Synthesis of epicatechin glucosides by a β-cyclodextrin glycosyltransferase

Article abstract of DOI:10.1016/j.molcatb.2011.07.013

Enzymatic synthesis of (-)-epicatechin (EC) glucosides was performed through the transglucosylation reaction catalyzed by the cyclodextrin glycosyltransferase (CGTase) from Paenibacillus sp. RB01. The enzyme showed the same product specificity for the three donor substrates, starch, β-cyclodextrin and maltoheptaose (G7). Using β-cyclodextrin as the glucosyl donor, several EC glucoside products were obtained at an overall minimal yield of 18.1%. The structures of the four main products were elucidated by MS and NMR techniques as (-)-EC-3′-O-α-d-glucopyranoside (EC3A), (-)-EC-3′-O-α-d-diglucopyranoside (EC3B), (-)-EC-3′-O- α-d-triglucopyranoside (EC3C) and (-)-EC-4′-O-α-d- glucopyranoside (EC4A). Of these, EC3A was the major product while EC4A, unique for this CGTase, was formed in the lowest amount. The water solubility and stability against UV irradiation of EC3A were significantly higher than that of EC. Although the antioxidant activity was 1.5-fold lower, the advantage of the enhanced solubility and stability makes the EC3A glucoside more beneficial as food ingredient than its parent EC.

Full text of DOI:10.1016/j.molcatb.2011.07.013

Journal of Molecular Catalysis B: Enzymatic 73 (2011) 27–34
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Synthesis of epicatechin glucosides by a ␤-cyclodextrin glycosyltransferase
Pornpun Aramsangtienchaia, Warinthorn Chavasirib, Kazuo Itoc, Piamsook Pongsawasdia,∗
a Starch and Cyclodextrin Research Unit, Department of Biochemistry, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand
b Natural Products Research Unit, Department of Chemistry, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand
c Laboratory of Enzyme Chemistry, Graduate School of Science, Osaka City University, Osaka 558-8585, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 May 2011
Received in revised form 9 July 2011
Accepted 22 July 2011
Available online 30 July 2011
Enzymatic synthesis of (−)-epicatechin (EC) glucosides was performed through the transglucosy-
lation reaction catalyzed by the cyclodextrin glycosyltransferase (CGTase) from Paenibacillus sp.
RB01. The enzyme showed the same product specificity for the three donor substrates, starch,
␤-cyclodextrin and maltoheptaose (G7). Using ␤-cyclodextrin as the glucosyl donor, several EC glu-
coside products were obtained at an overall minimal yield of 18.1%. The structures of the four main
products were elucidated by MS and NMR techniques as (−)-EC-3ꢀ-O-␣-d-glucopyranoside (EC3A),
(−)-EC-3ꢀ-O-␣-d-diglucopyranoside (EC3B), (−)-EC-3ꢀ-O-␣-d-triglucopyranoside (EC3C) and (−)-EC-4ꢀ-
O-␣-d-glucopyranoside (EC4A). Of these, EC3A was the major product while EC4A, unique for this CGTase,
was formed in the lowest amount. The water solubility and stability against UV irradiation of EC3A were
significantly higher than that of EC. Although the antioxidant activity was 1.5-fold lower, the advantage
of the enhanced solubility and stability makes the EC3A glucoside more beneficial as food ingredient than
its parent EC.
Keywords:
Cyclodextrin glycosyltransferase
Epicatechin
Transglucosylation
Epicatechin glucosides
␤-Cyclodextrin
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
platelet aggregation [5] and prevents gastric cancer [6]. Although
EC, and indeed catechins in general, have potential health bene-
Catechins are polyhydroxylated flavonoids [1] that are found
in various plant species, especially in the tea-plant Camellia sinen-
sis [2], and in other plant derived human dietary components,
such as fruits, vegetables, chocolate and red wine [3]. The green
tea catechins are composed of six kinds of catechin and its
derivatives: (+)-catechin, (+)-gallocatechin (GC), (−)-epicatechin
(EC), (−)-epicatechin gallate (ECg), (−)-epigallocatechin (EGC) and
(−)-epigallocatechin gallate (EGCg). The composition and level
of catechins in each plant may vary considerably between cul-
tivars and cultivation conditions. For example, the catechins
in red wine, which are mainly catechin and EC, are found
at a variable ratio and total amount depending on the grape
cultivars [4].
fits and are being used widely in food and drinks, they are poorly
soluble in water, have a bitter taste, and are easily oxidized and
degraded by light irradiation [7]. Therefore, they have limitations
upon their use as a natural food additive or medicine [8]. However,
these problems are mostly overcome by modifying them with sugar
molecules. Glucosylation of catechins, through enzyme catalysis
using members of the glycosyltransferase, sucrose phosphorylase
or glucansucrase families, has been reported [7,9–12]. (+)-Catechin
3ꢀ-O-␣-d-glucopyranoside was synthesized via transglucosylation
using the CGTase from Bacillus macerans (␣-CGTase) [9]. While
EGCg-4ꢀꢀ-monoglucopyranoside, EGCg-7,4ꢀꢀ-diglucopyranoside and
EGCg-4ꢀ, 4ꢀꢀ-di-O-␣-d-glucopyranoside were all synthesized by
transglucosylation using the glucansucrase from Leuconostoc
mesenteroid [12]. The most common glucoside derivatives of
catechins prepared by biotransformation reactions are the ␣-d-
glucosides [9], while, in contrast, very few reports on the enzymatic
synthesis of the ␤-d-glucosides exist, the most notable exception
being by catalysis of enzymes from plant extracts, e.g. eucalyptus,
barley and lentil [13,14]. Under these conditions the obtained struc-
ture and yield of the catechin glucosides were dependent on the
type of enzyme and the conditions used in the transglucosylation
reaction.
EC (cis-3,3ꢀ,4ꢀ,5,7-pentahydroxyflavane, an epimer of catechin,
Fig. 1), a strong antioxidant, has an insulin mimic action and
improves heart health. It also reduces lipid peroxidation, inhibits
Abbreviations: ␤-CD, ␤-cyclodextrin; CGTase, cyclodextrin glycosyltransferase;
DPPH, 2,2-diphenyl-1-picryhydrazyl; EC, (−)-epicatechin; EC3A, (−)-epicatechin-
3ꢀ-O-␣-d-glucopyranoside; EC3B, (−)-epicatechin-3ꢀ-O-␣-d-diglucopyranoside;
EC3C, (−)-epicatechin-3ꢀ-O-␣-d-trigluco-pyranoside; EC4A, (−)-epicatechin-
4ꢀ-O-␣-d-glucopyranoside; EGCg, (−)-epigallocatechin gallate; IC, inhibitory
concentration; Rt, retention time.
Cyclodextrin glycosyltransferase (CGTase, E.C. 2.4.1.19)
catalyzes the production of cyclodextrin from starch and related ␣-
(1,4)-linked glucose polymers or oligomers via a transglycosylation
∗
Corresponding author. Tel.: +66 2 218 5419; fax: +66 2 218 5418.
E-mail address: piamsook.p@chula.ac.th (P. Pongsawasdi).
1381-1177/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2011.07.013

28
P. Aramsangtienchai et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 27–34
2.2. Enzyme preparation
Paenibacillus sp. RB01 screened from the hot-spring area in
Ratchburi Province of Thailand [21] was grown in an enriched
Medium I [18] at 37 ◦C for 18 h, before transferring into Horikoshi’s
medium [22] and further cultivation at 40 ◦C for 72 h. Cells were
then removed by centrifugation (3000 × g for 30 min) and the cul-
ture broth, as a crude CGTase source, was partially purified by starch
adsorption. The CGTase activity at each stage was determined by
starch degrading assay [23].
2.3. Synthesis of EC glucosides
2.3.1. Donor specificity
Selection of the appropriate donor was performed by incubating
0.5% (w/v) EC and CGTase (30 U/ml) in 50 mM phosphate buffer (pH
6.0) with 1.8% (w/v) of the different glucosyl donors: starch, ␤-CD or
G7, in a reaction volume of 1 ml at 40 ◦C for 24 h. The products were
analyzed using TLC and HPLC resolution (see below) with reference
to each type of catechins as standard.
2.4. Detection of products
2.4.1. Thin layer chromatography (TLC)
The products were analyzed by silica gel 60 (Merck, Co.) based
TLC using a 3:1:1 (v/v) ratio of ethyl acetate: acetic acid: water as
the mobile phase solvent. Spots were detected by spraying with 1:
2 (v/v) ratio of sulfuric acid: methanol as reported [9], followed by
heating at 120 ◦C for 20 min.
2.4.2. High performance liquid chromatography (HPLC)
Glucosylated products were detected using a C18-reverse phase
HPLC column (Shimadzu, Japan; 5.0 ␮m; 4.6 mm × 250 mm) with
22% (v/v) methanol as the mobile (solvent) phase at a flow rate of
0.5 ml/min and at 35 ◦C. The eluted products were monitored by UV
detection at 279 nm.
Fig. 1. Structures of (−)-epicatechin (EC) and (+)-catechin.
reaction. CGTase catalyzes four related reactions: cyclization, cou-
pling, disproportination and hydrolysis [15–18]. Several reports on
the transglycosylation of interesting compounds to make useful
glucosides by the action of CGTase are available. For example,
the industrial production of glucosyl stevioside, a sweetener from
a plant stevioside with a very bitter taste, is catalyzed by the
CGTase from Bacillus stearothermophilus FERM-P No. 2222 [19]. In
addition, the 3G-␣-d-glucopyranosyl neohesperidin, synthesized
by transglucosylation with the CGTase from an alkalophilic Bacillus
species, was found to be 1500-fold more soluble and 10-fold
less bitter than its parent neohesperidin, a unique flavanoid of
oranges [20].
2.4.3. Determination of the transglucosylation efficiency
The efficiency of the transglucosylation reaction was judged by
evaluating the transglucosylation product yield, determined from
the HPLC profile by two methods. Firstly, the yield was calcu-
lated from the difference between the amount of EC at each time
point and the initial amount added [24]. Secondly, the product
yield was measured directly from the ratio of the peak area of
each product relative to that of the initial EC concentration added
[7].
The present work aims to use a ␤-CGTase, derived from Paeni-
bacillus sp. RB01, which was isolated from a hot spring area in
Thailand, to enzymatically synthesize EC glucosides, to obtain new
product and to improve properties of EC that are of beneficial use
for industrial applications. This is the first report on the synthesis
of EC glucosides using the enzyme CGTase, as well as their charac-
terization.
2.5. Optimization of transglucosylation reaction
The optimum conditions for the synthesis of EC glucosides were
considered in terms of obtaining the highest percentage yield of
the transglucosylated products, as analyzed by HPLC. The effects of
varying the pH (5.0–8.0), enzyme concentration (20–120 U/ml), EC
concentration (0.06–2.0% (w/v)), ␤-CD concentration (0.125–1.8%
(w/v)), temperature (30–70 ◦C) and incubation time (12–60 h) were
investigated.
2. Materials and methods
2.1. Materials
2.6. Larger scale preparation and isolation of glucosylated
products
(+)-Catechin, (−)-epicatechin (EC), ␤-cyclodextrin (␤-CD) and
maltoheptaose (G7) were purchased from Sigma. Soluble starch
(potato), glucoamylase (A. niger), ␣-glucosidase (S. cerevisiae),
Sephadex-LH-20 and 2,2-diphenyl-1-picryhydrazyl (DPPH) were
obtained from Merck Co. All other chemicals used were commer-
cially available and of analytical grade.
The synthesis of EC glucosides was then performed under the
optimized conditions obtained above in a reaction volume of 50 ml.
After completion, the reaction mixture was concentrated by rotary
evaporation to ∼5 ml and applied onto a Sephadex LH-20 column
(2.5 cm × 55 cm) equilibrated with distilled water. The glucoside

P. Aramsangtienchai et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 27–34
29
products were separated from the sugars and enzyme by elution
with distilled water (2 l, 0.5 ml/min). Thereafter, EC was then eluted
by changing the mobile phase to 50% (v/v) methanol. The fractions
containing the EC-glucosides were concentrated and further sep-
arated by HPLC. Each glucoside-containing peak was collected for
further characterization.
2.7. Characterization of glucoside products
2.7.1. Enzyme treatment
Glucoamylase from A. niger was added to the reaction mix-
ture (18.3 U/ml) after the CGTase catalyzed synthesis was finished,
and incubated at 40 ◦C for 1 h. After that, to one half of this reac-
tion mixture, ␣-glucosidase (18.3 U/ml) from S. cerevisiae was then
added and further incubated at 40 ◦C for 3 h. The products of all
three enzyme treatments (CGTase, CGTase – glucoamylase and
CGTase – glucoamylase – ␣-glucosidase) were then analyzed by
TLC.
Fig. 2. TLC chromatogram of the reaction products catalyzed by CGTase between EC
and starch, ␤-CD and G7 as donors. Standards of (a) EC, (b) glucose and (c) maltose
compared to the (d) control (starch + CGTase) (e and f) starch as donor, 0 and 24 h
incubation, (g) control (beta-CD + CGTase), (h and i) beta-CD as donor, 0 and 24 h
incubation, (j) control (G7 + CGTase) and (k and l) G7 as donor, 0 and 24 h incubation.
2.7.2. Mass spectrometry
Mass analysis of the major glucoside products was performed on
an Electrospray Ionization-MicroTOF Mass spectrometry (ESI-TOF
MS).
inhibitory concentration at which the absorbance was reduced by
50%.
2.7.3. Nuclear magnetic resonance
For structural elucidation, 1H-, 13C- and two-dimensional
(homonuclear proton correlation COSY and heteronuclear HSQC
and HMBC) NMR spectra were performed in CD3OD on a Varian
Gemini 400 MHz spectrometer. The chemical shifts in ppm were
expressed with respect to tetramethylsilane as an internal refer-
ence.
3. Results and discussion
3.1. Donor specificity
The donor specificity of the CGTase from Paenibacillus sp. RB01
was evaluated using starch, ␤-CD and G7 as the glucosyl donors in
the transglucosylation reaction with EC as the acceptor. The prod-
ucts were then resolved and analyzed by TLC (Fig. 2). All three
different glucosyl donors gave the same three major products (Glu
I, Glu II and Glu III in Fig. 2), as well as two minor products (X and
Y in Fig. 2), which were not observed at the initial time of incuba-
tion. The products were likely to be glucoside derivatives with the
relative intensities of Glu I > Glu II > Glu III > X > Y. The transgluco-
sylation yield of the main products Glu I to Glu III was determined
from the ratio of disappearance of the amount of EC acceptor to
the initial amount, as analyzed by HPLC. The obtained yields were
22.6, 22.4 and 21.4% for starch, ␤-CD and G7 as donors, respec-
tively. This suggests that these substrates were equally effective as
a glucosyl donor for this CGTase from Paenibacillus sp. RB01. For
all subsequent experiments, ␤-CD was accordingly chosen as the
appropriate donor since it is the specific substrate of the CGTase.
The reaction products of EC and ␤-CD when catalyzed by the
CGTase from Paenibacillus sp. RB01 were analyzed by HPLC (Fig. 3),
revealing at least four EC glucoside products with a Rt of 27, 30,
45 and 48 min, respectively. Surprisingly, two of the main gluco-
sides, with a Rt of 45 and 48 min, were eluted after the less polar EC
at 38 min. In addition, the elution of the standard catechin, which
appeared to be less polar than EC when resolved by either TLC (data
not shown) or Sephadex-LH20 column chromatography (Fig. 4),
had a Rt of 19 min in the HPLC peak. After 24 h of incubation, the
catechin peak at Rt 19 min was increased due to epimerization from
EC. The epimerization of EC to catechin was observed at an alkaline
pH, high temperature and longer incubation time [26], and brings
about a decrease in the EC substrate level and consequently a lower
yield of EC glucosides. Small product peaks around Rt 15–23 min
might be glucosides of catechin as by-products. Therefore, opti-
mization of the reaction conditions for effective transglucosylation
of EC is required.
2.8. Determination of the properties of EC glucosides
2.8.1. Solubility in water
Excess amounts of EC and EC glucosides were mixed with 200 ␮l
of water, vortexed for 5 min at room temperature to disperse into
solution, and then each sample was incubated at 30 ◦C for 15 min
to allow the solvation equilibrium (saturation) to be attained. The
sampleswerethenfilteredthrougha 0.45 ␮m membraneto remove
insoluble particles and subjected to quantitative reverse phase C18
HPLC analysis in order to determine the soluble EC and EC-glucoside
concentrations.
2.8.2. Browning resistance
Browning resistance was examined in an aqueous solution
(1.0 ml) containing 0.1% (w/v) of either EC or the respective EC
glucoside. The samples were then exposed to UV irradiation from
a 30 W UV lamp (254 nm) at a distance of 35 cm for 10 h. Fol-
lowing this treatment, the absorbance at 460 nm of the solutions
were then determined to evaluate the concentration of EC or
EC glucoside remaining, and so from this its photostability, as
a % of that remaining compared to that initially present, was
evaluated.
2.8.3. Antioxidant activity
The antioxidant activity of EC and each EC glucoside was eval-
uated using the DPPH radical scavenging reaction [25]. Each of
the samples was dissolved, or diluted, in the required volume of
ethanol to yield the final EC or EC glucoside concentration range
of 0–800 ␮M, then DPPH was added to a final of 100 ␮M. After
10 min in the dark, the absorbance of the mixture was measured
at 517 nm. The DPPH radical scavenging activity was evaluated by
analyzing the percentage decrease in the absorbance of the sam-
ple compared to the blank (ethanol). The IC50 value designates the

30
P. Aramsangtienchai et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 27–34
Fig. 3. HPLC chromatogram of the reaction products of EC and ␤-CD catalyzed by the CGTase from Paenibacillus sp. RB01 after (i) 0 and (ii) 24 h incubation. The asterix (*)
marks the EC-glucosides.
Fig. 4. Sephadex LH-20 column chromatography elution profile of the EC and ␤-CD reaction products, catalyzed by CGTase from Paenibacillus sp. RB01. Two ml fractions
were collected from the column (55 cm × 2.5 cm), eluted with water initially and then (arrow) with 50% (v/v) methanol at a flow rate of 30 ml/h. Fractions labeled I–V are EC
glucosides, and are discussed in the text, while EC and C are epicatechin and catechin, respectively.
3.2. Optimization of transglucosylation reaction
3.3. Isolation of EC glucoside products
In order to increase the product yield, the transglucosylation
reaction was optimized by sequential independent variable anal-
ysis (data not shown). The optimal conditions obtained were
incubation of 0.1% (w/v) ␤-CD and 0.5% (w/v) EC with 80 U/ml of
CGTase in 10 mM acetate buffer (pH 6.0) at 50 ◦C for 24 h. Under
these conditions, the product yield, which was calculated from
peak area of the product to that of EC at time t0, increased over
2.3-fold to 18.1% (Table 1). This increased yield was found for all
four EC glucoside products and, especially for that of the main
product at a Rt of 48, which showed a 2.7-fold increase in yield
up to a 8.33% yield. It was found that the total yield of EC gluco-
sides obtained in this study (18.1%) was approximately the same as
that reported previously for catechin glucosides (18.3%) when using
the Bacillus macerans ␣-CGTase to catalyze the transglucosylation
[27].
The reaction products were separated by Sephadex LH-20 col-
umn chromatography on the basis of size and polarity differences
[28]. The collected fractions were detected spectrophotometrically
for EC at A279 (Fig. 4). When the column was eluted with distilled
water, the enzyme CGTase (the first sharp peak) and five main sepa-
rate product peaks (peaks I–V) were obtained. Subsequent elution
by 50% (v/v) methanol yielded two more peaks with a high A279
absorbance that were subsequentially identified by comparative
HPLC analysis to be EC and catechin, respectively.
From the HPLC analysis, peaks I–V were composed of more than
one product, as summarized in Table 2. The main EC glucoside prod-
ucts, with a Rt of 48 and 45 min, were eluted in peaks IV and III,
respectively. Peak IV also contained a minor product with a Rt of
27. The minor product at a Rt of 30 min existed in both peaks II and
V, and were named 30(1) and 30(2), respectively, and later clarified
Table 1
Yields of EC glucoside products before and after optimization. Data are shown as the mean 1 SD and are derived from three independent repeats. Means within a column
with a different letter are significantly different (p < 0.05).
Product yield (%)
Rt 27
Total yield (%)
Rt 30
Rt 45
Rt 48
Before optimization
After optimization
0.47 0.07a
2.17 0.20b
2.01 0.40a
3.39 0.17b
2.04 0.21a
4.18 0.31b
3.13 0.09a
8.33 0.19b
7.65 0.47a
18.1 0.48b

P. Aramsangtienchai et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 27–34
31
Table 2
Identification of peaks I–V, resolved by Sephadex LH-20 column chromatography,
by their relative retention time in the C18 reverse phase HPLC column.
Peak
Retention time (Rt)
I
II
III
IV
V
15, 18, 22
15, 20, 30(1)a
15, 23, 45a
18, 27a, 48a
20, 30(2)a
a
−EC glucoside products (see Fig. 3).
by TLC (Fig. 5). Other peaks at Rt 15–23 min (see Fig. 3) should be
products of catechin. Each peak eluted from the Sephadex LH-20
column was then pooled and repeatedly purified by reverse phase
C18 HPLC.
3.4. Characterization of EC glucosides
To correlate the four main glucoside products from the HPLC
peaks with a Rt of 27, 30, 45 and 48 min, respectively, to the prod-
ucts resolved as TLC spots, the HPLC-resolved EC products at each of
these respective Rt values were collected and then analyzed by TLC
(Fig. 5). It was found that the peaks with a Rt of 48, 45 and 30(1) cor-
responded to Glu I, Glu II and Glu III, respectively. Due to the close
migration with the major products with a Rt of 48 and 45 min, Glu
I might also contain the compound with a Rt of 30(2), while Glu II
might contain that with a Rt of 27 min. The minor products X and
Y observed in lane b were not matched with any of the HPLC peaks
tested. They might be those products with Rt of 15–23 min which
were proposed to be glucosides of catechin.
Fig. 6. TLC chromatogram of the reaction products of EC and ␤-CD incubated with
Paenibacillus sp. RB01 CGTase for 24 h. Standard (a) glucose and (b) EC, plus the
reaction products after 24 h (c) untreated, (d) treated with glucoamylase and (e)
treated with glucoamylase and then ␣-glucosidase.
II and Glu III are both EC-oligoglucosides. To further hydrolyze
the glucoside products, ␣-glucosidase was added to the reaction
after the glucoamylase, and this resulted in no glucoside spots
being detected (Fig. 6), indicating that the first glucosyl residue
attached to the EC is ␣-linked. The result agrees with the fact that
the CGTase enzyme is specific for ␣-1,4-linkage transfer [29]. It was
also observed that the unidentified product Y was disappeared in
lane d and the X was not seen in lane e, this suggests that Y should
be an oligoglucoside and X monoglucoside of catechin.
3.4.1. Enzyme treatment
Product characterization was preliminary investigated by
enzymic treatment of the reaction products with glucoamylase
and ␣-glucosidase followed by TLC resolution of the final deglu-
cosylated products. Prior to glucoamylase treatment, three main
glucoside products were observed (Glu I, II and III), but after treat-
ment with glucoamylase the Glu II and III spots disappeared while
the intensities of the Glu I and glucose spots increased (Fig. 6).
This result suggests that Glu I is an EC-monoglucoside while Glu
3.4.2. Mass spectrometry and nuclear magnetic resonance
From the mass spectra, the molecular weight of the EC glu-
coside products with a Rt of 48 and 30(2) min corresponded to
the expected size for an EC monoglucoside, while those at a Rt of
45 and 30(1) min were an EC diglucoside and an EC-triglucoside,
respectively (Table 3).
NMR analysis was used to postulate the structures of the EC-
glucoside products (Table 4). From the proton [H1] signal of the
compound with a Rt of 48 min, a doublet signal at 7.31 ppm and
a doublet-doublet signal at 6.96 ppm were assigned to the H-2ꢀ
and H-6ꢀ of the EC moiety, respectively. Due to the glucosylation
of EC, these proton signals showed downfield shifts by 0.44 and
0.26 ppm, respectively, when compared with the standard EC. This
supports the presence of a glycosidic bond on the 3ꢀ-hydroxyl group
of the catechol ring of the EC moiety [7]. The signal for the anomeric
proton of the sugar moiety appeared at 5.28 ppm with a coupling
constant (J) of 3.6 Hz. This glucosyl residue was, therefore, con-
nected to the EC by an ␣-linkage. Integrated with the product size
analyzed by MS, the compound with a Rt of 48 min was identi-
fied as (−)-EC-3ꢀ-O-␣-d-glucopyranoside (EC3A) (Fig. 7i). Similarly,
Table 3
Molecular weight of the four main resolved EC glucoside products by mass
spectrometry.
Glucoside product
[M + Na]+ at m/z
MW (Da)
Rt 48
Rt 45
Rt 30(1)
Rt 30(2)
475
637
798
475
452
614
775
452
Fig. 5. TLC chromatogram of EC glucoside products separated by HPLC. (a) Standard
glucose, (b) reaction products after 24 h, and fractions (peaks) from the HPLC sepa-
ration of reaction products with Rt of (c) 27, (d) 30, (e) 45 and (f) 48 min (referred
to as Rt27, Rt30 (1) and Rt30 (2), Rt45 and Rt48, respectively, in the text).

32
P. Aramsangtienchai et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 27–34
Table 4
1H, 13C NMR data of the EC-glucosides.
δ1H (ppm, in CD3OD); J (Hz)
δ
13C (ppm, in CD3OD); J (Hz)
Position
Rt30(1) min
(EC3C)
Rt30(2) min
(EC4A)
Rt30(2) min
(EC4A)
EC
Rt48 min (EC3A)
Rt45 min (EC3B)
EC
Aglycone
2
3
4.72 (s)
4.08 (s)
4.78 (s)
4.06 (m)
4.87 (s)
4.19 (s)
4.88 (s)
4.09 (m)
4.85 (s)
4.19 (s)
78.2
78.2
66.0
27.8
98.6
95.0
94.4
65.8
27.7
98.4
94.3
93.8
4
2.63 (dd 16.8, 2.6)
2.76 (dd 16.7, 4.5)
5.82 (d 2.2)
2.64 (dd 16.8, 2.3)
2.77 (dd 16.6, 4.5)
5.82 (d 2.0)
2.74 (dd 16.8, 2.0)
2.86 (dd 16.7, 4.4)
5.92 (d 1.9)
2.73 (dd 16.8, 2.2)
2.86 (dd 17.2, 4.6)
5.91 (d 2.17)
5.93 (d 1.65)
2.72 (dd 16.9, 1.5)
2.87 (dd 16.5, 4.1)
5.92 (d 5.7)
4a
6
8
5.84 (d 2.2)
5.85 (d 2.2)
5.94 (d 2.0)
5.92 (d 5.7)
2′
6.87 (d 1.5)
6.66 (d 8.1)
7.31 (d 1.31)
6.75 (d 8.3)
7.42 (d 1.4)
6.84 (d 8.2)
7.41 (s)
7.04 (s)
114.1
114.3
117.8
114.5
117.2
117.8
5′
6.84 (d 8.1)
7.23 (d 8.4)
6.89 (d 8.3)
6′
6.70 (dd 8.1, 1.6)
6.96 (dd 8.1, 1.2)
7.06 (dd 8.2, 1.3)
7.05 (dd 8.1, 1.7)
Glycoside
1″
2″
3″
4″
5″
6″
5.28 (d 3.6)
5.35 (d 3.6)
5.21 (d 3.4)
5.32 (d 3.4)
100.0
72.0
73.4
69.9
73.0
60.9
3.35 - 3.76 (m)
3.39 - 3.90 (m)
3.34 - 3.91 (m)
3.40 - 3.89 (m)
s, singlet; d, doublet; dd, doublet of doublet; m, multiplet.
due to the downfield shift of the proton signals at the H-2ꢀ and H-
6ꢀ and also the MS result, the compounds with Rt of 45 and 30(1)
min were identified as (−)-EC-3ꢀ-O-␣-d-diglucopyranoside (EC3B))
(Fig. 7ii) and (−)-EC-3ꢀ-O-␣-d-triglucopyranoside (EC3C)) (Fig. 7iii),
respectively. All the structures with the 3ꢀ-O linkage were con-
firmed by comparison with literature data [7,11,27]. For the product
at Rt 30(2) min, a downfield shift of 0.57 ppm at the H-5ꢀ of the EC
moiety with a J value of the anomeric proton (5.32 ppm) of 3.4 Hz
was observed. This product was identified as a monoglucoside
from the MS analysis, the proposed structure was (−)-EC-4ꢀ-O-␣-
d-glucopyranoside (EC4A) (Fig. 7iv). Since this product was new for
CGTase reaction, analysis by 13C- and two-dimension NMR (COSY,
HSQC and HMBC) was performed. The result confirmed the signal
shift of 2.9 ppm at the C-5ꢀ position of EC (Table 4).
The structures of these four EC-glucoside products showed
either the 3ꢀ-O- (EC3A, EC3B and EC3C) or a 4ꢀ-O-linkage (EC4A).
Since the characteristics of the glucosylated products of EC have
never been reported before, here we compare them with the
reported glucoside products of catechin, which is an epimer of EC.
Catechin-3ꢀ-O-␣-d-glucopyranoside has been synthesized by tran-
glucosylation using (i) the ␣-CGTase from B. macerans [27], (ii)
the sucrose phosphorylase from L. mesenteroids [9], and (iii) by
transferase activity from X. campestris WU-9701 [11], while cat-
echin 4ꢀ-O-␣-d-glucopyranoside has been synthesized using the
glucosyltransferase from Streptococcus sp. [10] and S. sobrinus [30].
However, the formation of a 4ꢀ-O-glucoside by the action of a
CGTase has never previously been reported. Thus, our unique find-
ing of a 4ꢀ-O-glucoside derivative of EC shows that Paenibacillus
sp. RB01 CGTase, a ␤-CGTase, acts differently from the established
␣-CGTase of B. macerans in its transglucosylation of EC and catechin.
As mentioned, the catechin glucoside products obtained from
enzymatic transglucosylation reactions were mostly in the ␣-
configuration, whereas the ␤-glucosides are mainly found in the
natural glucosides from plants. For example, the catechin 7-O-␤-
d-glucopyranoside isolated from barley (Hordeum vulgare L.) [31]
and buckwheat (Fagopyrum) [32], while 4ꢀ-O-␤, 5-O-␤ and 3-O-␤
isomers also coexisted in those plants.
3.5. Properties of EC3A glucoside
3.5.1. Solubility, browning resistance and antioxidant activity
The water solubility of EC was 4.9 mg/ml (0.017 mol/l)
whereas that of EC3A was found to be significantly higher at
216.6 mg/ml (0.479 mol/l). Likewise, the solubility of catechin-3ꢀ-
O-␣-d-glucopyranoside has been reported to be between 50- and
100-fold higher than that of catechin [8,11], and the solubility of 4ꢀꢀ
EGCg-monoglucoside was 69 times higher than that of EGCg [12].
Although the discrepancy between these solubility properties may
come from the different structures of the catechin derivatives, it
may also be a result of variation in the experimental methods used
to determine their solubility.
The browning of EC was found to increase with increasing expo-
sure time to the UV irradiation, while under the same conditions the
EC3A derivative showed resistance to browning, even after 10 h of
irradiation (Fig. 8A). The EC3A glucoside was thus more stable than
its parent EC, and thus suggests that the glucosyl moiety helps pro-

P. Aramsangtienchai et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 27–34
33
Fig. 7. Proposed structure of the products (i) (−)-EC-3ꢀ-O-␣-d-glucopyranoside (EC3A), (ii) (−)-EC-3ꢀ-O-␣-d-diglucopyranoside (EC3B), (iii) (−)-EC-3ꢀ-O-␣-d-
triglucopyranoside (EC3C) and (iv) (−)-EC-4ꢀ-O-␣-d-glucopyranoside (EC4A).
Fig. 8. (A) Browning resistance to UV irradiation. A solution containing 0.1% (w/v) EC (ꢀ) or EC3A (×) was exposed to UV irradiation (254 nm) for the indicated periods before
the absorbance at 460 nm was monitored. (B) DPPH radical-scavenging activities. EC (ꢀ) and EC3A (×) at various concentrations mixed with 100 ␮M DPPH and left for 10 min
in the dark were then monitored for their absorbance at 517 nm. Data are shown as the mean 1 SD and are derived from three independent repeats.
tect the EC catechol rings from polymerization, which is the cause of
the browning phenomena [8]. Catechin was also reported to be eas-
ily degraded with rapid browning activated by UV irradiation [7]. In
addition, catechin solution was shown to turn brown after stand-
ing for 24 h in contact with air at 20 ◦C while the monoglucoside
solution showed no color change [11].
tea catechins was later confirmed [34]. The antioxidative molecu-
lar mechanism of catechins through NMR approach was presented
whereby the same mechanism for catechin and epicatechin was
proved [35]. Our result corresponds well with previous report on
the decrease in the antioxidant activity when catechins were in the
glycosylated form [36].
The free-radical scavenging activity was determined using the
stable free radical, DPPH where the IC50 (inhibitory concentration
at which the absorbance at 517 nm was reduced by 50%) of EC and
EC3A were 76.5 and 115 ␮M, respectively, making the IC50 of EC3A
1.5-fold higher than that of the parental EC (Fig. 8B). This reduced
scavenging activity for the EC3A glucoside might be due to the pres-
ence of an ortho-dihydroxyl group in the flavonoid B ring that is
important for achieving a high radical scavenging activity [33]. The
relationship between scavenging ability and the ring B structure of
4. Summary
The overall results demonstrated that the CGTase from Paeni-
bacillus sp. RB01 can be used to synthesize four to five different
EC glucosides, as analyzed by HPLC, which after optimization
had a total yield of 18.1%. The main product (EC3A) was a 3ꢀ-
O-monoglucoside with a higher water solubility and browning
resistance to UV irradiation than EC, but its antioxidant property

34
P. Aramsangtienchai et al. / Journal of Molecular Catalysis B: Enzymatic 73 (2011) 27–34
was decreased. The other products whose structures were analyzed
by NMR were EC3B and EC3C (3ꢀ-O-␣-di- and tri-glucosides, respec-
tively), plus EC4A. EC4A is a unique product with the monoglucosyl
␣-linked to EC at the 4ꢀ-position of the catechol ring. Although it is
produced in a relatively low amount, its synthesis has never been
previously reported for other CGTases.
[11] T. Sato, H. Nakagawa, J. Kurosu, K. Yoshida, T. Tsugane, S. Shimura, K. Kirimura,
K. Kino, S. Usami, J. Biosci. Bioeng. 90 (6) (2000) 625–630.
[12] Y.H. Moon, G. Kim, J.H. Lee, X.J. Jin, D.W. Kim, D. Kim, J. Mol. Anal. 40 (2006)
1–7.
[13] S. Otani, Y. Kondo, Y. Asada, T. Furuya, H. Hamada, N. Nakajima, K. Ishihara, H.
Hamada, Plant Biotechnol. 21 (2004) 407–409.
[14] T. Raab, D. Barron, F.A. Vera, V. Crespy, M. Oliveira, G. Williamson, J. Agric. Food
Chem. 58 (4) (2010) 2138–2149.
[15] S. Kitahata, N. Tsuyama, S. Okada, Agric. Biol. Chem. 38 (1974) 387–393.
[16] Z.F. Li, J.Y. Zhang, Q. Sun, M. Wang, Z.B. Gu, G.C. Du, J. Wu, J. Chen, J. Agric. Food
Chem. 57 (2009) 8386–8391.
Acknowledgements
[17] C. Weijers, M.C.R. Franssen, G.M. Visser, Biotechnol. Adv. 26 (2008) 436–456.
[18] W. Wongsangwattana, J. Kaulpiboon, K. Ito, P. Pongsawasdi, Process Biochem.
45 (2010) 947–953.
[19] M. Malsagov, T. Tugan, S. Timur, H.A. Varuzhan, US Patent Application Publica-
tion, US2007/0082102 A1 (2007).
[20] T. Kometani, Y. Terada, T. Nishimura, H. Takii, S. Okada, Biosci. Biotechnol.
Biochem. 58 (1994) 1990–1994.
[21] J. Kaulpiboon, V. Rimphanitchayakit, P. Pongsawasdi, J. Basic Microbiol. 44
(2010) 374–382.
Special thanks go to Dr. Robert Butcher of the Publication Coun-
seling Unit of the Faculty of Science for editing the manuscript.
Financial support from the Asahi Glass Foundation of Japan and
from Chulalongkorn University are appreciated. We also acknowl-
edge the support from the Higher Education Research Promotion
and National Research University Project of Thailand, Office of the
Higher Education Commission (FW650B) and the Thai Government
Stimulus Package 2 (TKK 2555) under the Project PERFECTA.
[22] K. Horikoshi, J. Agric. Biol. Chem. 35 (1971) 1783–1791.
[23] J. Kaulpiboon, W. Prasong, V. Rimphanitchayakit, S. Murakami, K. Aoki, P. Pong-
sawasdi, J. Basic Microbiol. 50 (2010) 1–9.
[24] D.C. Park, T.K. Kim, Y.H. Lee, Enzyme Microb. Technol. 22 (1998) 217–222.
[25] N. Abe, A. Nemoto, Y. Tsuchiya, H. Hojo, A. Hirota, Biosci. Biotechnol. Biochem.
64 (2000) 306–313.
References
[26] Y.L. Su, L.K. Leung, Y. Huang, Z.Y. Chen, Food Chem. 83 (2003) 189–195.
[27] M. Funayama, H. Arakawa, R. Yamamoto, T. Nishino, T. Shin, S. Marao, Biosci.
Biotechnol. Biochem. 58 (1994) 817–821.
[1] B.L. Silvina, G.F. Cesar, Biol. Res. 33 (2000) 151–157.
[2] T. Nagata, S. Sakai, Jpn. J. Breed. 34 (4) (1984) 459–467.
[3] L. Bravo, Nutr. Rev. 56 (11) (1998) 317–333.
[28] P.A. Drewes, A.J. Koaralski, Clin. Chem. 20 (1974) 1451–1453.
[29] A. Nakamura, K. Haga, K. Yamane, FEBS Lett. 337 (1994) 66–70.
[30] K. Nakahara, M. Kontari, H. Ono, T. Kodama, T. Tanaka, T. Ooshima, S. Hamada,
Appl. Environ. Microbiol. 61 (7) (1995) 2768–2770.
[4] D.M. Goldberg, A. Karumanchiri, E. Tsang, G.J. Soleas, Am. J. Enol. Viticult. 49
(1998) 23–34.
[5] C.L. Keen, R.R. Holt, P.I. Oteiza, C.G. Fraga, H.H. Schmitz, Am. J. Clin. Nutr. 81 (1)
(2005) 298S–303S.
[31] W. Friedrich, R. Galensa, Eur. Food Res. Technol. 214 (2002) 388–393.
[32] M. Watanabe, J. Agric. Food Chem. 46 (1998) 839–845.
[33] W. Bars, W. Heller, C. Michel, M. Saran, Methods Enzymol. 186 (1990) 343–355.
[34] F. Nanjo, K. Goto, R. Seto, M. Suzuki, M. Sakai, Y. Hara, Free Radical Biol. Med.
21 (1996) 895–902.
[35] Y. Sawai, K. Sakata, J. Agric. Food Chem. 46 (1998) 111–114.
[36] G.W. Plumb, S. De Pascual-Teresa, C. Santos-Buelga, V. Cheynier, G. Williamson,
Free Radical Res. 29 (1998) 351–358.
[6] V.J. Higdon, B. Frei, Crit. Rev. Food Sci. Nutr. 43 (1) (2003) 89–143.
[7] S. Kitao, T. Ariga, T. Matsudo, H. Sekine, Biosci. Biotechnol. Biochem. 57 (1993)
2010–2015.
[8] T. Ishizu, K. Kintsu, H. Yamamoto, J. Phys. Chem. 103 (1999) 8992–8997.
[9] M. Funayama, T. Nishimo, A. Hirota, S. Murao, S. Takenishi, H. Nakano, Biosci.
Biotechnol. Biochem. 57 (1993) 1666–1669.
[10] H.G. Meulenbeld, H. Zuilhof, A. Veldhuizen, R. Heuvel, S. Hartmans, Appl. Env-
iron. Microbiol. 65 (9) (1999) 4141–4147.