62% conversion of CDCA was achieved (Table
1, entry 11). At the substrate loading of 50 mM,
addition of CDCA in batches exerted remarkable
effect on the reaction rate (Table 1, entries 10 and 12).
Under this reaction condition, doubling the
concentration of co-substrate FMN shortened the
reaction time from 6 to 5 h (Table 1, entries 12 and
13). When the substrate concentration was further
increased to 75 mM (30 g/L), the reaction was
finished in 16 h. When the cell-free extracts were
used as the biocatalyst, the oxidation reaction
proceeded faster, shortening the reaction time
dramatically (Table 2, entry 14). The whole-cell
transformation was conducted in 100 mL at the
concentration of 50 mM CDCA. The reaction was
completed in 8 h, and 7-keto-LCA was obtained with
94% yield by adjusting the pH of the reaction mixture.
In this case, only 0.0044 equivalent of co-substrate
FMN was added with its turnover number being 227,
and 58.8 g·L-1·d-1 space-time yield was attained
(Table 1, entry 15).
Screening of 7α-HSDHs and 7-HSDHs
Many 7α-HSDHs have been identified as the
biocatalyst for the oxidation of CDCA to 7-keto-LCA.
For example, 7α-HSDH (Ec-7α-HSDH) from
Escherichia coli HB101 showed good catalytic
capacity toward CA or CDCA with NAD+ as cofactor
[12]. Recently, an increasing number of reports on the
bioconversion with 7β-HSDH has also been appeared
[3]. In order to search for more effective 7α/β-HSDHs,
Ec-7α-HSDH or Cae-7β-HSDH from Collinsella
aerofaciens ATCC 25986 was used as the template,
respectively, for searching new candidates and
neighbor-joining trees were constructed with MEGA
software (Fig. S1). Seven 7α-HSDH genes and four
new together with four reported 7β-HSDH genes were
selected and overexpressed in E. coli (Fig. S2).
The activity of crude 7α-HSDHs and 7β-HSDHs were
measured, and the results are shown in Table S1. Bs-
7α-HSDH from Brevundimonas sp. and Cm-7-
HSDH from Clostridium sp. Marseille showed the
highest enzyme activity, and thus selected for the
following studies. The recombinant Bs-7α-HSDH and
Cm-7-HSDH together with MgFR and TbADH were
purified (Tables S2-S5, Fig. S3-S6). The pH,
temperature dependence and kinetic parameters of the
purified Bs-7α-HSDH and Cm-7-HSDH enzymes
were determined (Table S6 and Fig. S7-S8).
Flavin oxidoreductase-catalyzed regeneration of
NAD+ for the oxidation of CDCA
Table 1. Optimization of the reaction conditions for the
FRs catalyze the reduction of free flavins (riboflavin,
FMN, or FAD) by using reduced pyridine nucleotides,
preparation of 7-keto-LCA a.
NADPH, or NADH. MgFR from Mycobacterium
Entry Substrate NAD+/FMN Conv. Reaction
[26]
goodii X7B
showed strict specificity toward
loading
(mM)
10
10
10
10
10
10
(mM)
(%)
time (h)
NADH with high enzyme activity (678 U/mg), and
thus was tested for the regeneration of NAD+ in the
Bs-7α-HSDH-catalyzed oxidation of CDCA. E. coli
(BsMg) cells co-expressing Bs-7α-HSDH and MgFR
were used as biocatalyst in the following studies (Fig.
S9).
1
2
3
4
5
6
7
8
9
0.15/10
0.15/0.88
0.15/0.44
0.15/0.22
0/0.22
> 99
> 99
> 99
> 99
> 99
16
> 99
> 99
94
0.5
0.5
1
1
6
24
6
4
The amount of FMN required for the effective cycling
of NAD+ was examined, and the results are
summarized in Table 1. The oxidation reaction was
finished in 30 min when 10 mM CDCA, 10 mM FMN
and 0.15 mM NAD+ were used (Table 1, entry 1). The
results showed that even 0.022 equivalent of FMN
was used, the oxidation reaction proceeded
successfully in 1 h (Table 1, entry 4), suggesting that
it is not necessary to add stoichiometric amount of
FMN, due to the fast auto-oxidation of reduced FMN
[28]. When the oxidation of CDCA was carried out at
75 mM substrate concentration with shaking or
stirring in an open reaction system, the dissolved O2
was enough for the oxidation step. Without addition
of extra NAD+ into the system the reaction time was
prolonged from 1 h to 6 h (Table 1, entry 5), implying
that the endogenous NAD+ of E. coli could be used as
the cofactor. Without extra addition of FMN, the
reaction proceeded with low conversion (Table 1,
entry 6).
0.15/0
25
0.15/0.22
0.15/0.22
0/0.22
25 b
25 b
24
(> 99) (30)
10
11
12
13
14
15
50
50
50
50
75
50 d
0.15/0.22
0/0.22
0.15/0.22
0.15/0.44
0.15/0.44
0.15/0.22
> 99
62
12
30
> 99
> 99
> 99
> 99
(94 e)
6 (4 c)
5
16 (10 c)
8
a
b
10 mg/mL wet cells, CDCA was added once, pH 8.0;
c
CDCA was added in batches; cell-free extracts of 10
mg/mL wet cells; d the reaction was conducted at 100 mL
volume; e isolated yield.
Compared with the previous reported cofactor
regeneration systems used for the preparation of 7-
keto-LCA (Table 2), we used recombinant E. coli
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
cells co-expressing 7-HSDH and FR genes instead Sequential and concurrent one-pot preparation of
of partial or purified enzyme as the biocatalyst, and UDCA from CDCA
only catalytic amount of FMN was required. Our
results demonstrate that the FR/FMN is an effective From the above results, it could be seen that FR
cofactor regeneration system for the alcohol effectively mediated the regeneration of NAD+ by
dehydrogenase-catalyzed reactions.
using catalytic amount of FMN and O2 as the
oxidizing agent, leading to efficient conversion of
CDCA to 7-keto-LCA by coupling with a highly
active NAD+-dependent 7α-HSDH. The reduction of
7-keto-LCA to UDCA was achieved by using 7-
HSDH and TbADH/IPA as cofactor recycling system.
Therefore, the oxidation and reduction reactions were
performed in sequential and concurrent one-pot
fashion with an aim to realize the complete
epimerization of CDCA to UDCA in a single step.
The E. coli (BsMg) and E. coli (CmTb) whole cells
were used as the biocatalysts, and results are
summarized in Table 4.
Table 2 Comparison of biotransformation conditions of 7α-
HSDH-catalyzed oxidations
Entry Substrate
loading
Co-substrate
(mM)
Catalyst
form
(mM)
1
CA
α-ketoglutarate
partial
12.5
50
purified
enzyme [30]
crude
2
3
4
CDCA
10
CDCA
50
CDCA
50
sodium pyruvate
12
Meldola’s blue
0.3
FMN
0.22
enzyme [4b]
purified
enzyme [20]
recombinant
E. coli (This
work)
Table 4. Sequential and concurrent one-pot preparation of
UDCA from CDCA a.
Entry Substrate
loading (mM)
Conv. (%)
oxidation/reduct time (h)
ion
Reaction
Optimization of the reduction reaction conditions
TbADH/i-propanol (IPA) was selected for the
regeneration of NADPH in the Cm-7-HSDH-
catalyzed reduction of 7-keto-LCA. E. coli (CmTb)
cells co-expressing Cm-7-HSDH and TbADH genes
were used as the biocatalyst for the reduction of 7-
keto-LCA (Fig. S10).
Different concentrations of cosubstrate IPA (0.1-1 M)
were examined for the reduction (Table 3, entry 1-4),
and 1 M IPA was chosen for the reaction. The
endogenous NADP+ of E. coli was sufficient for the
1
2
3
4
5
6
7
8
10
> 99/> 99
> 99/65 (70)d
> 99/85 (95)d
> 99/> 99
> 99/> 99
> 99/60
> 99
0.5/1.5
4/24 (30)
3/24 (30)
1/7
4/20
8/24
2
10 b
10 c
20
30
40
10
20
50
24
9
12.5
> 99
12
a
20 mg/mL wet cells, 0.15 mM NAD+, 0.15 mM NADP+,
complete reduction of 10 mM 7-keto-LCA, although 0.44 mM FMN, 1.0 M IPA were added, 37C, pH 7.0.
longer reaction time was needed. When the Entries 1-6, the reactions were carried out sequentially,
b
concentration of 7-keto-LCA was increased to 40 mM, entries 7-9, the reactions were performed concurrently.
the reduction was completed in 20 h. The reaction Without addition of external NAD+ and NADP+. 0.075
c
time was shortened to 14 h when 7-keto-LCA was mM NAD+ and NADP+. d The conversion of reduction step
added in batches. The reduction of 7-keto-LCA was after 30 h.
carried out at laboratory preparative scale, and UDCA
was isolated with 93% yield by adjusting the pH of
the reaction mixture (Table 3, entry 8).
For the one-pot sequential reaction, 30 mM CDCA
was completely transformed to 7-keto-LCA, which
was then completely reduced to UDCA (Table 4,
entry 5). When the concentration of CDCA was
increased to 40 mM, CDCA was completely oxidized
to 7-keto-LCA in 8 h. However, the reduction
reaction could not proceed to completion, resulting in
a 60% conversion of 7-keto-LCA to UDCA (Table 4,
entry 6). When the amount of cofactors was reduced
to 0.075 mM or no external cofactor was added, the
oxidation proceeded to completion in a longer time,
but the reduction was not completed in 30 h even at
10 mM substrate concentration (Table 4, entries 2 and
3). It seems that the reduction requires higher cofactor
concentration than the oxidation. For the one-pot
concurrent reaction with whole cells of BsMg and
CmTb being added at the same time, 12.5 mM (5 g/L)
of CDCA was completely epimerized into UDCA
(Table 4, entry 9). Addition of 20 mM CDCA into the
reaction mixture resulted in an incomplete
Table 3. Optimization of the reaction conditions for the
preparation of UDCA from 7-keto-LCA a.
Entry Substrate NADP+
Conv. (%)
Reaction
time (h)
loading
(mM)
10
10
10
10
10
20
(mM)/
IPA (M)
0.15/0.1
0.15/0.25
0.15/0.5
0.15/1
0/1
0.15/1
0.15/1
0.15/1
1
2
3
4
5
6
7
> 99
> 99
> 99
> 99
> 99
> 99
> 99
2
1.5
1
1
16
4
40
20
8
40 b
> 99 (93 c) 14
a
50 mg/mL wet cells, 7-keto-LCA was added once, 37C,
b
c
pH 7.0; 7-keto-LCA was added in batches; 50 mL
reaction was conducted, isolated yield.
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
epimerization with 50% conversion, and a mixture of for new enzymes from the Nature, and/or by
substrate CDCA, intermediate 7-keto-LCA and removing the product UDCA from the reaction
product UDCA was obtained (Table 4, entry 8). The mixture, that are currently pursued in our laboratory.
sequential and concurrent one-pot preparation of
UDCA from CDCA were performed at 50 mL scale.
The results showed that in sequential reaction mode
Experimental Section
30 mM CDCA was transformed to UDCA with 92%
isolated yield, and in concurrent reaction mode 12.5
mM CDCA was epimerized to UDCA with 78%
isolated yield.
General
All chemicals including CA, CDCA and 7-keto-LCA
were obtained from commercial sources and the
cofactors were obtained from Roche. The enzyme
activities toward the reduction of ketones were
measured using a SpectraMax M2 microplate reader
It has been reported that cholic acid and its
derivatives exert dramatic inhibition effect on the
enzyme activity of 7α-HSDH at mM concentration [12,
31]. We thus proposed that the low conversion of
CDCA to UDCA at higher concentration in the one-
pot process might be due to the inhibition effects of
the bile acids on these enzymes. As such, the
influence of all the reaction components (FMN, H2O2,
CDCA, 7-keto-LCA, UDCA, IPA, and acetone) in the
one-pot reaction mixture on the enzyme activity of
MgFR, Bs-7α-HSDH, TbADH, and Cm-7-HSDH
were investigated. FMN (2 mM), H2O2 (100 mM),
IPA (1.0 M) showed no dramatic influence on the
activity of these enzymes. CDCA, 7-keto-LCA,
UDCA exerted inhibition effects on the tested
enzymes. Especially, when 20 mM UDCA was added
into the reaction mixture, only 17%, 16%, 20%
remaining enzyme activities of MgFR, Cm-7β-HSDH,
TbADH were observed, respectively (Fig. S11).
Therefore, the inhibition of UDCA on the enzyme
activity is responsible for the incomplete
epimerization at higher substrate concentration.
Further studies are thus required to address this
inhibition issue.
1
(Molecular Devices). The H NMR spectra were
measured on Brucker Avance using CD3OD as the
solvent. Materials used for culture media including
peptone, yeast extract, and agar were purchased from
Becton, Dickinson, and Company (BDX).
Gene cloning and expression
In order to identify more potential 7α-HSDHs and 7-
HSDHs for the synthesis of UDCA from CDCA, Ec-
7α-HSDH (GenBank: P0AET8) and Cae-7β-HSDH
(GenBank: WP_006236005.1) were used as the
templates for BLASTP search in NCBI at default
settings
(http://blast.ncbi.nlm.nih.gov/Blast.cgi).
Phylogenetic tree and alignment of amino acid
sequences of 7α-HSDHs and 7-HSDHs were
presented in Fig. S1. The optimized gene sequences
of 7α-HSDHs and 7-HSDHs from the selected
strains were synthesized and cloned into NdeI and
XhoI sites of pET21a vector. The plasmids were
introduced into E. coli BL21(DE3), resulting in
recombinant strains.
The flavin oxidoreductase (MgFR) gene (GenBank:
EU154996.1) was amplified using genomic DNA
from Mycobacterium goodie X7B as template and
primers with recognition sites for the restriction
enzymes NdeI and XhoI were used for the
Conclusion
Because
HSDHs
catalyze
the
reversible construction of the enzyme with C-terminal 6×His-tag
oxidation/reduction reactions, it is inevitable that 7α- in pRSFDuet-1. The constructed vector was named
HSDH and 7-HSDH would interfere each other in pRSFDuet-MgFR. The DNA sequence were confirmed
the one-pot enzymatic epimerization of CDCA to by sequencing. The plasmid was introduced into E.
UDCA if they do not have different co-factor coli BL21 (DE3), resulting in recombinant strain
specificities. As such, Bs-7α-HSDH and Cm-7- pRSFDuet-MgFR-BL21 (DE3). The recombinant E.
HSDH, which are exclusively selective for NAD+ or coli BL21(DE3) carrying alcohol dehydrogenase
NADP+, respectively, were selected in this study. TbADH was preserved in our laboratory.
MgFR with strict specificity toward NADH was used Recombinant E. coli BL21(DE3) was propagated in
for the regeneration of NAD+, and NADPH-specific Luria-Bertani medium containing 100 μg/mL
TbADH was chosen for the recycling of NADPH. It ampicillin or 50 μg/mL kanamycin. When the OD600
should be noted that only a catalytic amount of FMN reached a value between 0.6 and 0.8, the production
is needed for an effective regeneration of NAD+, of the recombinant proteins was induced by the
avoiding the high cost of FMN mediator. Therefore, addition of isopropyl β-D-1-thiogalactopyranoside
the complete conversion of CDCA to UDCA was (IPTG) to a final concentration of 0.1 mM. The
achieved in one-pot epimerization either in sequential cultures were shaken for 20 h at 25°C and the cells
or concurrent mode. Although the inhibition of were harvested by centrifugation.
UDCA on the enzyme activity prevents the efficient Activity assay of crude enzymes
epimerization of CDCA to UDCA at high substrate Recombinant E. coli cells were resuspended in
concentration, the present study offers an approach to potassium phosphate buffer (50 mM, pH 8.0) and
enabling the epimerization proceeding to completion disrupted by ultrasonic. Cell debris was precipitated
in the one-pot setting. The inhibition issue of the CA by centrifuge at 10,000 × g for 20 min. The
derivatives on the enzyme activity should be supernatants (cell-free extracts) were used for activity
addressable by engineering the enzymes or searching assays with the corresponding substrates and the
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
expressed proteins were analyzed by SDS-PAGE. The monitored from time to time with TLC and HPLC.
enzyme activity of reduction/oxidation was assayed After the reaction was finished, pH of the reaction
by spectrophotometrically measuring the absorbance mixture was adjusted to 9.0. The mixture was
of NAD(P)H at 340 nm (ε=6.22 mM–1 cm–1). The centrifuged to remove the E. coli (BsMg) cells, and
specific activity U was defined as the number of the pH of the resulting solution was adjusted to 6.0.
micromoles of NAD(P)+ or NAD(P)H converted in 1 The precipitate was collected by centrifugation, and
min by 1 mg of enzyme (µmol∙min-1∙mg-1).
washed with 0.01 N HCl. The product was dried and
For 7α-HSDH, the assay mixture (0.2 mL) contained obtained with 94% yield (1.9 g), characterized by
2.5 mM CDCA, 0.5 mM NAD+ and appropriate NMR analysis. NMR data were consistent with
amount of crude enzyme in 50 mM potassium reference
phosphate buffer (pH 8.0).
[8]
1H NMR (CD3OD, 600 MHz) δ (ppm)
0.70 (s, 3H), 0.95 (d, J = 3.0 Hz, 3H), 1.22 (s, 3H),
For MgFR, the assay mixture (0.2 mL) contained 0.3 1.01-1.26 (m, 4H), 1.27-1.37 (m, 2H), 1.38-1.68 (m,
mM FMN, 0.5 mM NADH and appropriate amount of 6H), 1.77-1.88 (m, 4H), 1.88-1.97 (m, 2H), 2.04 (m,
enzyme in potassium phosphate buffer (50 mM, pH 1H), 2.18 (m, 2H), 2.32 (m, 1H), 2.53 (t, J = 9.6 Hz,
8.0).
1H), 2.98 (q, J = 6.0, 12.0 Hz, 1H), 3.51 (m, 1H). 13 C
For 7-HSDH, the assay mixture (0.2 mL) contained NMR (CD3OD, 125 MHz) δ (ppm) 215.1, 178.2, 71.9,
2.5 mM 7-keto-LCA, 0.5 mM NADPH and 56.4, 50.7, 50.5, 47.6, 46.5, 44.5, 43.9, 40.4, 38.3,
appropriate amount of crude enzyme in 50 mM 36.6, 36.4, 35.3, 32.4, 32.1, 30.7, 29.4, 25.9, 23.6,
potassium phosphate buffer (pH 8.0).
22.9, 18.9, 12.6.
For TbADH, the assay mixture (0.2 mL) contained Optimization for the reduction of 7-keto-LCA
100 mM IPA, 0.5 mM NADP+ and appropriate Into 5 mL potassium phosphate buffer containing E.
amount of crude enzyme in 50 mM potassium coli (CmTb) cells (about 250 mg), 0.1 g/L (0.15 mM)
phosphate buffer (pH 8.0).
NADP+, different concentration of cosubstrate IPA,
The influence of the reaction components on the and different concentrations of substrate CDCA were
enzyme activity of Bs-7α-HSDH, MgFR, Cm-7- added in one-time or in several batches (Table 3). The
HSDH and TbADH
reaction
were
monitored
by
thin-layer
For the influence of the reaction components, chromatography (TLC) or HPLC.
different final concentrations (0-20 mM) of CDCA, 7- Preparation of UDCA from 7-keto-LCA
keto-LCA, UDCA, (0-2 mM) FMN, (0-100 mM) Into 50 mL potassium phosphate buffer (pH 7.0)
H2O2, (0-1.2 M) IPA, and (0-100 mM) acetone were containing cell-free extract of wet strain (BsMg, 0.5
respectively added into the assay mixture of Bs-7α- g), 0.1 g/L NADP+, 6% v/v IPA, 12 g/L (30 mM) 7-
HSDH, MgFR, Cm-7-HSDH and TbADH, the keto-LCA was added in 4 h, the reaction was
activity of was determined by the assay procedure performed at 37C, and monitored from time to time
mentioned above.
Construction of co-expression vector and UDCA was purified as mentioned above with 93%
cultivation of strains yield (0.55 g), and characterized by NMR analysis.
with TLC and HPLC. After the reaction was finished,
1
pACYCDuet™-1 vector carrying chloramphenicol NMR data were consistent with reference [3d]. H
resistance gene (CmR) was designed for the co- NMR (CD3OD, 600 MHz) δ (ppm) 0.62 (s, 3H), 0.84-
expression of two target ORFs . The MgFR and Bs- 0.89 (m, 6H), 0.90-1.25 (m, 8H), 1.27-1.52 (m, 10H),
7α-HSDH gene was placed in the first and second 1.61-1.93 (m, 6H), 2.01-2.26 (m, 2H), 3.85 (m, 1H),
MCS, respectively. The TbADH and Cm-7-HSDH 4.42 (m, 1H). 13 C NMR (CD3OD, 100 MHz) δ (ppm)
genes were placed in the first and second MCS of 178.2, 72.7, 72.0, 57.6, 56.7, 44.9, 44.6, 44.1, 40.8,
pRSFDuet™-1 vector (kanaR), respectively. Thus 38.7, 37.1, 36.8, 36.2, 35.3, 32.5, 32.1, 31.1, 29.7,
two binary expression vectors pACYCDuet-MgFR- 28.0, 24.1, 22.5, 19.1, 12.8.
Bs-7α-HSDH (BsMg) and pRSFduet-TbADH-Cm-
7-HSDH (CmTb) were obtained. The plasmids were One-pot sequential preparation of UDCA from
introduced into E. coli BL21 (DE3). The cultivation CDCA
of E. coli BL21 was conducted as mentioned above.
Into 5 mL potassium phosphate buffer (pH 7.0)
Optimization for the oxidation of CDCA
containing E. coli (BsMg) cells (about 100 mg), 0.15
Into 5 mL broth (pH 8.0) containing E. coli (BsMg) mM NAD+, 0.44 mM FMN, and different
cells (about 50 mg), 0.1 g/L (0.15 mM) NAD+, concentrations of substrate CDCA were added, the
different concentration of cosubstrate FMN, and reaction were performed at 37C, and monitored by
different concentrations of substrate CDCA were TLC. After CDCA were completely transformed to 7-
added in one-time or in several batches. The reactions keto-LCA, E. coli (CmTb) cells (about 100 mg), 1.0
were monitored by thin-layer chromatography (TLC) M IPA, 0.15 mM NADP+ were added to the reaction,
or HPLC at 206 nm using methanol/water (75/25) as and monitored by TLC and HPLC.
mobile phase with Agela® ODS-4 column (250 mm × Into 50 mL potassium phosphate buffer (pH 7.0)
4.6 mm ID, 5 μm).
Preparation of 7-keto-LCA
containing E. coli (BsMg) cells (about 1.0 g), 0.15
mM NAD+, 0.44 mM FMN were added, 30 mM
Into 100 mL broth (pH 8.0) containing E. coli (BsMg) CDCA was added in three batches, after CDCA were
cells, 0.1 g/L (0.22 mM) FMN, 20 g/L (50 mM) completely transformed to 7-keto-LCA (4 h), E. coli
CDCA was added in 6 h, the reaction conversion was (CmTb) cells (about 1.0 g), 1.0 M IPA, 0.15 mM
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
NADP+ were added to the reaction, and monitored by
TLC. After 20 h, the reaction was finished. UDCA
was purified as mentioned above, with 92% yield
(0.54 g).
One-pot concurrent reaction of preparation of
UDCA from CDCA
Into 5 mL phosphate buffer (pH 7.0) containing E.
coli (BsMg) cells (about 100 mg), E. coli (CmTb)
cells (about 100 mg), 0.15 mM NAD+, 0.15 mM
NADP+, 0.44 mM FMN, 1.0 M IPA and different
concentrations of substrate CDCA were added, the
reaction were performed at 37C, and monitored by
TLC and HPLC.
Into 50 mL phosphate buffer (pH 7.0) containing E.
coli (BsMg) cells (about 1.0 g), E. coli (CmTb) cells
(about 1.0 g), 0.15 mM NAD+, 0.15 mM NADP+,
0.44 mM FMN, 1.0 M IPA and 12.5 mM CDCA were
added, the reaction were performed at 37C, and
monitored by TLC and HPLC. After 20 h, CDCA was
completely transformed to UDCA. UDCA was
purified as mentioned above in 78% yield (0.19 g).
Otten, I. W. C. E. Arends, ChemSusChem 2018, 11, doi:
10.1002/cssc.201801862.
[4] 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) 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.
[5] D. Monti, E. E. Ferrandi, I. Zanellato, L. Hua, F.
Polentini, G. Carrea, S. Riva, Adv. Synth. Catal. 2009,
351, 1303-1311.
[6] a) C. V. Voss, C. C. Gruber, W. Kroutil, Angew. Chem.
Int. Ed. 2008, 47, 741-745; b) C. V. Voss, C. C. Gruber,
K. Faber, T. Knaus, P. Macheroux, W. Kroutil, J. Am.
Chem. Soc. 2008, 130, 13969-13972.
[7] a) L. F. Fieser, S. Rajagopalan, J. Am. Chem. Soc. 1949,
71, 3935-3938; b) K. Königsgerger, H.-P. Chen, J.
Vivelo, G. Lee, J. Fitt, J. McKenna, T. Jenson, K.
Prasad, O. Repič, Org. Process Res. Dev. 2002, 6, 665-
669; c) A. Iuliano, A. Ruffini, Tetrahedron: Asymmetry
2005, 16, 3820-3828; d) Z. Jiang, Q. Dou, Synthesis
2015, 48, 588-594.
[8] P. S. Dangate, C. L. Salunke, K. G. Akamanchi,
Acknowledgements
Steroids 2011, 76, 1397-1399.
This work was financially supported by the Science and
Technology Service Network Initiative (No. KFJ-STS-ZDTP-65)
and Key Research Program (No. KFZD-SW-212) of Chinese
Academy of Sciences.
[9] D. Yu, D. L. Mattern, B. M. Forman, Steroids 2012, 77,
1335-1338.
[10] C. D'Amore, F. S. Di Leva, V. Sepe, B. Renga, C. Del
Gaudio, M. V. D'Auria, A. Zampella, S. Fiorucci, V.
Limongelli, J. Med. Chem. 2014, 57, 937-954.
References
[11] Y. T. Han, H. Yun, Org. Prep. Proced. Int. 2016, 48,
55-61.
[1] a) M. Podda, M. Zuin, P. M. Battezzati, C. Ghezzi, C.
De Fazio, M. L. Dioguardi, Gastroenterology 1989, 96,
222-229; b) B. Combes, R. L. Carithers, W. C.
Maddrey, D. Lin, M. F. McDonald, D. E. Wheeler, E. H.
Eigenbrodt, S. J. Muñoz, R. Rubin, G. Garcia-Tsao, G.
F. Bonner, A. B. West, J. L. Boyer, V. A. Luketic, M. L.
Shiffman, A. S. Mills, M. G. Peters, H. M. White, R. K.
Zetterman, S. S. Rossi, A. F. Hofmann, R. S. Markin,
Hepatology 1995, 22, 759-766; c) C. Colombo, P. M.
[12] I. A. Macdinald, C. N. Williams, D. E. Mahony,
Biochim Biophys Acta 1973, 309, 243-253.
[13] J. D. Sutherland, I. A. Macdonald, T. P. Forrest, Prep.
Biochem. 1982, 12, 307-321.
[14] S. Ueda, M. Oda, S. Imamura, M. Ohnishi, J. Biol.
Macromol. 2004, 4, 33-38.
Battezzati, M. Podda, N. Bettinardi, A. Giunta, [15] P. P. Giovannini, A. Grandini, D. Perrone, P. Pedrini,
Hepatology 1996, 23, 1484-1490. G. Fantin, M. Fogagnolo, Steroids 2008, 73, 1385-1390.
[2] a) F. Tonin, I. W. C. E. Arends, Beilstein J. Org. Chem. [16] S. F. Baron, C. V. Franklund, P. B. Hylemon, J.
2018, 14, 470-483; b) T. Eggert, D. Bakonyi, W.
Hummel, J. Biotechnol. 2014, 191, 11-21.
Bacteriol. 1991, 173, 4558-4569.
[17] P. B. Hylemon, E. J. Stellwag, Biochem. Bioph. Res.
Commun. 1976, 69, 1088-1094.
[3] a) L. Liu, A. Aigner, R. D. Schmid, Appl. Microbiol.
Biotechnol. 2011, 90, 127-135; b) E. E. Ferrandi, G. M.
Bertolesi, F. Polentini, A. Negri, S. Riva, D. Monti,
Appl. Microbiol. Biotechnol. 2012, 95, 1221-1233; c)
J.-Y. Lee, H. Arai, Y. Nakamura, S. Fukiya, M. Wada,
A. Yokota, J. Lipid Res. 2013, 54, 3062-3069; d) M.-M.
Zheng, R.-F. Wang, C.-X. Li, J.-H. Xu, Process
Biochem. 2015, 50, 598-604; e) J. Shi, J. Wang, L. Yu,
L. Yang, S. Zhao, Z. Wang, J. Ind. Microbiol.
Biotechnol. 2017, 4, 1073-1082; f) M.-M. Zheng, F.-F.
Chen, H. Li, C.-X. Li, , J.-H. Xu, ChemBioChem 2018,
19, 347-353; g) Q. Ji, B. Wang, C. Li, J. Hao, W. Feng,
RSC Adv. 2018, 8, 34192-34201; h) F. Tonin, L. G.
[18] a) S. Riva, R. Bovara, P. Pasta, G. Carrea, J. Org.
Chem. 1986, 51, 2902-2906; b) O. Ertl, N. Stauning, M.
Sut, B. Mayer, WO2013117584A1 2013.
[19] A. Gupta, A. Tschentscher, M. Bobkova,
EP2004838B1 2004.
[20] E. E. Ferrandi, D. Monti, I. Patel, R. Kittl, D. Haltrich,
S. Riva, R. Ludwig, Adv. Synth. Catal. 2012, 354,
2821-2828.
[21] a) H.-J. Park, C. O. A. Reiser, S. Kondruwei, H.
Erdmann, R. D. Schmid, M. Sprinzl, Eur. J. Biochem.
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
1992, 205, 881-885; b) R. Jiang, A. S. Bommarius,
Tetrahedron: Asymmetry 2004, 15, 2939-2944.
[22] a) B. R. Riebel, P. R. Gibbs, W. B. Wellborn, A. S.
Bommarius, Adv. Synth. Catal. 2002, 344, 1156-1168;
b) B. R. Riebel, P. R. Gibbs, W. B. Wellborn, A. S.
Bommarius, Adv. Synth. Catal. 2003, 345, 707-712.
[23] M. Rauch, S. Schmidt, I. W. C. E. Arends, K. Oppelt,
S. Karac, F. Hollmann, Green Chem. 2017, 19, 376-379.
[24] Q. Li, J. Feng, C. Gao, F. Li, C. Yu, L. Meng, Z.
Zhang, C. Ma, L. Gu, G. Wu, P. Xu, Process Biochem.
2012, 47, 1144-1149.
[25] F. Fieschi, V. Nivière, C. Frier, J.-L. Décout, M.
Fontecave, J. Biol. Chem. 1995, 270, 30392-30400.
[26] a) F. Li, P. Xu, J. Feng, L. Meng, Y. Zheng, L. Luo, C.
Ma, Appl. Environ. Microbiol. 2005, 71, 276-281; b) B.
Yu, F. Tao, F. Li, J. Hou, H. Tang, C. Ma, P. Xu, J.
Biotechnol. 2015, 212, 56-57.
[27] J. Li, J. Feng, Q. Li, C. Ma, B. Yu, C. Gao, G. Wu, P.
Xu, Bioresource Technol. 2009, 100, 2594-2599.
[28] Q. H. Gibson, H. J. W., Biochem. J. 1962, 83, 368-377.
[29] M. Peretz, O. Bogin, S. Tel-Or, A. Cohen, G. Li, J.-S.
Chen, Y. Burstein, Anaerobe 1997, 3, 259-270.
[30] R. Bovara, G. Carrea, S. Riva, F. Secundo, Biotechnol.
Lett. 1996, 18, 305-308.
[31] J. A. Sherrod, P. B. Hylemon, Biochim Biophys Acta
1977, 486, 351-358.
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201900111
Advanced Synthesis & Catalysis
FULL PAPER
Flavin Oxidoreductase-Mediated Regeneration
of Nicotinamide Adenine Dinucleotide with
Dioxygen and Catalytic Amount of Flavin
Mononucleotide for One-Pot Multi-Enzymatic
Preparation of Ursodeoxycholic Acid
Adv. Synth. Catal. Year, Volume, Page – Page
Xi Chen, Yunfeng Cui, Jinhui Feng, Yu Wang,
Xiangtao Liu, Qiaqing Wu, Dunming Zhu* and
Yanhe Ma
9
This article is protected by copyright. All rights reserved.