One-Pot Synthesis of Ginsenoside Rh2 and Bioactive Unnatural Ginsenoside by Coupling Promiscuous Glycosyltransferase from Bacillus subtilis 168 to Sucrose Synthase

Article abstract of DOI:10.1021/acs.jafc.8b00597

Ginsenosides, the major effective ingredients of Panax ginseng, exhibit various biological properties. UDP-glycosyltransferase (UGT)-mediated glycosylation is the last biosynthetic step of ginsenosides and contributes to their immense structural and functional diversity. In this study, UGT Bs-YjiC from Bacillus subtilis 168 was demonstrated to transfer a glucosyl moiety to the free C3-OH and C12-OH of protopanaxadiol (PPD) and PPD-type ginsenosides to synthesize natural and unnatural ginsenosides. In vitro assays showed that unnatural ginsenoside F12 (3-O-β-d-glucopyranosyl-12-O-β-d-glucopyranosyl-20(S)-protopanaxadiol) exhibited remarkable activity against diverse human cancer cell lines. A one-pot reaction by coupling Bs-YjiC to sucrose synthase (SuSy) was performed to regenerate UDP-glucose from sucrose and UDP. With PPD as the aglycon, an unprecedented high yield of ginsenosides F12 (3.98 g L^-1) and Rh2 (0.20 g L^-1) was obtained by optimizing the conversion conditions. This study provides an efficient approach for the biosynthesis of ginsenosides using a UGT-SuSy cascade reaction.

Full text of DOI:10.1021/acs.jafc.8b00597

Article
pubs.acs.org/JAFC
Cite This: J. Agric. Food Chem. XXXX, XXX, XXX−XXX
One-Pot Synthesis of Ginsenoside Rh2 and Bioactive Unnatural
Ginsenoside by Coupling Promiscuous Glycosyltransferase from
Bacillus subtilis 168 to Sucrose Synthase
Longhai Dai,†,⊥ Can Liu,‡,⊥ Jiao Li,†,# Caixia Dong,§ Jiangang Yang,† Zhubo Dai,† Xueli Zhang,*,†
and Yuanxia Sun*,†
†National Engineering Laboratory for Industrial Enzymes, Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences,
Tianjin 300308, China
‡Key Laboratory of Urban Agriculture (North) of Ministry of Agriculture, Beijing University of Agriculture, Beijing, China
#University of Chinese Academy of Sciences, Beijing 100049, China
§Tianjin Key Laboratory on Technologies Enabling Development of Clinical Therapeutics and Diagnosis, School of Pharmacy,
Tianjin Medical University, Tianjin 300070, China
S
* Supporting Information
ABSTRACT: Ginsenosides, the major effective ingredients of Panax ginseng, exhibit various biological properties. UDP-
glycosyltransferase (UGT)-mediated glycosylation is the last biosynthetic step of ginsenosides and contributes to their immense
structural and functional diversity. In this study, UGT Bs-YjiC from Bacillus subtilis 168 was demonstrated to transfer a glucosyl
moiety to the free C3-OH and C12-OH of protopanaxadiol (PPD) and PPD-type ginsenosides to synthesize natural and
unnatural ginsenosides. In vitro assays showed that unnatural ginsenoside F12 (3-O-β-D-glucopyranosyl-12-O-β-D-
glucopyranosyl-20(S)-protopanaxadiol) exhibited remarkable activity against diverse human cancer cell lines. A one-pot reaction
by coupling Bs-YjiC to sucrose synthase (SuSy) was performed to regenerate UDP-glucose from sucrose and UDP. With PPD as
the aglycon, an unprecedented high yield of ginsenosides F12 (3.98 g L−1) and Rh2 (0.20 g L−1) was obtained by optimizing the
conversion conditions. This study provides an efficient approach for the biosynthesis of ginsenosides using a UGT-SuSy cascade
reaction.
KEYWORDS: microbial UDP-glycosyltransferase, glycosylation, protopanaxadiol, unnatural ginsenosides, sucrose synthase,
UDP-glucose recycle
Rh2, F2, Rg3, Rd, and 3-O-β-D-glucopyranosyl-12-O-β-D-
glucopyranosyl-20(S)-protopanaxadiol, have been biosynthe-
sized through the introduction of an engineered PPD-
producing pathway and ginseng or microbial UGTs into yeast
cell factories;10,14−17 however, the titers of PPD-type ginseno-
sides (ca. 9−300 mg L−1) produced by yeast cell factories are
still considerably low due to the poor catalytic ability of UGTs,
the synthetic ability of yeast cells, and the inhibitory effects of
the synthesized ginsenosides.
The chemical approach for glycosylation of PPD to
synthesize PPD-type ginsenosides is fairly complicated by
disadvantages, such as poor stereopecificity, side reactions, low
efficiency, and environmental pollution.18 Although these
disadvantages can be alleviated by in vitro enzymatic
glycosylation using effective UGTs, UGT-catalyzed glycosyla-
tion of PPD requires the costly UDP-glucose (UDPG) as a
glucosyl donor, thereby restricting the practical applications of
UGTs.19 Sucrose synthase (SuSy) is a versatile biocatalyst that
can efficiently convert cheap sucrose and UDP into UDPG and
fructose.20,21 Furthermore, the coupling of SuSy and UGT in a
INTRODUCTION
■
Ginseng, the roots and rhizomes of Panax ginseng C. A Meyer,
is a well-known and highly valuable medicinal and dietary
plant.1 This perennial herb has been used to enhance immunity,
reduce fatigue, prevent aging, and provide nutrition in eastern
Asia for more than 2000 years.2−4 Ginsenosies are the major
pharmacologically active components of ginseng.5 These
natural products form a group of glycosylated triterpene
saponins that exhibit various pharmacological effects, such as
anticancer, antitumor, antistress, antiaging, anti-inflammatory,
and immune-system-enhancing activities.6−9
Protopanaxadiol (PPD)-type ginsenosides represent one of
the two major groups of ginsenosides with a structure
consisting of a PPD skeleton and one or more sugar moieties
attached to the C3-OH and/or C20-OH of PPD by β-
glycosidic linkage.10 Glycosylation mediated by UDP-glycosyl-
transferase (UGT), the last biosynthetic step of PPD-type
ginsenosides, contributes to their immense structural and
functional diversity. Most of the key genes involved in the
biosynthetic pathway of PPD-type ginsenosides have currently
been functionally characterized; an engineered PPD-producing
yeast that can accumulate high amounts of PPD has been
constructed in our previous study.11−13 Diverse natural and
unnatural PPD-type ginsenosides, including ginsenosides CK,
Received: January 31, 2018
Accepted: February 27, 2018
Published: February 27, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jafc.8b00597
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
unit of enzyme activity was defined as the amount of Bs-YjiC that
glycosylated 1 μmol of PPD per minute.
one-pot reaction can create a UDP recycle, which could
instantly regenerate UDPG in the presence of a catalytic
amount of UDP, thereby making sucrose the expedient glycosyl
donor for the glycosylation of natural products.22,23
Enzymatic assays of AtSuSy were conducted in a 0.3 mL volume
containing 0.5 mM UDP, 300 mM sucrose, 50 mM Tris-HCl (pH
7.5), and various amounts of purified AtSuSy for 30 min at 40 °C.
AtSuSy activity was measured by determining the amount of fructose
released from sucrose and analyzed using the bicinchoninic acid
method.31 A single unit of enzymatic activity was defined as the
amount of enzyme that released 1 μmol of fructose from sucrose per
min.
Kinetic Analysis of Bs-YjiC. Kinetic studies (300 μL) of Bs-YjiC
toward PPD and ginsenoside Rh2 were conducted with purified Bs-
YjiC (2 μg for PPD and 0.5 μg for ginsenoside Rh2), 50 mM Tris-HCl
(pH 8.0), 10 mM MgCl2, 10 mM UDPG, and varying concentrations
of PPD or ginsenoside Rh2 (50−800 μM). The reactions were
incubated at 40 °C for 20 min and terminated by adding an equal
volume of methanol. All the subsequent steps were performed as
described above. The kinetic parameters were calculated by nonlinear
regression analysis using GraphPad Prism 5.0 software. The kcat values
were calculated using the predicted molecular mass of 4.5 × 104 g
mol−1 for Bs-YjiC.
Optimizing the Conditions for the Bs-YjiC/AtSuSy Cascade
Reaction. Standard Bs-YjiC/AtSuSy cascade reactions containing 1
mM PPD, 0.5 mM UDP, 50 mM Tris-HCl (pH 7.5), 500 mM sucrose,
10% DMSO (v/v), 160 mU mL−1 Bs-YjiC, and 200 mU mL−1 AtSuSy
were incubated at 35 °C for 30 min. The optimal pH was determined
using 50 mM Na2HPO4−NaH2PO4 (pH 6.5−7.5) and 50 mM Tris-
HCl buffer (pH 7.5−8.5). The optimal temperature was determined
ranging from 25 to 45 °C. The optimal concentrations of DMSO,
UDP, sucrose, and PPD were individually obtained using different
concentrations of DMSO (0−20%), UDP (0.05−0.3 mM), sucrose
(0.2−1 M), and PPD (0.5−2 mM) in the standard assay. The optimal
ratio of Bs-YjiC and AtSuSy was determined by adding various ratios
of Bs-YjiC and AtSuSy in the standard assay.
Fed-Batch Synthesis of Ginsenosides Rh2 and F12 by the
Bs-YjiC/AtSuSy Cascade Reaction. The reaction mixtures (10 mL)
contained 1 mM PPD, 0.2 mM UDP, 50 mM Tris-HCl (pH 8.0), 0.4
M sucrose, 5% DMSO, 160 mU mL−1 Bs-YjiC, and 200 mU mL−1
AtSuSy. The reaction was performed at 40 °C and 150 rpm. In
addition, 50 μL of reactant was collected and quenched by adding an
equal volume of methanol. After centrifugation at 12000g for 10 min,
the supernatant was filtered through 0.22 μm filters and analyzed by
HPLC. A 50 μL amount of PPD (200 mM) was periodically added to
the reaction mixtures at 1, 2, 4, 6, and 8 h. Fresh enzymes (160 mU
mL−1 Bs-YjiC and 200 mU mL−1 AtSuSy) were added at 6 h.
HPLC and HPLC-ESI-MS Analysis of the Glycosylated
Products. HPLC and HPLC-ESI-MS analyses of the reactants were
conducted according to our previous study with some minor
changes.32 The reactants (20 μL) were subjected to an XB-C18
column (4.6 × 250 mm, 5 μm particles, Welch, Shanghai, China) with
a 1 mL/min flow rate and 203 nm UV wavelength. The column was
eluted with solvents A (water with 0.1% formic acid) and B
(acetonitrile with 0.1% formic acid) using a gradient program of
25−85% B in 0−25 min and 85% B in 25−45 min. The electrospray
ionization (ESI) probe was operated in a positive ion mode.
Structural Analysis of the Glycosylated Products. For the
structural analysis of ginsenosides Rh2 and F12, the Bs-YjiC/AtSuSy
cascade reaction (200 mL) was prepared as described above. The
reaction was terminated by adding an equal volume of methanol. The
reaction mixtures were subsequently concentrated under reduced
pressure distillation and resuspended in chromatographic methanol
(15 mL). After centrifugation at 10000g for 10 min and filtration
through a 0.22 μm filter, the glycosylated products were purified using
an Agilent 1200 preparative HPLC system coupled with a reverse-
phase Ultimate C18 column (21.2 × 250 mm, 5 μm particles, Welch,
Shanghai, China). The purified products were dissolved in DMSO-d6
or methanol-d4. 1D NMR (1H NMR and 13C NMR) and 2D NMR
spectra (heteronuclear multiple-bond correlation spectroscopy
[HMBC], heteronuclear singular quantum correlation [HSQC], and
Compared with plant UGTs, some naturally occurring
microbial UGTs show high aglycon promiscuity, poor
regiospecificity, and high catalytic proficiency.24−28 Addition-
ally, microbe-derived UGTs are easier to highly express in the
engineered bacteria than plant UGTs.29 Thus, microbial UGTs
are promising biocatalysts for the in vitro semibiosynthesis of
ginsenosides or even novel bioactive ginsenosides. Bs-YjiC from
Bacillus subtilis 168 is a promiscuous UGT toward a
considerable number of structurally diverse compounds.30 Bs-
YiC can also transfer a glucosyl moiety to the free C3-OH, C6-
OH, and C12-OH of protopanaxatriol (PPT) to synthesize
natural and unnatural PPT-type ginsenosides.31 In the present
study, the regio- and stereospecificity of Bs-YjiC toward PPD
and PPD-type ginsenosides was elucidated by analyzing the
glycosylated products using high-performance liquid chroma-
tography−electrospray ionization mass spectrometry (HPLC-
ESI-MS) and NMR spectra. Furthermore, a one-pot reaction by
coupling Bs-YjiC to SuSy from Arabidopsis thaliana (AtSuSy)
was established and applied to synthesize PPD-type ginseno-
sides by using PPD and sucrose as the aglycon and sugar donor,
respectively.
MATERIALS AND METHODS
■
Strains and Chemicals. Authentic PPD and PPD-type ginseno-
sides were purchased from Chengdu Biopurify Phytochemicals Ltd.
(Chengdu, Sichuan, China) and dissolved in dimethyl sulfoxide
(DMSO) for use. UDP-glucose (UDPG), UDP-N-acetylglucosamine
(UDP-GlcNAc), UDP-galactose (UDP-Gal), and UDP-glucuronic
acid (UDP-GlcA) were obtained from Sigma-Aldrich (St. Louis, MO,
USA). Escherichia coli strains were incubated at 37 °C in Luria−Bertani
(LB) medium (5 g L−1 yeast extract, 10 g L−1 NaCl, and 10 g L−1
tryptone) with 100 mg L−1 ampicillin or 50 mg L−1 kanamycin.
Preparation of Recombinant Enzymes. The coding regions of
Bs-YjiC (NP_389104) and AtSuSy (NM_001036838) were amplified
from the genomic DNA of B. subtilis 168 and cDNA of A. thaliana,
respectively. Bs-YjiC and AtSuSy were inserted into the BamHI and
SalI sites of vector pET28a and NcoI and XhoI sites of vector pET32a
to construct the plasmids pET28a-Bs-YjiC and pET32a-AtSuSy,
respectively. For the preparation of recombinant proteins, pET28a-
Bs-YjiC and pET32a-AtSuSy were transformed into E. coli BL21
(DE3). The recombinant E. coli BL21 (DE3) strains were cultured at
37 °C and 200 rpm in an LB medium until the absorbance (OD600
)
reached 0.6−0.8. Isopropyl-β-D-thiogalactopyranoside was added to
the medium to a final concentration of 0.2 mM, and the recombinant
E. coli BL21 (DE3) strains were further incubated at 16 °C for
approximately 16−18 h.
Recombinant E. coli BL21 (DE3) strains were harvested by
centrifugation, resuspended in a lysis buffer (25 mM imidazole, 50
mM Tris-HCl, pH 7.5, and 500 mM NaCl), and disrupted using a
French press. The cell extracts were centrifuged at 20000g for 1 h. The
supernatants containing the target proteins were loaded onto a nickel−
nitrilotriacetic acid−agarose column connected to an AKTA Purifier
system (GE Healthcare, Piscataway, NJ, USA) and purified using a
25−200 mM imidazole gradient.
Enzyme Activity Assays. Enzymatic assays (0.3 mL) of Bs-YjiC
toward PPD or PPD-type ginsenosides were conducted with 1 mM
PPD or PPD-type ginsenosides, 5 mM UDPG, UDP-GlcNAc, UDP-
Gal, or UDP-GlcA, 50 mM Tris-HCl (pH 8.0), 10 mM MgCl2, and 5
μg of purified Bs-YjiC for 0.5 h at 40 °C. The reactions were
terminated by adding equal volumes of methanol. The reactants were
subsequently centrifuged at 10000g for 5 min, filtered through a 0.22
μm filter, and directly analyzed by HPLC or HPLC-ESI-MS. A single
B
DOI: 10.1021/acs.jafc.8b00597
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 1. Proposed glycosylation patterns of Bs-YjiC toward PPD and PPD-type ginsenosides. The solid arrows indicated that the synthesized
ginsenosides were confirmed by NMR spectra; the dotted arrows signified proposed glycosylation steps.
homonuclear correlation spectroscopy [COSY]) were obtained using a
Bruker DMX-600 NMR spectrometer.
UGTs involved in the biosynthesis of ginsenosides from
previous studies.15,16
Cell Viability Assays. Cell viability assays of unnatural ginsenoside
F12 were determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphe-
nyltetrazolium bromide (MTT) colorimetric assay with ginsenosides
Rh2 and Rg3 as the positive controls.33 Cancer cells were seeded at a
density of 1.0 × 104 cells mL−1 into a 96-well plate in triplicate. After
incubation for 24 h, the cells were treated with different concentrations
of ginsenoside Rh2, F12, or Rg3 for 72 h. After incubation, 15 μL of
MTT solution (5 mg mL−1) was added to each well, and the cells were
incubated for another 4 h. The growth medium was removed, and 150
μL of DMSO was added. The absorbance value of the resulting
reaction solution was measured spectrophotometrically at 570 nm
using an ELISA reader.
PPD is the common triterpene skeleton of PPD-type
ginsenosides (Figure 1). Therefore, PPD was selected as a
probe for the in vitro glycosylation of PPD-type ginsenosides
with UDPG as the sugar donor. Two new products (1 and 2)
from the Bs-YjiC-catalyzed reaction were identified by HPLC,
whereas no new products were detected in the control reaction
using the total lysate from induced E. coli BL21 (DE3) that
expressed an empty vector (Figure 2A). Further mass analysis
of product 1 ([M + H]+ m/z+ ∼623.4497]+, [M + H − H2O]+
m/z+ ∼605.4417]+, and [M + H − 2H2O]+ m/z+ ∼587.4311]+)
and product 2 ([M + H]+ m/z+ ∼785.5033]+ and [M + H −
H2O]+ m/z+ ∼767.4947]+) confirmed that these products were
monoglucoside and diglucoside derivatives of PPD (C30H52O3,
calculated molecular weight, [M + H]+ m/z+ ∼461.3989)
(Figure 2B). Microbial UGTs are usually more flexible than
plant-derived UGTs toward both the sugar donors and aglycon
acceptors.34 Thus, various reactions by incubating PPD with
UDP-Gal, UDP-GlcA, and UDP-GlcNAc were also performed;
however, no new products were obtained (data not shown).
Thus, Bs-YjiC can distinguish the sugar moiety of UDP-sugars.
RESULTS AND DISCUSSION
■
Glycosylation of PPD with Bs-YjiC. Bs-YjiC was
heterologously expressed in E. coli BL21 (DE3) as an N-
terminal His6-tagged protein. Notably, Bs-YjiC was expressed
mainly as a soluble protein within 16−37 °C (Figure S1). The
expression level of Bs-YjiC was approximately 600 mg L−1
culture medium, which was much higher than those of ginseng
C
DOI: 10.1021/acs.jafc.8b00597
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
anosyl-20(S)-protopanaxadiol, and ginsenoside F12. Thus,
further studies should be performed to determine the key
amino acids of Bs-YjiC involved in the regiospecific
glycosylation of PPD and to the synthesize a specific
ginsenoside (e.g., ginsenoside Rh2 or 12-O-β-D-glucopyrano-
syl-20(S)-protopanaxadiol) using an engineered Bs-YjiC.
The Km values of Bs-YjiC for PPD and ginsenoside Rh2 were
163.00 and 139.50 μM, respectively (Figure S12) (Table 1).
Table 1. Kinetic Parameters of Bs-YjiC toward PPD and
Ginsenoside Rh2
substrate
Km (μM)
kcat (s−1
)
kcat/Km(s−1 M−1
)
PPD
Rh2
163.00 16.76
139.50 21.09
1.31 0.04
5.68 0.27
0.34 × 104
4.07 × 104
The Km values of Bs-YjiC for PPD and ginsenoside Rh2 were
compared with those of other UGTs involved in the
biosynthesis of ginsenosides.10,14,15,17 Nevertheless, the kcat
values of Bs-YjiC for PPD (1.31 s−1) and Rh2 (5.68 s−1)
were considerably high, thereby resulting in a higher catalytic
efficiency (0.34 × 104 s−1 M−1 for PPD and 4.07 × 104 s−1 M−1
for ginsenoside Rh2) than those of PgUGT74AE2 from P.
ginseng and engineered UGT51 from S. cerevisiae.15,17 The high
catalytic efficiencies of Bs-YjiC toward PPD and ginsenoside
Rh2 and its remarkable acceptor plasticity were in agreement
with the previous notion that naturally occurring UGTs with
high catalytic proficiencies generally show broad acceptor
tolerance.30,35 In addition, the higher catalytic efficiency of Bs-
YjiC toward ginsenoside Rh2 than toward PPD was consistent
with the results in which ginsenoside F12 was the major
product when PPD was used as the substrate, and F12 was
synthesized via a continuous two-step glycosylation of PPD.
In Vitro Cytotoxicity of Ginsenoside F12. Glycosylation
of C3-OH and/or C20-OH of PPD considerably influences the
pharmacological properties of PPD-type ginsenosides.10
Ginsenoside F12 synthesized by UGT109A1was recently
found to have higher effectiveness against NCI-H460 lung
cancer cells than that of ginsenoside Rg3.14 In the present
study, the bioactivity of ginsenoside F12 was further evaluated
using Lovo colon cancer cells, DMS53 lung cancer cells,
HepG2 liver cancer cells, and SNU719 gastric cancer cells,
which were compared with the most active ginsenosides,
ginsenosides Rh2 and Rg3. Ginsenoside F12 can inhibit the
proliferation of Lovo colon cancer cells, HepG2 liver cancer
cells, DMS53 lung cancer cells, and SNU719 gastric cancer cells
with IC50 values of 49.2, 58.8, 55.9, and 44.8 μM, respectively
(Table 2). Although both ginsenosides Rg3 and F12 were
diglucoside derivatives of PPD, the in vitro cytotoxicity of
ginsenoside F12 against cancer cell lines was much higher than
that of ginsenoside Rg3. These results indicated that the
position of glucosyl moieties attached to PPD remarkably
influenced the anticancer activities of PPD-type ginsenosides.
Notably, ginsenoside F12 exhibited the highest cytotoxicity
against DMS53 lung cancer cells, with an IC50 of 55.9 μM,
whereas the IC50 values of ginsenosides Rh2 and Rg3 against
DMS53 lung cancer cells were 107.5 and >400 μM,
respectively. However, ginsenoside Rh2 was still the most
active ginsenoside against Lovo colon cancer cells, HepG2 liver
cancer cells, and SNU719 gastric cancer cells, with IC50 values
of 24.1, 23.3, and 12.0 μM, respectively. The present study was
consistent with previous findings that ginsenosides with a low
Figure 2. HPLC-ESI-MS analysis of the glycosylated products of PPD.
(A) HPLC chromatograms of ginsenoside standards and Bs-YjiC-
catalyzed reactant. (B) MS spectra for products 1 (i) and 2 (ii).
Products 1 and 2 were purified using preparative HPLC.
Their structures were confirmed by 1D NMR (1H NMR and
13C NMR) and 2D NMR (HMBC, HSQC, and COSY)
spectrometry (Supporting Information, Figures S2−S11). The
1H and 13C NMR spectra of product 1 were consistent with
those of Rh2 (Table S1). The observation of significant
downfield 13C shift (∼8 ppm, “glycosylation shift”) of the C3
carbon confirmed that a glucosyl moiety was attached to the
C3-OH of PPD.7,15 Furthermore, detailed HMBC correlations
among the C3 of PPD and the anomeric proton of the glucosyl
moiety, as well as the large anomeric proton-coupling constant
(δH 4.14, J = 7.7 Hz), indicated that product 1 was the 3-O-β-
glucoside, thereby confirming an inverting mechanism of Bs-
1
YjiC. The H and 13C NMR spectra of product 2 were highly
similar to those of product 1 (Table S2). A notably significant
downfield 13C-shift (“glycosylation shift”) at δ 90.86 (∼12 ppm,
C3) and δ 79.66 (∼9 ppm, C12) indicated that a glucosyl
moiety was attached to the C3-OH and C12-OH of PPD,
respectively. In the HMBC, long-range correlations between C3
of the PPD backbone and C1′ and C12 of the PPD backbone
and C1″ further confirmed that product 2 was 3,12-O-β-
diglucoside of PPD (an unnatural ginsenoside named F12 in
this study). The coupling constants (J = 8.0 Hz) of the
anomeric protons indicated that both of the glucosyl moieties
of ginsenoside F12 adopted the β-configuration. Thus, Bs-YjiC
can glycosylate both the C3-OH and C12-OH of PPD and
PPD-type ginsenosides with regio- and stereospecificity.
Furthermore, the absence of ginsenoside CK or F2 in Bs-
YjiC-catalyzed reactants using PPD as an acceptor indicated
that Bs-YjiC could not glycosylate the free C20-OH of PPD or
ginsenoside Rh2 (Figure 2). Only ginseonosides Rh2 and F12
were detected; however, 12-O-β-D-glucopyranosyl-20(S)-proto-
panaxadiol was not observed in the Bs-YjiC-catalyzed reaction
mixture (Figure 2). These results indicated that ginsenoside
F12 was a C12-glycosylation product of ginsenoside Rh2.
UGT109A1 was recently isolated from B. subtilis CTCC
63501.14 The deduced amino acid sequences of UGT109A1
exhibited 94% identity with those of Bs-YjiC; however,
UGT109A1 could glycosylate the free C3-OH and C12-OH
of PPD to synthesize ginsenoside Rh2, 12-O-β-D-glucopyr-
D
DOI: 10.1021/acs.jafc.8b00597
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 2. Cancer Cell Line Cytotoxicity (μM)
a
b
c
d
compound
Lovo
HepG2
DMS53
>400
SNU719
Rg3
Rh2
F12
>400
364.1 20.2
23.3 1.8
58.8 3.9
359.4 30.1
12.0 0.6
44.8 2.8
24.1 1.5
49.2 4.4
107.5 10.5
55.9 3.2
a
b
c
d
Lovo, colon cancer cells. HepG2, liver cancer cells. DMS53, lung cancer cells. SNU719, gastric carcinoma cells.
degree of glycosylation show stronger inhibitory activity against
cancer cells than ginsenosides bearing more sugar moieties.36
Glycosylation of PPD-Type Ginsenosides with Bs-YjiC.
We were interested in the glycosylation of other PPD-type
ginsenosides using Bs-YjiC. Thus, the reactions of PPD-type
ginsenosides, including ginsenosides Rh2, CK, F2, Rg3, and Rd,
were carried out under conditions that were identical to those
of PPD (Figure 3). With ginsenoside Rh2, only one diglucoside
be the same derivative of F2 with an additional glucosyl moiety
attached to the free C12-OH, namely, 3-O-β-D-glucopyranosyl-
12-O-β-D-glucopyranosyl-20-O-β-D-glucopyranosyl-20(S)-pro-
topanaxadiol (Figure 1). With ginsenosides Rg3 and Rd, which
only contained a free C12-OH, only one new glucoside was
detected in each reaction mixtures, as follows: product 7 (tR =
12.6 min, [M + H]+ m/z+ ∼947.5639) and product 8 (tR = 11.8
min, [M + H]+ m/z+ ∼1109.6080) (Figures S16 and S17).
Based on the regiospecificity of Bs-YjiC toward PPD and the
speculated glycosylation patterns of Bs-YjiC toward ginseno-
sides Rh2, CK, and F2, products 7 and 8 should be novel
unnatural ginsenosides with an additional glucosyl moiety
attached to the free C12-OH of ginsenosides Rg3 and Rd,
respectively (Figure 1). Considering that glycosylation is the
prominent biological mechanism for the structural and
functional diversity of ginsenosides, the unnatural PPD-type
ginsenosides 5 or 6, 7, 8, and 9 with a glucosyl moiety attached
to the C12-OH might have various novel biological and
pharmacological activities that are similar to those of ginseno-
side F12 and other PPD-type ginsenosides.36
Optimizing the Parameters for One-Pot Synthesis of
Ginsenosides Rh2 and F12. An engineered yeast cell factory
that can produce substantial amounts of PPD has been
constructed in our previous study, which makes PPD a widely
available precursor for the biosynthesis of PPD-type ginseno-
sides.11 The high expression level and catalytic efficiency of Bs-
YjiC toward PPD suggested that Bs-YjiC was a promising
biocatalyst for industrial applications. Thus, the AtSuSy gene
was amplified from the cDNA of A. thaliana and heterologously
expressed in E. coli BL21 (DE3) as that of Bs-YjiC (Figure S1).
A one-pot cascade reaction by coupling Bs-YjiC to AtSuSy was
established to create an in situ UDPG regeneration and was
applied to synthesize ginsenosides Rh2 and F12 with PPD and
sucrose as the aglycon and sugar donor, respectively.
Figure 3. HPLC analysis of the glycosylated products of ginsenosides
PPD could be transformed into ginsenosides Rh2 and F12
through a Bs-YjiC-AtSuSy cascade reaction as an enzymatic
approach using UDPG as the sugar donor (Figure S18). This
reaction revealed that AtSuSy could effectively synthesize
UDPG. To determine the optimal reaction conditions, the
effects of ratios of Bs-YjiC and AtSuSy on the one-pot synthesis
of ginsenosides Rh2 and F12 were initially determined (Table
3). Ginsenoside F12 production increased by 237% when
increasing the amount of Bs-YjiC from 40 mU mL−1 to 160 mU
mL−1; however, only a trace amount of ginsenoside Rh2 (∼0.04
mM) was detected, which was similar to the Bs-YjiC-catalyzed
reaction using UDPG as the sugar donor. When the amount of
AtSuSy was increased from 50 mU mL−1 to 200 mU mL−1,
ginsenoside F12 production was increased by 373%, whereas
ginsenoside Rh2 was decreased from 0.11 mM to 0.05 mM.
The molar ratios of ginsenoside Rh2/ginsenoside F12
decreased from 0.58 to 0.08 when the amount of AtSuSy was
increased from 50 mU mL−1 to 200 mU mL−1. Considering
that AtSuSy was responsible for the formation of UDPG, the
present study suggested that ginsenoside Rh2 was favorably
produced in the presence of limited UDP-glucose concen-
Rh2, CK, F2, Rg3, and Rd catalyzed by Bs-YjiC.
(product 3, [M + H]+ m/z+ ∼785.5042) with the same
retention time (tR = 17.1 min) as that of F12 was produced, as
confirmed by HPLC-ESI-MS (Figure S13). Product 3 should
be ginsenoside F12 because Bs-YjiC catalyzed a continuous
two-step glycosylation reaction of PPD (Figure 1). With
ginsenoside CK containing the free C3-OH and C12-OH, one
diglucoside (product 4, tR = 15.9 min, [M + H]+ m/z+
∼785.5017) and one triglucoside (product 5, tR = 13.9 min,
[M + H]+ m/z+ ∼947.5689) were confirmed by HPLC-ESI-MS
(Figure S14). Considering that Bs-YjiC could only glycosylate
the free C3-OH and C12-OH of PPD and the tR of product 4
was identical with that of ginsenoside F2, we speculated that
product 4 should be ginsenoside F2 (Figure 1). For ginsenoside
F2, which only contained a free C12-OH, only one triglucoside
(product 6, tR = 13.9 min, [M + H]+ m/z+ ∼947.5655) was
detected (Figure S15). Notably, the tR values of products 5 and
6 were identical to each other and were not consistent with
those of Rd. Thus, we speculated that products 5 and 6 should
E
DOI: 10.1021/acs.jafc.8b00597
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
when the concentration of UDP was increased from 0.05 mM
to 0.2 mM and was slightly increased when the concentration
of UDP continued to increase (Figure 4d). The highest
conversion of PPD was observed in the presence of 0.4 M
sucrose and was slightly changed when the concentration of
sucrose continued to increase (Figure 4e). The conversion of
PPD was decreased from 98% to 32% when the concentration
of PPD was increased from 0.5 mM to 2 mM, and maximal
ginsenoside F12 production (0.67 mM) was obtained in the
presence of 1.0 mM PPD (Figure 4f). In summary, 40 °C, pH
8.0, 5% DMSO, 0.2 mM UDP, 0.4 M sucrose, and 1.0 mM
PPD were used for the fed-batch reaction.
One-Pot Synthesis of Ginsenosides Rh2 and F12 by
Fed-Batch Reaction. Based on the optimal parameters
presented above, a fed-batch reaction by periodic addition of
PPD was conducted to avoid the inhibition of a high
concentration of PPD. As shown in Figure 5, the glycosylation
Table 3. Optimization of the Enzyme Concentrations for the
Bs-YjiC/AtSuSy Cascade Reaction
Bs-YjiC
AtSuSy
F12
Rh2
entry
(mU/mL)
(mU/mL)
(mM)
(mM)
1
2
3
4
5
6
7
8
40
200
200
200
200
50
0.27
0.41
0.64
0.72
0.19
0.36
0.63
0.71
0.04
0.04
0.05
0.05
0.11
0.12
0.05
0.04
80
160
320
160
160
160
160
100
200
300
trations and further reconfirmed that ginsenoside F12 was a
C12-glycosylation product of ginsenoside Rh2. Ginsenoside
F12 production increased slightly (∼13%) when the amount of
Bs-YjiC or AtSuSy was doubled. Thus, the optimal ratio of Bs-
YjiC and AtSuSy was 160:200 mU mL−1.
The conversion rates of PPD were determined with different
pH values (pH 6.0−8.5), temperatures (30−45 °C), DMSO
(0−20%), UDP (0.05−0.3 mM), sucrose (0.2−1 M), and PPD
(0.5−2 mM) to further determine the optimal reaction
conditions (Figure 4). The highest conversion of PPD was
Figure 5. One-pot synthesis of ginsenosides Rh2 and F12 by fed-batch
reaction. One mM PPD was added to the reaction mixtures at 1, 2, 4,
6, and 8 h. Fresh enzymes (160 mU mL−1 Bs-YjiC and 200 mU mL−1
AtSuSy) were added at 6 h.
of PPD was relatively fast, and 1 mM PPD could be completely
converted into 0.02 mM ginsenoside Rh2 and 0.92 mM F12 in
1 h (Figure 5). Only a trace amount of ginsenoside Rh2 was
detected in the first 3 h, and ginsenoside F12 was the major
product, which was consistent with the in vitro enzymatic
reaction using UDPG as the glucosyl donor. Nevertheless, the
molar ratios of ginsenosides Rh2/F12 increased with the
extension of reaction times. After 12 h, 0.32 mM (0.20 g L−1)
ginsenoside Rh2 and 5.07 mM ginsenoside F12 (3.98 g L−1)
were obtained through the periodic feeding of PPD with the
conversion of PPD by 90%. A maximum number of 27.0 was
obtained for the UDPG regeneration cycles. To the best of our
knowledge, the titer of ginsenoside F12 produced in the
present study was much higher than those of other PPD-type
ginsenosides synthesized by yeast cell factories.14,16,17,37 This is
the first report for biosynthesis of triterpene saponins using the
UGT-SuSy cascade system. Considering that a PPD-producing
yeast cell factory can produce substantial amounts of PPD,11 an
in vitro UGT-SuSy cascade reaction could be exploited as a
novel promising alternative for semisynthesis of highly valuable
PPD-type ginsenosides by using PPD as the precursor.
Figure 4. Optimization of the reaction conditions for the Bs-YjiC/
AtSuSy cascade reaction.
achieved at 40 °C and pH 8.0 (Figure 4a and b). Although the
poor solubility of hydrophobic PPD can remarkably inhibit Bs-
YjiC-catalyzed reaction and the addition of DMSO can
significantly improve the availability of PPD in the reaction
system, the conversions of PPD were slightly increased when
DMSO was increased from 0% to 5%. When the concentration
of DMSO exceeded 10%, the conversion of PPD rapidly
decreased, and only 32% of PPD was glycosylated in the
presence of 20% DMSO (Figure 4c). UDP is the most
expensive cofactor in the UGT-SuSy reaction mixtures and has
intricate effects on the UGT-SuSy reaction due to its
importance during the formation of UDPG and inhibiting the
activity of UGT. The conversion of PPD increased by 2.3-fold
In summary, Bs-YjiC from B. subtilis 168 can regio- and
stereospecifically transfer a glucosyl moiety to the free C3-OH
and C12-OH of PPD and PPD-type ginsenosides to synthesize
natural and unnatural PPD-type ginsenosides. These findings
provided insights into the important roles of microbial UGTs
for the biosynthesis of ginsenosides. This study suggested that
F
DOI: 10.1021/acs.jafc.8b00597
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
(8) Seo, J. Y.; Ju, S. H.; Oh, J.; Lee, S. K.; Kim, J. S. Neuroprotective
and Cognition Enhancing Effects of Compound K Isolated from Red
Ginseng. J. Agric. Food Chem. 2016, 64, 2855−2864.
the in vitro UGT-SuSy cascade reaction could be exploited as
an effective approach for biosynthesis of ginsenosides. In vivo
and in vitro biological activities of ginsenosides F12 and other
unnatural ginsenosides synthesized in this study should be
investigated in future studies.
(9) Wang, C.; He, H.; Dou, G.; Li, J.; Zhang, X.; Jiang, M.; Li, P.;
Huang, X.; Chen, H.; Li, L. Ginsenoside 20(S)-Rh2 Induces Apoptosis
and Differentiation of Acute Myeloid Leukemia Cells: Role of Orphan
Nuclear Receptor Nur77. J. Agric. Food Chem. 2017, 64, 7687−7697.
(10) Jung, S. C.; Kim, W.; Park, S. C.; Jeong, J.; Park, M. K.; Lim, S.;
Lee, Y.; Im, W. T.; Lee, J. H.; Choi, G. Two ginseng UDP-
glycosyltransferases synthesize ginsenoside Rg3 and Rd. Plant Cell
Physiol. 2014, 55, 2177−88.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jafc.8b00597.
(11) Dai, Z.; Liu, Y.; Zhang, X.; Shi, M.; Wang, B.; Wang, D.; Huang,
L.; Zhang, X. Metabolic engineering of Saccharomyces cerevisiae for
production of ginsenosides. Metab. Eng. 2013, 20, 146−156.
(12) Han, J. Y.; Kim, H. J.; Kwon, Y. S.; Choi, Y. E. The Cyt P450
enzyme CYP716A47 catalyzes the formation of protopanaxadiol from
dammarenediol-II during ginsenoside biosynthesis in Panax ginseng.
Plant Cell Physiol. 2011, 52, 2062−73.
(13) Han, J. Y.; Kwon, Y. S.; Yang, D. C.; Jung, Y. R.; Choi, Y. E.
Expression and RNA interference-induced silencing of the dammar-
enediol synthase gene in Panax ginseng. Plant Cell Physiol. 2006, 47,
1653−62.
1H and 13C NMR spectral data for products 1 and 2
(Tables S1 and S2), some experimental results (Figures
S1, S2, and S18), and HPLC-Q-TOF/ESI-MS and NMR
analysis (Figures S2−S17) (PDF)
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail (X. Zhang): zhang_xl@tib.cas.cn. Tel:
+862284861983.
(14) Liang, H.; Hu, Z.; Zhang, T.; Gong, T.; Chen, J.; Zhu, P.; Li, Y.;
Yang, J. Production of a bioactive unnatural ginsenoside by
metabolically engineered yeasts based on a new UDP-glycosyltransfer-
ase from Bacillus subtilis. Metab. Eng. 2017, 44, 60−69.
(15) Wang, P.; Wei, Y.; Yun, F.; Liu, Q.; Wei, W.; Yang, C.; Lei, Z.;
Zhao, G.; Yue, J.; Xing, Y. Production of bioactive ginsenosides Rh2
and Rg3 by metabolically engineered yeasts. Metab. Eng. 2015, 29, 97−
105.
(16) Yan, X.; Fan, Y.; Wei, W.; Wang, P.; Liu, Q.; Wei, Y.; Zhang, L.;
Zhao, G.; Yue, J.; Zhou, Z. Production of bioactive ginsenoside
compound K in metabolically engineered yeast. Cell Res. 2014, 24,
770−3.
(17) Zhuang, Y.; Yang, G.-Y.; Chen, X.; Liu, Q.; Zhang, X.; Deng, Z.;
Feng, Y. Biosynthesis of plant-derived ginsenoside Rh2 in yeast via
repurposing a key promiscuous microbial enzyme. Metab. Eng. 2017,
42, 25−32.
(18) Atopkina, L. N.; Malinovskaya, G. V.; Elyakov, G. B.; Uvarova,
N. I.; Woerdenbag, H. J.; Koulman, A.; Pras, N.; Potier, P. Cytotoxicity
of natural ginseng glycosides and semisynthetic analogues. Planta Med.
1999, 65, 30−34.
*E-mail (Y. Sun): sun_yx@tib.cas.cn. Tel: +862284861960.
ORCID
Longhai Dai: 0000-0001-5662-9252
Caixia Dong: 0000-0002-4749-9510
Jiangang Yang: 0000-0001-9463-8862
Author Contributions
⊥L. Dai and C. Liu contributed equally to this work.
Funding
This work was supported by the National Natural Science
Foundation of China (no. 21702226), the Key Research
Program of the Chinese Academy of Sciences (no. KFZD-SW-
215), and the Technology Planning Project of Tianjin (no.
16YFXTNC00160).
Notes
The authors declare no competing financial interest.
REFERENCES
■
(19) Gutmann, A.; Lepak, A.; Diricks, M.; Desmet, T.; Nidetzky, B.
Glycosyltransferase cascades for natural product glycosylation: Use of
plant instead of bacterial sucrose synthases improves the UDP-glucose
recycling from sucrose and UDP. Biotechnol. J. 2017, 12, 1600557.
(1) Choi, J.; Choi, T.; Lee, M.; Kim, T. Ginseng for Health Care: A
Systematic Review of Randomized Controlled Trials in Korean
Literature. PLoS One 2013, 8, 697−702.
(2) Dong, W. W.; Xuan, F. L.; Zhong, F. L.; Jiang, J.; Wu, S.; Li, D.;
Quan, L. H. Comparative Analysis of the Rats’ Gut Microbiota
Composition in Animals with Different Ginsenosides Metabolizing
Activity. J. Agric. Food Chem. 2016, 65, 327.
(3) Kang, A.; Xie, T.; Zhu, D.; Shan, J.; Di, L.; Zheng, X. Suppressive
Effect of Ginsenoside Rg3 against Lipopolysaccharide-Induced
Depression-Like Behavior and Neuroinflammation in Mice. J. Agric.
Food Chem. 2017, 65, 6861−6869.
(4) Wang, Z.; Hu, J.; Yan, M.; Xing, J.; Liu, W.; Li, W. Caspase-
mediated anti-apoptotic effect of Ginsenoside Rg5, a main rare
ginsenoside, on Acetaminophen-induced Hepatotoxicity in Mice. J.
Agric. Food Chem. 2017, 65, 9226.
(5) Dai, Z.; Wang, B.; Liu, Y.; Shi, M.; Wang, D.; Zhang, X.; Liu, T.;
Huang, L.; Zhang, X. Producing aglycons of ginsenosides in bakers’
yeast. Sci. Rep. 2015, 4, 3698.
(6) Li, L.; Shin, S. Y.; Lee, S. J.; Moon, J. S.; Im, W. T.; Han, N. S.
Production of ginsenoside F2 by using Lactococcus lactis with enhanced
expression of β-glucosidase gene from Paenibacillus mucilaginosus. J.
Agric. Food Chem. 2016, 64, 2506.
(7) Pei, J.; Xie, J.; Yin, R.; Zhao, L.; Ding, G.; Wang, Z.; Xiao, W.
Enzymatic transformation of ginsenoside Rb1 to ginsenoside 20(S)-
Rg3 by GH3 β-glucosidase from Thermotoga thermarum DSM 5069 T.
J. Mol. Catal. B: Enzym. 2015, 113, 104−109.
(20) Schmolzer, K.; Gutmann, A.; Diricks, M.; Desmet, T.; Nidetzky,
̈
B. Sucrose synthase: A unique glycosyltransferase for biocatalytic
glycosylation process development. Biotechnol. Adv. 2015, 34, 88−111.
(21) Schmolzer, K.; Lemmerer, M.; Gutmann, A.; Nidetzky, B.
̈
Integrated process design for biocatalytic synthesis by a Leloir
glycosyltransferase: UDP-glucose production with sucrose synthase.
Biotechnol. Bioeng. 2017, 114, 924−928.
(22) Dewitte, G.; Walmagh, M.; Diricks, M.; Lepak, A.; Gutmann, A.;
Nidetzky, B.; Desmet, T. Screening of recombinant glycosyltrans-
ferases reveals the broad acceptor specificity of stevia UGT-76G1. J.
Biotechnol. 2016, 233, 49−55.
(23) Gutmann, A.; Nidetzky, B. Unlocking the Potential of Leloir
Glycosyltransferases for Applied Biocatalysis: Efficient Synthesis of
Uridine 5′-Diphosphate-Glucose by Sucrose Synthase. Adv. Synth.
Catal. 2016, 358, 3600−3609.
(24) Chiu, H. H.; Shen, M. Y.; Liu, Y. T.; Fu, Y. L.; Chiu, Y. A.;
Chen, Y. H.; Huang, C. P.; Li, Y. K. Diversity of sugar acceptor of
glycosyltransferase 1 from Bacillus cereus and its application for
glucoside synthesis. Appl. Microbiol. Biotechnol. 2016, 100, 1−13.
(25) Gantt, R. W.; Goff, R. D.; Williams, G. J.; Thorson, J. S. Probing
the Aglycon Promiscuity of an Engineered Glycosyltransferase. Angew.
Chem., Int. Ed. 2008, 47, 8889−92.
G
DOI: 10.1021/acs.jafc.8b00597
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
(26) Pandey, R. P.; Parajuli, P.; Koirala, N.; Park, J. W.; Sohng, J. K.
Probing 3-hydroxyflavone for in vitro glycorandomization of flavonols
by YjiC. Appl. Environ. Microbiol. 2013, 79, 6833−8.
(27) Pandey, R. P.; Tai, F. L.; Kim, E. H.; Yamaguchi, T.; Yong, I. P.;
Kim, J. S.; Sohng, J. K. Enzymatic Synthesis of Novel Phloretin
Glucosides. Appl. Environ. Microbiol. 2013, 79, 3516−21.
(28) Zhou, M.; Hou, Y.; Hamza, A.; Pain, C.; Zhan, C. G.; Bugni, T.
S.; Thorson, J. S. Probing the regiospecificity of enzyme-catalyzed
steroid glycosylation. Org. Lett. 2012, 14, 5424−7.
(29) Bo, F.; Chen, T.; Zhang, S.; Wu, B.; He, B. Mining of efficient
microbial UDP-glycosyltransferases by motif evolution cross plant
kingdom for application in biosynthesis of salidroside. Sci. Rep. 2017,
7, 463.
(30) Dai, L.; Li, J.; Yao, P.; Zhu, Y.; Men, Y.; Zeng, Y.; Yang, J.; Sun,
Y. Exploiting the aglycon promiscuity of glycosyltransferase Bs-YjiC
from Bacillus subtilis and its application in synthesis of glycosides. J.
Biotechnol. 2017, 248, 69−76.
(31) Dai, L.; Li, J.; Yang, J.; Zhu, Y.; Men, Y.; Zeng, Y.; Cai, Y.; Dong,
C.; Dai, Z.; Zhang, X.; Sun, Y. Promiscuous glycosyltransferase from
Bacillus subtilis 168 for the enzymatic synthesis of novel protopanax-
atriol-type ginsenosides. J. Agric. Food Chem. 2018, 66, 943.
(32) Dai, L.; Liu, C.; Zhu, Y.; Zhang, J.; Men, Y.; Zeng, Y.; Sun, Y.
Functional Characterization of Cucurbitadienol Synthase and
Triterpene Glycosyltransferase Involved in Biosynthesis of Mogrosides
from Siraitia grosvenorii. Plant Cell Physiol. 2015, 56, 1172.
(33) Liu, C.; Zeng, Y.; Dai, L. H.; Cai, T. Y.; Zhu, Y. M.; Dou, D. Q.;
Ma, L. Q.; Sun, Y. X. Mogrol represents a novel leukemia therapeutic,
via ERK and STAT3 inhibition. Am. J. Cancer Res. 2015, 5, 1308.
(34) Song, M. C.; Kim, E.; Ban, Y. H.; Yoo, Y. J.; Kim, E. J.; Park, S.
R.; Pandey, R. P.; Sohng, J. K.; Yoon, Y. J. Achievements and impacts
of glycosylation reactions involved in natural product biosynthesis in
prokaryotes. Appl. Microbiol. Biotechnol. 2013, 97, 5691−704.
(35) Oberthur, M.; Leimkuhler, C.; Kruger, R. G.; Lu, W.; Walsh, C.
̈
T.; Kahne, D. A systematic investigation of the synthetic utility of
glycopeptide glycosyltransferases. J. Am. Chem. Soc. 2005, 127, 10747−
10752.
(36) Park, C.-S.; Yoo, M.-H.; Noh, K.-H.; Oh, D.-K. Biotransforma-
tion of ginsenosides by hydrolyzing the sugar moieties of ginsenosides
using microbial glycosidases. Appl. Microbiol. Biotechnol. 2010, 87, 9−
19.
(37) Wei, W.; Wang, P.; Wei, Y.; Liu, Q.; Yang, C.; Zhao, G.; Yue, J.
Characterization of Panax ginseng UDP-Glycosyltransferases Catalyz-
ing Protopanaxatriol and Biosyntheses of Bioactive Ginsenosides F1
and Rh1 in Metabolically Engineered Yeasts. Mol. Plant 2015, 8,
1412−1424.
H
DOI: 10.1021/acs.jafc.8b00597
J. Agric. Food Chem. XXXX, XXX, XXX−XXX