An Ambidextrous Polyphenol Glycosyltransferase PaGT2 from Phytolacca americana

Article abstract of DOI:10.1021/acs.biochem.0c00224

The glycosylation of small hydrophobic compounds is catalyzed by uridine diphosphate glycosyltransferases (UGTs). Because glycosylation is an invaluable tool for improving the stability and water solubility of hydrophobic compounds, UGTs have attracted attention for their application in the food, cosmetics, and pharmaceutical industries. However, the ability of UGTs to accept and glycosylate a wide range of substrates is not clearly understood due to the existence of a large number of UGTs. PaGT2, a UGT from Phytolacca americana, can regioselectively glycosylate piceatannol but has low activity toward other stilbenoids. To elucidate the substrate specificity and catalytic mechanism, we determined the crystal structures of PaGT2 with and without substrates and performed molecular docking studies. The structures have revealed key residues involved in substrate recognition and suggest the presence of a nonconserved catalytic residue (His81) in addition to the highly conserved catalytic histidine in UGTs (His18). The role of the identified residues in substrate recognition and catalysis is elucidated with the mutational assay. Additionally, the structure-guided mutation of Cys142 to other residues, Ala, Phe, and Gln, allows PaGT2 to glycosylate resveratrol with high regioselectivity, which is negligibly glycosylated by the wild-type enzyme. These results provide a basis for tailoring an efficient glycosyltransferase.

Full text of DOI:10.1021/acs.biochem.0c00224

Subscriber access provided by ECU Libraries
Article
An ambidextrous polyphenol glycosyltransferase
PaGT2 from Phytolacca americana
rakesh maharjan, Yohta Fukuda, Naomichi Shimomura, Taisuke Nakayama, Yuta Okimoto,
Koki Kawakami, Toru Nakayama, Hiroki Hamada, Tsuyoshi Inoue, and Shin-ichi Ozaki
Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.0c00224 • Publication Date (Web): 11 Jun 2020
Downloaded from pubs.acs.org on June 18, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 19
Biochemistry
1
2
3
4
5
6
An ambidextrous polyphenol glycosyltransferase PaGT2 from
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Phytolacca americana
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
,
†,
‡
§
‡
⊥
⊥
Rakesh Maharjan, Yohta Fukuda, Naomichi Shimomura, Taisuke Nakayama, Yuta Okimoto,
#
,
,†,
⊥
,‡
∥
∥
Koki Kawakami, Toru Nakayama, Hiroki Hamada,* Tsuyoshi Inoue,* Shin-ichi Ozaki*
†Department of Applied Chemistry, Graduate School of Engineering, Osaka University, Suita City, Osaka 565-0871, Japan
⊥Graduate School of Pharmaceutical Science, Osaka University, Suita City, Osaka 565-0871, Japan
‡Department of Biological Chemistry, Graduate School of Science and Technology for Innovations, Yamaguchi University,
Yamaguchi 753-8515, Japan
§National Institute of Biomedical Innovation, Health and Nutrition, Saito-Asagi, Ibaraki City, Osaka 567-0085, Japan
#Department of Biomolecular Engineering, Graduate School of Engineering, Tohoku University, Sendai, Miyagi 980-8579,
Japan
∥Department of Life Science, Faculty of Science, Okayama University of Science, Okayama 700-0005, Japan
ABSTRACT: The glycosylation of small hydrophobic compounds is catalyzed by uridine diphosphate glycosyltransferases (UGTs). Since
glycosylation is an invaluable tool for improving stability and water solubility of hydrophobic compounds, UGTs have gained attention for
their application in food, cosmetics, and pharmaceutical industries. However, the ability of UGTs to accept and glycosylate a wide range of
substrates is not clearly understood due to the existence of a large number of UGTs. PaGT2, a UGT from Phytolacca americana can regi-
oselectively glycosylate piceatannol but has low activity towards other stilbenoids. To elucidate the substrate specificity and catalytic mech-
anism, we determined the crystal structures of PaGT2 with and without substrates and performed molecular docking studies. The structures
have revealed key residues involved in substrate recognition and suggest the presence of a non-conserved catalytic residue (His81) in addition
to the highly conserved catalytic histidine in UGTs (His18). The role of the identified residues in substrate recognition and catalysis is
elucidated with the mutational assay. Additionally, the structure-guided mutation of Cys142 to other residues, Ala, Phe, and Gln, allows
PaGT2 to glycosylate resveratrol with high regioselectivity, which is negligibly glycosylated by the wildtype enzyme. These results provide
a basis for tailoring an efficient glycosyltransferase.
OleI24 from Streptomyces antibioticus, and others have provided a
basis for the understanding of the glycosylation mechanism in plant
UGTs. Besides, site-directed mutagenesis and directed evolution of
UGTs have been studied for enhancing the enzyme’s activity
and/or changing the regioselectivity of glycosylated products.
Plant UGTs are characterized by the presence of a highly conserved
plant secondary product glycosyltransferase (PSPG) motif that is
involved in the recognition of the UDP-sugar donor. Although the
residues involved in recognition of sugar-acceptor in UGTs are less
conserved, a highly conserved histidine-aspartate (His-Asp) pair
has been identified to catalyze the glycosylation.
pair is involved in the removal of a proton from the glycosylation
site to generate a nucleophile that attacks the C1 carbon atom of the
INTRODUCTION
Glycosyltransferases (GTs) catalyze the transfer of a sugar moiety
1
from activated sugar donor to acceptors. GTs play a major role in
2
5–29
cellular localization, storage, or metabolism of metabolites and xe-
nobiotic compounds, thereby maintaining the cellular homeosta-
2
sis. Natural products such as polyphenols are usually glycosyl-
3
–5
30
ated. Glycosylation improves water solubility and chemical sta-
6
bility of polyphenols, preserving their bioactivity. Polyphenols
and their glycosides are known to have several health benefits such
as antioxidant, anti-aging, anti-estrogenic, anticancer, anti-inflam-
3
1–34
The catalytic
7
–10
matory, and cardio-protective effects.
Though glycosylation of
polyphenols can be achieved by chemical synthesis, these methods
are limited by poor regioselectivity and involve tedious steps of
3
4
sugar moiety. However, UGTs with non-conserved catalytic res-
1
1
35,36
32
protection and deprotection of functional groups. Contrary to this,
enzyme-mediated glycosylation is more efficient and highly regi-
idue
or without catalytic pair are known to catalyze glycosyl-
ation. This necessitates the exploration of more UGTs for improv-
ing our understanding of enzymatic glycosylation.
12
oselective.
Glycosylation of small hydrophobic compounds in plants is cata-
lyzed mainly by uridine diphosphate glycosyltransferases (UGTs),
classified as the GT1 family in the carbohydrate-active enzyme
Glucosides of compounds such as kaempferol, quercetin, and triter-
pene saponins have been extracted from various parts of a toxic
plant Phytolacca americana. The cell suspension of P. americana
is also able to glucosylate exogenous compounds such as resvera-
trol, perillyl alcohol, and, 2,2,5,7,8-pentamethyl-6-chromanol.
PaGT2, a glycosyltransferase from P. americana regioselectively
glycosylates piceatannol (3, 5, 3’, 4’-tetrahydroxystilbene), a stil-
benoid polyphenol, while it has negligible activity with resveratrol
3
7
1
3
(
CAZy) database. Recently, several plant and bacterial UGTs are
38
characterized for their ability to glycosylate different polyphe-
14–19
nols.
The determination of crystal structures of UGTs such as
2
0
21
22
VvGT1 from Vitis vinifera, UGT85H2 , UGT71G1 from Medi-
2
3
cago truncatula, UGT78K6 from Clitoria ternatae, OleD and
ACS Paragon Plus Environment

Biochemistry
Page 2 of 19
(3, 5, 4'-tetrahydroxystilbene) which lacks a hydroxyl group. This
Site-directed mutagenesis. Site-directed mutagenesis was per-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bias in the substrate specificity of PaGT2 could not be explained as
UGT structures with stilbenoids are not available. Herein, we de-
termined the crystal structures of PaGT2 with and without stil-
benoid substrates. Together with molecular docking, we identified
the key residues for substrate recognition and found that PaGT2
catalyzes glycosylation through two individual catalytic residues.
Additionally, the structure-guided mutagenesis allowed PaGT2 to
glycosylate resveratrol with high regioselectivity, which was neg-
ligibly glycosylated by the wildtype enzyme.
formed using the whole plasmid pCold-PaGT2 and specific oligo-
nucleotide primers listed in Table S1. Briefly, the whole plasmid
was linearized with mutagenesis primers by PCR. Amplified PCR
products were treated with DpnI (New England Biolabs) and puri-
®
fied using NucleoSpin Gel and PCR Clean-up (Macherey-Nagel),
following the manufacturer’s protocol. The purified PCR product
was ligated using T4-polynucleotide kinase (Toyobo co.) and liga-
tion high ver. 2.0 (Toyobo co.), and transformed into E. coli DH5α.
The desired mutation of PaGT2 was confirmed by DNA sequenc-
ing.
MATERIALS AND METHODS
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Enzyme assay. Wildtype PaGT2 and all mutants for enzyme assay
were expressed and purified as mentioned above. The glucosylation
reactions were performed at 37°C in a 200 µl total reaction volume.
The structure of the substrates used and the HPLC profile of the
reaction mixture after the glucosylation reaction is shown in Figure
Expression and purification of PaGT2. PaGT2 cDNA and pCold
vector were amplified by polymerase chain reaction (PCR) using
following
primers:
pCold
forward
5'-
TAGGTAATCTCTGCTTAAAAGCACAG-3', pCold reverse 5'-
ACCCTGGAAATAAAGATTCTCC-3', PaGT2 forward 5'-
CTTTATTTCCAGGGTATGGAAATGGAAGCACCACTC-3',
1
. The reaction solution containing 50 mM potassium phosphate
buffer (pH 7.4), 50 µM acceptor substrates, 100 µM UDP-Glucose,
and 5 µM enzyme was incubated at 37°C for 10 minutes. The reac-
tion was stopped by adding 1.5% trifluoroacetic acid, centrifuged
at 12,000 rpm for 1 minute, and filtered using a dismic-13HP filter.
HPLC analysis of the reaction mixtures was performed on an Im-
tact US-C18 (2.0 × 150 mm) reverse-phase column at a flow rate
of 0.2 mL/min. Piceatannol and resveratrol glucosylation mixtures
were analyzed by isocratic elution, starting with 15% acetonitrile
and 85% water for 20 minutes followed by 100% acetonitrile for
and
PaGT2
reverse
5'-
AGCAGAGATTACCTAGCTTTTGCATTGGCTCCATTTAG-3'.
Amplified PaGT2 cDNA was cloned into the pCold vector using
the Infusion kit (Takara Bio USA, Inc.) following the manufactur-
er's protocol. The construct contained N-terminal 6x histidine tag
followed by a TEV protease recognition site. A single colony of E.
coli BL21 Star (DE3), transformed with pCold PaGT2, was inocu-
lated into 2 ml LB medium supplemented with 100 μg/ml ampicil-
lin and grown at 37 °C for about 8 hours as a starter culture. 200 μl
of the starter culture was introduced into a 200 ml LB medium with
the same concentration of antibiotics and grown overnight at 37 °C.
This culture was used to inoculate 1 L culture in the same medium
in the next morning and the cells were continued to grow at 37 °C.
When the OD600 was ~0.4, the temperature was decreased to 15 °C.
Expression was induced with isopropyl-β-D-thiogalactopyranoside
(IPTG) at 15 °C and OD600 ~0.6-0.8. After 24 hours, cells were
harvested by centrifugation at 9000 × g for 10 min at 4 °C. Har-
vested cells were frozen with liquid nitrogen and stored at -80 °C
until its use.
1
0 minutes. Rhapontigenin glucosylation mixtures were also ana-
lyzed by isocratic elution, starting with 50% acetonitrile and 50%
water for 30 minutes followed by 100% acetonitrile for 10 minutes.
Kaempferol glucosylation products were eluted by a linear gradient
of acetonitrile starting from 10% acetonitrile and 90% water to 30%
acetonitrile and 70% water in 20 minutes followed by 100% ace-
tonitrile for 10 minutes. Pterostilbene glucosylation products were
also eluted using a linear gradient of 50% acetonitrile and 50% wa-
ter to 100% acetonitrile in 30 minutes followed by 100% acetoni-
trile for 10 minutes. The glucoside products were identified by
comparison of the HPLC chromatograms of the reaction mixtures
with the chromatograms of standard samples of respective com-
The cell pellet was re-suspended in buffer A (20 mM Tris-HCl pH
3
9
pounds. However, the obtained kaempferol product glucosylated
by PaGT2, kaempferol 3-O-β-glucoside, and kaempferol 4'-O-β-
glucoside were eluted at the nearly same time (Figure S1A). Thus,
the regioselectivity of kaempferol glucoside products by wildtype
PaGT2 was determined by a nuclear magnetic resonance (NMR)
method (Figure S1B). 1H NMR spectral data were recorded using
8
.5, 100 mM NaCl, 5 mM DTT) including 1 tablet of a protease
inhibitor cocktail (Roche). Cells were lysed by sonication on ice
with the following pulse sequence: 15-sec burst, 15-sec rest for a
total burst time of 10 min at a power output of 80. Lysed cells were
subjected to centrifugation at 20,000 ×g for 30 min at 4 °C. The
obtained supernatant was filtered using a 0.45 μm membrane sy-
ringe filter. The filtered supernatant was loaded on to a Ni-NTA
column (HisTrap HP 5 ml) equilibrated with buffer A. The column
was washed with 10 column volume (CV) of buffer A. bound pro-
tein was eluted with a 0-50% gradient of buffer A supplemented
with 300 mM imidazole. Fractions containing PaGT2 were pooled,
mixed with TEV protease, and dialyzed overnight against buffer A
to remove the histidine tag. The dialyzed sample was diluted 10
times using 20 mM Tris-HCl pH 8.5 and loaded on to a HiTrap Q
6
a JEOL 400 MHz NMR spectrometer in DMSO-d and compared
with known chemical shift values of authentic kaempferol 3-O-β-
glucoside or kaempferol 4'-O-β-glucoside. The chemical shift and
coupling constants are expressed in δ (ppm) and Hz, respectively.
The anomeric proton signal of the kaempferol product appeared at
δ 5.20 (1H, d, J=7.2 Hz). The observed chemical shift value is sim-
ilar to δ 5.25 (1H, d, J=7.6 Hz) obtained with standard kaempferol
3-O-β-glucoside and different from δ 5.03 (1H, d, J=7.0 Hz) re-
4
0
ported for kaempferol 4'-O-β-glucoside. To quantitate glucoside
products in the reaction mixture, standard curves were generated.
For enzyme kinetic studies, acceptor concentrations were varied
(5 ml) column equilibrated with 20 mM Tris-HCl pH 8.5. PaGT2
was eluted with a linear gradient ranging from 0 to 1 M NaCl in 20
mM Tris-HCl pH 8.5. Fractions containing PaGT2 were pooled,
concentrated using a 30 kDa MWCO Vivaspin tube (Sartorius), and
loaded on to a Hiload 16/60 Superdex 200 pg size exclusion chro-
matography (SEC) column equilibrated with an SEC buffer (20
mM Tris-HCl pH 8.0, 100 mM NaCl, 5 mM DTT). Fractions con-
taining PaGT2 were pooled, concentrated, and stored at -80 °C.
m
(25-150 µM). To determine the K values of less than 30 µM in
Table 1, we varied the acceptor concentration from 7.5-150 µM.
The final concentration of piceatannol and kaempferol was deter-
mined based on extinction coefficients reported previously
-
1
-1
41
(
2
piceatannol 26,350 M cm at 304 nm in ethanol , kaempferol
-
1
-1
42
1,242 M cm at 365 nm in methanol ). The hyperbolic depend-
ence of glucoside production rates on acceptor concentrations was
2
ACS Paragon Plus Environment

Page 3 of 19
Biochemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1: Structure of aglycones utilized for glucosylation assay and HPLC analysis chromatogram. Aglycones for glycosylation A) Piceatannol,
B) Resveratrol, C) Pterostilbene, D) Rhapontigenin, and E) Kaempferol and the HPLC chromatograms (below) after the glycosylation reaction by PaGT2
utilizing donor UDP-glucose. For piceatannol and kaempferol, black and blue lines indicates the elution profile of standard compound and the products
of glycosylation by PaGT2, respectively. For resveratrol, pterostilbene, and rhapontigenin, HPLC chromatograms of the reaction mixture is shown. The
names of aglycones are colored green, orange, and red indicating good, poor, and not substrates for PaGT2, respectively.
fitted using the Michaelis–Menten equation to determine the kcat
and K values.
(PDB: 2VCH) as a search model using Molrep45 in CCP4.46 The
structures of PaGT2 complexes were solved using the solved apo-
PaGT2 structure as a search model. Model building and refinement
m
4
7
48
Protein crystallization. Crystallization screening of PaGT2 with
and without resveratrol and UDP-2FGlc were performed by the sit-
ting drop vapor diffusion method from 1:1 mixture of 100 nL pro-
tein stock (15 mg/ml in 20 mM Tris-HCl pH 8.0, 100 mM NaCl, 5
mM DTT) and 100 nL well solution. Apo PaGT2 was crystallized
using well solution containing 0.1 M magnesium formate, 0.1 M
MOPS pH 7.0, and 17% w/v PEG 3350 at 20 °C. Optimal crystals
for diffraction were achieved by micro-seeding and hanging drop
vapor diffusion from 1:1 mixture of protein (1 μl) and reservoir so-
lution (1 μl). Crystals were harvested in the same reservoir solution
supplemented with 15% ethylene glycol and flash cooled in liquid
nitrogen.
were performed using COOT and refmac5. Figures were pre-
4
9
50
pared using PyMol and LigPlot+.
Molecular Docking. Molecular docking of piceatannol and
kaempferol were performed using an automatic docking program
51
PyRx Virtual Screening tool. The crystal structure of PaGT2 in
complex with UDP-2FGlc and resveratrol was used as a reference
after removing the ligands. Ligands for docking- piceatannol, and
kaempferol were obtained from the COOT library and saved in
the .pdb format. Both protein and ligands were opened in the PyRx
program. PaGT2 chosen as the macromolecule and piceatan-
nol/kaempferol as ligand were automatically converted to
the .pdbqt format. The center of the grid box was set to (X= 25.09,
Y= 48.81, and Z=72.93) and the dimension of the grid box was set
to (6.64×15.65×5.26) together with exhaustiveness 8. Docking was
Complexes of PaGT2 with UDP-2FGlc and resveratrol/pterostil-
bene were obtained by co-crystallization in the presence of 5 mM
UDP-2FGlc and 2 mM resveratrol/pterostilbene in ethanol. Crys-
tals were grown by the hanging-drop vapor diffusion method in 1:1
mixture of protein solution and the reservoir solution containing
52
performed with AutoDock Vina in PyRx. Molecular docking of
resveratrol and pterostilbene were also performed using the same
grid box parameters to confirm the accuracy of molecular docking.
To evaluate the binding affinity of acceptors in the active-site in
His18/81Ala double mutant, a model of PaGT2 with His18/81Ala
mutation was prepared using COOT. Molecular docking of all four
acceptors was performed also performed with the PaGT2
His18/81Ala mutant model. PyMol was used for visualization of
the results and preparation of figures. Figures of molecular docking
show the binding modes of acceptors with high possibilities.
0
.11 M potassium citrate, 0.06 M lithium citrate, 0.11 M sodium
phosphate, and 23-25% w/v PEG 6000. Crystals were harvested in
the same reservoir solution supplemented with 15% xylitol and
flash cooled in liquid nitrogen.
Data collection and crystal structure determination. Diffraction
data were collected on a beamline BL44XU at SPring-8 with an
MX300HE CCD detector (Rayonix, LLC) and an EIGER X 16M
detector (Dectris). The data for apo-PaGT2 was collected to 2.30
Å. The data of PaGT2 complexed with UDP-2FGlc acceptors
resveratrol or piceatannol were collected to 2.60 Å and 2.65 Å, re-
spectively. X-ray diffraction data were indexed and scaled using
RESULTS AND DISCUSSION
4
3
44
HKL2000 or XDS. The structure of apo PaGT2 was solved by
molecular replacement using Arabidopsis thaliana UGT72B1
Glucosylation of stilbenoids by PaGT2. Stilbenoid glycosylation
ability of PaGT2 was screened against a panel of stilbenoids using
3
ACS Paragon Plus Environment

Biochemistry
Page 4 of 19
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Kinetic parameters of PaGT2 and mutants
Enzyme (PaGT2) Acceptor (µM)
-
1
-1 -1
K
m
k
cat (s )
m
kcat/K (M s )
-
2
-3
-4
-3
-3
-3
WT
Piceatannol
31 ±1*
17 ± 3
52 ± 3
40 ± 3
49 ± 2
N.D.
1.49×10 ± 1.41×10
480.64 ± 45
217.64 ± 28
166.66 ± 36
108.33 ± 30
442.17 ± 38
N.D.
-3
H18A
H81A
E82A
C142A
H18A/H81A
WT
3.70×10 ± 6.08×10
-
3
8.66×10 ± 1.83×10
-
3
4.33×10 ± 1.16×10
-
2
2.16×10 ± 1.66×10
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
N.D.
Kaempferol†
33 ± 7
25 ± 2
64 ± 5
40 ± 2
41 ± 2
N.D.
3.66×10 ± 3.33×10
-3
-4
111.11 ± 25
137.24 ± 8
96.35 ± 27
157.99 ± 7
127.65 ± 4
N.D.
-
3
-4
H18A
H81A
E82A
C142A
H18A/H81A
WT
3.40×10 ±2.11×10
-
3
-3
6.16×10 ± 1.66×10
-
3
-4
6.17×10 ±1.36×10
-
3
-4
5.16×10 ±1.17×10
N.D.
N.D.
N.D.
N.D.
Resveratrol
N.D.
N.D.
WT
Pterostilbene
Rhapointigenin
N.D.
N.D.
WT
N.D.
N.D.
* Data are presented by means ± SEM (standard error of the mean, n = 3).
† Kinetic parameters for kaempferol determined for overall glucoside products. The H18A, E82A, and C142A mutants produced a mixture of glucosides
including kaempferol 3-O-β-glucoside.
N.D.: no detectable activity
Crystal structure of PaGT2. The crystal structure of apo-PaGT2
was solved by molecular replacement using the structure of Ara-
bidopsis thaliana UGT72B1 (PDB: 2VCH) and refined to 2.30 Å
resolution (Figure 2A, Table S2). The asymmetric unit (a.u.) con-
tains two PaGT2 molecules that are highly similar to each other
with overall root mean square deviation (rmsd) of 0.46 Å for over-
lap of 450 Cα atoms. The structure of PaGT2 belongs to GT-B fold
uridine diphosphate glucose (UDP-Glc) as the glucose donor (Fig-
ure 1, Table 1). The products were determined by comparing the
31
39
HPLC chromatograms with that of the standard samples. Among
the stilbenoids, PaGT2 was highly specific towards piceatannol
and exclusively formed piceatannol 4'-O-β-glucoside. Glycosyla-
tion of resveratrol produced a trace amount of both resveratrol 3-
O-β-glucoside and resveratrol 4'-O-β-glucoside. Although the
structures of pterostilbene and rhapontigenin are comparable to
resveratrol and piceatannol, respectively, no peaks for their gluco-
sylated products were observed. A previous study shows that
PaGT2 actively glycosylates different flavonoids and iso-flavo-
34
made up of two Rossmann (β/α/β) domains and a kinked C-ter-
minal helix that crossed over to the N-terminal domain , which is
53
a characteristic feature of UGTs with GT-B fold structures. The
flexible linker connecting N-terminal (Ala5-Ser243) and C-termi-
nal (Ser252-Gln466) domains is disordered in the structure. Alt-
hough PaGT2 is a monomer in solution (Figure S2), the two
PaGT2 molecules present in the a.u. formed a dimer through the
insertion of a long loop from the opposite molecule into the sub-
strate-binding pocket.
1
5,38
noids.
To elucidate the difference between the mechanism of
stilbenoid and flavonoid glycosylation, we performed glycosyla-
tion of kaempferol as a representative flavonoid. PaGT2 regiose-
lectively produced only one kaempferol glucoside (Figure 1, Table
1
). Comparison of the HPLC chromatogram with those of standard
kaempferol glucosides indicated that PaGT2 synthesized
kaempferol 3-O-β-glucoside or kaempferol 4'-O-β-glucoside. The
product was further analyzed by a nuclear magnetic resonance
(NMR) technique and determined to be kaempferol 3-O-β-gluco-
side (Figure S1).
The structural homology search using the DALI server54 shows that
the amino acid sequence and the structure of PaGT2 are highly sim-
ilar to an indoxyl glucosyltransferase PtUGT1 (PDB: 5NLM; se-
quence identity ~56%; rmsd ~1.9 Å) and a bifunctional N- and O-
55
31
glucosyltransferase AtUGT72B1 (PDB: 2VCH; sequence identity
~59%; rmsd ~2.3 Å). These UGTs are involved in glucosylation of
non-polyphenol molecules such as indoxyl sulfate and trichloro-
phenol, respectively. PaGT2 also shares high structural homology
with other plant UGTs whose structures have been solved. PaGT2
has sequence similarity of ~40% with a C-glycosyltransferase
TcCGT (PDB: 6JTD; rmsd ~3.0 Å), ~31% with a multifunctional
triterpene/flavonoid UGT71G1 (PDB: 2ACV; rmsd ~2.5 Å),
~32% with a multifunctional (iso)flavonoid glycosyltransferase
UGT85H2 (PDB: 2PQ6; rmsd ~2.4 Å), and ~24% with a flavo-
The K
be 29 µM and 33 µM, respectively. A comparison of the catalytic
efficiency (kcat/K ) showed that PaGT2 could glycosylate piceatan-
m
values for piceatannol and kaempferol were determined to
m
nol ~4.5 times more efficiently than kaempferol at identical reac-
tion conditions. Interestingly, both piceatannol and kaempferol
have four possible glycosylation sites. However, PaGT2 regiose-
lectively glycosylated only the 4' position in piceatannol and 3 po-
sitions in kaempferol. To elucidate the molecular basis for this in-
triguing reactivity of PaGT2, we performed structural analysis as
follows.
56
57
21
20
noid 3-O-glycosyltransferase VvGT1 (PDB: 2C1Z; rmsd ~3.0 Å).
4
ACS Paragon Plus Environment

Page 5 of 19
Biochemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2: Structures of PaGT2. (A) Molecules of apo-PaGT2 in the a.u. of the crystal structure. The N-terminal domain (light-green), C-terminal domain
(light-blue), and the loop inserted into the active-site (magenta, highlighted by a red oval) is indicated in one of the molecules of the a.u. The other molecule
in the a.u. is illustrated by light orange. (B) Resveratrol and UDP-2FGlc bound to one of the molecules of PaGT2 in the a.u. (C) Substrate induced domain
movements are seen from the superposition of apo-PaGT2 (light-brown) and ternary complex PaGT2 (green). Phe309 and Phe312 from the opposite
molecule of apo-PaGT2, occupying the ring structures of substrates, are also shown.
binding of UDP-2FGlc. These sugar-donor binding residues are
highly conserved among plant UGTs and their role in the recogni-
Crystal structure of PaGT2 with substrates. To understand the
catalytic mechanism in PaGT2, we obtained PaGT2 crystals with
donor analog UDP-2-fluoro-glucose (UDP-2FGlc) and stilbenoids,
tion of UDP-sugar donor has been extensively studied else-
5
7,20,32,35
where.
resveratrol, or pterostilbene (Figure 2b, Figure S3). We also per-
formed crystallization of PaGT2 with piceatannol and kaempferol;
however, diffraction quality crystals were not obtained. The struc-
ture of PaGT2 with UDP-2FGlc and resveratrol was refined to 2.60
Å resolution, and with pterostilbene to 2.65 Å resolution (Table S2).
The a.u. of each ternary complex consists of three independent
PaGT2 molecules. The comparison of apo and ternary complex
PaGT2 structures shows that in the apo-PaGT2, the substrate-bind-
ing sites are occupied by the loop from opposite molecules in the
a.u. It is seen that the sidechains of Phe309 and Phe312 from the
opposite molecule mimicked the ring structures of the substrates.
They are respectively placed at the position occupied by resor-
cinol/dimethylresorcinol (ring A) and 2-fluoro-glucose moieties of
UDP-2FGlc (Figure 2C, 3A, 3B).
The aglycones bind to the N-terminal domains, similar to the bind-
20,23
ing of acceptors in other UGTs (Figure S3).
The acceptor bind-
ing sites show the electron densities for resveratrol and pterostil-
bene in the respective ternary complexes. However, strong electron
density maps are observed only for the resorcinol or resorcinol di-
methyl ether moieties (ring A) of the acceptors. The electron den-
sity maps for the phenol moieties (ring B) are weaker in all PaGT2
molecules. The sugar-acceptors are stabilized mainly through hy-
drophobic residues, Ile85, Leu116, Phe117, Phe136, Leu146,
Val181, Pro183, Leu195, and Ala384 (Figure 3B). Polar residues,
His18, His81, and Glu82, are present at the entrance, and Cys142
and Ser138 are placed at the interior of the active site. However,
the distance between the polar residues of the enzyme and the OH
or the OCH3 groups of the acceptors is large for strong interactions
In the ternary complex of PaGT2, the C-terminal domain of each
molecule shows a clear electron density for UDP-2FGlc which in-
dicates the binding location of the sugar-donor (Figure S3). UDP-
(
Table S3). As suggested by the electron density maps, ring A of
the substrates are oriented towards the interior of the acceptor bind-
ing pocket, while ring B is exposed to solvent. Ring A of the sub-
strates is surrounded by more PaGT2 residues and stabilized by
stronger hydrophobic interactions compared to ring B. The poor
stabilization of ring B of the substrates by PaGT2 resulted in their
weak electron density maps in the crystallographic data.
5
8
2
FGlc is stabilized in PaGT2 mainly by residues from the PSPG
motif that extends from Trp343 to Gln386 (Figure 3A, Figure S4).
The uracil ring of UDP-2FGlc is placed between the sidechains of
Trp343 and Gln346 and forms hydrogen bonds with the main chain
nitrogen and oxygen atoms of Ala344. Gln346 and Glu369 form
hydrogen bonds with the OH groups on the ribose moiety. The di-
phosphate is stabilized by interaction with His361, Asn365, and
Ser366. The sidechain of Asn365 is observed to shift, making space
to bind UDP-2FGlc. Tyr383, Gln386, Glu385, and Trp364 form
hydrogen bonds and stabilize the 2-fluoro-glucose moiety. In the
structure of PaGT2, some residues outside of the PSPG motif,
Gly17, Thr137, Gln242, and Ser270, are also involved in the stabi-
lization of the sugar-donor analog. The sugar-donor binding resi-
dues are shifted towards the sugar-donor analog, induced by the
As seen in the crystal structures and from comparison with other
UGT structures, His18 is recognized as the catalytic base in PaGT2
which forms a catalytic dyad with Asp115. The catalytic histidine
assists in the deprotonation of the hydroxyl group to form an ac-
ceptor nucleophile, which is considered as the main step in glyco-
3
1,21,35,20
sylation by UGTs.
In the crystal structures of PaGT2, the
average distances from the putative glycosylation site (4'OH) on
the stilbenoid acceptor to His18 is unexpectedly long (5.5 Å). Con-
trary, we found that another histidine residue (His81) is located 4.3
5
ACS Paragon Plus Environment

Biochemistry
Page 6 of 19
We also performed molecular docking of resveratrol and pterostil-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bene to confirm the accuracy of docking (Figure S5). Molecular
docking of resveratrol and pterostilbene produced orientations of
these substrates comparable to those in the crystal structures along
with other possible orientations, which predicts the binding possi-
bility of substrates in opposite orientations than observed in the
crystal structures. The possibilities of binding of acceptors in the
active-site other than those observed in the crystal structures have
2
0,57
been predicted in UGT71G1 and VvGT1.
The crystal structures
of UGT76G1 from Stevia rebaudiana have shown that its acceptor
5
9
can bind in two opposite orientations in the active site. The bind-
ing affinity of acceptors in the observed binding modes are compa-
rable and are summarized in Table S4. Although, piceatannol and
kaempferol can also bind to the active-site in different orientations
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Figure S6), the formation of single glucoside products of each
compound with high regioselectivity indicates that the binding ori-
entations as shown in Figure 4 should be among the most suitable
binding mode of these compounds in the wildtype PaGT2.
The molecular docking explains the specificity of PaGT2 towards
piceatannol and its low activity with other stilbenoids (Figure 4A).
Although piceatannol has a similar backbone structure with
resveratrol and pterostilbene, it contains an extra OH group at the
3' position. The 3'OH group on piceatannol is seen to interact with
the sidechain of Glu82. The interaction of the 3'OH group on
piceatannol and Glu82 plays an important role in its stabilization in
the active site. Both resveratrol and pterostilbene lack such interac-
tion with Glu82 due to the absence of a hydroxyl or polar group at
the 3' position, which results in poor stabilization of these com-
pounds. In the molecular docking of piceatannol, His81 is placed
closer to the glycosylation site (the 4'OH of piceatannol) than the
conserved catalytic histidine (His18), which is similar to the obser-
vations from the crystal structures. Thus, the piceatannol docking
also indicates the possibility of both His18 and His81 as the cata-
lytic residues in PaGT2.
In the molecular docking of kaempferol, His18 lies in the proximity
to the 3OH group of kaempferol (Figure 4B). This is consistent
with the glycosylation of kaempferol which exclusively forms
kaempferol 3-O-β glucoside. The molecular docking of kaempferol
also shows that, aside from Glu82, His81 is also involved in the
stabilization of kaempferol. Glu82 and His81 interact with the O1
atom and the 4'OH group on kaempferol, respectively (Figure 4B).
Similar to His81 or Glu82, polar residues at the entrance of the ac-
2
0
55
tive site in VvGT1 (Gln84) and PtUGT1 (Glu88) are reported to
stabilize their respective substrates (Figure S7). Meanwhile, the
main-chain carbonyl oxygen of Pro78 stabilizes the substrate in
Figure 3: Substrates binding in PaGT2. (A) Residues involved in
recognition of sugar donor are shown from both apo-PaGT2 (light-
brown) and ternary complex PaGT2 (green). Donor analog in crystal
structure UDP-2FGlc (grey) and Phe312 (magenta) from the opposite
molecule in apo-crystal are shown. Hydrogen bonds are indicated with
dotted lines. (B) Resveratrol in the acceptor binding site and residues
of PaGT2 from both apo (light-brown) and complex (green) structures.
Distances between the polar sidechains and nearest hydroxyl group in
resveratrol is indicated. The corresponding glycosylation site in
piceatannol (4'OH) is indicated by *.
2
3
CtUGT78K6. These earlier studies suggest that the residues lo-
cated around the entrance of the acceptor binding site have an im-
portant role in the stabilization of substrates. Though it is inferred
from the case of piceatannol that His81 could also be involved in
catalysis to produce kaempferol 4'-O-β-glucoside, the analysis of
kaempferol glucosylation product by NMR showed only the for-
mation of kaempferol 3-O-β-glucoside, implying the non-catalytic
role of His81 in kaempferol glucosylation by WT-PaGT2.
Å away from the 4'OH group. Although the distances between both
histidines and the 4'OH groups on stilbenoids are long, His81 is
closer to the 4'OH group than His18. This suggests that His81 in
PaGT2 can be involved in catalysis in addition to the conserved
histidine (His18).
Molecular docking. Since the attempts to get the crystal structures
of PaGT2 with good acceptors piceatannol or kaempferol were not
successful, we performed molecular docking of these compounds.
6
ACS Paragon Plus Environment

Page 7 of 19
Biochemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4: Molecular docking and Mutational assay. Polar residues in the acceptor binding site of PaGT2 and (A) piceatannol and (B) kaempferol from
the molecular docking. Right-side to each figure, the glycosylation activity of PaGT2 mutants is presented in graph along with the HPLC chromatograms.
+
++
+++
( *piceatannol 4'-O-β-glucoside, **piceatannol, kaempferol 3-O-β-glucoside, kaempferol 7-O-β-glucoside, and other kaempferol glucoside. The peak
for starting material kaempferol is not shown in the HPLC chromatograms (B), which is eluted by column washing.) The glucosylation was performed
under standard reaction condition, the acceptor 50 µM, UDP-glucose 100 µM, enzyme 5 µM at 37℃ for 30min. The product concentration was divided
by enzyme concentration 5 µM and reaction time 30 min to calculate the average rate in 30 min.
sidechain being exposed to solvent allows it to function as also a
gate-keeper and prevent the escape of acceptors before glucosyla-
Site-directed mutagenesis. For further exploration of the catalytic
mechanism in PaGT2, we generated the site-directed mutants of
tion. The absence of His81 in the His81Ala mutant should have al-
the polar residues in the acceptor binding pocket and conducted en-
lowed the acceptors to easily escape from the active-site which is
zyme assays. To determine the glycosylation ability of mutants, re-
m
reflected with the increase in the K value. However, the His18Ala
action mixtures containing 50 µM acceptor, 100 µM UDP-glucose,
and 5 µM mutant enzymes in 50 mM potassium phosphate buffer
(pH 7.4) were incubated at 37 ℃ for 30 min. The reaction mixtures
were analyzed by HPLC to determine the formation of products
mutant can glycosylate piceatannol, although the catalytic effi-
ciency (kcat/K ) dropped by 45%. Moreover, the catalytic efficiency
m
of the His18Ala mutant for kaempferol glycosylation is unaffected,
though the mutant enzyme loses the regioselectivity and forms a
mixture of kaempferol glucosides including kaempferol 3-O-β glu-
coside. It is noteworthy that the substitution of His18 with Ala does
not impair the catalytic activity of PaGT2, unlike that the mutation
of the conserved histidine in many UGTs completely inactivates
(Figure 4). The enzyme kinetic studies were conducted by varying
the concentrations of acceptors. The list of mutants and their kinetic
constants are listed in Table 1. Kinetic traces for piceatannol and
kaempferol glucosylation by wildtype PaGT2 and the mutants in
the study are shown in Figure S8. A liner relationship between re-
action time and product formation was observed in 20 min.
5
7,55,20,33
the enzyme.
Therefore, we concluded that His18 is not an
only catalytic residue and there should be the second catalytic res-
idue in PaGT2. The reason the H18A mutant does not show regi-
oselectivity for kaempferol is that the mutation at the active site
with a shorter sidechain allows the binding of the acceptor in dif-
ferent conformations compared to that in the WT enzyme, which
resulted in the formation of a mixture of kaempferol glucosides.
Based on the obtained crystal structures and the mutational assay
results, we hypothesize that His81 is another catalytic residue in
PaGT2.
To clarify the role of His18 and His81 in catalysis, we substituted
each histidine with alanine (Ala) and examined their glycosylation
activity. The mutation of His81 to Ala results in an increase in the
K
m
values for both piceatannol and kaempferol by ~2 times. Be-
sides, the catalytic efficiency (kcat/K ) of the His81Ala mutant is
m
reduced. The mutant enzyme, which still maintains the activity,
shows that the conserved histidine His18 function as the catalytic
residue in PaGT2. Therefore, His81 is a key residue for substrate
binding rather than a catalytic residue. The location of His81 at the
opening of the acceptor-binding site and the flexibility of its
To further confirm the catalytic role of His81, we prepared a
His18Ala/His81Ala double mutant and a mutant which lacked 31
7
ACS Paragon Plus Environment

Biochemistry
Page 8 of 19
N-terminal residues (Δ31) including the conserved catalytic histi-
form a mixture of kaempferol glucosides including kaempferol 3-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
dine (His18). PaGT2 is completely impaired by the introduction of
His18Ala/His81Ala double mutation and does not form any detect-
able glucosides with both piceatannol and kaempferol (Figure 4).
The possibility of Glu82, Ser138, and Cys142 in the active-site is
ruled out by this result. Although these residues could be able to
generate a nucleophile by abstracting a proton from the acceptor,
the nucleophile would be too far from the C1 carbon atom on the
glucose moiety of UDP-glucose, which could have prevented the
formation of glucosylated products. In contrast, Δ31 retains the
ability to glycosylate piceatannol, although its activity is highly di-
minished (Figure S9). The deletion of 31 N-terminal residues pos-
sibly distorted the enzyme’s 3D structure and affected its glycosyl-
ation activity. The glucosylation ability of the Δ31 mutant shows
that the mutant enzyme is still able to bind the substrates albeit lack
of some secondary structures. Moreover, we also performed mo-
lecular docking of the acceptors with a PaGT2 His18/81Ala mutant
model-structure. The molecular docking results show the binding
affinity of the compounds is comparable between the wildtype
PaGT2 and the His18/81Ala PaGT2 (Table S4). Thus, we infer that
the inactivation of His18/81Ala mutant enzyme is due to the lack
of catalytic residue(s) and not due to the inability of the
His18/81Ala mutant enzyme to bind the acceptors. These results
confirm both His18 and His81 as catalytic residues in PaGT2. Fur-
thermore, the ability to form glycoside products by both His18Ala
and His81Ala mutants shows that both histidines can independently
catalyze glycosylation. We propose that PaGT2 can generate the
acceptor nucleophile in two different ways as shown in scheme S1.
Since both histidines in PaGT2 can catalyze glycosylation, we
called PaGT2 as an ambidextrous glycosyltransferase.
O-β-glucoside which is comparable to the formation of a mixture
of kaempferol glucosides by His18Ala mutant. It should be noted
that the His18Ala mutant and the Glu82Ala mutants lost the regi-
oselectivity for kaempferol glucosylation but not in the case with
the His81Ala mutant. His18 and Glu82 are located inside the ac-
tive-site, whose replacement with Ala with shorter sidechain should
have increased the volume. This could have increased the fre-
quency of binding kaempferol in different orientations rather than
the stable binding mode in the wildtype enzyme. Thus, in the
His18Ala mutant different OH groups are placed closer to His81 in
different binding modes and in the Glu82Ala mutant, in addition to
the previous possibility, both histidine could have catalyzed gluco-
sylation to form a mixture of kaempferol glucosides. Whereas, the
size of the active-site is less affected by the mutation of His81 since
it is placed at the opening of the acceptor binding pocket, which
resulted in a single product comparable to the wildtype enzyme.
The possible mechanism of nucleophile generation in kaempferol
glycosylation is shown in scheme S2.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Structure-based modification of PaGT2 for resveratrol glyco-
sylation. Resveratrol, a stilbenoid polyphenol, has a broad range of
biological activities and is a compound of interest for cosmetics,
6
0–62
food and pharmaceutical industries.
While the low water solu-
bility limits its uptake and complicates the development of thera-
peutics, the water-soluble glucoside derivatives of resveratrol are
known to have similar effects as the parent compound.1 Resvera-
trol can primarily form two glucosides, namely resveratrol 3-O-β-
glucoside and resveratrol 4'-O-β-glucoside. These glucosides show
different pharmaco-kinetics and have advantages over each
0,63
1
0,6
other. Hence, it is preferable to have an enzyme that produces
the resveratrol glucosides with high regiospecificity.
The presence of an additional catalytic residue has been predicted
in an isoflavonoid 7-O-glucosyltransferase GmIF7GT from Gly-
3
5
cine max . GmIF7GT lacking 49 N-terminal residues including the
conserved catalytic histidine is reported to form glycoside products
with the substrates. Furthermore, UGTs with non-conserved cata-
lytic residue and without the catalytic dyad has been reported. For
example, glycosylation in CsUGT75L12, a flavonoid 7-O-glyco-
syltransferase from Camellia sinensis, is catalyzed by Gln54 in-
Unfortunately, the poor stabilization of resveratrol in the active site
results in the low activity of PaGT2 towards it. Besides, it is seen
in the crystal structure that none of the hydroxyl group on resvera-
trol is close enough to both catalytic histidine for effective interac-
tion. It is well known that the mutation of amino acids in the active
site can change the size of the active-site. Thus, the mutation of
residues in the acceptor-binding pocket in PaGT2 can change the
position of resveratrol in the active-site and place one of the hy-
droxyl groups close to either of the catalytic histidines. In the crys-
tal structure, it is seen that the sidechain of Cys142 protrudes into
the active-site (Figure 5). The 3OH group of resveratrol is located
at an average distance of 4.03 Å, which is suitable for the formation
of a hydrogen bond between Cys142 and resveratrol (Table S3).
However, this interaction between Cys142 and resveratrol could be
one of the reasons that place resveratrol at an unsuitable distance,
3
6
stead of the conserved histidine (His56) . Similarly, a single histi-
dine catalyzes rhamnosylation in UGT89C1 from Arabidopsis tha-
3
2
liana , showing the His-Asp catalytic dyad is not always required
for glycosylation in some UGTs. In PaGT2 the conserved His-Asp
catalytic pair could be important for catalysis in the wildtype en-
zyme and the His81Ala mutant. Conversely, in the His18Ala mu-
tant, aspartate from the conserved His-Asp pair is less likely to have
a role in the catalysis because this aspartate residue lies far and does
not make direct contact with acceptors. Thus, the formation of gly-
cosylated products by the His18Ala mutant supports these previous
reports, where His81 is a non-conserved residue and does not re-
quire another residue to function as a catalytic pair. However, in
contrast to these previous reports, PaGT2 is the first UGT having
two active catalytic residues at the same time and probably uses
them separately depending on the cases.
We also performed the mutation of Glu82 that was determined to
be important for the stabilization of acceptors in the active site. Mu-
tation of Glu82 to Ala reduces the catalytic efficiency for piceatan-
nol glycosylation by ~5 times compared to the wild type enzyme.
The interaction between the 3'OH group and Glu82 is, therefore,
necessary for piceatannol glycosylation as previously indicated by
molecular docking. For overall kaempferol glycosylation, Glu82
does not seem to be important. However, the interaction between
Glu82 and the O1 atom of kaempferol is necessary for regioselec-
tive kaempferol glycosylation. The Glu82Ala mutant is seen to
Figure 5: Position of Cys142 in the active-site. (A) Cys142 is placed
at the rear side of acceptor binding pocket. Cys142 was mutated to
other residues for changing the size of acceptor binding pocket.
8
ACS Paragon Plus Environment

Page 9 of 19
Biochemistry
as observed in the crystal structure, from either of the catalytic his-
formed resveratrol 3-O-β-glucoside and resveratrol 4'-O-β-gluco-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tidines in the active-site for glycosylation. We assumed that the mu-
tation of Cys142 into amino-acids with varying sidechain length
will alter the position of resveratrol in the active-site. Thus, we pro-
ceeded with the substitution of Cys142 with other amino acid resi-
dues. We replaced Cys142 with Ala, Phe, or Gln, which has differ-
ent sidechain lengths. The mutation of Cys142 dramatically in-
creased the glycosylation activity of PaGT2 with resveratrol and
endowed the enzyme with the ability to form resveratrol glucosides
(Table 2). Among the three mutants, the Cys142Ala mutant had the
highest activity with resveratrol and primarily formed resveratrol
side, respectively with high regioselectivity. Thus, our results pro-
vide a basis for the modification of UGTs for efficient and regiose-
lective glycosylation of compounds of interest.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS
Publications website.
3
-O-β-glucoside with high regioselectivity. The Cys142Phe mutant
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The PDB codes for apo-PaGT2, PaGT2 with UDP-2Flc and
resveratrol, and PaGT2 with UDP-2FGlc and pterostilbene are
also had high regioselectivity but formed resveratrol 4'-O-β-gluco-
side as a major product. The Cys142Gln mutant had the lowest ac-
tivity with resveratrol as well as low regioselectivity and formed
both resveratrol 3-O-β-glucoside and resveratrol 4'-O-β-glucoside
at a nearly equal ratio. Possibly, these mutations allow resveratrol
to bind in different positions and orientations compared to its bind-
ing in wild type PaGT2, such that one of the OH group on resvera-
trol is suitably placed close to either of the catalytic histidines for
glucosylation.
6
JEL,
6JEM,
and
6JEN,
respectively.
AUTHOR INFORMATION
Corresponding Author
*hamada@dls.ous.ac.jp
*
t_inoue@phs.osaka-u.ac.jp
Table 2. Resveratrol glycosylation activity of PaGT2 Cys142
mutants
*ozakis@yamaguchi-u.ac.jp
Resveratrol 4'
glucoside
Resveratrol 3
glucoside (turn- ver/min)
Sum (turno-
ORCID
Hiroki Hamada: https://orcid.org/0000-0002-7877-3367
Tsuyoshi Inoue: https://orcid.org/0000-0001-5337-5066
Shin-ichi Ozaki:
(turnover/min) over/min)
Wild type
0
0
0
PaGT2
Author Contributions
C142Q
PaGT2
0.036 ± 0.007
0.072 ± 0.014
0.020 ± 0.004
0.007 ± 0.001
0.307 ± 0.036
0.05 ± 0.01
0.08 ± 0.014
0.31 ± 0.041
R.M., Y.F., H.H., T.I., and S.O. designed research; R.M., N.S., and
S.O. performed the assay and analyzed the results; K.K. and H.H.
performed NMR analysis. Taisuke N. established protocols for pro-
tein purification and performed the preliminary crystallographic
analysis; Toru N. contributed cloning of PaGT2; R.M. purified and
crystallized proteins; R.M. and Y.F. collected and analyzed crys-
tallographic data; R.M., Y.F., and S.O. wrote the paper with input
from all authors.
C142F
PaGT2
C142A
PaGT2
0.003 ±
0.0004
We also evaluated the effect of the Cys142 mutations on piceatan-
nol glycosylation (Table S5). The Cys142 mutations show a slight
increase in the specific activity but do not affect regioselectivity for
piceatannol glycosylation, meaning that we achieved engineering
of PaGT2 that regioselectively glucosylate both resveratrol and
piceatannol. However, the regioselectivity of kaempferol glycosyl-
ation is affected by the Cys142Ala mutant. We infer that the muta-
tion of Cys142 in PaGT2 can be utilized to modulate the size of the
active site which may be useful for changing the regioselectivity of
glycosylation of acceptors.
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
The authors wish to thank Dr. K. Fujimoto, Fuji molecular planning
co., ltd. for the synthesis of UDP-2FGlc. The authors thank the
beamline staff for their support during data collection on BL44XU
at SPring-8 under proposal Nos. 2017A6745, 2017B6745,
CONCLUSION
2
018A6844, and 2018B6844. This work was partly supported by
We characterized a polyphenol glycosyltransferase, PaGT2 from P.
americana, by determining its crystal structures, performing mo-
lecular docking, and mutagenesis studies. The structures provide
insights into the stilbenoid glycosylation mechanism. We recog-
nized a non-conserved catalytic residue (His81) is present in addi-
tion to the conserved catalytic residue (His18) in PaGT2 and show
that both catalytic residues in PaGT2 can independently catalyze
glycosylation. Due to the ability of PaGT2 to utilize both catalytic
histidines independently for glycosylation, we called it as ambidex-
trous glycosyltransferase. Additionally, the structure-guided mu-
tants of Cys142 endowed PaGT2 with the ability to regioselec-
tively glycosylate resveratrol. Cys142Ala and Cys142Phe mutants
Grant-in-Aid for Young Scientists (B) 17K17862 for YF, Grant-in-
Aid for Scientific Research (B) 18H02004 for TI, and Grant-in-Aid
for Scientific Research (C) 17K05933 for SO.
REFERENCES
1
. Lairson, L. L., Henrissat, B., Davies, G. J., and Withers, S. G.
(2008) Glycosyltransferases: structures, functions, and mecha-
nisms, Annual review of biochemistry 77, 521–555.
2. Bowles, D., Isayenkova, J., Lim, E. K., and Poppenberger, B.
(2005) Glycosyltransferases: managers of small molecules, Cur-
rent opinion in plant biology 8, 254–263.
9
ACS Paragon Plus Environment

Biochemistry
Page 10 of 19
3
. Härtl, K., Huang, F.-C., Giri, A. P., Franz-Oberdorf, K.,
17. Gurung, R. B., Kim, E.-H., Oh, T.-J., and Sohng, J. K. (2013)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Frotscher, J., Shao, Y., Hoffmann, T., and Schwab, W. (2017)
Glucosylation of Smoke-Derived Volatiles in Grapevine (Vitis vi-
nifera) is Catalyzed by a Promiscuous Resveratrol/Guaiacol Glu-
cosyltransferase, Journal of agricultural and food chemistry 65,
Enzymatic synthesis of apigenin glucosides by glucosyltransfer-
ase (YjiC) from Bacillus licheniformis DSM 13, Molecules and
cells 36, 355–361.
18. Pandey, R. P., Parajuli, P., Koirala, N., Park, J. W., and
Sohng, J. K. (2013) Probing 3-hydroxyflavone for in vitro gly-
corandomization of flavonols by YjiC, Applied and environmental
microbiology 79, 6833–6838.
5
681–5689.
4. Yoshimura, M., Amakura, Y., Tokuhara, M., and Yoshida, T.
(2008) Polyphenolic compounds isolated from the leaves of Myr-
tus communis, Journal of natural medicines 62, 366–368.
19. Zhou, M., and Thorson, J. S. (2011) Asymmetric enzymatic
glycosylation of mitoxantrone, Organic letters 13, 2786–2788.
5. Rahman, M. A. A., and Moon Surk-Sik (2007) Antioxidant
Polyphenol Glycosides from the Plant Draba nemorosa, Bulletin
of the Korean Chemical Society 28, 827–831.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
0. Offen, W., Martinez, C. F., Yang, M., Lim, E. K., Davis, B.
G., Tarling, C. A., Ford, C. M., Bowles, D. J., and Davies, G. J.
(2006) Structure of a flavonoid glucosyltransferase reveals the ba-
sis for plant natural product modification, EMBO J. 25, 1396–
1405.
6. Lepak, A., Gutmann, A., Kulmer, S. T., and Nidetzky, B.
(2015) Creating a Water-Soluble Resveratrol-Based Antioxidant
by Site-Selective Enzymatic Glucosylation, Chembiochem : a Eu-
ropean journal of chemical biology 16, 1870–1874.
21. Li, L., Modolo, L. V., Escamilla-Trevino, L. L., Achnine, L.,
Dixon, R. A., and Wang, X. (2007) Crystal structure of Medicago
truncatula UGT85H2--insights into the structural basis of a multi-
functional (iso)flavonoid glycosyltransferase, Journal of Molecu-
lar Biology 370, 951–963.
7. Akinwumi, B. C., Bordun, K.-A. M., and Anderson, H. D.
(2018) Biological Activities of Stilbenoids, International journal
of molecular sciences 19.
8. Ross, J. A., and Kasum, C. M. (2002) Dietary flavonoids: bioa-
vailability, metabolic effects, and safety, Annual review of nutri-
tion 22, 19–34.
22. He, X.-Z., Wang, X., and Dixon, R. A. (2006) Mutational
analysis of the Medicago glycosyltransferase UGT71G1 reveals
residues that control regioselectivity for (iso)flavonoid glycosyla-
tion, The Journal of biological chemistry 281, 34441–34447.
9. Quideau, S., Deffieux, D., Douat-Casassus, C., and Pouységu,
L. (2011) Plant polyphenols: chemical properties, biological activ-
ities, and synthesis, Angewandte Chemie (International ed. in
English) 50, 586–621.
23. Hiromoto, T., Honjo, E., Noda, N., Tamada, T., Kazuma, K.,
Suzuki, M., Blaber, M., and Kuroki, R. (2015) Structural basis for
acceptor-substrate recognition of UDP-glucose: anthocyanidin 3-
O-glucosyltransferase from Clitoria ternatea, Protein science : a
publication of the Protein Society 24, 395–407.
10. Hosoda, R., Kuno, A., Hori, Y. S., Ohtani, K., Wakamiya, N.,
Oohiro, A., Hamada, H., and Horio, Y. (2013) Differential cell-
protective function of two resveratrol (trans-3,5,4'-trihydroxystil-
bene) glucosides against oxidative stress, The Journal of pharma-
cology and experimental therapeutics 344, 124–132.
24. Bolam, D. N., Roberts, S., Proctor, M. R., Turkenburg, J. P.,
Dodson, E. J., Martinez-Fleites, C., Yang, M., Davis, B. G., Da-
vies, G. J., and Gilbert, H. J. (2007) The crystal structure of two
macrolide glycosyltransferases provides a blueprint for host cell
antibiotic immunity, Proceedings of the National Academy of Sci-
ences of the United States of America 104, 5336–5341.
11. Dai, L., Li, J., Yao, P., Zhu, Y., Men, Y., Zeng, Y., Yang, J.,
and Sun, Y. (2017) Exploiting the aglycon promiscuity of glyco-
syltransferase Bs-YjiC from Bacillus subtilis and its application in
synthesis of glycosides, Journal of biotechnology 248, 69–76.
2
5. Gantt, R. W., Peltier-Pain, P., Singh, S., Zhou, M., and Thor-
12. Lim, E.-K. (2005) Plant glycosyltransferases: their potential as
novel biocatalysts, Chemistry (Weinheim an der Bergstrasse, Ger-
many) 11, 5486–5494.
son, J. S. (2013) Broadening the scope of glycosyltransferase-cat-
alyzed sugar nucleotide synthesis, Proceedings of the National
Academy of Sciences of the United States of America 110, 7648–
13. Lombard, V., Golaconda Ramulu, H., Drula, E., Coutinho, P.
M., and Henrissat, B. (2014) The carbohydrate-active enzymes
database (CAZy) in 2013, Nucleic acids research 42, D490-D495.
7
653.
2
6. Gantt, R. W., Goff, R. D., Williams, G. J., and Thorson, J. S.
(
2008) Probing the aglycon promiscuity of an engineered glyco-
14. Wen, C., Huang, W., Zhu, X. L., Li, X. S., Zhang, F., and
Jiang, R. W. (2018) UGT74AN1, a Permissive Glycosyltransfer-
ase from Asclepias curassavica for the Regiospecific Steroid 3-O-
Glycosylation, Organic letters 20, 534–537.
syltransferase, Angewandte Chemie (International ed. in English)
4
7, 8889–8892.
2
7. Ma, W., Zhao, L., Ma, Y., Li, Y., Qin, S., and He, B. (2020)
Oriented efficient biosynthesis of rare ginsenoside Rh2 from PPD
by compiling UGT-Yjic mutant with sucrose synthase, Interna-
tional journal of biological macromolecules 146, 853–859.
15. Noguchi, A., Kunikane, S., Homma, H., Liu, W., Sekiya, T.,
Hosoya, M., Kwon, S., Ohiwa, S., Katsuragi, H., Nishino, T.,
Takahashi, S., Hamada, H., and Nakayama, T. (2009) Identifica-
tion of an inducible glucosyltransferase from Phytolacca ameri-
cana L. cells that are capable of glucosylating capsaicin, Plant Bi-
otechnology 26, 285–292.
2
8. Williams, G. J., Zhang, C., and Thorson, J. S. (2007) Expand-
ing the promiscuity of a natural-product glycosyltransferase by di-
rected evolution, Nature chemical biology 3, 657–662.
2
9. Kim, T.-S., Le, T. T., Nguyen, H. T., Cho, K. W., and Sohng,
1
6. Noguchi, A., Horikawa, M., Fukui, Y., Fukuchi-Mizutani, M.,
J. K. (2018) Mutational analyses for product specificity of YjiC
towards α-mangostin mono-glucoside, Enzyme and microbial
technology 118, 76–82.
Iuchi-Okada, A., Ishiguro, M., Kiso, Y., Nakayama, T., and Ono,
E. (2009) Local differentiation of sugar donor specificity of flavo-
noid glycosyltransferase in Lamiales, The Plant cell 21, 1556–
1
572.
1
0
ACS Paragon Plus Environment

Page 11 of 19
Biochemistry
3
0. Lim, E. K., and Bowles, D. J. (2004) A class of plant glyco-
42. R. Telange, D., T. Patil, A., Tatode, A., and Bhoyar, B. (2014)
Development and Validation of UV Spectrophotometric Method
for the Estimation of Kaempferol in Kaempferol: Hydrogenated
Soy PhosphatidylCholine (HSPC) Complex, Phm 5, 34–38.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
syltransferases involved in cellular homeostasis, The EMBO jour-
nal 23, 2915–2922.
3
1. Brazier-Hicks, M., Offen, W. A., Gershater, M. C., Revett, T.
J., Lim, E.-K., Bowels, D. J., Davies, G. J., and Edwards, R.
(2007) Characterization and engineering of the bifunctional N-
and O-glucosyltransferase involved in xenobiotic metabolism in
plants, Proceedings of the National Academy of Sciences of the
United States of America 104, 20238–20243.
43. Otwinowski, Z., and Minor, W. (1997) 20 Processing of X-ray
diffraction data collected in oscillation mode, Methods in enzy-
mology 276, 307–326.
4
4. Kabsch, W. (2010) XDS, Acta crystallographica. Section D,
Biological crystallography 66, 125–132.
5. Vagin, A., and Teplyakov, A. (2010) Molecular replacement
32. Zong, G., Fei, S., Liu, X., Li, J., Gao, Y., Yang, X., Wang, X.,
and Shen, Y. (2019) Crystal structures of rhamnosyltransferase
UGT89C1 from Arabidopsis thaliana reveal the molecular basis
of sugar donor specificity for UDP-β-l-rhamnose and rhamnosyla-
tion mechanism, The Plant journal : for cell and molecular biol-
ogy 99, 257–269.
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
with MOLREP, Acta crystallographica. Section D, Biological
crystallography 66, 22–25.
4
6. Winn, M. D., Ballard, C. C., Cowtan, K. D., Dodson, E. J.,
Emsley, P., Evans, P. R., Keegan, R. M., Krissinel, E. B., Leslie,
A. G. W., McCoy, A., McNicholas, S. J., Murshudov, G. N.,
Pannu, N. S., Potterton, E. A., Powell, H. R., Read, R. J., Vagin,
A., and Wilson, K. S. (2011) Overview of the CCP4 suite and cur-
rent developments, Acta crystallographica. Section D, Biological
crystallography 67, 235–242.
3
3. Ozaki, S.-i., Imai, H., Iwakiri, T., Sato, T., Shimoda, K., Na-
kayama, T., and Hamada, H. (2012) Regioselective glucosidation
of trans-resveratrol in Escherichia coli expressing glucosyltrans-
ferase from Phytolacca americana, Biotechnology letters 34, 475–
4
81.
47. Emsley, P., Lohkamp, B., Scott, W. G., and Cowtan, K.
(2010) Features and development of Coot, Acta crystallographica.
Section D, Biological crystallography 66, 486–501.
3
4. Wang, X. (2009) Structure, mechanism and engineering of
plant natural product glycosyltransferases, FEBS letters 583,
3
303–3309.
48. Kovalevskiy, O., Nicholls, R. A., Long, F., Carlon, A., and
Murshudov, G. N. (2018) Overview of refinement procedures
within REFMAC5: utilizing data from different sources, Acta
crystallographica. Section D, Structural biology 74, 215–227.
3
5. Noguchi, A., Saito, A., Homma, Y., Nakao, M., Sasaki, N.,
Nishino, T., Takahashi, S., and Nakayama, T. (2007) A UDP-
glucose:isoflavone 7-O-glucosyltransferase from the roots of soy-
bean (glycine max) seedlings. Purification, gene cloning, phyloge-
netics, and an implication for an alternative strategy of enzyme
catalysis, The Journal of biological chemistry 282, 23581–23590.
49. Schrödinger, L. L.C. (2015) The PyMOL Molecular Graphics
System, Version 1.8.
5
0. Wallace, A. C., Laskowski, R. A., and Thornton, J. M. (1995)
3
6. Dai, X., Zhuang, J., Wu, Y., Wang, P., Zhao, G., Liu, Y.,
LIGPLOT: a program to generate schematic diagrams of protein-
ligand interactions, Protein engineering 8, 127–134.
Jiang, X., Gao, L., and Xia, T. (2017) Identification of a Flavo-
noid Glucosyltransferase Involved in 7-OH Site Glycosylation in
Tea plants (Camellia sinensis), Scientific reports 7, 5926.
5
1. Dallakyan, S., and Olson, A. J. (2015) Small-molecule library
screening by docking with PyRx, Methods in molecular biology
(Clifton, N.J.) 1263, 243–250.
3
7. Wang, L., Bai, L., Nagasawa, T., Hasegawa, T., Yang, X., Sa-
kai, J.-i., Bai, Y., Kataoka, T., Oka, S., Hirose, K., Tomida, A.,
Tsuruo, T., and Ando, M. (2008) Bioactive triterpene saponins
from the roots of Phytolacca americana, Journal of natural prod-
ucts 71, 35–40.
5
2. Trott, O., and Olson, A. J. (2010) AutoDock Vina: improving
the speed and accuracy of docking with a new scoring function,
efficient optimization, and multithreading, Journal of computa-
tional chemistry 31, 455–461.
3
8. Iwakiri, T., Mase, S., Murakami, T., Matsumoto, M., Hamada,
5
3. Gibson, R. P., Tarling, C. A., Roberts, S., Withers, S. G., and
H., Nakayama, T., and Ozaki, S.-i. (2013) Glucosylation of hy-
droxyflavones by glucosyltransferases from Phytolacca ameri-
cana, Journal of Molecular Catalysis B: Enzymatic 90, 61–65.
Davies, G. J. (2004) The donor subsite of trehalose-6-phosphate
synthase: binary complexes with UDP-glucose and UDP-2-deoxy-
2-fluoro-glucose at 2 A resolution, The Journal of biological
chemistry 279, 1950–1955.
3
9. Hamada, H., Shimoda, K., Shimizu, N., Shimizu, Y., and Ak-
agi, M. (2014) Synthesis of Glycosides of Resveratrol, Pterostil-
bene, and Piceatannol by Glucosyltransferase from Phytolacca
americana Expressed in Bacillus subtilis and their Chemopreven-
tive Activity Against Cancer, Allergic, and Alzheimer’s Diseases,
Glycobiology Insights 4, 1–6.
5
4. Holm, L. (2019) Benchmarking Fold Detection by DaliLite
v.5, Bioinformatics 35, 5326–5327.
55. Hsu, T. M., Welner, D. H., Russ, Z. N., Cervantes, B.,
Prathuri, R. L., Adams, P. D., and Dueber, J. E. (2018) Employing
a biochemical protecting group for a sustainable indigo dyeing
strategy, Nature chemical biology 14, 256–261.
4
0. Yim, S. H., Lee, Y. J., Park, K. D., Lee, I. S., Shin, B. A.,
Jung, D. W., Williams, D. R., and Kim, H. J. (2015) Phenolic
Constituents from the Flowers of Hamamelis japonica Sieb. et
Zucc, Nat Prod Sci 21, 162–169.
56. He, J. B., Zhao, P., Hu, Z. M., Liu, S., Kuang, Y., Zhang, M.,
Li, B., Yun, C. H., Qiao, X., and Ye, M. (2019) Molecular and
Structural Characterization of a Promiscuous C ‐Glycosyltransfer-
ase from Trollius chinensis, Angew. Chem. 58, 11513–11520.
4
1. Rhayem, Y., Thérond, P., Camont, L., Couturier, M.,
Beaudeux, J.-L., Legrand, A., Jore, D., Gardés-Albert, M., and
Bonnefont-Rousselot, D. (2008) Chain-breaking activity of
resveratrol and piceatannol in a linoleate micellar model, Chemis-
try and physics of lipids 155, 48–56.
1
1
ACS Paragon Plus Environment

Biochemistry
Page 12 of 19
5
7. Shao, H., He, X., Achnine, L., Blount, J. W., Dixon, R. A.,
61. Ligt, M. de, Timmers, S., and Schrauwen, P. (2015) Resvera-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and Wang, X. (2005) Crystal structures of a multifunctional triter-
pene/flavonoid glycosyltransferase from Medicago truncatula,
The Plant cell 17, 3141–3154.
trol and obesity: Can resveratrol relieve metabolic disturbances?,
Biochimica et biophysica acta 1852, 1137–1144.
6
2. Stivala, L. A., Savio, M., Carafoli, F., Perucca, P., Bianchi, L.,
58. Lim, E.-K., and Bowles, D. J. (2004) A class of plant glyco-
syltransferases involved in cellular homeostasis, The EMBO jour-
nal 23, 2915–2922.
Maga, G., Forti, L., Pagnoni, U. M., Albini, A., Prosperi, E., and
Vannini, V. (2001) Specific structural determinants are responsi-
ble for the antioxidant activity and the cell cycle effects of
resveratrol, The Journal of biological chemistry 276, 22586–
5
9. Yang, T., Zhang, J., Ke, D., Yang, W., Tang, M., Jiang, J.,
2
2594.
Cheng, G., Li, J., Cheng, W., Wei, Y., Li, Q., Naismith, J. H., and
Zhu, X. (2019) Hydrophobic recognition allows the glycosyltrans-
ferase UGT76G1 to catalyze its substrate in two orientations, Na-
ture communications 10, 3214.
6
3. Shimoda, K., Kubota, N., Uesugi, D., Hamada, H., Tanigawa,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
M., and Hamada, H. (2015) Synthesis and pharmacological evalu-
ation of glycosides of resveratrol, pterostilbene, and piceatannol,
Annals of the New York Academy of Sciences 1348, 141–149.
60. Dirks-Hofmeister, M. E., Verhaeghe, T., De Winter, K., and
Desmet, T. (2015) Creating Space for Large Acceptors: Rational
Biocatalyst Design for Resveratrol Glycosylation in an Aqueous
System, Angew. Chem. 127, 9421–9424.
1
2
ACS Paragon Plus Environment

Page 13 of 19
Biochemistry
Insert Table of Contents artwork here
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
ACS Paragon Plus Environment

Biochemistry
Page 14 of 19
1
2
3
4
5
6
7
8
9
1
1
1
1
1
0
1
2
3
4
ACS Paragon Plus Environment

Page 15 of 19
Biochemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
7
ACS Paragon Plus Environment

Biochemistry
Page 16 of 19
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
ACS Paragon Plus Environment

Page 17 of 19
Biochemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
ACS Paragon Plus Environment

Biochemistry
Page 18 of 19
1
2
3
4
5
6
7
8
ACS Paragon Plus Environment

Page 19 of 19
Biochemistry
1
2
3
4
5
6
7
ACS Paragon Plus Environment