Regioselective formation of quercetin 5-O-glucoside from orally administered quercetin in the silkworm, Bombyx mori

Article abstract of DOI:10.1016/j.phytochem.2007.11.009

The cocoons of some races of the silkworm, Bombyx mori, have been shown to contain 5-O-glucosylated flavonoids, which do not occur naturally in the leaves of their host plant, mulberry (Morus alba). Thus, dietary flavonoids could be biotransformed in this insect. In this study, we found that after feeding silkworms a diet rich in the flavonol quercetin, quercetin 5-O-glucoside was the predominant metabolite in the midgut tissue, while quercetin 5,4′-di-O-glucoside was the major constituent in the hemolymph and silk glands. UDP-glucosyltransferase (UGT) in the midgut could transfer glucose to each of the hydroxyl groups of quercetin, with a preference for formation of 5-O-glucoside, while quercetin 5,4′-di-O-glucoside was predominantly produced if the enzyme extracts of either the fat body or silk glands were incubated with quercetin 5-O-glucoside and UDP-glucose. These results suggest that dietary quercetin was glucosylated at the 5-O position in the midgut as the first-pass metabolite of quercetin after oral absorption, then glucosylated at the 4′-O position in the fat body or silk glands. The 5-O-glucosylated flavonoids retained biological activity in the insect, since the total free radical scavenging capacity of several tissues increased after oral administration of quercetin.

Full text of DOI:10.1016/j.phytochem.2007.11.009

Available online at www.sciencedirect.com
PHYTOCHEMISTRY
Phytochemistry 69 (2008) 1141–1149
www.elsevier.com/locate/phytochem
Regioselective formation of quercetin 5-O-glucoside from
orally administered quercetin in the silkworm, Bombyx mori
a,
b
a
a
*
Chikara Hirayama , Hiroshi Ono , Yasumori Tamura , Kotaro Konno ,
a
Masatoshi Nakamura
a
National Institute of Agrobiological Sciences, Tsukuba, Ibaraki 305-8634, Japan
b
National Food Research Institute, Tsukuba, Ibaraki 305-8642, Japan
Received 27 July 2007; received in revised form 22 October 2007
Available online 31 December 2007
Abstract
The cocoons of some races of the silkworm, Bombyx mori, have been shown to contain 5-O-glucosylated flavonoids, which do not
occur naturally in the leaves of their host plant, mulberry (Morus alba). Thus, dietary flavonoids could be biotransformed in this insect.
In this study, we found that after feeding silkworms a diet rich in the flavonol quercetin, quercetin 5-O-glucoside was the predominant
0
metabolite in the midgut tissue, while quercetin 5,4 -di-O-glucoside was the major constituent in the hemolymph and silk glands. UDP-
glucosyltransferase (UGT) in the midgut could transfer glucose to each of the hydroxyl groups of quercetin, with a preference for for-
0
mation of 5-O-glucoside, while quercetin 5,4 -di-O-glucoside was predominantly produced if the enzyme extracts of either the fat body or
silk glands were incubated with quercetin 5-O-glucoside and UDP-glucose. These results suggest that dietary quercetin was glucosylated
0
at the 5-O position in the midgut as the first-pass metabolite of quercetin after oral absorption, then glucosylated at the 4 -O position in
the fat body or silk glands. The 5-O-glucosylated flavonoids retained biological activity in the insect, since the total free radical scaveng-
ing capacity of several tissues increased after oral administration of quercetin.
Ó 2007 Elsevier Ltd. All rights reserved.
Keywords: Silkworm; Bombyx mori; Mulberry; Morus alba; Moraceae; Flavonoid; Flavonol; Quercetin; UDP-glucosyl transferase; Glucosylation;
Antioxidant; Regioselectivity
1
. Introduction
from the cocoon shell. However, these compounds were
not present in mulberry leaves, in which flavonol glycosides
with a sugar group at the 3-O position in the C ring such as
isoquercitrin (quercetin 3-O-glucoside), rutin (quercetin 3-
O-rutinoside), quercetin 3-O-(6-malonylglucoside), and
astragalin (kaempferol 3-O-glucoside) are naturally occur-
ring (Doi et al., 2001; Katsube et al., 2006; Onogi et al.,
1994). Thus, we can infer that flavonoids absorbed from
their diet are modified in the insect for using these com-
pounds to increase fitness. In insects, the formation of glu-
coside is the predominant pathway for dietary flavonoids
(Hopkins and Ahmad, 1991; Lahtinen et al., 2006; Salmi-
nen et al., 2004; Wiesen et al., 1994), and the glucosylation
of polyphenolics in insects is catalyzed by UDP-glucosyl-
transferase (UGT) (Ahmad and Hopkins, 1993; Rausell
et al., 1997; Real et al., 1991), suggesting the possibility
Uptake and utilization of dietary flavonoids is wide-
spread in insects, in particular in the Lepidoptera. It has
been reported that some insects sequester plant flavonoids
into their body cuticles for protection against natural ene-
mies, or into their wings to increase attractiveness to mates
(
Simmonds, 2003). Larvae of the silkworm Bombyx mori
sequester flavonoids into their cocoons from the leaves of
their host plant, the mulberry tree (Morus alba) (Fujimoto
et al., 1959). Recently, Tamura et al. (2002) identified three
flavonol glucosides, quercetin-5-O-glucoside (5), quercetin
0
0
5
,4 -di-O-glucoside (2), and quercetin 5,7,4 -tri-O-glucoside
*
Corresponding author. Tel.: +81 29 838 6087; fax: +81 29 838 6028.
E-mail address: hirayam@nias.affrc.go.jp (C. Hirayama).
0
031-9422/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2007.11.009

1142
C. Hirayama et al. / Phytochemistry 69 (2008) 1141–1149
that a UGT enzyme that can transfer a glucose moiety to
the C-5 position of the flavonols is functioning in B. mori.
In the present study, we identified flavonoids distributed
in the tissues of the silkworms fed a diet supplemented with
flavonol quercetin (10), and showed that the flavonoids
helped increase the anti-oxidative state of the tissues. Fur-
ther, we determined the in vitro activity of UGT in trans-
ferring glucose to each OH group of quercetin (10) in the
midgut, fat body, and silk glands, in order to establish
the metabolic pathway of dietary quercetin (10) in the
insect. This is the first study on the metabolism of flavonoid
through the regiospecific glycosylation pathway in insects
and we demonstrate the first example of glycosylation of
quercetin (10) by a UGT enzyme with the preferred 5-O
regioselectivity.
R
1
,
3
R
2
,
4
B
R
5
8
a
O
7
A
C
3
4
a
R3
5
O
R
4
1
2
R1 = R2 = R4 = OH, R3 = R5 = O-Glucose
R1 = R3 = R5 = OH, R2 = R4 = O-Glucose
3 R2 = R4 = R5 = OH, R1 = R3 = O-Glucose
R2 = R3 = R5 = OH, R1 = R4 = O-Glucose
R1 = R2 = R3 = R5 = OH, R4 = O-Glucose
R1 = R2 = R3 = R4 = OH, R5 = O-Glucose
R1 = R2 = R4 = R5 = OH, R3 = O-Glucose
R1 = R3 = R4 = R5 = OH, R2 = O-Glucose
R2 = R3 = R4 = R5 = OH, R1 = O-Glucose
4
5
6
7
8
9
2
. Results and discussion
2
.1. Structural elucidation of cocoon shell flavonoids
Tamura et al. (2002) demonstrated that the cocoon shell
1
0 R1 = R2 = R3 = R4 = R5 = OH
1 R1 = H, R3 = R5 = OH, R2 = R4 = O-Glucose
2 R1 = H, R2 = R3 = R5 = OH, R4 = O-Glucose
1
1
of the silkworm (the race ‘‘Multi-Bi”) contained quercetin
0
Fig. 1. Structure of flavonoids either isolated or detected from the
5
-O-glucoside (5), quercetin 5,4 -di-O-glucoside (2), and
silkworm, Bombyx mori.
0
quercetin 5,7,4 -tri-O-glucoside. In the present study, we
isolated 9 flavonoids (compounds 1–6, 10–12) from the
cocoon shell of the race ‘‘Pure-Mysore” reared on mulberry
leaves. LC–MS analysis of the isolated flavonoids showed
0
di-O-glucoside (2), quercetin 3,3 -di-O-glucoside (3), quer-
cetin 5,3 -di-O-glucoside (4), quercetin 5-O-glucoside (5),
quercetin 7-O-glucoside (6), quercetin (10), kaempferol
5,4 -di-O-glucoside (11), and kaempferol 5-O-glucoside
0
ꢀ
deprotonated ions [MꢀH] with peaks at m/z 625 (com-
0
pounds 1–4), 609 (11), 463 (5 and 6), 447 (12), and 301
(
10), respectively. In addition, reaction with b-glucosidase
(12) (Table 1, Fig. 1). Among these flavonoids, quercetin
0
0
gave quercetin (10) or kaempferol, suggesting that com-
pounds 1–4 were quercetin diglucosides, 5 and 6 were quer-
cetin monoglucosides, 11 was kaempferol diglucoside, and
5,3 -di-O-glucoside (4) and kaempferol 5,4 -di-O-glucoside
(11) are novel natural products. To confirm the chemical
structures of these novel compounds, their structures were
1
13
1
2 was kaempferol monoglucoside.
To identify the glycosylation sites of the flavonoids, the
characterized by H NMR, C NMR, and high-resolution
ESI mass spectroscopic analyses. The negative HR-
FTICR-MS spectrum of 4 showed a prominent peak at
UV–Vis spectra were analyzed in the presence of the shift
reagents as reported by Day et al. (2000). Obtained data
identified quercetin 3,7-di-O-glucoside (1) quercetin 5,4 -
ꢀ
m/z 625 [MꢀH] (observed 625.1424, calculated 625.1410
0
for C H O ); this value corresponds to a diglucoside
2
7
29 17
Table 1
Properties of flavonoids isolated from the cocoon shell of the silkworm, Bombyx mori (Race: Pure Mysore)
a
[MꢀH]ꢀb (m/z)
Cocoon flavonoid
Peak wavelength (nm)
Aglycone
MeOH
+NaOMe
+AlCl
3
+AlCl
3
/HCl
+NaOAc
c
1
2
3
4
5
6
1
1
1
Quercetin 3,7-di-O-glucoside
257, 358
251, 360
254, 365
250, 364
253, 368
256, 373
256, 372
259, 358
259, 362
266, 402
269, 317, 394
272, 431
272, 324, 414
271, 323, 411
Decomposed
Decomposed
275, 397
276, 436
263, 420
265, 423
262, 425
272, 451
269, 441
272, 450
266, 418
268, 424
272, 406
263, 421
265, 425
262, 425
264, 429
266, 432
267, 432
265, 419
267, 423
263, 415
273, 382
262, 383
274, 394
267, 392
259, 384
273, 385
272, 380
273, 388
Q
625
625
625
625
463
463
301
609
447
0
Quercetin 5,4 -di-O-glucoside
Q
Q
Q
Q
Q
0
0
Quercetin 3,3 -di-O-glucoside
Quercetin 5,3 -di-O-glucoside
Quercetin 5-O-glucoside
Quercetin 7-O-glucoside
0 Quercetin
0
d
1 Kaempferol 5,4 -di-O-glucoside
K
2 Kaempferol 5-O-glucoside
375, 323, 405
K
a
3
UV spectral data were recorded in methanol and after addition of various shift reagents NaOMe, AlCl3, AlCl +HCl, NaOAc.
Deprotonated ion.
Quercetin.
b
c
d
Kaempferol.

C. Hirayama et al. / Phytochemistry 69 (2008) 1141–1149
1143
derivative of quercetin (10). The D-glucosyl moiety and the
60
1
A
5
structure of the quercetin moiety were identified by H
1
3
NMR, and C NMR spectroscopic analyses (see data
40
below). Moreover, the cross peak from C-5 (d 159.74) to
5
-O-Glc-H-1 (d 4.88) of glucose and the cross peak from
0
0
C-3 (d 146.81) to 3 -O-Glc-H-1 (d 4.87) of glucose in the
20
HMBC spectrum indicated that a glucose moiety was
attached to the 5-O position and another to the 3 -O posi-
tion of the quercetin (10) (data not shown).
2
2
7
10
0
4
0
60
The structure of 11 was also characterized by the same
method used in the analysis of 4. The negative HR-
FTICR-MS spectrum of 11 showed a prominent peak at
B
C
40
ꢀ
m/z 609 [MꢀH] (observed 609.1470, calculated 609.1461
for C H O ), indicating that 11 was a diglucosyl conju-
2
7
29 16
20
gate of kaempferol. The HMBC analysis confirmed that
two glucopyranosyl moieties were conjugated to the 5-O
1
4
3
5
0
0
800
and 4 -O positions of kaempferol. From these results, we
0
concluded that 4 was quercetin 5,3 -di-O-b-D-glucopyrano-
0
side, and 11 was kaempferol 5,4 -di-O- b-D-
2
glucopyranoside.
600
4
2
00
00
0
2.2. Identification and quantification of flavonoid metabolites
in the silkworm tissues after feeding with a quercetin (10)
diet
1
3 4
10
5
Using the flavonoids commercially available and
authentic compounds purified from the cocoon shell, fla-
vonoid metabolites in the tissues was analyzed by HPLC.
LC-MS were also used to confirm that the metabolites
were quercetin (10) or kaempferol glucosides based on
0
5
15
20
Retention Time (min)
Fig. 2. Representative HPLC chromatograms of tissues obtained from the
silkworm fed a quercetin-supplemented diet. A; midgut; B; hemolymph; C;
silk glands. Detection was performed at 365 nm: 1, quercetin 3,7-di-O-
glucoside; 2, quercetin 5,4 -di-O-glucoside; 3, quercetin 3,3 -di-O-gluco-
side; 4, quercetin 5,3 -di-O-glucoside; 5, quercetin 5-O-glucoside; 7,
ꢀ
the detected deprotonated ions [MꢀH] at m/z 463,
0
0
6
25, 447, 609. Furthermore, decrease of the metabolite
0
peaks with concomitant formation of either quercetin
10) or kaempferol was not observed when tissue samples
quercetin 3-O-glucoside; 10, quercetin.
(
were incubated with glucuronidase and sulfatase, suggest-
ing that no conjugates other than glucosides existed in
the silkworm tissues (data not shown). In the midgut tis-
0
0
sues, quercetin 5,4 -di-O-glucoside (2), quercetin 5,3 -di-
O-glucoside (4), quercetin 5-O-glucoside (5), quercetin
Table 2
Quantification of quercetin metabolites in the silkworm reared on a
quercetin (10) supplemented diet
3
(
1
-O-glucoside (7), and quercetin (10) were detected
Fig. 2A). Total flavonoid content in the midgut was
68 nmol/g fresh tissue, and 1 g fresh midgut tissue con-
a
Quercetin
metabolite
Midgut (nmol/g Silk glands (nmol/ Hemolymph
fresh tissue)
g fresh tissue)
nmol/ml
tained 124 nmol quercetin 5-O-glucoside (5) as the main
metabolite (Table 2).
1
2
3
4
5
7
1
Quercetin 3,7-di-
–
57.9 ± 6.4
7.5 ± 2.2
O-glucoside
Quercetin 5,4 -
In the hemolymph, quercetin 3,7-di-O-glucoside (1),
0
11.0 ± 1.8
–
1829 ± 277
42.2 ± 4.1
82.1 ± 8.8
27.5 ± 16.7
–
59.3 ± 12.9
1.0 ± 0.8
4.5 ± 1.2
3.0 ± 0.8
–
0
0
quercetin 5,4 -di-O-glucoside (2), quercetin, 3 -di-O-gluco-
di-O-glucoside
0
0
Quercetin 3,3 -
side (3), quercetin 5,3 -di-O-glucoside (4), and quercetin
di-O-glucoside
5
-O-glucoside (5) were detected, with the most abundant
0
Quercetin 5,3 -
5.9 ± 1.8
124.4 ± 63.8
9.8 ± 1.6
0
metabolite being quercetin 5,4 -di-O-glucoside (2)
di-O-glucoside
Quercetin 5-O-
glucoside
Quercetin 3-O-
glucoside
(
Fig. 2B). Whole silk glands showed a very similar flavo-
noid profile to that of hemolymph, with ca. 90% of the
0
total flavonoids in the tissue being quercetin 5,4 -di-O-glu-
coside (2) (Fig. 2C, Table 2). These results suggest a second
0 Quercetin
16.8 ± 5.8
–
–
stage of biotransformation in the internal tissues with
0
Total
168.0 ± 74.8
2039 ± 308
75.3 ± 17.8
subsequent glucosylation at the 4 -O position of quercetin
a
5
-O-glucoside (5), which is produced in the midgut as the
Values are the means ± SD (n = 5). Metabolite quantity is expressed as
quercetin 5-O-glucoside (5) equivalents.
first-pass metabolite of quercetin (10) after oral absorption.

1144
C. Hirayama et al. / Phytochemistry 69 (2008) 1141–1149
Kaempferol glycosides were not detected in the tissues
A
examined after feeding with a quercetin diet. In addition,
the cocoon shells of the silkworms reared with mulberry
leaves had much lower amounts of kaempferol glycosides
than quercetin glycosides, probably because the mulberry
leaves contained smaller amounts of the former. However,
10
5
6
7
8
9
0
we confirmed that kaempferol 5,4 -di-O-glucoside (11) and
B
C
kaempferol 5-O-glucoside (12) were the main metabolites
when the insects were fed a diet containing kaempferol only
(
data not shown).
2
.3. Antioxidant capacity in the silkworm tissues
The tissue antioxidative status of silkworms fed a quer-
cetin-supplemented diet was compared with that of insects
fed a control diet (non-quercetin diet) (Table 3). The mid-
gut extract of the quercetin-supplemented insects exhibited
a strong antioxidant property; 1 g of fresh midgut con-
tained an antioxidant equivalent to 3 lmol Trolox, about
twice that of the control-diet silkworms. In addition, the
larvae fed a quercetin diet also had stronger antioxidant
activity in the silk glands than those fed a control diet.
No significant difference between the groups was found in
the antioxidative capacity of the hemolymph. It has been
reported that the position of flavonoid glycosylation signif-
icantly affects their biological activity and potential benefits
to humans (Day et al., 2000; Teixeira et al., 2005). For
example, quercetin glucuronidated at the 3-O position
has a protective effect against cardiovascular diseases
5
D
E
2
4
(
Yoshizumi et al., 2002), and has an anti-oxidant effect
on copper-ion-induced lipid peroxidation in human plasma
low-density lipoprotein (Moon et al., 2001). In contrast,
0
5
10
15
20
Retention Time (min)
0
0
substitution at the 3 -O and 4 -O positions results in a
marked decrease in antioxidant activity (Yamamoto
et al., 1999). Our study demonstrated that quercetin (10)
exerted its anti-oxidant activity, at least in part, even after
it was glucosylated at the 5-O position. However, total
anti-oxidant activity in the hemolymph did not change,
probably because the hemolymph mainly contained metab-
Fig. 3. HPLC chromatograms of silkworm UDP-glucosyltransferase
assays: A, incubation mixture of midgut membrane fraction with quercetin
10) and UDP-glucose; B, incubation mixture of midgut membrane
fraction with quercetin (10) without UDP-glucose; C, incubation mixture
of midgut soluble fraction with quercetin (10) and UDP-glucose; D,
incubation mixture of middle silk gland membrane fraction with querce-
tin-5-glucoside (5) and UDP-glucose; E, incubation mixture of middle silk
glands membrane fraction with quercetin-5-glucoside (5) without UDP-
glucose. Detection was performed at 365 nm: 2, quercetin 5,4 -di-O-
glucoside; 4, quercetin 5,3 -di-O-glucoside; 5, quercetin 5-O-glucoside; 6,
quercetin 7-O-glucoside; 7, quercetin 3-O-glucoside; 8, quercetin 4 -O-
glucoside; 9, quercetin 3 -O-glucoside; 10, quercetin.
(
0
olites, such as quercetin 5,4 -di-O-glucoside (2), whose cat-
0
echol moiety in the B-ring was glucosylated.
0
0
0
2.4. Determination and tissue distribution of UDP-glucosyl
transferase (UGT) activity
Table 3
Total antioxidant activity in tissues of the silkworms fed an experimental
a
diet
Fig. 3A–C indicate typical HPLC chromatograms of
quercetin (10) incubation with the midgut extracts. When
the midgut membrane extract was incubated with UDPG
and quercetin, five metabolites (5–9) were formed
b
Diet group
Control
TEAC (lmol/g fresh weight) Hemolymph TEAC
c
(
lmol/ml)
Midgut
Silk glands
1.84 ± 0.27
2.32 ± 0.36
6.73 ± 0.66
2.60 ± 0.26
2.69 ± 0.28
(Fig. 3A). Incubation of the sample with b-glucosidase
Quercetin (10) 0.5% 3.45 ± 0.60
resulted in the disappearance of the metabolite peaks, with
concomitant formation of quercetin (10), indicating that
these metabolites were glucoside derivatives of quercetin
(10) (data not shown). These metabolites were identified
a
Values are means ± SD (n = 5).
TEAC, Trolox-equivalent capacity. Data are expressed as lmol of
b
Trolox equivalent/g fresh weight.
c
Data are expressed as lmol of Trolox equivalent/ml.

C. Hirayama et al. / Phytochemistry 69 (2008) 1141–1149
Table 5
1145
based on their co-elution with authentic reference
Glucosyltransferase activity toward quercetin 5-O-glucoside (5) in the
silkworm tissues
standards (5: quercetin 5-O-glucoside, 6: quercetin 7-O-glu-
coside, 7: quercetin 3-O-glucoside, 8: quercetin 4 -O-gluco-
side, 9: quercetin 3 -O-glucoside) (Fig. 3A). However, no
a
0
0
Tissue
Specific activity (pkat/mg protein)
0
0
glucosylation of quercetin (10) occurred in the absence of
UDP-glucose (Fig. 3B), nor did glucosides form when the
cytosolic extract was used instead of the membrane frac-
tion (Fig. 3C). The membrane fractions from other tissues
also had UDP-glucose-dependent activity that glucosylated
the OH groups of quercetin (10), while the cytosolic frac-
tions had no activity (data not shown). The results of the
present study thus agree with the report by Ahmad and
Hopkins (1992) that UGTs were always concentrated in
the heavy cellular particles, probably associated with
microsomes. UDP-glycosytransferases are a large family
of enzymes capable of transferring the glycosyl group from
the substrate donor (UDP-glucose, UDP-glucuronic acid,
UDP-galactose, etc.) to wide ranges of substrate acceptor
molecules (Bowles et al., 2005; Bowles et al., 2006; Coutin-
ho et al., 2003). In general, glucosylation seems to be a
more important pathway for the detoxication of plant
phenolics in insects (Ahmad and Hopkins, 1993; Rausell
et al., 1997; Real et al., 1991), rather than glucuronidation,
which is the predominant pathway in mammals. We con-
firmed that the silkworm UGTs use only UDP-glucose as
the substrate donor (data not shown).
3-O-Glc
7-O-Glc
3 -O-Glc
4 -O-Glc
Midgut
Fat body
–
–
–
–
–
–
–
–
0.5 ± 0.2
8.8 ± 2.3
3.7 ± 1.0
4.7 ± 1.3
4.0 ± 1.7
47.0 ± 8.2
109 ± 22
169 ± 26
Posterior silk glands
Middle silk glands
a
Values are means ± SD (n = 5).
tion of quercetin 5-O-glucoside (5) occurred in the absence
of UDP-glucose (Fig. 3E). The UGT activity toward quer-
cetin 5-O-glucoside (5) in each tissue is demonstrated in
Table 5. The rate of glucosylation for quercetin 5-O-gluco-
side (5) was very low with the midgut membrane extract,
although the activity of glucosylation toward quercetin
aglycone was high (see Table 4). The rate of glucosylation
toward quercetin 5-O-glucoside (5) was very high in the fat
body and in both the posterior and middle silk glands, in
0
which the dominant product, quercetin 5,4 -di-O-glucoside
(2), was produced along with a minor product, quercetin
0
5,3 -di-O-glucoside (4). Interestingly, no products were
0
formed other than quercetin-5,4 -diglucoside (2) and quer-
cetin-5,3 -diglucoside (4), irrespective of the tissues investi-
0
gated. Thus, the in vitro study indicates that glucosylation
of quercetin (10) at the 5-O position occurs preferably in
the intestinal tissue followed by subsequent glucosylation
at the 4 -O and 3 -O positions in the fat body and silk gland
tissues.
UGT activity toward quercetin (10) in each tissue was
calculated, and the values, shown in Table 4, indicate that
the rate of glucosylation of OH positions of quercetin var-
ied between tissues. UGT in the midgut produced 5-, 3-, 7-,
0
0
0
0
3
-, and 4 -O-glucoside, with a preference for the formation
Recently, there has been increasing interest in the bio-
catalytic synthesis of glycosides with potential medicinal
properties using regioselective UGTs (Kramer et al.,
2003; Lim, 2005; Weis et al., 2006; Willits et al., 2004),
because the chemical synthesis of specific glycosides is quite
complicated. For example, to produce any single monog-
lucoside of quercetin (10), four other hydroxyl groups
should be protected in each step of the reactions, but this
may reduce the yield of the target product. The regioselec-
tivity of UGTs can offer a potential solution to overcome
the problems of chemical synthesis of specific glycosides.
In order to produce specific quercetin glucosides, a large
multigene family of UGTs in Arabidopsis thaliana has been
expressed as recombinant enzymes in Escherichia coli by
Lim et al. (2004). They analyzed the activity of 91 recombi-
nant enzymes for in vitro activity toward quercetin (10)
and found that 29 enzymes were capable of glucosylating
the compound, and that different monoglucosides of quer-
cetin (10) (3-, 7-, 3 -, and 4 -O-glucosides) were synthesized
by using these UGTs. However, significant activity toward
5-O position of flavonols has not been observed by any of
the enzymes assayed. Similarly, studies of the regioselectiv-
ity of glucuronidation of quercetin (10) demonstrated that
neither human liver and intestinal cell free extracts nor
individual UGT isoforms generate 5-glucuronosyl metabo-
lites of quercetin (10) (Boersma et al., 2002; Day et al.,
2000). The OH group at the C-5 position of flavone is
of the 5-glucoside. The fat body had strong activity of
0
UGT that glucosylated quercetin at the 3-OH and 4 -OH
groups. UGT in both the posterior and middle silk glands
showed a regiospecificity toward the glucosylation at the 7-
O position of quercetin (10). Subsequently, the capacity of
the silkworm membrane extract of each tissue to transfer
glucose from UDP-glucose to quercetin-5-O-glucoside (5)
was investigated. Fig. 3D shows that two metabolites (2:
0
0
quercetin 5,4 -di-O-glucoside and 4: quercetin 5,3 -di-O-
glucoside) were formed when the membrane extract of
the middle silk glands was incubated with UDP-glucose
and quercetin 5-O-glucoside (5). However, no glucosyla-
Table 4
a
Glucosyltransferase activity toward quercetin (10) in the silkworm tissues
Tissue
Specific activity (pkat/mg protein)
0
0
0
0
3-O-Glc
5-O-Glc
7-O-Glc
3 -O-Glc 4 -O-Glc
Midgut
3.0 ± 1.0
38.7 ± 4.7
3.2 ± 0.8 0.5 ± 0.2
1.3 ± 0.5
Fat body 61.4 ± 20.2 18.0 ± 3.8 18.0 ± 3.8 5.3 ± 0.8 63.5 ± 12.7
Posterior
silk
–
–
133 ± 21 0.5 ± 0.3
8.7 ± 1.5
glands
Middle
silk
–
–
186 ± 27
–
10.8 ± 2.2
glands
a
Values are means ± SD (n = 5).

1146
C. Hirayama et al. / Phytochemistry 69 (2008) 1141–1149
the most inert site for glucuronidation, since the carbonyl
group at C-4 could produce steric hindrance for the adja-
cent OH-group at C-5. In addition, strong intra-molecular
hydrogen bonding of OH at C-5 with the carbonyl group at
C-4 also prevents glycosylation of OH at C-5 (Zhang et al.,
the pale yellow cocoon shell as described in a previous
report (Tamura et al., 2002) with some modifications.
Briefly, MeOH–H O (7:3, v/v) extract from the cocoon
2
shell was concentrated by evaporation and diluted with dis-
tilled H O. This aqueous solution was applied to a solid
2
2
006). Therefore, the inert glycosylation activity of OH-
phase extraction cartridge (Oasis HLB, 35 ml, Waters, Mil-
group at C-5 could explain why 5-O-glycosylated flavones
and flavonols are very rare compounds as reported by
Iwashina et al. (1995). We consider that the unique silk-
worm midgut UGT with a regiospecificity toward the gly-
cosylation of the 5-O position of quercetin (10) would be
useful as a novel biocatalyst for the production of potential
pharmaceuticals.
ford, MA, USA). After washing with distilled H O, the col-
2
umn was eluted with MeOH–H O (1:1, v/v) followed by
2
MeOH. The MeOH–H O (1:1, v/v) fraction contained
2
mainly flavonoid diglucosides, and the MeOH fraction
contained monoglucosides and aglycones. Each fraction
was concentrated by evaporation and loaded on a 2.6 ꢁ
60 cm Toyopearl HW-40F column (TOSOH, Tokyo,
Japan) eluted with a linear gradient of MeOH:H O:H-
2
CO H (40:59.9:0.1, v/v/v) to MeOH:HCO H (99.9:0.1,v/
2
2
3
. Conclusions
v). Fractions containing flavonoids were further purified
by preparative reversed-phase HPLC (Nova-Pak C18,
19 ꢁ 300 mm, Waters).
In conclusion, our results demonstrated for the first time
that a UGT with a preferred 5-O regioselectivity for quer-
cetin (10) glycosylation was operative in the silkworm;
however, the physiological function of the ‘‘unusual” regi-
oselectivity of the enzyme remains to be established.
Besides the antioxidant potential, 5-O-glucosylated flavo-
noids might act as antibiotic and antiviral agent, or for
protection against predators. Further research to obtain
more information about the role of 5-O-glucosylated flavo-
noids in the insect and the isolation of the UGT enzyme are
necessary to understand the functional significance of the
unique glucosylation pathway.
4.3. Deconjugation of isolated flavonoids
Hydrolysis with b-glucosidase was carried out to iden-
tify the flavonoids isolated from the cocoon shell. Each fla-
vonoid (ca 2.5 lg) was treated with 1.3 U b-glucosidase in
1 ml 25 mM acetate buffer (pH 5.0) at 37 °C for 1 h. After
incubation, four volumes of MeOH were added to stop the
enzyme reaction. The aglycones formed were identified by
HPLC using commercially available standards.
4.4. Identification of conjugation positions by UV spectra
4
. Experimental
Information on the conjugation positions of flavonoids
was determined indirectly by UV spectra according to
Markham (1982). Standards of flavonols commercially
available (quercetin (10), quercetin 3-O-glucoside (7), quer-
4
.1. Chemicals
0
0
Quercetin (10), uridine-5 -diphospho (UDP)-glucose,
cetin 4 -O-glucoside (8), kaempferol, kaempferol 3-O-glu-
rutin, and b-glucosidase from almond were purchased from
Sigma Chemical Co. (St. Louis, MO, USA). Quercetin 3-O-
glucoside (7) and quercetin 4 -O-glucoside (8), kaempferol
coside), and each isolated flavonoid glucoside were
dissolved in MeOH (20 ml) and diluted until the absor-
bance of the maxima was between 0.6 and 0.7. Shift
reagents were added sequentially to the freshly prepared
stock solution (2.5 ml), and the UV spectra were recorded
on a UV-2500 PC spectrometer (Shimadzu, Tokyo, Japan).
0
3
-O-glucoside, and kaempferol were purchased from
Extrasynthase (Genay, France). Trolox (6-hydroxy-2,
,7,8-tetramethylchroman-2-carboxylic acid), a water-solu-
5
ble analogue of vitamin E, was from EMD Biosciences (La
0
Jolla, CA, USA). 2,2 -Azinobis- (3-ethylbenzothiazoline-6-
4.5. Structural analysis for novel flavonol glucosides
sulfonic acid) (ABTS) was obtained from Wako Pure
Chemical Industries (Osaka, Japan). All other chemicals
and solvents used were of either analytical or HPLC grade.
High-resolution ESI mass analysis for novel glucosyl
conjugates purified from the cocoon shell were carried
out using an Apex II 70e Fourier-transform ion cyclo-
tron resonance (FTICR) mass spectrometer (Bruker,
Gothenburg, Sweden). These compounds were character-
4.2. Isolation of flavonoids from the cocoon shell of the
silkworm
1
13
ized by H NMR and C NMR spectroscopic analyses
using either a Bruker AVANCE 500 spectrometer
To obtain authentic flavonoids available for the identifi-
1
13
cation and quantification of quercetin metabolites in the
silkworm tissues, flavonoids were isolated from the cocoon
shell of the silkworm race ‘‘Pure-Mysore”. These insects
were reared on mulberry leaves (M. alba) throughout the
larval stage. Flavonoids were extracted and purified from
(500 MHz for H; 125 MHz for
C) or a Bruker
1
AVANCE 800 spectrometer (800 MHz for
H;
1
3
200 MHz for
C) in CD OD. Chemical shifts were
3
expressed in ppm relative to tetramethylsilane (TMS) as
an internal standard.

C. Hirayama et al. / Phytochemistry 69 (2008) 1141–1149
1147
0
4
.6. Chemical synthesis of quercetin 3 -glucoside
Frozen midgut and whole silk glands were homogenized
in three volumes (v/w) of MeOH–H O (7:3, v/v) at 4 °C.
2
0
Among quercetin monoglucosides, quercetin 3 -O-gluco-
After centrifugation at 20,000g for 10 min, the precipitates
side (9) was neither commercially available nor abundant
in the cocoon shell of the silkworm. To use quercetin 3 -
were re-homogenized in MeOH–H O (7:3, v/v). After cen-
2
0
trifugation, the second extract was combined with the first
supernatant. This step was performed again. The combined
extract was diluted with MeOH to a constant volume and
filtered, and then the supernatant was used for HPLC anal-
ysis and for determination of the total antioxidant activity.
O-glucoside (9) as an authentic standard, we synthesized
0
it from quercetin 5,3 -di-O-glucoside (4) purified from the
0
cocoon shell. Quercetin 5,3 -di-O-glucoside (4) (1 mg) was
dissolved in MeOH–H O (20 ml, 7:3, v/v), and 4 ml 6 M
2
HCl was added to the solution. Then the acidic solution
was heated at 100 °C for 5 min. Acid hydrolysis at the 5-
O position of quercetin (10) occurred much more rapidly
4.9. HPLC analysis
0
0
than that at 3 -O, predominantly giving quercetin 3 -O-glu-
Flavonoids in the tissues were identified and quantified
with a Shimadzu HPLC system equipped with an LC-7A
pump, a CTO-10A column oven, and an SPD-7AV UV–
Vis detector monitoring at 365 nm. Samples were loaded
onto a Nova-Pak C18 column (150 ꢁ 3.9 mm i.d.). Elution
was performed using H O:HCO H (99.8:0.2 v/v) as solvent
coside (9). The hydrolysate was diluted with distilled H O
and applied to a solid phase extraction cartridge (Oasis
2
HLB, 20 ml, Waters). After washing with distilled H O,
2
the column was eluted with MeOH. The eluent was concen-
trated by evaporation and loaded to a preparative
reversed-phase HPLC (Nova-Pak C18, 300 ꢁ 19 mm i.d.,
2
2
A and CH CN as solvent B at a flow rate of 1.0 ml/min at
3
0
Waters) to yield quercetin 3 -O-glucoside (9). The glucosy-
40 °C. Flavonoids were separated out with linear gradient
from 7% B to 40% B for 20 min and then to 100% for
10 min. Quercetin (10) metabolites were identified based
on their co-elution with either commercially available stan-
dards or authentic reference samples purified from the
cocoon shell. Flavonoids were quantified using quercetin
5-O-glucoside (5) as a reference compound.
lation site of the obtained compound was confirmed by the
pattern of UV spectra produced by the shift reagents as
described above.
4.7. Dietary administration of quercetin (10) and collection
of silkworm tissues
The larvae of Pure-Mysore were reared on a diet of mul-
berry leaf powder from hatching to the 4th ecdysis (Hiray-
ama et al., 1997). Female larvae just after the 4th ecdysis
were randomly divided into two groups. The control group
received a semi-synthetic diet for 5 days. The semi-syn-
thetic diet had a composition similar to that of the diet used
during the young larval stage, except that the mulberry leaf
powder was omitted and the contents of soybean protein
and cellulose powder were increased to 40 g/100 g and
4.10. LC–MS analysis
LC–MS analyses of flavonoids in the cocoon shells and
tissues were carried out by an HP 1100 series HPLC (Agilent
Technologies, Santa Clara, CA, USA) equipped with an HP
1100MSD mass spectrometer. An aliquot (10 ll) of a sample
was injected into the LC–MS and separated by a C18
reversed phase column, 150 mm ꢁ 2.0 mm i.d. (Nova-Pak
C18, Waters), at a flow rate of 0.3 ml/min. The column tem-
perature was maintained at 40 °C. The mobile phase con-
1
5.3 g/100 g, respectively. The test group received the
semi-purified diet supplemented with 0.5 g quercetin (10)
per 100 g for 5 days. After the larvae fed on the respective
diets ad libitum for 5 days, the hemolymph was collected
from the larvae by cutting prolegs. Two volumes (v/v) of
MeOH were added to the collected hemolymph, and dena-
tured protein was removed by centrifugation at 20,000g for
sisted of solvents A (0.2% aq. HCO H) and B (0.2%
2
HCO H in CH CN). Metabolites were separated with a lin-
2
3
ear gradient from 7% B to 40% B over 20 min and then to
100% for 5 min. UV detection was carried out using an HP
1100 photodiode array detector to facilitate peak assign-
ment. UV–Vis spectra were recorded in the 190–450 nm
range, and the chromatograms were acquired at 190, 365,
and 420 nm. The eluent was ionized by negative electrospray
ionization. The electrospray mass spectrometer conditions
were as follows: negative ion mode; fragmentor voltage,
70 V; capillary voltage, 3500 V; drying gas (nitrogen) flow,
1
0 min, then immediately frozen at ꢀ80 °C. After the
hemolymph was drawn, the larvae were quickly dissected
and the silk glands and midgut were collected. These tissues
were well rinsed with an ice-cold 0.85% KCl solution, then
rapidly frozen and stored at ꢀ80 °C until analysis.
10.0 L/min; nebulizer pressure, 25 psig; drying gas tempera-
4.8. Preparation of tissue samples for flavonoid analyses and
antioxidant activity
ture, 350 °C. The flavonoid conjugates were identified on the
basis of their retention times, UV–Vis spectra, and electro-
spray mass spectra.
The MeOH extracts of hemolymph stored at ꢀ80 °C
were thawed at room temperature and centrifuged at
4.11. In vitro assay of UDP-glucosyltransferase (UGT)
2
0,000g for 10 min, after which the precipitates were
removed. The supernatant was passed through a 0.45 lm
filter and used for analysis.
Female larvae after the 4th ecdysis were fed semi-syn-
thetic diet for 5 days, after which the larvae were dissected

1148
C. Hirayama et al. / Phytochemistry 69 (2008) 1141–1149
and the midgut, posterior and middle silk glands, and fat
body were collected. These tissues were well rinsed with
an ice-cold 0.85% KCl solution, then rapidly frozen and
stored at ꢀ80 °C until analysis.
standard solution (final concentration of 0–0.5 mM) in
EtOH–H O (4:1, v/v) was prepared and assayed under
2
the same conditions. The results were expressed in terms
of trolox equivalent antioxidant capacity.
Frozen organs or tissues were homogenized in four vol-
umes (v/w) of 0.1 M potassium phosphate buffer (pH 7.0)
and centrifuged at 23,000g for 20 min at 4 °C. The superna-
tants were saved for measuring the fractional distribution
of UGT. The precipitates were re-homogenized in 0.1 M
potassium phosphate buffer (pH 7.0) containing 0.8%
sodium cholate at 4 °C. After extraction with shaking for
0
4.13. Quercetin 5,3 -di-O-b-D-glucopyranoside (compound
4)
1
H NMR (CD OD): d 3.434ꢀ3.48 (3H, m, 5-O-Glc-H-4,
3
0
5-O-Glc-H-5, 3 -O-Glc-H-4), 3.49–3.51 (3H, m, 5-O-Glc-H-
3, 3 -O-Glc-H-3, 3 -O-Glc-H-5), 3.55 (1H, dd, J = 7.8 and
9.3 Hz, 3 -O-Glc-H-2), 3.63 (1H, dd, J = 7.8 and 9.3 Hz,
0
0
0
3
2
0 min at 4 °C, the homogenates were centrifuged at
3,000g for 20 min. The supernatant was used as the
5-O-Glc-H-2), 3.75 (1H, dd, J = 5.2 and 12.1 Hz, 5-O-
0
enzyme solution of the membrane fraction.
Glc-H-6a), 3.79 (1H, dd, J = 5.3 and 12.1 Hz, 3 -O-Glc-
UGT was assayed as described by Ahmad and Hopkins
1993). Briefly, the incubation mixtures consisted of 0.1 M
potassium phosphate (pH 7.0), 5 mM UDPG, 25 mM
H-6a), 3.94 (1H, dd, J = 2.0 and 12.1 Hz, 5-O-Glc-H-6b),
0
(
3.97 (1H, dd, J = 2.1 and 12.1 Hz, 3 -O-Glc-H-6b), 4.87
0
(1H, d, J = 7.8 Hz, 3 -O-Glc-H-1), 4.88 (1H, d,
MgCl , 5 mM D-glucuronolactone (as an inhibitor of b-glu-
J = 7.8 Hz, 5-O-Glc-H-1), 6.72 (1H, d, J = 2.1 Hz, H-8),
6.790 (1H, d, J = 2.1 Hz, H-6), 6.97 (1H, d, J = 8.6 Hz,
H-5 ), 7.90 (1H, dd, J = 2.0 and 8.6 Hz, H-6 ), 8.14 (1H,
2
cosidase), 0.1 mM quercetin (10) or quercetin 5-O-gluco-
side (5), and enzyme preparation in a final incubation
volume of 0.1 ml. The enzyme reactions were started by
addition of UDP-glucose, then incubated for 30 min at
0
0
0
d, J = 2.0 Hz, H-2 ).
1
3
0
C NMR (CD OD): d 62.48 (3 -O-Glc-C-6), 62.51 (5-
3
0
3
7 °C. The reaction was stopped by the addition of MeOH
O-Glc-C-6), 71.22 (5-O-Glc-C-4), 71.28 (3 -O-Glc-C-4),
74.76 (5-O-Glc-C-2), 74.85 (3 -O-Glc-C-2), 77.34 (5-O-
Glc-C-3), 77.59 (3 -O-Glc-C-3), 78.47 (3 -O-Glc-C-5),
78.59 (5-O-Glc-C-5), 98.89 (C-8), 104.35 (3 -O-Glc-C-1),
0
(
0.2 ml). Precipitated protein was removed by centrifuga-
0
0
0
tion at 20,000g for 10 min, then the supernatant was used
for determination of glucosides by HPLC. Control incuba-
tions were run, in which the enzyme source, substrates, or
UDP-glucose was omitted. The glucosyl metabolites
formed by tissue extracts were measured on the basis of
the peak area in the HPLC chromatogram at 365 nm, using
commercially available standards or authentic reference
samples purified from the cocoon shell, as described above.
The protein concentration was measured by a commercial
assay kit (Coomasie Plus, Pierce, Rockford, IL, USA),
using bovine serum albumin as a standard.
104.45 (C-6), 104.93 (5-O-Glc-C-1), 107.42 (C-4a), 117.18
0
0
0
0
0
(C-5 ), 117.77 (C-2 ), 124.23 (C-1 ), 124.89 (C-6 ), 138.91
0
(C-3), 145.05 (C-2), 146.81 (C-3 ), 150.26 (C-4 ), 159.51
(C-8a), 159.74 (C-5), 165.10 (C-7), 173.84 (C-4), HR-
ꢀ
FTICR-MS m/z 625.1424 [MꢀH]
(calculated for
C H O 625.1410).
2
7
29 17
0
4.14. Kaempferol 5,4 -di-O-b-D-glucopyranoside (compound
11)
1
0
4.12. Quantification of antioxidant activity of the silkworm
tissues
H NMR (CD OD): d 3.39ꢀ3.42 1H, m, 4 -O-Glc-H-4),
3
0
0
0
3.48–3.51 (3H, m, 4 -O-Glc-H-2, 4 -O-Glc-H-3, 4 -O-Glc-
H-5), 3.4 (1H, dd, J = 8.3 and 9.8 Hz, 5-O-Glc-H-4), 3.48
(1H, ddd, J = 2.0, 5.2 and 9.8 Hz, 5-O-Glc-H-5), 3.50
(1H, dd, J = 8.3 and 9.8 Hz, 5-O-Glc-H-3), 3.63 (1H, dd,
J = 7.8 and 9.3 Hz, 5-O-Glc-H-2), 3.71 (1H, dd, J = 5.8
and 12.2 Hz, 4 -O-Glc-H-6a), 3.76 (1H, dd, J = 5.2 and
12.1 Hz, 5-O-Glc-H-6a), 3.92 (1H, dd, J = 2.2 and
12.2 Hz, 4 -O-Glc-H-6b), 3.943 (1H, dd, J = 2.0 and
12.1 Hz, 5-O-Glc-H-6b), 4.87 (1H, d, J = 7.8 Hz, 5-O-
Aqueous methanolic extracts of hemolymph, midgut,
and silk glands were tested for their ability to quench the
ABTS radical as previously reported (Re et al., 1999).
The ABTS stock solution (7.0 mM) was prepared and
reacted with 2.45 mM potassium persulfate (final concen-
tration) to generate the radical cation. After the mixture
was kept in the dark at room temperature for 12–16 h to
allow the completion of radical generation, the concentra-
tion of the ABTS radical solution was adjusted with MeOH
to an absorbance of 0.7 ± 0.02 (mean ± SD) at 734 nm. To
determine the scavenging activity, 1 ml of diluted ABTS
solution was added to 10 ll sample solution, and the absor-
bance was measured at 734 nm 5 min after the initial mix-
ing, using MeOH as blank. The percentage inhibition was
0
0
0
Glc-H-1), 5.01–5.03 (1H, m, 4 -O-Glc-H-1), 6.66 (1H, d,
J = 2.1 Hz, H-8), 6.791 (1H, d, J = 2.1 Hz, H-6), 7.22
(2H, d, J = 9.0 Hz, H-3 and 5 ), 8.18 (2H, d, J = 9.0 Hz,
0
0
0
0
H-2 and 6 ).
1
3
0
C NMR (CD OD): d 62.5 0 4 -O-Glc-C-6), 62.52 (5-O-
3
0
Glc-C-6), 71.22 (5-O-Glc-C-4), 71.34 (4 -O-Glc-C-4), 74.76
(5-O-Glc-C-2), 74.88 (4 -O-Glc-C-2), 77.33 (5-O-Glc-C-3),
77.98 (4 -O-Glc-C-3), 78.28 (4 -O-Glc-C-5), 78.59 (C-5 ),
98.76 (C-8), 101.83 (4 -O-Glc-C-1), 104.40 (C-6), 104.90
(5-O-Glc-C-1), 107.49 (C-4a), 117.47 (2C, C-3 and 5 ),
0 0 0
126.44 (C-1 ), 130.19 (2C, C-2 and 6 ), 139.20 (C-3),
0
0
0
00
calculated by the equation% inhibition = (Ac ꢀ A )/
S
0
A ꢁ 100 where A is the absorbance of the control and
c
c
0 0
As is the absorbance of the samples. The Trolox (6-
hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic
acid)

C. Hirayama et al. / Phytochemistry 69 (2008) 1141–1149
1149
0
1
1
[
44.86 (C-2), 160.14 (C-4 ), 159.59 (C-8a), 159.81 (C-5),
Lahtinen, M., Kapari, L., Kentta, J., 2006. Newly hatched neonate larvae
can glycosylate: the fate of Betula pubescens bud flavonoids in first
instar Epirrita autumnata. J. Chem. Ecol. 32, 537–546.
Lim, E.K., 2005. Plant glycosyltransferases – their potential as novel
biocatalysts. Chem.-Eur. J. 11, 5486–5494.
65.14 (C-7), 173.95 (C-4), HR-FTICR-MS m/z 609.1470
ꢀ
MꢀH] (calculated for C H O 609.1461).
2
7
29 16
Lim, E.K., Ashford, D.A., Hou, B.K., Jackson, R.G., Bowles, D.J., 2004.
Arabidopsis glycosyltransferases as biocatalysts in fermentation for
regioselective synthesis of diverse quercetin glucosides. Biotechnol.
Bioeng. 87, 623–631.
Markham, K.R., 1982. Techniques of Flavonoid Identification. Academic
press, London.
Moon, J.H., Tsushida, T., Nakahara, K., Terao, J., 2001. Identification of
quercetin 3-O-b-D-glucuronide as an antioxidative metabolite in rat
plasma after oral administration of quercetin. Free Radic. Biol. Med.
Acknowledgments
We would like to thank Ms. Ikuko Maeda and Dr. Tak-
ashi Murata (Instrumental Analysis Center for Food
Chemistry) for their technical help with the NMR and
MS measurements. We also thank Ms. Mayumi Hazeyama
(
National Institute of Agrobiological Sciences) for her
3
0, 1274–1285.
Onogi, A., Osawa, K., Yasuda, H., Sakai, A., Morita, H., Itokawa, H.,
994. Flavonol glycosides from the leaves of Morus alba L. Sho-
technical assistance.
1
yakugaku Zasshi 47, 423–425.
References
Rausell, C., Llorca, J., Real, M.D., 1997. Separation by FPLC chromato-
focusing of UDP-glucosyltransferases from three developmental stages
of Drosophila melanogaster. Arch. Insect Biochem. Physiol. 34, 347–
Ahmad, S.A., Hopkins, T.L., 1992. Phenol b-glucosyltransferase and b-
glucosidase activities in the tobacco hornworm larva Manduca sexta
3
58.
Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., Rice-
Evans, C., 1999. Antioxidant activity applying an improved ABTS
radical cation decolorization assay. Free Radic. Biol. Med. 26, 1231–
(L) – properties and tissue localization. Arc. Insect Biochem. Physiol.
21, 207–224.
Ahmad, S.A., Hopkins, T.L., 1993. b-Glucosylation of plant phenolics by
phenol b-glucosyltransferase in larval tissues of the tobacco horn-
worm, Manduca sexta (L). Insect Biochem. Mol. Biol. 23, 581–589.
Boersma, M.G., van der Woude, H., Bogaards, J., Boeren, S., Vervoort,
J., Cnubben, N.H.P., van Iersel, M., van Bladeren, P.J., Rietjens, I.,
1
237.
Real, M.D., Ferre, J., Chapa, F.J., 1991. UDP-glucosyltransferase activity
toward exogenous substrates in Drosophila melanogaster. Anal. Bio-
chem. 194, 349–352.
Salminen, J.-P., Lahtinen, M., Lempa, K., Kapari, L., Haukioja, E.,
Pihlaja, K., 2004. Metabolic modifications of birch leaf phenolics by an
herbivorous insect: detoxification of flavonoid aglycones via glycosyl-
ation. Z. Naturforsch. (C) 59, 437–444.
Simmonds, M.S.J., 2003. Flavonoid–insect interactions: recent advances
in our knowledge. Phytochemistry 64, 21–30.
Tamura, Y., Nakajima, K.-I., Nagayasu, K.-i., Takabayashi, C., 2002.
Flavonoid 5-glucosides from the cocoon shell of the silkworm, Bombyx
mori. Phytochemistry 59, 275–278.
Teixeira, S., Siquet, C., Alves, C., Boal, I., Marques, M.P., Borges, F.,
Lima, J.L., Reis, S., 2005. Structure-property studies on the antiox-
idant activity of flavonoids present in diet. Free Radic. Biol. Med. 39,
2
002. Regioselectivity of phase II metabolism of luteolin and quercetin
by UDP-glucuronosyl transferases. Chem. Res. Toxicol. 15, 662–670.
Bowles, D., Isayenkova, J., Lim, E.K., Poppenberger, B., 2005. Glyco-
syltransferases: managers of small molecules. Curr. Opin. Plant Biol. 8,
254–263.
Bowles, D., Lim, E.K., Poppenberger, B., Vaistij, F.E., 2006. Glycosyl-
transferases of lipophilic small molecules. Annu. Rev. Plant Biol. 57,
567–597.
Coutinho, P.M., Deleury, E., Davies, G.J., Henrissat, B., 2003. An
evolving hierarchical family classification for glycosyltransferases. J.
Mol. Biol. 328, 307–317.
Day, A.J., Bao, Y.P., Morgan, M.R.A., Williamson, G., 2000. Conjuga-
tion position of quercetin glucuronides and effect on biological activity.
Free Radic. Biol. Med. 29, 1234–1243.
Doi, K., Kojima, T., Makino, M., Kimura, Y., Fujimoto, Y., 2001.
Studies on the constituents of the leaves of Morus alba L. Chem.
Pharm. Bull. 49, 151–153.
Fujimoto, N., Hayashiya, K., Nakajima, K., 1959. Studies on the pigments
of cocoon. (IV) The formation and translocation of the pigments of
green cocoon in the silkworm larvae. J. Seric. Sci. Jpn.. 28, 30–32.
Hirayama, C., Konno, K., Shinbo, H., 1997. The pathway of ammonia
assimilation in the silkworm, Bombyx mori. J. Insect Physiol. 43, 959–
1
099–1108.
Weis, M., Lim, E.K., Bruce, N., Bowles, D., 2006. Regioselective
glucosylation of aromatic compounds: screening of a recombinant
glycosyltransferase library to identify biocatalysts. Angew. Chem.-Int.
Ed.. 45, 3534–3538.
Wiesen, B., Krug, E., Fiedler, K., Wray, V., Proksch, P., 1994.
Sequestration of host-plant-derived flavonoids by lycaenid butterfly
Polyommatus icarus. J. Chem. Ecol. 20, 2523–2538.
Willits, M.G., Giovanni, M., Prata, R.T., Kramer, C.M., De Luca, V.,
Steffens, J.C., Graser, G., 2004. Bio-fermentation of modified flavo-
noids: an example of in vivo diversification of secondary metabolites.
Phytochemistry 65, 31–41.
Yamamoto, N., Moon, J.H., Tsushida, T., Nagao, A., Terao, J., 1999.
Inhibitory effect of quercetin metabolites and their related derivatives
on copper ion-induced lipid peroxidation in human low-density
lipoprotein. Arch. Biochem. Biophys. 372, 347–354.
964.
Hopkins, T.L., Ahmad, S.A., 1991. Flavonoid wing pigments in
grasshoppers. Experientia 47, 1089–1091.
Iwashina, T., Matsumoto, S., Nishida, M., Nakaike, T., 1995. New and
rare flavonol glycosides from Asplenium trichomanes ramosum as stable
chemotaxonomic markers. Biochem. Syst. Ecol. 23, 283–290.
Katsube, T., Imawaka, N., Kawano, Y., Yamazaki, Y., Shiwaku, K.,
Yamane, Y., 2006. Antioxidant flavonol glycosides in mulberry
Yoshizumi, M., Tsuchiya, K., Suzaki, Y., Kirima, K., Kyaw, M., Moon,
J.H., Terao, J., Tamaki, T., 2002. Quercetin glucuronide prevents
VSMC hypertrophy by angiotensin II via the inhibition of JNK and
AP-1 signaling pathway. Biochem. Biophys. Res. Commun. 293, 1458–
(
Morus alba L.) leaves isolated based on LDL antioxidant activity.
Food Chem. 97, 25–31.
1
465.
Kramer, C.M., Prata, R.T.N., Willits, M.G., De Luca, V., Steffens, J.C.,
Graser, G., 2003. Cloning and regiospecificity studies of two flavonoid
glucosyltransferases from Allium cepa. Phytochemistry 64, 1069–1076.
Zhang, L., Lin, G., Zuo, Z., 2006. Position preference on glucuronidation
of mono-hydroxylflavones in human intestine. Life Sci. 78, 2772–2780.