Species-specific glucosylation of DIMBOA in larvae of the rice armyworm

Article abstract of DOI:10.1271/bbb.80903

DIMBOA [2,4-dihydroxy-7-methoxy-2H-1,4-benzoxazin-3(4H)-one] is a benzoxazinoid (Bx), part of the chemical defense system of graminaceous plants such as maize, wheat, and rye. When Bombyx mori larvae were fed artificial diets containing DIMBOA, they died

Full text of DOI:10.1271/bbb.80903

Biosci. Biotechnol. Biochem., 73 (6), 1333–1338, 2009
Species-Specific Glucosylation of DIMBOA in Larvae of the Rice Armyworm
*
*
*
Hiroaki SASAI, Masahiro ISHIDA, Kenjiro MURAKAMI, Naoko TADOKORO,
y
Atsushi ISHIHARA, Ritsuo NISHIDA, and Naoki MORI
Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University,
Sakyo, Kyoto 606-8502, Japan
Received December 22, 2008; Accepted February 25, 2009; Online Publication, June 7, 2009
[doi:10.1271/bbb.80903]
DIMBOA
[2,4-dihydroxy-7-methoxy-2H-1,4-benz-
the detoxification of various xenobiotics derived from
plants.4) Elucidation of the detoxification mechanisms in
insects is very important to understand the interactions
between plants and herbivores, because tolerance to
potentially poisonous plant secondary metabolites is
crucial to the determination of the host range of a
species. We report in this study the detoxification
mechanism of the rice armyworm, Mythimna separata
(Lepidoptera: Noctuidae), which feeds on graminaceous
plants containing benzoxazinoids (Bxs).
oxazin-3(4H)-one] is a benzoxazinoid (Bx), part of the
chemical defense system of graminaceous plants such as
maize, wheat, and rye. When Bombyx mori larvae were
fed artificial diets containing DIMBOA, they died in
three days. In contrast, Mythimna separata larvae, a
serious pest of rice, maize, sorghum, wheat etc., grew
well on the same diets. Three kinds of glucosides [1-(2-
hydroxy-4-methoxyphenylamino)-1-deoxy-ꢀ-glucopyra-
noside-1,2-carbamate (methoxy glucoside carbamate),
2-O-ꢀ-glucopyranosyl-4-hydroxy-7-methoxy-2H-1,4-benz-
oxazin-3(4H)-one (DIMBOA-2-O-Glc), and 2-O-ꢀ-
glucopyranosyl-7-methoxy-2H-1,4-benzoxazin-3(4H)-one
(HMBOA-2-O-Glc)] were identified by LC-MS and
NMR analyses from the frass of M. separata that had
been fed on a DIMBOA-containing diet. Furthermore,
the incubation of DIMBOA with a midgut tissue suspen-
sion of M. separata in the presence of UDP-D-glucose
generated DIMBOA-2-O-Glc. These findings strongly
suggest that glucosylation by UDP-glucosyltransferase(s)
was important for detoxification to circumvent the
defenses of host plants against M. separata larvae.
The major Bx in maize and wheat is DIMBOA [2,4-
dihydroxy-7-methoxy-2H-1,4-benzoxazin-3(4H)-one] (2).
This compound is stored as a non-toxic glucoside [2-O-
ꢀ-D-glucopyranosyl-4-hydroxy-7-methoxy-2H-1,4-benz-
oxazin-3(4H)-one, DIMBOA-2-O-Glc] (3) in the plants.
When plant tissues are injured, the aglycone is re-
leased through the action of the hydrolytic enzyme
ꢀ-glucosidase.5–9) DIMBOA (2) has exhibited many
different biological activities: it reduced the survival and
reproductive rates of aphids10) and had a feeding-
deterrent effect on the European corn borer.11) It has
also been reported that DIMBOA (2) inhibited the
activities of trypsin and chymotripsin in the larval
midgut of Ostrinia nubilalis (Lepidoptera: Pyralidae),12)
and those of digestive proteases and detoxification
enzymes in the larval midgut of Sesamia nonagrioides
(Lepidoptera: Noctuidae).13) It has been suggested that
the toxicity of DIMBOA (2) resulted from the electro-
philicity of its open-chain tautomer which reacts with
nucleophilic groups of biomolecules.14) Several other
Bxs including their glucosides are known today, their
chemical structures and biological activities having been
reviewed by Niemeyer.11) DIBOA [2,4-dihydroxy-2H-
1,4-benzoxazin-3(4H)-one] is a demethoxylated deriva-
tive of DIMBOA (2). BOA [benzoxazolin-2(3H)-one]
and MBOA [6-methoxy-benzoxazolin-2(3H)-one] (5)
are respectively formed in a ring-contraction reaction
from released DIBOA and DIMBOA.15) MBOA has
been isolated from maize roots,16) and reported to inhibit
the germination of conidia and the growth of germ tubes
of some fungi.17) HMBOA [2-hydroxy-7-methoxy-2H-
1,4-benzoxazin-3(4H)-one] and its glucoside (4) have
been identified from seedlings of maize.18)
Key words: 2,4-dihydroxy-7-methoxy-2H-1,4-benz-
oxazin-3(4H)-one (DIMBOA); defense sub-
stance; Graminaceae; Mythimna separata;
glucosylation
Plants produce various secondary metabolites, some
of which provide defense against herbivorous insects,
like nitrogen compounds (including alkaloids, cyano-
gens glycosides, and glycosinolates), terpenoids, and
phenolics. These chemicals are either toxic to insects or
affect their development and reproduction as reviewed
1)
´
by Despres et al. To overcome these chemical
defenses, insects use multiple mechanisms including
contact and ingestion avoidance, excretion, sequestra-
tion, degradation of toxins, and target-site mutations.1,2)
Among the mechanisms, the biochemical alteration of
defense compounds is a major counter-plan that insects
have evolved against host plants. For example, the
hydroxylation reactions catalyzed by cytochrome P450
monooxygenases have important roles in plant-insect
interactions and have been intensively studied.3) Con-
jugation by glutathione S-transferases is also involved in
We demonstrate here that larvae of Mythimna
separata that fed on graminaceous species in a natural
y
*
To whom correspondence should be addressed. Tel: +81-75-753-6307; Fax: +81-75-753-6312; E-mail: mokurin@kais.kyoto-u.ac.jp
Sasai H, Ishida M, and Murakami K contributed equally to this work.
Abbreviations: BOA, benzoxazolin-2(3H)-one; Bx, benzoxazinoid; DIBOA, 2,4-dihydroxy-2H-1,4-benzoxazin-3(4H)-one; DIMBOA, 2,4-
dihydroxy-7-methoxy-2H-1,4-benzoxazin-3(4H)-one; ESI, electrospray ionization; HMBOA, 2-hydroxy-7-methoxy-2H-1,4-benzoxazin-3(4H)-one;
MBOA, 6-methoxybenzoxazolin-2(3H)-one; PDA, photodiode array

1334
H. SASAI et al.
habitat were resistant to DIMBOA (2), in comparison
with larvae of Bombyx mori that did not feed on Bx-
containing plants. In view of the possibility that
M. separata has evolved a specific detoxification system
for Bxs, we further focused on the identification of
metabolites derived from DIMBOA (2) in frass of the
larvae. The identification of methoxy glucoside carba-
mate (1), DIMBOA-2-O-Glc (3), and HMBOA-2-O-Glc
(4) reveals that the detoxification of DIMBOA (2) in
M. separata larvae occurred through O- and N-gluco-
sylation, most likely involving UDP-D-glucosyltransfer-
ase activity.
A.D. containing DIMBOA
A.D.
600
300
0
M. separata
1
3
5
7
Days
A.D. containing DIMBOA
Unfed
A.D.
Results
200
100
0
B. mori
The growth rates of M. separata and B. mori are
shown in Fig. 1. When B. mori larvae were fed on an
artificial diet containing DIMBOA, they did not grow
well, and the rate of mortality was very high. All of the
larvae died within three days. When B. mori larvae were
left unfed, all of the larvae died within five days. The
rates of mortality for the larvae fed with a DIMBOA-
containing diet and the larvae left unfed made it clear
that DIMBOA was highly toxic to B. mori larvae. On
the other hand, there were no significant differences in
growth rate between larvae of M. separata fed on an
artificial diet and those fed on an artificial diet with
DIMBOA. These results suggest that B. mori larvae
were highly sensitive to DIMBOA, while M. separata
larvae, a notorious pest species to monocotyledonous
plants, some of which produce benzoxazinoids including
DIMBOA as defense substances, have developed a
counteradaptation that enables them to feed on chemi-
cally defended plants.
Typical chromatograms of extracts from the frass of
M. separata larvae that had been fed on artificial diets
both containing and not containing DIMBOA are shown
in Fig. 2. Five distinct peaks (1–5) were observed in the
frass of M. separata larvae fed on the diet containing
DIMBOA, whereas they were absent in the frass of
M. separata larvae that had been fed without DIMBOA.
By comparing the chromatographic behavior in the
HPLC analyses and ESI mass spectra with those of
authentic compounds, compounds 2 (tR 21.0 min) and 5
(tR 31.7 min) were identified as DIMBOA (m=z 164,
½M ꢀ H ꢀ CH2O2ꢁꢀ) and MBOA (m=z 164, ½M ꢀ Hꢁꢀ),
respectively (Fig. 3). The other compounds (1, 3 and 4)
were isolated for structural determination.
1
3
5
Days
Fig. 1. Comparison of the Caterpillar Growth of M. separata and
B. mori Larvae Fed on an Artificial Diet (A.D.) with (triangular
shape) or without (rhomboid shape) DIMBOA.
When B. mori larvae were kept unfed, they died within five days
(square shape). Each point represents the mean ꢂ SEM of 6
replications.
resulted in a coupling constant of about 9 Hz for each,
indicating a trans-configuration. Similarly, 1H-1H COSY
showed correlations from H-30 to H-40 and from H-40 to
H-50 with coupling constants of about 9 Hz, and from
H-50 to H-60 with coupling constants of 12.0, 5.6, and
2.0 Hz. Furthermore, the HMBC spectral data showed
correlations between H-10 (ꢁ 5.24) and NCOO (ꢁ 156.1)
and C-1 (ꢁ 124.1). All spectral data corresponded with
those of 1-(2-hydroxy-4-methoyphenylamino)-1-deoxy-
ꢀ-glucopyranoside-1,2-carbamate.20)
Compound 3 (10 mg, tR 22.7 min in Fig. 2) was
identified as DIMBOA-2-O-Glc (Fig. 3). The molecular
weight of 3 was determined to be 373 on the basis of the
molecular-related ions detected at m=z 372 ½M ꢀ Hꢁꢀ
and m=z 408 ½M þ Clꢁꢀ in the negative mode and at m=z
396 ½M þ Naꢁþ in the positive mode in ESI-MS analy-
ses. In the 1H-NMR spectrum, two signals at 6.78
(2H, d, J ¼ 2:4 Hz, H-6 and 8) and 7.31 (1H, d,
J ¼ 9:2 Hz, H-5) in the aromatic region, and two signals
at 3.78 (3H, s, –OCH3) and 5.98 (1H, s, H-2) were
observed. The anomeric proton of the sugar at 4.83 (1H,
d, J ¼ 8:0 Hz, H-10) gave a doublet with a coupling
constant of 8.0 Hz which is typical of the axial position
(ꢀ-configuration). In the HMBC spectrum, additionally,
a correlation from H-10 (ꢁ 4.83) to C-2 (ꢁ 93.2) was
observed which was caused by three-bond coupling
from the glucose to the C-2 position of DIMBOA. All
spectral data corresponded well with those of authentic
2-O-ꢀ-D-glucopyranosyl-4-hydroxy-7-methoxy-2H-1,4-
benzoxazin-3(4H)-one.21)
Isolated compound 1 (7.5 mg, tR 13.8 min in Fig. 2)
was identified as methoxy glucoside carbamate (Fig. 3).
The molecular weight of 1 was determined by ESI-LC-
MS to be 327 on the basis of molecular-related ions at
m=z 326 ½M ꢀ Hꢁꢀ and m=z 362 ½M þ Clꢁꢀ in the
negative mode and at m=z 350 ½M þ Naꢁþ in the positive
1
mode. The H-NMR spectrum was very similar to that
of a glucoside carbamate derivative [1-(2-hydroxylphen-
ylamino)-1-deoxy-ꢀ-glucopyranoside-1,2-carbamate],
which was obtained from 3-ꢀ-glucopyranosyl-benz-
oxazolin-2(3H)-one (BOA-N-ꢀ-D-glucoside) through
ring transformation.19) In the glycoside moiety, a ꢀ-
configuration at the anomeric carbon (C-10) was identi-
fied from the coupling constant of 9.2 Hz of the anomeric
proton (ꢁ 5.24) with its vicinal proton H-20. Proton H-20
was coupled with vicinal protons H-10 and H-30, which
Compound 4 (10 mg, tR 23.4 min in Fig. 2) was
identified as HMBOA-2-O-Glc (Fig. 3). The molecular
weight of 4 was determined to be 357 on the basis of the
molecular-related ions observed at m=z 356 ½M ꢀ Hꢁꢀ
and m=z 392 ½M þ Clꢁꢀ in the negative mode and m=z

Glucosylation of DIMBOA in Mythimna separata
1335
60
A
B
3
1
4
2
5
0
100
0
0
10
20
30
40
50
Retention time (min)
Fig. 2. HPLC Chromatograms (with UV detection at 264 nm) of the Frass Extracts from M. separata Larvae Fed on an Artificial Diet Containing
DIMBOA (A) and on a DIMBOA-Free Diet (B).
Five peaks (1–5) were observed at tR 13.8 (compound 1), 21.5 (compound 2), 22.7 (compound 3), 23.4 (compound 4), and 31.3 min
(compound 5).
3’ OH 4’
2’
O
O
O
OH
O
HO
2
3
O
OH
6’
O
N
1
OH
1’
5’
N
6
OH
4
O
O
5
2
1
OH
OH
HO
O
HO
2
O
OH
OH
OH
OH
O
8
5
O
O
8a
4a
2
O
O
9
7
O
8a
4a
O
7
O
3
3
N
H
6
6
N
O
N
H
O
5
OH
4
5
3
Fig. 3. Structures of Compounds 1–5 Identified in M. separata Larvae.
1-(2-hydroxy-4-methoxyphenylamino)-1-deoxy-ꢀ-glucopyranoside-1,2-carbamate (methoxy glucoside carbamate) (1); 2,4-dihydroxy-7-
methoxy-2H-1,4-benzoxazin-3(4H)-one (DIMBOA) (2); 2-O-ꢀ-glucopyranosyl-4-hydroxy-7-methoxy-2H-1,4-benzoxazin-3(4H)-one (DIMBOA-
2-O-Glc) (3); 2-O-ꢀ-glucopyranosyl-7-methoxy-2H-1,4-benzoxazin-3(4H)-one (HMBOA-2-O-Glc) (4); 6-methoxybenzoxazolin-2(3H)-one
(MBOA) (5).
400
380 ½M þ Naꢁþ in the positive mode in ESI-MS analy-
ses. This molecular weight was 16 mass units smaller
than that of DIMBOA-2-O-Glc, suggesting that one
oxygen atom in DIMBOA-2-O-Glc was absent in 4. The
300
1H-NMR and 13C-NMR spectra of compound 4 were
similar to those of DIMBOA-2-O-Glc, except for the
200
signals of a proton at 6.95 (1H, d, J ¼ 8:8 Hz, H-5) and a
100
carbonyl carbon at 160.5 (C-3), indicating the absence
of the oxygen atom at the 4-position in DIMBOA. A
0
correlation from H-10 (ꢁ 4.84) to C-2 (ꢁ 91.4) was
observed in the HMBC spectrum. All these results
1
2
3
4
5
enabled 4 to be concluded as HMBOA-2-O-Glc. All
spectral data corresponded well with those of authentic
2-O-ꢀ-D-glucopyranosyl-7-methoxy-2H-1,4-benzoxazin-
3(4H)-one.21)
Fig. 4. Quantitative Analyses of DIMBOA Analogues Identified
from Frass of M. separata Larvae Fed on DIMBOA.
Each point represents the mean ꢂ SEM of
6 replications.
Numbers represent the same compounds as those in Fig. 3.
The amounts of methoxy glucoside carbamate,
DIMBOA, DIMBOA-2-O-Glc, HMBOA-2-O-Glc and
MBOA in the frass of M. separata larvae that had been
fed on DIMBOA were determined to be 125:1 ꢂ 6:0,
5:8 ꢂ 0:5, 321:9 ꢂ 33:3, 111:4 ꢂ 12:5, and 4:2 ꢂ 0:5
mg/larva, respectively (mean ꢂ SEM, n ¼ 6, Fig. 4).
Based on the administered amount of DIMBOA (3 mg,
14.2 mmol), the molar percentages of the compounds
were 2:7 ꢂ 0:5, 0:2 ꢂ 0:1, 6:1 ꢂ 0:6, 2:2 ꢂ 0:2, and
0:2 ꢂ 0:1%, respectively (mean ꢂ SEM, n ¼ 6). The

1336
H. SASAI et al.
Discussion
I. S.
A
Bxs play a defensive role in wheat and maize plants
against insects.10–13) In this study, it was clearly
demonstrated that the major Bx, DIMBOA, was highly
toxic to B. mori larvae but not to M. separata larvae, as
shown in Fig. 1. This finding suggests that M. separata
larvae had developed counteradaptations that enabled
them to feed on plants defended with DIMBOA. One
plausible target site of DIMBOA is digestive enzymes
such as chymotrypsin and trypsin. However, the chy-
motrypsin-like activity from M. separata larvae was
susceptible to DIMBOA in vitro (data not shown). This
result suggests that M. separata had not evolved
digestive enzymes resistant to DIMBOA, but instead
biochemical systems to alleviate the toxicity of the
compound. We thus analyzed the metabolism of
DIMBOA in M. separata larvae.
DIMBOA-2-O-Glc
m/z 408
m/z 312
0
5.0
10.0
Retention time (min)
100
80
60
40
20
0
B
We identified three kinds of glucosides, methoxy
glucoside carbamate (1), DIMBOA-2-O-Glc (3) and
HMBOA-2-O-Glc (4), from the frass of M. separata
larvae that had been fed on an artificial diet containing
DIMBOA. As far as we know, this is the first report
on the identification of Bxs-glucosides from insects.
Furthermore, a midgut suspension of M. separata
showed stronger activity to convert DIMBOA into the
corresponding glucoside in the presence of UDP-D-
glucose than that of B. mori. This indicates the presence
of UDP-D-glucosyltransferase(s) that could metabolize
Bxs in M. separata larvae. The metabolism of Bxs to the
glucosides is potentially beneficial for the survival of
M. separata that feeds on Bx-accumulating plants. The
toxicity of DIMBOA has been proposed to be attribut-
able to the electrophilicity of its open-chain tautomer.14)
If this is the case, the glucosylation of DIMBOA at the
2-position would prevent the formation of the open-
chain tautomer, and thus would reduce its toxicity. In
addition, glucosylation modified various physicochem-
ical and biochemical properties of Bxs. It is plausible
that the increased hydrophilicity and preferred recog-
nition by transporters for glucosides facilitated the
elimination and isolation of Bxs from metabolically
active sites in insect tissues, analogous to the function of
glucuronidation in phase II xenobiotic metabolism in
mammals.
Glucose conjugates of Bxs have also been reported in
plants. In Bx-accumulating species, Bxs are sequestered
in the vacuole as 2-O-glucosides, and when the tissue
is damaged by pathogen infection or by feeding by
herbivores, aglycones (DIMBOA, DIBOA and HMBOA
etc.) that have some biological activities including
strong antifungal and antifeeding effects10–13,17) are
released.5–9) Bxs are stored as glucosides, probably
because the aglycones are also toxic to the plants them-
selves. Furthermore, it has been reported that BOA and
MBOA were metabolized by glucose conjugation. When
Avena sativa, Hordeum vulgare, and Zea mays were
incubated with BOA or MBOA, glucoside carbamate or
methoxy glucoside carbamate were identified in the
roots.19,20) These respective glucoside carbamate deriv-
atives are thought to originate from BOA-N-Glc and
MBOA-N-Glc through spontaneous isomerization. It
has been reported that the glucoside carbamte compound
exhibited dramatically reduced phytotoxicity relative
M. separata
B. mori
Fig. 5. Typical Chromatogram of DIMBOA-2-O-Glc and Compar-
ison of the Glucosylation Activity between M. separata and B. mori
Midgut Tissues.
(A) LC-MS-selected ion chromatograms of produced DIMBOA-
2-O-Glc (m=z 408) as ½M þ Clꢁꢀ when suspended midgut tissue
preparations were incubated with DIMBOA and UDP-D-glucose.
BTEE (m=z 312, ½M ꢀ Hꢁꢀ) was used as an internal standard.
(B) The in vitro glucosylation activity was evaluated by the amount
of produced DIMBOA-2-O-Glc (mg)/wet weight of used midgut
tissue (g).
total of analyzed compounds in the frass represented
approximately 10% of DIMBOA that had been admin-
istered to the larvae. Although the fate of the rest (90%)
of DIMBOA remains unknown, further analysis should
be done to investigate the presence of DIMBOA-related
compounds in the hymolymph and fat body of the larvae.
We incubated the suspended midgut preparation from
M. separata larvae with synthesized DIMBOA in the
presence of UDP-D-glucose to detect the UDP-D-
glucosyltransferase activity. The incubated mixture
was analyzed by LC-MS with selected ion monitoring
of the ½M þ Clꢁꢀ ions at m=z 408 for DIMBOA-2-O-
Glc. As shown in Fig. 5A, a peak corresponding to the
glucoside was clearly detected. The midgut preparation
did not contain a detectable amount of the glucoside,
and no glucoside had been formed without the midgut
preparation or UDP-D-glucose in the incubated mixture.
These results demonstrate that DIMBOA-2-O-Glc had
been produced by ꢀ-glucosyltransferase(s), and that
UDP-D-glucose had been used as a glucose donor in this
reaction. We determined the in vitro glucosylation
activity for DIMBOA in midgut preparations from
M. separata and B. mori. The activity of M. separata
and B. mori was 91:1 ꢂ 6:1 and 12:7 ꢂ 5:6 mg/g, re-
spectively (Fig. 5B). These results suggest that the
metabolism of Bxs by glucosylation could be reasonably
deduced to be one of the capabilities that M. separata
had acquired during co-evolution to adapt to feeding on
graminaceous species that accumulate Bxs at a high
concentration.

Glucosylation of DIMBOA in Mythimna separata
1337
to that of BOA-6-O-Glc.22) The glucose conjugation
of BOA has also been reported in dicotyledonous
species. In Portulaca oleracea (Portulacaceae), treat-
ment with BOA resulted in the formation of unknown
detoxification products resembling the known BOA-6-
OH compounds.20) Arabidopsis seedlings are able to
detoxify exogenously supplied BOA and to produce
BOA-6-OH, BOA-6-O-Glc, and glucoside carbamate.22)
Although it is unknown to what extent the toxicity of
these Bxs and their metabolites can be seen in insects, it
is of interest that plants and insects have found the same
way to deal with the same compounds through the
convergent evolution of a biochemical system.
OPN, Nacalai Tesque, Kyoto, Japan) was added to the solution and
then filtered off to remove any hydrophobic impurities. Finally, the
solution was analyzed by HPLC [a Shimadzu chromatograph equipped
with a PDA detector and a reversed-phase column (Cosmosil 5C18-AR-
II, 150 ꢄ 4:6 mm ID, Nacalai Tesque, Kyoto, Japan)], eluting with
H2O containing 0.1% acetic acid (eluent A, EA) and MeOH containing
0.088% acetic acid (eluent B, EB). The following gradient was used:
0–30 min, 10–20% EB and 30–50 min, 20–100% EB in EA, at a flow
rate of 1 ml/min.
Extraction and isolation of DIMBOA metabolites. The frass (8.2 g)
of M. separata, which had been fed on DIMBOA as already described,
was homogenized in MeOH (82 ml), and extracted for 10 min two
times. The combined extract was filtered and evaporated to dryness.
The residue was dissolved in 5% MeOH/H2O (7.5 ml). The solution
Glucosylation in insects is known to be involved in
cuticle formation, pigmentation, and olfaction.23,24) The
ecdysone-glucoside has also been identified in insects
was applied to a reversed phase column (17.5 g of Cosmosil 140C18
-
OPN, Nacalai Tesque, Kyoto, Japan), eluting (80 ml each) in sequence
with MeOH/H2O (v/v, 5/95; 1/9; 1/4; 2/3; 3/2). The 20% fraction
was then subjected to preparative HPLC, using a reversed-phase
column (YMC-Pack Pro C18, 250 ꢄ 20 mm ID, YMC, Kyoto, Japan)
and eluting with 40% EB in EA at a flow rate of 5 ml/min, with UV
detection at 264 nm. Compound 1 was isolated at tR 16.5 min from the
20% fraction. The 40% fraction was also injected into the preparative
HPLC system just described, eluting with a solvent gradient of
30–100% EB in EA over 50 min at a flow rate of 5 ml/min, with UV
detection at 264 nm. Compounds 3 and 4 were isolated at tR 21.7 and
23.6 min, respectively.
infected with a baculovirus.25) As mentioned by Despres
´
et al., only limited information is currently available
about the roles of UDP-glycosyltransferases in the
detoxification of plant allelochemicals.1) In fact, most
of the compounds studied as substrates for insect
glucosyltransferases have been plant and synthetic
phenolics.26–28) In Drosophila melanogaster, variations
in the UDP-glucosyltransferase activity have been
analyzed by using the xenobiotic compounds, 1-naphtol
and 2-naphtol.29) In this study, however, we showed that
M. separata larvae were able to transform such Bxs
such as DIMBOA, HMBOA and MBOA to O-gluco-
sides and methoxy glucoside carbamate as detoxification
products. Although the total recovery ratio was approx-
imately 10% based on the administered amount of
DIMBOA in the feeding experiment, these findings
might be a good example showing that UDP-glucosyl-
transferases play an important role in the detoxification
of plant allelochemicals by insects, and provide a wider
range of detoxification facilities to circumvent the
chemical defenses of host plants than we expected.
Analytical instruments. 1H-NMR (400 MHz), 13C-NMR (100 MHz),
and 2D-NMR spectra were recorded with TMS as an internal standard
in CDCl3 and methanol-d4, or with acetone in D2O by using a Bruker
AC400 spectrometer. Chemical shifts are given in ꢁ values.
ESI mass spectra (positive and negative mode) were recorded with
an LC-MS-2010A instrument (Shimadzu, Kyoto, Japan) combined
with an HPLC system (LC-10AD VP pump, CTO-10A VP column
oven and SIL-H8C system controller; Shimadzu, Kyoto, Japan). The
CDL temperature was 250 ꢃC, the voltage was 1.5 kV, and the
nebulizer gas flow was 1.5 l/min. A reversed-phase column (Cosmosil
5C18-MS-II, 50 ꢄ 2:5 mm ID, Nacalai Tesque, Kyoto, Japan) was
eluted with EA and CH3CN containing 0.088% acetic acid (eluent C,
EC). The following gradient was used: 0–5 min, 10–20% EC; 5–
10 min, 20–100% EC in EA; and 10–15 min, 100% EC; at a flow rate
of 0.2 ml/min. The column temperature was maintained at 35 ꢃC in the
oven (CTO-10Avp column oven, Shimadzu, Kyoto, Japan).
Experimental
Quantification of DIMBOA and its metabolites. The frass
(<375 mg) obtained from each last instar larva fed on an artificial
diet containing DIMBOA (3 mg/g) for 3 d was collected and extracted
in 3.75 ml of MeOH, with 12.5 mg of BTEE (N-benzoyl-L-tyrosine
ethyl ester) as an internal standard. The extract was centrifuged at
12,000 g for 5 min, and then the supernatant (5 ml) was analyzed by
LC-MS in the selected ion mode, as already shown. All compounds
except DIMBOA were identified by comparing the retention time and
characteristic ½M ꢀ Hꢁꢀ or ½M þ Clꢁꢀ species. The ½M ꢀ Hꢁꢀ species
of MBOA, HMBOA-2-O-Glc and BTEE, and ½M þ Clꢁꢀ of methoxy
glucoside carbamate and DIMBOA-2-O-Glc were m=z 164, 356, 312,
362, and 408, respectively. DIMBOA was identified by its retention
time and characteristic m=z 164 ½M ꢀ H ꢀ CH2O2ꢁꢀ in comparison
with a synthetic sample. The amounts of DIMBOA and its metabolites
were calculated by the area ratio of their characteristic ions and are
also shown as metabolic rates in molar percentages to the given
amount of DIMBOA. Authentic MBOA was purchased from Wako
Pure Chemical Industries Co. (Osaka, Japan).
Insect rearing and feeding experiments. Larvae of the rice army-
worm, Mythimna separata (Lepidoptera: Noctuidae), were reared on
an artificial diet (Insecta-LFS; Nihon Nosan Kogyo, Yokohama, Japan)
at 24 ꢃC under a 16 h/8 h (light/dark) cycle. Commercially available
Bombyx mori (Lepidoptera: Bombycidae; Kinsyu ꢄ Shouwa) was
purchased from Mukin Yosan System Institute (Kyoto, Japan).
In the assay, the fifth and third instar larvae of M. separata and
B. mori, respectively, were fed on an artificial diet containing
DIMBOA (3 mg/g fresh weight), about two times more than the
maximum concentration of Bxs in graminaceous plants,30) or on a
DIMBOA-free artificial diet for 7 d, and weighed every day.
Synthesized DIMBOA was prepared as a solution in EtOH (1.5%,
w/v) which was added to the artificial diet. The diets were dried to
evaporate the EtOH, and an adequate amount of water was added. A
control diet was prepared in the same way, but without DIMBOA. The
larvae of M. separata fed on both diets and B. mori fed on the
DIMBOA-free artificial diet developed normally to the pupal stage.
Synthesis of DIMBOA. DIMBOA was synthesized as reported.31,32)
In vitro glucosylation. Last instar larvae of M. separata were
anesthetized by immersion in tepid water for 10 min and then dissected
in saline. Midgut tissues were removed from the larvae. Four midgut
tissues were homogenized with a 25 mM CHES buffer (4 ml, pH 9.0,
0 ꢃC). After vigorously vortexing, the suspended tissue samples
(440 ml) were incubated with 45 ml of UDP-D-glucose (10 mM in
H2O, Wako Pure Chemical Industries, Osaka, Japan) and 15 ml of
synthesized DIMBOA (10 mM in DMSO). After incubating for 90 min
at 30 ꢃC, all samples were boiled for 20 min to stop the enzymatic
The identity of the compound was confirmed by 1H-NMR spectroscopy.
Analysis of DIMBOA metabolites in frass. The frass (242–317 mg,
n ¼ 4) of an M. separata larva fed on an artificial diet containing
DIMBOA or the DIMBOA-free artificial diet for 7 d was homogenized
in MeOH (2.4–3.2 ml) and extracted for 10 min. The extract was
filtered, and the filtrate was diluted by adding H2O to make 50%
MeOH/H2O (v/v) solutions. ODS gel (240–317 mg, Cosmosil 140C18
-

1338
H. SASAI et al.
reactions and then added to 500 ml of acetonitrile. Each homogenized
sample was centrifuged for 10 min at 14,000 g, and the supernatant was
analyzed by LC-MS as already described. To compare the in vitro
glucosylation activity, 5th instar larvae of B. mori were also treated in
the same way. The activity was evaluated by the amount of produced
DIMBOA-2-O-Glc (mg)/wet weight of used midgut tissue (g).
3) Feyereisen R, ‘‘Comprehensive Molecular Insect Science,’’ eds.
Gilbert LI, Iatrou K, and Gill SS, Elsevier, Oxford, pp. 1–77
(2005).
4) Francis F, Vanhaelen N, and Haubruge E, Arch. Insect Biochem.
Physiol., 58, 166–174 (2005).
5) Esen A, Plant Physiol., 98, 174–182 (1992).
6) Oikawa A, Ishihara A, Hasegawa M, Kodama O, and Iwamura
H, Phytochemistry, 56, 669–675 (2001).
1-(2-Hydroxy-4-methoxyphenylamino)-1-deoxy-ꢀ-glucopyranoside-
1,2-carbamate (methoxy glucoside carbamate) (1). 1H-NMR (CD3OD)
ꢁ (ppm): 3.44–3.55 (3H, m, H-30, 40, 50), 3.73 (1H, dd, J ¼ 12:0,
5.6 Hz, Ha-60), 3.79 (3H, s, –OCH3), 3.91 (1H, dd, J ¼ 12:0, 2.0 Hz,
Hb-60), 3.99 (1H, dd, J ¼ 9:2, 8.8 Hz, H-20), 5.24 (1H, d, J ¼ 9:2 Hz,
H-10), 6.78 (1H, dd, J ¼ 8:8, 2.4 Hz, H-5), 6.94 (1H, d, J ¼ 2:4 Hz,
H-3), 7.23 (1H, d, J ¼ 8:8 Hz, H-6); 13C-NMR (CD3OD) ꢁ (ppm): 56.7
(–OCH3), 63.0 (C-60), 71.4 (C-20), 71.6 (C-40), 79.0 (C-30), 81.4 (C-50),
85.9 (C-10), 98.4 (C-3), 110.9 (C-5), 113.1 (C-6), 124.1 (C-1), 145.1
(C-2), 156.1 (NCOO), 158.3 (C-4).
7) Oikawa A, Ishihara A, and Iwamura H, Phytochemistry, 61,
331–337 (2002).
8) Sue M, Ishihara A, and Iwamura H, Planta, 210, 432–438
(2000a).
9) Sue M, Ishihara A, and Iwamura H, Plant Sci., 155, 67–74
(2000b).
10) Givovich A, Sandstrom J, Niemeyer HM, and Pettersson J,
¨
J. Chem. Ecol., 20, 1923–1930 (1994).
11) Niemeyer HM, Phytochemistry, 27, 3349–3358 (1988).
12) Houseman JG, Campos F, Thie NMR, Philogene BJR, Atkinson
J, Morand P, and Arnason JT, J. Econ. Entomol., 85, 669–674
(1992).
2-O-ꢀ-Glucopyranosyl-4-hydroxy-7-methoxy-2H-1,4-benzoxazin-
3(4H)-one (DIMBOA-2-O-Glc) (3). 1H-NMR (D2O) ꢁ (ppm): 3.14 (1H,
dd, J ¼ 9:2, 8.0 Hz, H-20), 3.34 (1H, t, J ¼ 9:2 Hz, H-40), 3.45 (1H, t,
J ¼ 9:2 Hz, H-30), 3.48 (1H, ddd, J ¼ 9:2, 6.0, 2.0 Hz, H-50), 3.69 (1H,
dd, J ¼ 12:4, 6.0 Hz, Ha-60), 3.78 (3H, s, –OCH3), 3.87 (1H, dd,
J ¼ 12:4, 2.0 Hz, Hb-60), 4.83 (1H, d, J ¼ 8:0 Hz, H-10), 5.98 (1H, s,
H-2), 6.78 (2H, d, J ¼ 2:4 Hz, H-6, 8), 7.31 (1H, d, J ¼ 9:2 Hz, H-5);
13C-NMR (D2O) ꢁ (ppm): 54.8 (–OCH3), 59.6 (C-60), 68.4 (C-40), 71.4
(C-20), 74.3 (C-30), 75.3 (C-50), 93.2 (C-2), 97.5 (C-10), 102.8 (C-8),
108.1 (C-6), 113.7 (C-5), 120.1 (C-4a), 140.2 (C-8a), 155.3 (C-3),
155.9 (C-7).
13) Ortego F, Ruiz M, and Castan
˜era P, J. Insect Physiol., 44, 95–
101 (1998).
´
14) Perez FJ and Niemeyer HM, Phytochemistry, 28, 1831–1834
(1989).
15) Sicker D, Frey M, Schulz M, and Gierl A, ‘‘International
Review of Cytology-A Survey of Cell Biology’’ Vol. 198,
ed. Jeong KW, Academic Press, New York, pp. 319–346
(1999).
16) Chen CM and Chen MT, Phytochemistry, 15, 1997–1999
(1976).
17) Oikawa A, Ishihara A, Tanaka C, Mori N, Tsuda M, and
Iwamura H, Phytochemistry, 65, 2995–3001 (2004).
18) Tipton CL, Klun JA, Husted RR, and Pierson MD, Biochem-
istry, 6, 2866–2870 (1967).
2-O-ꢀ-Glucopyranosyl-7-methoxy-2H-1,4-benzoxazin-3(4H)-one
(HMBOA-2-O-Glc) (4). 1H-NMR (D2O) ꢁ (ppm): 3.14 (1H, dd,
J ¼ 9:2, 8.0 Hz, H-20), 3.33 (1H, t, J ¼ 9:2 Hz, H-40), 3.45 (1H, t,
J ¼ 9:2 Hz, H-30), 3.48 (1H, ddd, J ¼ 9:2, 6.0, 2.4 Hz, H-50), 3.69 (1H,
dd, J ¼ 12:4, 6.0 Hz, Ha-60), 3.76 (3H, s, –OCH3), 3.88 (1H, dd,
J ¼ 12:4, 2.4 Hz, Hb-60), 4.84 (1H, d, J ¼ 8:0 Hz, H-10), 5.82 (1H, s,
H-2), 6.70 (1H, dd, J ¼ 8:8, 2.8 Hz, H-6), 6.77 (1H, d, J ¼ 2:8 Hz,
H-8), 6.95 (1H, d, J ¼ 8:8 Hz, H-5); 13C-NMR (D2O) ꢁ (ppm): 54.8
(–OCH3), 59.6 (C-60), 68.4 (C-40), 71.5 (C-20), 74.4 (C-30), 75.2 (C-50),
91.4 (C-2), 97.6 (C-10), 103.0 (C-8), 108.4 (C-6), 116.0 (C-5), 117.5
(C-4a), 139.6 (C-8a), 155.3 (C-7), 160.6 (C-3).
19) Sicker D, Schneider B, Henning L, Knop M, and Schulz M,
Phytochemstry, 58, 819–825 (2001).
20) Hofmann D, Knop M, Hao H, Henning L, Sicker D, and Schulz
M, J. Nat. Prod., 69, 34–37 (2006).
21) Baumeler A, Hesse M, and Werner C, Phytochemistry, 53, 213–
222 (2000).
´
22) Baerson SR, Sanchez-Moreiras A, Pedrol-Bonjoch N, Schulz
M, Kagan IA, Agrawal AK, Reigosa MJ, and Duke SO, J. Biol.
Chem., 280, 21867–21881 (2005).
Acknowledgments
23) Hopkins TL and Kramer KJ, Annu. Rev. Entomol., 37, 273–302
(1992).
24) Wang Q, Hasan G, and Pikielny CW, J. Biol. Chem., 274,
10309–10315 (1999).
We thank Drs. Yasuhisa Kunimi and Kazuko
Nakanishi (Tokyo University of Agriculture and Tech-
nology) for supplying Mythimna separata. This study
was partly supported by grants-aid for scientific research
(nos. 15580090, 18580053 and 19580122) and by the
21st century COE program for Innovative Food and
Environmental Studies Pioneered by Entomomimetic
Sciences from the Ministry of Education, Culture,
Sports, Science and Technology of Japan.
25) O’reilly DR, Insect Biochem. Mol. Biol., 25, 541–550 (1995).
26) Ahmad SA and Hopkins TL, Arch. Insect Biochem. Physiol., 21,
207–224 (1992).
27) Ahmad SA and Hopkins TL, Insect Biochem. Mol. Biol., 23,
581–589 (1993).
28) Luque T, Okano K, and O’reilly DR, Eur. J. Biochem., 269,
819–825 (2002).
29) Rausell C, Llorca J, and Real MD, Arch. Insect Biochem.
Physiol., 34, 347–358 (1997).
30) Argandon˜a VH, Niemeyer HM, and Corcuera LJ, Phytochem-
istry, 20, 673–676 (1981).
References
´
1) Despres L, David J-P, and Gallet C, Trends Ecol. Evol., 22,
31) Hao H, Sieler J, and Sicker D, J. Nat. Prod., 65, 466–469
(2002).
298–307 (2007).
2) Dowd PF, Smith CM, and Sparks TC, Insect Biochem., 13, 453–
468 (1983).
32) Tays
K and Atkinson J, Synth. Commun., 28, 903–912
(1998).