Glucosylation of hydroxyflavones by glucosyltransferases from Phytolacca americana

Article abstract of DOI:10.1016/j.molcatb.2013.01.016

Cell suspension cultures of Phytolacca americana can glucosylate 6- and 7-hydroxyflavone, but not 5-hydroxyflavone. In order to identify the enzymes responsible for these transformations, glucosyltransferases (GTs) from P. americana were overexpressed in Escherichia coli and purified. The purified PaGT3 enzyme could glucosylate 6- and 7-hydroxyflavone when incubated with UDP-glucose, a glucosyl donor molecule, but PaGT2 could conjugate a glucose moiety only to 6-hydroxyflavone. E. coli cells expressing PaGT2 and 3 could also be utilized for the glucosylation of hydroxyflavones. The glucoside products which had accumulated in the medium of overnight E. coli cell cultures were isolated using hydrophobic resins. This methodology might be suitable for the glucosylation of aglycones with important health-related properties.

Full text of DOI:10.1016/j.molcatb.2013.01.016

Journal of Molecular Catalysis B: Enzymatic 90 (2013) 61–65
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Glucosylation of hydroxyflavones by glucosyltransferases from Phytolacca
americana
Tomoya Iwakiria, Shogo Masea, Tomonori Murakamia, Masahiro Matsumotob,
Hiroki Hamadab,∗, Toru Nakayamac, Shin-ichi Ozakia,∗∗
a Department of Biological Sciences, Faculty of Agriculture, Yamaguchi University, 1677-1 Yoshida, Yamaguchi 753-8515, Japan
b Department of Life Sciences, Faculty of Science, Okayama University of Sciences, 1-1 Ridai-cho, Kita-ku, Okayama 700-0005, Japan
c Department of Biomolecular Engineering, Graduate School of Engineering, Tohoku University, Sendai, Miyagi 980-8579, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Cell suspension cultures of Phytolacca americana can glucosylate 6- and 7-hydroxyflavone, but not
5-hydroxyflavone. In order to identify the enzymes responsible for these transformations, glucosyltrans-
ferases (GTs) from P. americana were overexpressed in Escherichia coli and purified. The purified PaGT3
enzyme could glucosylate 6- and 7-hydroxyflavone when incubated with UDP-glucose, a glucosyl donor
molecule, but PaGT2 could conjugate a glucose moiety only to 6-hydroxyflavone. E. coli cells expressing
PaGT2 and 3 could also be utilized for the glucosylation of hydroxyflavones. The glucoside products which
had accumulated in the medium of overnight E. coli cell cultures were isolated using hydrophobic resins.
This methodology might be suitable for the glucosylation of aglycones with important health-related
properties.
Received 28 September 2012
Received in revised form
27 December 2012
Accepted 16 January 2013
Available online 23 January 2013
Keywords:
Glucosyltransferase
Glucoside
Hydroxyflavone
Phytolacca americana
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
bacteria [7]. It is also known that plants produce a variety of gly-
cosylated metabolites. For example, glycoconjugates of aromatic
Glycosylation is one of the mechanisms for converting lipophilic
molecules into hydrophilic compounds in order to regulate cellu-
lar homeostasis [1,2]. When the gene encoding ZOG1, a Phaseolus
enzyme shown to O-glycosylate trans-zeatin, was overexpressed
in tobacco callus, much higher levels of supplementary trans-
zeatin were required for induction of shoot differentiation [3].
In Arabidopsis plants, the gene encoding UGT84B1, a glycosyl-
transferase for indol-3-acetic acid, was overexpressed, and the
resulting transgenic plants displayed a phenotype resembling
auxin deficiency [4]. The synthesis of glycosides is normally
performed by glucosyltransferases (GTs) catalyzing the trans-
fer of the glucose moiety of a nucleotide-activated sugar donor
molecule (e.g., UDP-glucose) to an aglycone [5]. In vertebrates,
the formation of water-soluble metabolites originating from the
exposure to drugs and xenobiotics appears to be a conven-
tional detoxification process [6]. In microorganisms, some GTs
are involved in the production of peptidoglycan, which is the
major component of the bacterial cell wall; therefore, GTs can
be potential targets for drugs against infections by pathogenic
volatiles have been reported in a number of fruits, including grape,
apricot, mango, passion fruit, and kiwi [8–12]. Odorous aglycones
might be released from the glycosylated volatiles during matu-
ration. The production of anthocyanin is another example of a
glucosylation reaction in plants [13]. The compound synthesized
in red grape is a colorant that determines flower and fruit color
[14,15]. Glucosylation leads to the transport of anthocyanin from
the cytosol to the vacuole as the fruit ripens. More than 10,000 GT
sequences from various organisms have now been collected and
stored in the Carbohydrate-active Enzymes Database [16].
Phytolacca americana (Pa) is a perennial plant native to North
America. Although the plant is toxic, it is used as a traditional
herbal medicine in China [17,18]. The following compounds
have been identified in the leaves of P. americana: kaempferol
3-O-␤-d-glucopyranoside, kaempferol 3-O-␤-d-xylopyranos-
yl (1→2)-␤-d-glucopyranoside, kaempferol 3-O-␣-l-rhamnop-
yranosyl (1→2)-␤-d-glucopyranoside, kaempferol 3-O-diglucoside
and quercetin 3-O-glucoside [19]. The glucosides of triterpene
saponins (e.g., esculentoside B, phytolaccoside E, and esculen-
toside G) were extracted from the root of P. americana, and
their structures were established on the basis of spectroscopic
analysis [18]. Since these triterpene saponin glucosides were also
obtained from cell cultures of P. americana, we examined if the cell
suspensions could transform exogenous aglycones into their glu-
cosides. Perillyl alcohol, 2,2,5,7,8-pentamethyl-6-chromanol,
∗
Corresponding author. Tel.: +81 86 256 9473; fax: +81 86 256 8468.
Corresponding author. Tel.: +81 83 933 5865; fax: +81 83 933 5820.
E-mail addresses: hamada@dls.ous.ac.jp (H. Hamada), ozakis@yamaguchi-u.ac.jp
∗∗
(S.-i. Ozaki).
1381-1177/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2013.01.016

62
T. Iwakiri et al. / Journal of Molecular Catalysis B: Enzymatic 90 (2013) 61–65
and resveratrol were converted to the corresponding
␤-glucosides by P. americana cell cultures [20–22]. Inter-
100 mL of MS liquid medium, and cultured with continuous shak-
ing at 25 ◦C. 5-, 6-, or 7-hydroxyflavone (20 ␮mol) was dissolved
in dimethyl sulfoxide (200 ␮L) and added to the flask. The cells
were incubated for an additional 2 days at 25 ◦C. Following incu-
bation, the cells were isolated by filtration, disrupted by sea sand,
and extracted with methanol (200 mL) for 24 h. The methanol frac-
tion was concentrated and partitioned between water (100 mL) and
ethyl acetate (300 mL), and the ethyl acetate fraction was analyzed
by high-performance liquid chromatography (HPLC). The yield was
estimated on the basis of peak area of hydroxyflavone and its glu-
coside.
estingly,
4-[4-(␤-d-glucopyranosyloxy)phenyl]-2-butanone,
4-[(3S)-3-hydroxybutyl]phenyl-␤-d-glucopyranoside, (2S)-4-(4-
hydroxyphenyl)but-2-yl-␤-d-glucopyranoside, 2-hydroxy-4-[(3S)
-3-hydroxybutyl]phenyl-␤-d-glucopyranoside, and 2-hydroxy-
5-[(3S)-3-hydroxybutyl]phenyl-␤-d-glucopyranoside
were
produced from 4-(4-hydroxyphenyl)butan-2-one (raspberry
ketone), suggesting that the cell suspension cultures of P. amer-
icana could glucosylate, reduce, and hydroxylate the exogenous
substrate (Supplementary Fig. 1) [23].
Supplementary material related to this article found, in the
online version, at http://dx.doi.org/10.1016/j.molcatb.2013.01.016.
In order to investigate the glucosyltransferases of P. amer-
icana, we isolated three glucosyltransferase cDNAs (PaGT1–3)
from the cultured cells, heterologously overexpressed PaGT3 in
Escherichia coli, and purified the enzyme [24]. It was found that
purified PaGT3 enzyme could glucosylate resveratrol, capsaicin,
kaempferol, quercetin, apigenin, genistein, and naringenin in the
presence of UDP-glucose (Supplementary Fig. 2) [22,24]. When
PaGT3-expressing E. coli cells were cultured in the presence of
resveratrol, the glucoside products of resveratrol were detected in
the medium of overnight culture [22]. The addition of UDP-glucose
was not required because glucosyl donor molecules were regener-
ated in E. coli cells.
2.3. Time course experiments
Callus tissue (5 g) was cultured in an Erlenmeyer flask contain-
ing 50 mL of MS liquid medium with continuous shaking at 25 ◦C
for 2 days. 5-, 6-, or 7-hydroxyflavone (10 ␮mol) was dissolved in
dimethyl sulfoxide (100 ␮L) and added to the flask. After incuba-
tion for 2, 6, 24, and 48 h, the cells were isolated by filtration and
disrupted by sea sand. The extraction and analysis procedures were
as described in Section 2.2. Products yield was determined based
on HPLC peak area and expressed as a relative percentage to the
total amount of the whole reaction product extracted.
2.4. Glucosylation of hydroxyflavones by purified PaGT enzymes
Supplementary material related to this article found, in the
online version, at http://dx.doi.org/10.1016/j.molcatb.2013.01.016.
The amino acid sequence of PaGT3 is homologous to that of
flavonol 3-O-glucosyltransferase from A. thaliana (identity 56%)
[25] and Fragaria x ananassa (identity 57%) [26], anthocyanin 3ꢀ-
O-glucosyltransferase from Gentiana triflora (identity 56%) [27],
and flavonol 7-O-glucosyltransferase from A. thaliana (identity
54%) [25]. Sequence identity between PaGT3 and PaGT2 is 27%,
and PaGT2 is related to hydroquinone glucosyltransferase from
Rauvolfia serpentine (identity 58%) [28] and coniferyl alcohol 4-O-
glucosyltransferase from A. thaliana (identity 39%) [29].
In this study, we explored the regiospecificity of glucosylation
reaction by PaGTs. Although apigenin (4ꢀ,5,7-trihydroxyflavone)
was known to be glucosylated by PaGT3 [24], the structure of
glucoside product was not determined at the moment. Thus, we
examined whether 5-, 6-, and 7-hydroxyflavone could be glucosy-
lated to produce ␤-O-glucosides by cell suspension cultures of P.
americana, E. coli expressing PaGTs, and the purified enzymes. The
PaGT3 enzyme could glucosylate 6- and 7-hydroxyflavone, while
the PaGT2 enzyme could transfer only the glucosyl group to 6-
hydroxyflavone. Neither PaGT3 nor PaGT2 could produce glucoside
from 5-hydroxyflavone. Since prodrugs, procosmetics, and prosup-
plements have recently attracted a great deal of attention, there is
a growing need for techniques that are capable of modifying func-
tional compounds. Glucosylation is a useful chemical modification
that can be used to stabilize biologically active compounds or to
activate physiological functions.
PaGT2 cDNA was previously cloned into pET32a, and the
resultant plasmid was transformed into BL21(DE3) E. coli cells;
however, the expression level was low [24]. In order to improve
the expression level, we amplified the coding sequence of
PaGT2 from PaGT2-pET32a by polymerase chain reaction (PCR)
using the primers, 5ꢀ-CGGCATGCATGGAAATGGAAGCAC-3ꢀ and 5ꢀ-
GCGTCGACTTAGCTTTTGCATTGG-3ꢀ. The amplified fragment was
digested with SalI and SphI and then ligated into pQE30 contain-
ing a 6×-His tag at the N-terminus. The resulting plasmid was
transformed into M15 E. coli cells. PaGT3-pQE30 constructed in our
previous study [24] was also transformed into M15 E. coli cells.
The transformants were cultured with continuous shaking at
30 ◦C in Terrific broth (TB) medium containing 50 ␮g/mL ampicillin.
TB medium contained tryptone 10 g, yeast extract 24 g, glycerol
4 g, KH2PO4 2.31 g, and K2HPO4 12.54 g per 1 L of deionized water.
At OD600 = 1, isopropyl ␤-d-1-thiogalactopyranoside (IPTG) was
added at 0.1 mM, followed by incubation at 30 ◦C for an additional
16 h. The harvested cells were resuspended in 15 mM potassium
phosphate containing 1 mM EDTA and 2 mM 2-mercaptoethanol.
The cells were lysed by sonication, and the PaGT enzymes were
purified from the supernatant using a His-Accept column (Nakalai).
The enzyme solution eluted with the buffer supplemented with
200 mM imidazole was dialyzed with 10 mM Tris–HCl containing
5 mM dithiothreitol and NaCl 100 mM (pH 7.2), concentrated, and
stored at −80 ◦C.
Glucosylation reactions were performed at 37 ◦C for 60 min in
0.5 mL of 50 mM potassium phosphate buffer (pH 7.2) supple-
mented with 50 ␮M hydroxyflavone, 200 ␮M UDP-glucose, and
5 ␮M enzyme. The incubation was stopped by adding 1.5% trifluo-
roacetic acid, and the reaction mixture was analyzed by HPLC to
detect the glucoside products. When the Km and kcat values were
determined, the concentration of hydroxyflavone was varied as fol-
lows: For glucosylation of 6-hydroxyflavone by PaGT3, 50–150 ␮M;
for glucosylation of 7-hydroxyflavone by PaGT3, 50–200 ␮M; for
glucosylation of 6-hydroxyflavone by PaGT2, 10–120 ␮M. The aver-
age values from more than three independent experiments were
used to construct Lineweaver–Burk plots.
2. Experimental
2.1. Materials
All chemicals and enzymes were purchased from Wako, Nakalai,
Sigma–Aldrich, and Toyobo.
2.2. Glucosylation of hydroxyflavones with cultured plant cells
Callus tissue from P. americana was prepared as described previ-
ously [22]. Briefly, callus tissue (30 g) was cultured under light for
4 weeks, transferred to an Erlenmeyer flask (300 mL) containing
To hydrolyze the glucosylated products, the reaction mixture
(500 ␮l) containing ␤-glucosidase from sweet almond (1 unit) and
the glucosides in 50 mM potassium phosphate buffer (pH 7.2) was

T. Iwakiri et al. / Journal of Molecular Catalysis B: Enzymatic 90 (2013) 61–65
63
Fig. 1. HPLC profiles of the products obtained from (a) 5-hydroxyflavone, (b) 6-hydroxyflavone, and (c) 7-hydroxyflavone following incubation with P. americana cells. Peak
1: 5-hydroxyflavone; Peak 2: 6-hydroxyflavone monoglucoside; Peak 3: 6-hydroxyflavone; Peak 4: 7-hydroxyflavone monoglucoside; Peak 5: product X (a monoglucoside
of product Y); Peak 6: 7-hydroxyflavone; and Peak 7: product Y. The structures of glucosylated products from 6-hydroxyflavone and 7-hydroxyflavone are shown in (d) and
(e), respectively.
incubated for 90 min at 37 ◦C. The reaction mixture was analyzed
by HPLC to examine the conversion of glucoside products to the
corresponding hydroxyflavones.
3. Results and discussion
3.1. Glucosylation of hydroxyflavones by cell suspension cultures
of P. americana
2.5. Glucosylation of hydroxyflavones in E. coli cultures
Cell suspension cultures of P. americana did not glucosylate
5-hydroxyflavone (Fig. 1(a)), but transformed 6-hydroxyflavone
into the corresponding glucoside with a yield of approximately
40% (Fig. 1(b)). Under reverse-phase HPLC conditions, the gluco-
side product (Fig. 1(b), peak 2) eluted faster than the substrate
(Fig. 1(b), peak 3), and the mass spectrum of the isolated
product showed a molecular ion [M+H]+ peak at m/z 400.8
(162(gluc) + 238(6-hydroxyflavone)), suggesting the monoglucosy-
lation of 6-hydroxyflavone. Furthermore, upon treatment of the
glucosylated product with ␤-glucosidase, the glucoside peak in
the HPLC profile completely disappeared, and the aglycone was
regenerated in the reaction mixture. These results clearly indi-
cate that the cell suspension cultures of P. americana produced the
␤-O-glucoside of 6-hydroxyflavone, 6-(␤-d-glucopyranosyloxy)-2-
phenyl-4H-1-benzopyran-4-one (Fig. 1(d)).
E. coli M15 cells bearing pQE-PaGT2 or pQE-PaGT3 were cul-
tured in 10 mL LB medium containing 1 mg hydroxyflavone and
0.5 mg ampicillin. Hydroxyflavone (1 mg) was previously dissolved
in 0.5 mL of dimethyl sulfoxide and added to the medium. At
16 h after induction with 0.1 mM IPTG, the cells and medium
were separated. Isolation of the accumulated glucosylated prod-
ucts in the medium was easily achieved by adsorption with Diaion
HP20, a polyaromatic adsorbent resin for hydrophobic compounds.
Diaion HP20 (1 g) was added to the medium and mixed gently
by shaking at room temperature for 60 min. The medium-HP20
mixture was then loaded onto the column. HP20 was washed
twice with 5 mL of water, and the glucoside products were eluted
with 7.5 mL of methanol. The methanol solution was analyzed
by HPLC to examine the production of glucosides. The yield
was estimated based on the peak area of hydoxyflavone and its
glucoside.
Somewhat interestingly, we obtained the following three prod-
ucts during the incubation of 7-hydroxyflavone: 7-hydroxyflavone
2.6. LC and LC–MS analysis
The substrates and glucosylated products were separated
using
a
Crestpak C18S (4.6 mm × 150 mm) or an eco-ODS
(4.6 mm × 150 mm). The ratio of water and acetonitrile was linearly
increased from 85:15 (v/v) at t = 0 min to 60:40 (v/v) at t = 40 min.
The flow rate was 1 mL/min, and the UV–visible detector was set at
300 nm. A Unison UK-C18 (2.0 mm × 150 mm) was used for liquid
chromatography–mass spectrometry (LC–MS) analysis with a flow
rate of 0.2 mL/min. The mass was performed in positive ion mode
on an AB Sciex 3200 Q TRAP. The ESI conditions were as follows:
declustering potential (DP), 30 V; entrance potential EP), 10 V; cur-
tain gas, 40 (arbitrary units); collision gas, high; ion spray voltage,
5.5 kV; source temperature, 350 ◦C; ion source gas 1, 50 (arbitrary
units); ion source gas 2, 80 (arbitrary units); and MS full scan range:
m/z 100–450.
Fig. 2. Time course of the biotransformation of (a) 6-hydroxyflavone and (b) 7-
hydroxyflavone by cultured cells of P. americana. Yield is expressed as a percentage
of the total amount of reaction products. Yields are shown by symbols as follows: 6-
hydroxyflavone monoglucoside (ꢁ); 7-hydroxyflavone monoglucoside (᭹); product
X (ꢀ); and product Y (ꢁ).

64
T. Iwakiri et al. / Journal of Molecular Catalysis B: Enzymatic 90 (2013) 61–65
was transformed into product Y after incubation for 2 h (Fig. 2(b)).
Then, 7-hydroxyflavone glucoside and product X accumulated
with a concomitant decrease in the amount of 7-hydroxyflavone
and product Y. These results indicate that product Y was formed
first and that further glucosylation yielded product X. Finally,
it should be noted that the use of cell suspension cultures is
advantageous for the synthesis of glucosides because they do not
require expensive UDP-glucose. P. americana cells can regenerate
nucleotide-activated sugars as glucosyl donor molecules.
3.2. Glucosylation of hydroxyflavone by purified PaGT2 and
PaGT3 enzymes
To identify the enzymes responsible for the glucosylation of
hydroxyflavone in cell suspension cultures, we performed an
in vitro assay using the purified enzymes. Previously, the three gly-
cosyltransferase genes encoding PaGT1, 2, and 3 were isolated from
P. americana, and PaGT3 expressed in E. coli was purified [24]. This
procedure, with some modifications, allowed us to obtain purified
PaGT2 as well as PaGT3 (Fig. 3). The typical yields were 5 mg/L of
E. coli culture.
Purified PaGT3 enzyme could catalyze the glucosylation of 6-
and 7-hydroxyflavone when UDP-glucose was used as the gluco-
syl donor. The extracts from the reaction mixtures were analyzed
by LC–MS, and mass ions corresponding to the respective glu-
cosides of hydroxyflavones were confirmed (Fig. 4). Since the
glucosides were hydrolyzed upon reaction with ␤-glucosidase, we
concluded that the products contained ␤-O-glucosidic bonds. The
Km and kcat values for 6-hydroxyflavone were 70 7.7 ␮M and
(1.4 0.2) × 10−2 s−1, respectively. When 7-hydroxyflavone was
used as an acceptor (i.e., an aglycone), we observed a 2-fold increase
in the Km value and a 1.5-fold decrease in the kcat value (kcat
(9.4 0.8) × 10−3 s−1, Km 140 10 ␮M). These results indicate that
6-hydroxyflavone is a better substrate than 7-hydroxyflavone for
PaGT3. The purified PaGT2 enzyme could also catalyze the glu-
cosylation of 6-hydroxyflavone, albeit inefficiently; the Km and
kcat values were 46 6.4 ␮M and (1.7 0.2) × 10−4 s−1, respec-
tively. The Km value did not differ significantly from that of PaGT3,
but the kcat value decreased by approximately 100-fold. When
7-hydroxyflavone and UDP-glucose were incubated with PaGT2,
the corresponding glucoside product was not detected. Further-
more, neither PaGT2 nor PaGT3 was able to conjugate glucose to
5-hydroxyflavone.
Fig. 3. Coomassie Brilliant Blue (CBB) staining of purified (1) PaGT2 and (2) PaGT3
in SDS-PAGE.
monoglucoside, product X, and product Y (Fig. 1(c), peaks 4, 5,
7). LC–MS analysis indicated the production of 7-hydroxyflavone
monoglucoside (Fig. 1(c), peak 4, a molecular ion [M+H]+ at
m/z 400.9, 162(gluc) + 238(7-hydroxyflavone)). We also confirmed
that the glucoside product was hydrolyzed by ␤-glucosidase.
The molecular ions [M+H]+ associated with product X (Fig. 1(c),
peak 5) and product Y (Fig. 1(c), peak 7) were detected at m/z
430.9 (162(gluc) + 238(7-hydroxyflavone) + 30) and 268.9 (238(7-
hydroxyflavone) + 30), respectively, in the mass spectra. Product X
appears to be a monoglucosylated adduct of product Y. It was pre-
viously reported that cell suspension cultures of P. americana could
reduce the carbonyl group and the hydroxylate phenyl ring of 4-(4-
hydroxyphenyl)butan-2-one [23]. The increase of m/z by 30 might
be related to these activities, but the exact structures of products X
and Y remain unknown.
The time course experiments indicated that approximately 40%
of 6-hydroxyflavone was converted into the glucoside product after
incubation for 24 h (Fig. 2(a)). On the other hand, 7-hydroxyflavone
Fig. 4. HPLC profiles of the glucosylated products obtained from (a) 6-hydroxyflavone and (b) 7-hydroxyflavone following incubation with purified PaGT3 enzyme. The
glucoside products (*) eluted faster than the hydroxyflavones. The mass spectra of the glucosides from (c) 6-hydroxyflavone and (d) 7-hydroxyflavone. The spectra show
molecular ions [M+H]+ at m/z 401.3 (162(gluc) + 238(6-hydroxyflavone)) in (c) and 400.9 (162(gluc) + 238(7-hydroxyflavone)) in (d).

T. Iwakiri et al. / Journal of Molecular Catalysis B: Enzymatic 90 (2013) 61–65
65
Our results on the synthesis of hydroxyflavone glucosides by the
purified enzymes suggest that PaGT2 and PaGT3 are responsible
for the biotransformation of 6-hydroxyflavone in P. americana. We
also found that PaGT3 can accept 7-hydroxyflavone as a glucose
acceptor, although the conversion was less efficient.
[2] T. Vogt, P. Jones, Trends Plant Sci. 5 (2000) 380–386.
[3] R. Martin, D. Mok, R. Smets, H. van Onckelen, M. Mok, In Vitro Cell. Dev. Biol.
Plant 37 (2001) 354–360.
[4] R.G. Jackson, M. Kowalczyk, Y. Li, G. Higgins, J. Ross, G. Sandberg, D.J. Bowles,
Plant J. 32 (2002) 573–583.
[5] X. Wang, FEBS Lett. 583 (2009) 3303–3309.
[6] R.H. Tukey, C.P. Strassburg, Annu. Rev. Pharmacol. Toxicol. 40 (2000)
581–616.
[7] A.L. Lovering, S.S. Safadi, N.C.J. Strynadka, Annu. Rev. Biochem. 81 (2012)
451–478.
3.3. Glucosylation of hydroxyflavones by E. coli expressing PaGTs
[8] F. Parada, C. Duque, HRC J. High Resolut. Chromatogr. 21 (1998) 577–581.
[9] F. Parada, C. Duque, Y. Fujimoto, J. Agric. Food Chem. 48 (2000)
6200–6204.
[10] G. Reineccius, Flavor Chemistry and Technology, 2nd ed., Taylor and Francis,
Boca Raton, FL, 2006.
[11] C.V. Garcia, S.Y. Quek, R.J. Stevenson, R.A. Winz, J. Agric. Food Chem. 59 (2011)
8358–8365.
[12] C.V. Garcia, S.Y. Quek, R.J. Stevenson, R.A. Winz, Food Chem. 134 (2012)
655–661.
[13] W. Offen, C. Martinez-Fleites, M. Yang, E. Kiat-Lim, B.G. Davis, C.A. Tarling, C.M.
Ford, D.J. Bowles, G.J. Davies, EMBO J. 25 (2006) 1396–1405.
[14] S. Shimada, Y.T. Inoue, M. Sakuta, Plant J. 44 (2005) 950–959.
[15] S. Shimada, K. Takahashi, Y. Sato, M. Sakuta, Plant Cell Physiol. 44 (2004)
1290–1298.
[16] P.M. Coutinho, E. Deleury, G.J. Davies, B. Henrissat, J. Mol. Biol. 328 (2003)
307–317.
[17] H. Takahashi, Y. Namikawa, M. Tanaka, Y. Fukuyama, Chem. Pharm. Bull. 49
(2001) 246–248.
We attempted to use E. coli expressing PaGTs to produce the
glucosides because E. coli grows faster than P. americana and can
regenerate UDP-glucose. When PaGT3-expressing E. coli were cul-
tured in the presence of hydroxyflavones, 6- and 7-hydroxyflavone
glucosides were detected in the medium of overnight cultures (Sup-
plementary Fig. 3(a) and (b)). In these incubations, endogenous
glucosyl donor molecules were sufficient for the reaction, and we
did not need to add UDP-glucose.
The conversion rates by PaGT3-expressing E. coli were 15% and
20% for 6- and 7-hydroxyflavone, respectively. 6-Hydroxyflavone
was also transformed by PaGT2-expressing E. coli with a yield of
16% (Supplementary Fig. 3(c)). However, 5-hydorxyflavone gluco-
side was not synthesized by PaGT-expressing E. coli. These results
were consistent with those of the incubations using the purified
enzymes.
[18] L. Wang, L. Bai, T. Nagasawa, T. Hasegawa, X. Yang, J.-I. Sakai, Y. Bai, T. Kataoka,
S. Oka, K. Hirose, A. Tomida, T. Tsuruo, M. Ando, J. Nat. Prod. 71 (2008)
35–40.
[19] W. Bylka, I. Matlawska, Acta Poloniae Pharmaceutica – Drug Res. 58 (2001)
69–72.
4. Conclusions
[20] K. Shimoda, S. Sakamoto, N. Nakajima, H. Hamada, H. Hamada, Chem. Lett. 37
(2008) 556–557.
[21] Y. Kondo, K. Shimoda, J. Takimura, H. Hamada, H. Hamada, Chem. Lett. 35 (2006)
324–325.
[22] S. Ozaki, H. Imai, T. Iwakiri, T. Sato, K. Shimoda, T. Nakayama, H. Hamada,
Biotechnol. Lett. 34 (2012) 475–481.
[23] K. Shimoda, T. Harada, H. Hamada, N. Nakajima, H. Hamada, Phytochemistry
68 (2007) 487–492.
[24] A. Noguchi, S. Kunikane, H. Homma, W. Liu, T. Sekiya, M. Hosoya, S. Kwon, S.
Ohiwa, H. Katsuragi, T. Nishino, S. Takahashi, H. Hamada, T. Nakayama, Plant
Biotechnol. 26 (2009) 285–292.
[25] E.K. Lim, D.A. Ashford, B. Hou, R.G. Jackson, D.J. Bowles, Biotechnol. Bioeng. 87
(2004) 623–631.
[26] M. Griesser, F. Vitzthum, B. Fink, M.L. Bellido, C. Raasch, J. Munoz-Blanco, W.
Schwab, J. Exp. Bot. 59 (2008) 2611–2625.
[27] M. Fukuchi-Mizutani, H. Okuhara, Y. Fukui, M. Nakao, Y. Katsumoto, K.
Yonekura-Sakakibara, T. Kusumi, T. Hase, Y. Tanaka, Plant Physiol. 132 (2003)
1652–1663.
[28] J. Arend, H. Warzecha, J. Stockigt, Phytochemistry 53 (2000)
187–193.
[29] A. Lanot, D. Hodge, R.G. Jackson, G.L. George, L. Elias, E.K. Lim, F.E. Vaistij, D.J.
Bowles, Plant J. 48 (2006) 286–295.
[30] A. Crozier, D. Del Rio, M.N. Clifford, Mol. Aspects Med. 31 (2010)
446–467.
[31] I. Erlund, Nutr. Res. 24 (2004) 851–874.
[32] B. Halliwell, J. Rafter, A. Jenner, Am. J. Clin. Nutr. 81 (2005)
268S–276S.
[33] J.A. Ross, C.M. Kasum, Annu. Rev. Nutr. 22 (2002) 19–34.
[34] K. Ishihara, H. Hamada, T. Hirata, N. Nakajima, J. Mol. Catal. B: Enzym. 23 (2003)
145–170.
Cell suspension cultures of P. americana could glucosylate 6- and
7-hydroxyflavone, but not 5-hydroxyflavone. Two glucosyltrans-
ferases, PaGT2 and 3, were overexpressed in E. coli and purified
successfully. Incubation with the purified enzymes clearly indi-
cated that PaGT3 could transform 6- and 7-hydroxyflavone into the
corresponding glucoside and that 6-hydroxyflavone was a better
glucose-acceptor molecule. In contrast, the PaGT2 enzyme could
only glucosylate 6-hydroxyflavone. We also proved that E. coli
expressing the PaGT enzymes could be utilized for the glucosy-
lation of hydroxyflavones. The glucoside products could be easily
isolated from the medium by adsorption on HP20 resin. An advan-
tage of using E. coli cultures and P. americana cultures is that the
addition of exogenous glucosyl donors is not required. E. coli and P.
americana can use endogenous UDP-glucose to convert aglycone
to glucosides. It is known that glycosylation of bioactive com-
pounds can increase their water solubility and biological half-life.
Since the biological effects of flavonoids, including their anti-tumor,
anti-inflammatory, and antimicrobial activity, are now recognized
[30–33], methods to endow flavonoids with health-related prop-
erties by conjugating a sugar moiety may lead to the development
of new technology [34–36].
[35] E.K. Lim, Chem. – Eur. J. 11 (2005) 5486–5494.
[36] J.W. Gavin, S.T. Jon, E.J. Toone, Adv. Enzymol. Relat. Areas Mol. Biol. 76 (2009)
55–119.
References
[1] E.K. Lim, D.J. Bowles, EMBO J. 23 (2004) 2915–2922.