Bioactivity of latifolin and its derivatives against termites and fungi

Article abstract of DOI:10.1021/jf900719p

Latifolin (1) and its derivatives were investigated with the aim of confirming the correlation between bioactivity (antitermite and antifungal activity) and chemical structure. Termite mortality in response to the derivatives 2′-O-methyllatifolin (2), lat

Full text of DOI:10.1021/jf900719p

J. Agric. Food Chem. 2009, 57, 5707–5712 5707
DOI:10.1021/jf900719p
Bioactivity of Latifolin and Its Derivatives against Termites
and Fungi
§
NOBUHIRO
S
EKINE,*,† TATSUYA
ATOSHI
A
SHITANI,‡ TETSUYA
‡
M
URAYAMA,‡ SAKAE
SHIBUTANI
,
‡
S
H
ATTORI
,
AND
K
OETSU
T
AKAHASHI
†The United Graduate School Agricultural Science, Iwate University, 3-18-34 Ueda, Morioka 020-8550,
Japan, ‡Faculty of Agriculture, Yamagata University, 1-23 Wakaba, Tsuruoka 997-8555, Japan, and
§Institute of Wood Technology, Akita Prefectural University, 11-1 Kaieizaka, Noshiro 016-0876, Japan
Latifolin (1) and its derivatives were investigated with the aim of confirming the correlation between
bioactivity (antitermite and antifungal activity) and chemical structure. Termite mortality in response
to the derivatives 20-O-methyllatifolin (2), latifolin dimethyl ether (4), and latifolin diacetate (5)
increased 2-fold compared to compound 1. The mortality rate from 5-O-methyllatifolin (3) was not
different from 1. The mass loss (feed consumption by termite) in response to compounds 3-5 was 3
times greater than compound 1, and the mass loss from compound 2 was twice as great as
compound 1. The mortality rate from compounds 4 and 5 increased sharply 7 days after initial
exposure. In assessing the antifungal activity of these compounds, it was found that the inhibition
rates of white- and brown-rot fungi in response to all derivatives were less than that for compound 1.
Our findings indicate that the phenolic hydroxyl group at C-5 of the A ring provides antitermite
activities (mortality and mass loss). In addition, both C-5 and C-20 phenolic hydroxyl groups in the A
and B rings have antifungal activity against white- and brown-rot fungi. In conclusion, the bioactivity
of compound 1 depends upon the position of phenolic hydroxyl groups.
KEYWORDS: Latifolin; neoflavonoid; derivative; bioactivity; antitermite activity; antifungal activity
INTRODUCTION
dermatitis agents. However, the biological function of neoflavo-
noids of the open type has been investigated to a lesser extent than
that of closed types of neoflavonoids, flavonoids, and isoflavo-
noids.
In the genus Dalbergia, approximately 10 species, including D.
melanoxylon and D. nigra, are used as wood products. D. latifolia,
which is found in India and Indonesia, is a highly regarded fancy
wood.
From the wood of D. latifolia, the open type of neoflavonoid
molecules latifolin (1) (6), 4-methoxydalbergione (7-9), and
obtusaquinol(10) have been identified. This wood is quite durable
and is reported to have termite antifeedant and antifungal activity
because of various extractives (10, 11).
Recently, 4-methoxydalbergione, dalbergiphenol, and com-
pound 1, all of which are classified as open-type neoflavonoids,
were isolated from the n-hexane extract of D. latifolia heatwood
and were identified as bioactive molecules (12). Of the three
isolated compounds, the bioactivity level of compound 1 was
found to be the greatest against Reticulitermes speratus Kolbe and
wood-decay fungi (Trametes versicolor and Fomitopusis palustris).
These findings suggested that the bioactivity of neoflavonoids
is affected by the presence of phenolic hydroxyl and methoxy
groups.
Flavonoids are a group of compounds that have a phenyl-
chromone structure (C6-C3-C6), and some of them exist as
glycosides. They are classified as flavonoids, isoflavonoids, neo-
flavonoids, flavanonols, flavanones, isoflavones, or anthocyani-
dins (1).
It is well-known that Fagus and Acacia trees contain flavonoids
and isoflavonoids, while Macherium and Dalbergia trees contain
isoflavonoids and neoflavonoids (1). There are fewer plants that
contain neoflavonoids than those that contain flavonoid and
isoflavonoid, and probably for that reason, neoflavonoids have
been less studied than flavonoids and isoflavonoids.
Most of neoflavonoids possess a 4-phenylchromone skele-
ton (1) and are classified by open or closed C rings. It has been
reported widely that the closed type of neoflavonoids are found in
Leguiminosae, Passifloraceae, Rubiaceae, Guttiferae, and Com-
positae (1). For example, kuhlmannin (2) and mesuol (3) are
closed types of neoflavonoids, and bioactivities of them were
reported, along with antioxidative activities (2), inhibition of
HIV-1 replication,(3) etc.
On the other hand, only a few reports have identified the open
type of neoflavonoids in Dalbergia and Machaerium of Leguimi-
nosae and Calophyllum of Guttiferae (1). Of the open type of
neoflavonoids, some dalbergiones (4,5) have been found to act as
Therefore, in this study, open-type neoflavonoids were isolated
from durable wood, such as Dalbergia and Machaerium, which
compound 1 and its derivatives were investigated, with the aim of
confirming the correlation between bioactivity and chemical
structure.
*To whom correspondence should be addressed. Telephone: +81-
235-2926. Fax: +81-235-2812. E-mail: abe10152@tds1.tr.yamagata-u.
ac.jp.
© 2009 American Chemical Society
Published on Web 06/05/2009
pubs.acs.org/JAFC

5708 J. Agric. Food Chem., Vol. 57, No. 13, 2009
Sekine et al.
MATERIALS AND METHODS
Preparation of Compound 1 and Its Derivatives. Identification
of Compound 1 and Its Derivatives. The identification of compound 1
and its derivatives was conducted by spectral analysis, gas-liquid chro-
matography (GLC)/mass spectroscopy (MS), nuclear magnetic resonance
(NMR), and optical rotation. GLC analysis was conducted with a
SHIMADZU QP-5000 GLC/MS system, using a DB-1 capillary column
(30 m ꢀ 0.32 mm inner diameter, 0.25 μm film thickness; J&W Scientific,
Folsom, CA). The column temperature was from 150 °C (1 min) to 320 °C
(5 min), heated at a rate of 5 °C/min. The injector temperature was 230 °C;
the detector temperature was 250 °C; and the acquisition mass range was
50-450 amu using helium as a carrier gas (1.7 mL/min). One-dimensional
(1D) and two-dimensional (2D) NMR were performed by a JEOL JNM-
EX400 (1H, 400 MHz; 13C, 100 MHz) spectrometer. Optical rotation of
compound 1 was determined by a Horiba SEPA-300 polarimeter. The
chemical structures of compounds 1-5 are shown in Figure 1.
Preparation of Compound 1. Compound 1, isolated from D.
latifolia wood in previous work (12), was used in this research.
R-(-)-latifolin (1). MS, m/z (relative intensity): 286 [M]+ (47),
269 (4), 255 (25), 240 (3), 227 (3), 211 (4), 193 (2), 180 (9), 167 (12), 154
(100), 139 (16), 133 (13), 131 (13), 115 (10), 107 (12), 105 (6), 91 (7), 77 (13),
69 (15), 65 (7), 51 (8). 1H NMR (400 MHz, CDCl3, ppm) δ: 3.84 (3H, s,
-OCH3), 3.86 (3H, s, -OCH3), 5.04 (1H, ddd, dCH2 trans, J = 16.55, 1.51,
Figure 1. Chemical structures of compounds 1-5: latifolin (1), 20-O-
methyllatifolin (2), 5-O-methylatifolin (3), latifolin dimethyl ether (4), and
latifolin diacetate (5).
1.51), 5.18 (1H, ddd, C-HA, J=5.86, 1.74, 1.74), 5.26 (1H, ddd,=CH2 cis
,
J=10.31, 1.40, 1.40), 6.32 (1H, ddd, C-HX, J=17.09, 10.35, 5.95), 6.51
(1H, s, C3-H), 6.74 (1H, s, C6-H), 6.80-7.25 (4H, m, B ring). 13C NMR
(100 MHz, CDCl3, ppm) δ: 40.11 (C-HA), 56.17 (2-OCH3), 57.19 (4-
OCH3), 97.21 (C3), 115.33 (C6), 116.30, (C30), 116.67 (dCH2), 120.61
(C50), 122.74 (C1), 127.71 (C60), 128.53 (C40), 129.42 (C10), 139.09 (C-
HX), 140.15 (C5), 145.60 (C4), 149.55 (C2), 153.79 (C20). [R]21.8D -27.12
(c 0.5, MeOH) (13).
Methylation. A total of 50 mg of compound 1 (12) was added to an
ether solution of diazomethane prepared by reacting KOH in EtOH with
p-tolylsulfonylmethylnitrosoamide in ether. The methylation was stopped
before proceeding completely to give partial methylated products. Methy-
lated compound 1 was isolated using column chromatography. The
absorbent used was silica gel (60 N, spherical 63-210 μm, neutral; Kanto
Chemical Co., Inc., Japan), and the column was eluted with a gradient of
n-hexane-EtOAc to obtain derivatives 4, 2, and 3.
Latifolin was methylated with diazomethane, and the products were
analyzed by GC/MS. Three compounds were detected by the gas chro-
matograph. The molecular weight of these compounds was analyzed, and
the purity was checked. One of the compounds was identified as latifolin
dimethyl ether (4). On the basis of 1D NMR (1H and 13C) and NOESY-
NMR (Figure 2), the other compounds were identified as 20-O-methylla-
tifolin (2) and 5-O-methylatifolin (3), respectively.
Figure 2. NOESY correlations of 2 and 3. Numbers 2 and 3 refer to the
compounds shown in Figure 1.
2-O-Methyllatifolin (2). MS, m/z (relative intensity): 300 [M]+
(100), 285 (8), 271 (16), 253 (8), 237 (7), 225 (8), 209 (8), 192 (19), 177 (16), 167
(25), 163 (24), 161 (22), 137 (38), 121 (42), 115 (41), 107 (17), 105 (17), 91 (64),
77 (37), 69 (51), 58 (42). 1H NMR (400 MHz, CDCl3, ppm) δ: 3.65 (3H, s,
-OCH3), 3.69 (3H, s, -OCH3), 3.80 (3H, s, -OCH3), 4.70 (1H, ddd,
dCH2 trans, J=17.08, 1.71, 1.71), 5.07 (1H, ddd, dCH2 cis, J=10.18, 1.59,
1.59), 5.36 (1H, brd, C-HA, J=5.92), 6.13 (1H, ddd, C-HX, J = 17.20,
10.41, 5.54), 6.44 (1H, s, C3-H), 6.60 (1H, s, C6-H), 6.77-7.13 (4H, m, B
ring). 13C NMR (100 MHz, CDCl3, ppm) δ: 41.56 (C-HA), 57.16 (-OCH3),
57.53 (-OCH3), 58.85 (-OCH3), 99.32 (C3), 111.52 (C1), 112.27 (C30)
116.75 (C6), 116.93 (dCH2), 119.17 (C10), 121.72 (C50), 128.69 (C40), 130.75
(C60), 140.64 (C5), 141.64 (C-HX), 146.36 (C4), 152.27 (C2), 158.50 (C20).
5-O-Methyllatifolin (3). MS, m/z (relative intensity): 300 [M]+
(27), 285 (1), 269 (11), 253 (2), 241 (2), 225 (3), 210 (2), 194 (5), 181 (12), 168
(100), 153 (32), 131 (14), 115 (8), 107 (13), 91 (9), 77 (15), 69 (19), 51 (8).
1H NMR (400 MHz, CDCl3, ppm) δ: 3.74 (3H, s, -OCH3), 3.84 (3H, s,
-OCH3), 3.85 (3H, s, -OCH3), 5.01 (1H, ddd, dCH2 trans, J=17.33, 1.68,
1.68), 5.20 (1H, brd, C-HA, J=5.68), 5.27 (1H, ddd, dCH2 cis, J=10.38,
1.65, 1.65), 6.32 (1H, ddd, C-HX, J=17.21, 10.04, 5.17), 6.53 (1H, s,
C3-H), 6.67 (1H, s, C6-H), 6.80-7.15 (4H, m, B ring). 13C NMR (100
MHz, CDCl3, ppm) δ: 41.68 (C-HA), 57.56 (-OCH3), 58.00 (-OCH3),
58.42 (-OCH3), 99.37 (C3), 114.64 (C6), 117.78 (C1), 118.19 (dCH2),
122.00 (C30), 122.74 (C10) 129.17 (C50) 129.80 (C40), 130.71 (C60), 140.51
(C-HX), 145.14 (C5), 149.92 (C2), 151.61 (C4), 155.27 (C20).
Latifolin Dimethyl Ether (4). MS, m/z (relative intensity): 314
[M]+ (100), 299 (15), 283 (19), 267 (6), 253 (5), 241 (5), 225 (6), 207 (7), 191
(11), 181 (25), 175 (22), 165 (15), 151 (32), 145 (31), 131 (23), 121 (43), 115
(31), 107 (14), 105 (14), 91 (54), 77 (25), 69 (33), 55 (9).
Acetylation. A total of 50 mg of 1, isolated in a previous study (12),
was dissolved in pyridine (2.0 mL) and added to acetic anhydride (2.0 mL).
The solution was stored overnight at room temperature and was then
extracted by partition extraction with EtOAc.
Latifolin Diacetate (5). MS, m/z (relative intensity): 370 [M]+
(15), 328 (100), 311 (1), 297 (5), 285 (47), 270 (11), 255 (81), 237 (4), 227 (6),
211 (6), 191 (5), 181 (9), 167 (21), 154 (34), 139 (10), 133 (12), 131 (22), 115
(11), 107 (11), 91 (9), 77 (11), 69 (21), 55 (7).
Antitermite Test. Termite. A colony of R. speratus Kolbe was
collected from Tsuruoka, Japan, in May of 2008. The colony was
maintained in a dark room at 27(1 °C and 70% relative humidity (RH)
until initiation of the test.
Mortality Rate and Mass Loss. These tests were carried out
according to previously described methods (14, 15). Test samples (com-
pound 1 and its derivatives) dissolved in acetone to make a 3% (w/w)
solution were applied to paper discs (thickness, 1.5 mm; 8 mm diameter;
Advantec, Tokyo, Japan). The paper discs were dried overnight in a
vacuumed desiccator and placed on 2.0 g of sterile sand in a Petri
dish (50 mm diameter). A total of 10 termites were placed into each
Petri dish (six replicates), and the sand was moistened to supply water.

Article
J. Agric. Food Chem., Vol. 57, No. 13, 2009 5709
Table 1. Statistical Analysis of Bioactivity of Compound 1 and Its Derivativesa
compounds
1
2
3
4
5
mortality
mass loss
percent mortality (mean ( SD)
mass loss of paper disk (mg ( SD)
26.7 ( 13.7 A
0.39 ( 0.13 A
15.0 ( 4.1 A
32.5 ( 5.3 A
ns
46.7 ( 13.7 B
0.82 ( 0.14 B
60.0 ( 1.3 BC
42.5 ( 1.7 B
ns
30.0 ( 8.9 A
1.29 ( 0.14 C
62.3 ( 2.5 C
52.3 ( 0.96 C
ns
56.7 ( 18.6 B
1.27 ( 0.24 C
60.8 ( 1.2 C
57.0 ( 5.9 C
ns
53.3 ( 5.2 B
1.13 ( 0.08 C
56.2 ( 5.4 B
51.3 ( 2.5 C
ns
termite
fungus
T. versicolor
F. palustris
R. oryzae
growth in diameter (mm ( SD)
C. cladosporioides
26.2 ( 2.6 A
30.0 ( 0.89 B
29.2 ( 1.2 B
29.3 ( 2.2 B
28.7 ( 0.52 B
a Numbers 1-5 refer to the compounds shown in Figure 1. The same letters are not significantly different. LSD, p < 0.05. ns = not significant.
The Petri dish with the test sample was placed in a dark room for 14 days.
Survival of termites was counted each day to determine mortality rates.
The mass loss (weight loss of the paper disk) was used to evaluate anti-
feedant activity of samples toward termite. The mass loss was determined
by subtracting post-treatment from pretreatment paper disk weights,
which had been vacuum-dried overnight. From the mortality rate and
mass loss of compound 1, the relative mortality and relative mass loss were
expressed, respectively. In control experiments, termites were exposed to
paper discs treated only with acetone.
elicited the least mass loss, followed by compound 2, 20-methy-
lated B ring, and other derivatives. The mass loss because of
dimethylated (4), diacetylated (5) derivatives and compound 3
was approximately 3-fold greater than compound 1.
Antifungal Activity of Compound 1 and Its Derivatives. The
antifungal activity of compound 1 and its derivatives is shown in
Figure 5. The inhibition rates of white-rot fungi (T. versicolor) for
compounds 1, 2, 3, 4, and 5 were 79.1, 16.5, 13.2, 15.3, and 21.8%,
respectively. The inhibition rates of brown-rot fungi (P. palustris)
for compounds 1, 2, 3, 4, and 5 were 37.5, 18.3, -0.5, -9.6, and
1.4%, respectively. Therefore, derivatives appear to have reduced
activity; in particular, the activity level of compound 2 decreased
50% in relation to compound 1, and the others (compounds 3, 4,
and 5) had negligible inhibiting effect. The antifungal activity
against R. oryzae and C. cladosporioides for compound 1 and its
derivatives was relatively low compared to white- and brown-rot
fungi.
Antifungal Test. Fungus. In this study, we used the following fungi,
provided by the Natural Institute of Technology and Evaluation Biolo-
gical Resource Center (NBRC) (Tokyo, Japan): T. versicolor (NBRC:
30340), F. palustris (NBRC:30339), Rhizopus oryzae (NBRC:31005), and
Cladosporium cladosporioides (NBRC:6348). These fungi are routinely
chosen for antifungal test in Japan Industrial Standard (JIS) K1571.
Growth Inhibition Rate of Fungi. The test was carried out
according to the method previously reported (16). Test samples (com-
pound 1 and its derivatives) were dissolved in acetone, and 5.0 μg/cm2 was
applied to the surface of potato glucose agar (PGA) media in Petri dish
(90 mm diameter) and then air-dried. The edge of each precultured fungus
colony was cut with a cork-borer (5 mm diameter), and the trimmed
colony section was placed in the center of the media.
Compound 1 showed the highest level of activity against all
fungi, while the activity levels of all derivatives were significantly
lower than that of compound 1.
The fungi were cultured in the dark in an incubator at 25.5 °C. The
cultivation period for the each species of fungi was 5 days (T. versicolor),
14 days (F. palustris), 36 h (R. oryzae), and 20 days (C. cladosporioides).
As for the control experiment, acetone (300 μL) alone was applied to
PGA media and this treatments (six replicates) was carried out in parallel
with the test sample treatments. The diameter of the mycerial growth was
measured periodically. The growth and inhibition rates were calculated as
the ratio between the relative growth to the control treatment sample.
Statistical Analysis. Test samples are compared using analysis of
variance (ANOVA), and means were separated using a protected Fisher
least-significant difference (LSD) test (p< 0.05; SPSS 10.0, SPSS, Inc.).
The LSD means separations test followed transformation to arcsine
square root percent mortality. The actual percent mortality is reported
in the Table 1.
DISCUSSION
Relationship of Neoflavonoids and the Bioactivity. It has been
reported that the termite mortality rate from isorhapontigenin
(stilbene group) is increased when the compound is methy-
lated (17). By comparing the activity of deacetylgedunin, 17-
hydroxyazadiradione, nimbandiol, and their derivatives, it has
been suggested that the hydroxyl group is responsible for the
termite antifeedant activity (mass loss) (18). Ohmura et al. (19)
observed that the level of termite (Coptotermes formosanus
Shiraki) antifeedant activity of some flavonoids (quercetin, taxi-
folin, and naringetin) was influenced by the position of hydroxyl
groups.
In this study, except for compound 3, the mortality rate of
compounds 2, 4, and 5 in the final days from exposure to the
methylated and acetylated derivatives were twice as high as
compared to that of compound 1. There was no significant
difference in the termite mortality level for derivatives with
methylation at C-5 of the A ring. On the other hand, a high
mortality rate was observed following methylation at C-20 of the
B ring. As previously described, the mortality of compounds 2, 4,
and 5 increased 2-fold compared to compound 1 (Figure 3).
However, a different pattern of mortality was observed here, in
which the mortality rate of compound 2 increased steadily, while
the mortality rate of compounds 4 and 5 increased sharply 7 days
after initiation of the test (particularly for compound 5). In other
words, performance in insecticidal activity was retarded on
dimethylated (4) and diacetylated (5) derivatives.
RESULTS
Antitermite Activity of Compound 1 and Its Derivatives. After 14
days, the mortality rate from compound 1 was 26.7%. Termite
mortality rates of each sample during the test period (14 days) are
expressed relative to the mortality for compound 1 (1.00) in
Figure 3. The relative mortality rates at the final days of this test
from exposure to compounds 2, 3, 4, and 5 were 1.75, 1.13, 2.13,
and 2.00, respectively. Therefore, among the derivatives, the
insecticidal activity of compounds 2, 4, and 5 were 2-fold larger
than compound 1. The activity of compound 3, however, was the
same level as compound 1. The mortality rate from dimethylated
(4) and diacetylated (5) derivatives rapidly increased from the
later stage compared to other samples in the test period.
The mass loss because of exposure to compound 1 and the
control was 0.39 and 7.67 mg, respectively. The mass loss of each
samples is also expressed relative to the mass loss for compound 1
(1.00) in Figure 4. The relative mass loss of compounds 2, 3, 4, and
5 were 1.98, 3.25, 3.42, and 2.89, respectively. Compound 1
The termite mass loss caused by derivatives was 2-fold greater
than compound 1. In comparing the mass loss from compounds 2
and 3, methylation at C-5 of the A ring induced greater mass loss
than methylation at C-20 of the B ring.

5710 J. Agric. Food Chem., Vol. 57, No. 13, 2009
Sekine et al.
Figure 3. Termite mortality rate of compound 1 and its derivatives. Numbers 1-5 refer to the compounds shown in Figure 1. Mortality rate (%) = number
of dead termites after 14 days of the tests/number of initial termites of the tests ꢀ 100, Relative mortality = mortality rate of each sample (%)/mortality rate of
compound 1 (%) ꢀ 100.
Figure 4. Mass loss of compound 1 and its derivatives. Numbers 1-5 refer to the compounds shown in Figure 1. Mass loss (%) = (pretreatment paper disk
weight (mg) - post-treatment paper disk weight (mg))/pretreatment paper disk weight (mg)ꢀ100. Relative mass loss = mass loss of each sample (%)/mass
loss of compound 1 (%) ꢀ 100. The error bar represents (standard deviation (SD), which was calculated from the mass loss (%).
Figure 5. Fungal growth inhibition rate of compound 1 and its derivatives. Numbers 1-5 refer to the compounds shown in Figure 1. Growth rate (%) =
mycelial growth in diameter of each sample (mm)/mycelial growth in diameter of the control (mm) ꢀ 100. Inhibition rate (%) = 100 - growth rate (%). The error
bars represent (SD.
In relating termite mortality and mass loss, it was found that,
as the mortality rate of compound 2 became larger, then the
mass loss was greater compared to compound 1. Thus, for
compound 2, the increased mortality rate corresponded well
with the increased mass loss. The mass loss from compound 3
was 3 times greater than compound 1. However, the mortality
rate from compound 3 did not increase considerably. That is,
compound 3 had lower toxicity and antifeedant activity for
termite than compound 2. This finding might be due to
methylation at C-5 of the A ring changing the antitermite
activity of compound 1. Therefore, it was suggested that the
C-5 hydroxyl group of compound 1 contributed to the anti-
feedant and insecticidal activities more than the C-20 hydroxyl
group.

Article
J. Agric. Food Chem., Vol. 57, No. 13, 2009 5711
Figure 6. Bioactivity of compound 1 and its derivatives. Numbers 1-5 refer to the compounds shown in Figure 1. Mor means the relative mortality of termites
in Figure 3: 0.5 < + < 1.5 and 1.5 < ++ < 2.5. Mas means the relative mass loss of termites in Figure 4: + < 1.5, 1.5 < ++ < 2.5, and 2.5 < +++. W-inh and B-inh
mean the inhibition rates of white-rot and brown-rot fungi in Figure 5: 10.0 < + < 30.0 (%), 30.0 < ++ < 50.0 (%), 50.0 < +++ < 70.0 (%), and 70.0 (%) < ++++.
On the other hand, the mortality rate from compounds 4 and 5
increased sharply on the 7th day of the test and the level of activity
became larger. Furthermore, the mass loss at the 14th day from
compounds 4 and 5 was 3 times greater than for compound 1. It is
known that during termite metabolism, syntrophic acetate-oxi-
dizing microbes in the termite body cause demethylation of
vanillin and syringil acid under anaerobic conditions (20-22).
We hypothesize that compounds 4 and 5 changed into compound
1 through demethylation in the termite body, such as vanillin and
syringil acid, after they were fed. This change eventually led to the
mortality of termites. This hypothesis is well-supported by the
results of compound 5. In addition, compounds 4 and 5 might be
more accumulated in the termite body compared to compound 1,
because compounds 4 and 5 had lower antifeedant activity than
compound 1. Therefore, it was considered that the results in the
case of compounds 4 and 5 were caused by demethylation or
deacetilation of the derivatives accumulated in the termite body.
Studies of antifungal activity of stilbenes and its derivatives
have shown that the activity against white-rot fungi (Coriolus
versicolor) and brown-rot fungi (Gloeophyllum trabeum and Poria
placenta) is related to the hydrophobicity (23,24). In this study, it
appears that the bioactivity is affected by the structure of the
compounds, particularly relative to the presence and position of
hydroxyl and methoxy groups, as well as their hydrophobicity.
In assessing the antifungal activity of these compounds, we
found that the inhibition rates for white-rot fungi from all
derivatives were lower than the rate for compound 1. There was
no difference in the inhibition rate from the monomethylated
derivatives (at C-5 and C-20), indicating that the growth inhibition
is mediated equally by hydroxyl groups at C-5 and C-20. More-
over, the inhibition rate of brown-rot fungi was reduced in
response to the 5-methylated A ring relative to the 20-methylated
B ring, indicating that the activity was affected by the hydroxyl
group at C-5 of the A ring.
The summarized bioactivity of compound 1 and its derivatives
is shown in Figure 6, and the statistical analysis of antitermite and
antifungal activity is shown in Table 1. The difference in bioac-
tivity among compound 1 and its derivatives was significant.
In conclusion, the bioactivity of compound 1, which is a
neoflavonoid, depends upon the position of phenolic hydroxyl
groups. This position is a necessary and effective property in
estimating the performance of compound 1 because the hydroxyl
group at C-5 of the A ring has been shown to have antitermite
activities (mortality and mass loss). In addition, both C-5 and C-
20 hydroxyl groups in the A and B rings have antifungal activity
against white- and brown-rot fungi. Further work is needed for
other functional groups of latifolin to thoroughly determine any
correlation between the bioactivity and chemical structure.
LITERATURE CITED
(1) Baxter, H.; Harborne, J. B. In The Handbook of Natural Flavonoids;
Baxter, H., Harborne, J. B., Eds; John Wiley and Sons, Inc.:
Chichester, U.K., 1999; Vol. 1 and 2, pp 689-707.
(2) Whang, W. K.; Park, H. S.; Ham, I.; Oh, M.; Namkoong, H.; Kim,
H. K.; Hwang, D. W.; Hur, S. Y.; Kim, T. E.; Park, Y. G.; Kim,
J.-R.; Kim, J. W. Natural compounds, fraxin and chemicals struc-
turally related to fraxin protect cells from oxidative stress. Exp. Mol.
Med. 2005, 37, 436–446.
ꢀ
ꢀ ꢀ
(3) Marquez, N.; Sancho, R.; Bedoya, L. M.; Alcamı, J.; Lopez-Perez,
´
~
J. L.; Feliciano, A. S.; Fiebich, B. L.; Munoz, E. Mesuol, a natural
occurring 4-phenylcoumarin, inhibits HIV-1 replication by targeting
the NF-κB pathway. Antiviral Res. 2005, 66, 137–145.
ꢀ
(4) Rojas-Hijazo, B.; Lezaun, A.; Hausen, B. M.; Segura, N.; Garces,
ꢀ
M.; Colas, C. Airborne contact dermatitis in gaitas (flageolets)
constructors after exposure to sawdust of caviuna. Contact Derma-
titis 2007, 56, 274–277.

5712 J. Agric. Food Chem., Vol. 57, No. 13, 2009
Sekine et al.
(5) Schulz, K. H.; Garbe, I.; Hausen, B. M.; Simatupang, M. H. The
sensitizing capacity of naturally occurring quinones experimental
studies in guinea pigs. II. Benzoquinones. Arch. Derm. Res. 1979,
264, 275–286.
(17) Shibutani, S.; Samejima, M.; Doi, S. Effects of stilbenes from bark of
Picea glehnii (Sieb. et Zucc) and their related compounds against
feeding behaviour of Reticulitermes speratus (Kolbe). J. Wood Sci.
2004, 50, 439–444.
(6) Balakrishna, S.; Rao, M. M.; Seshadri, T. R. Constitution of O-
dimethyl latifolin. Tetrahedron 1962, 18, 1503–1507.
(7) Eyton, W. B.; Ollis, W. D.; Sutherland, I. O.; Gottlieb, O. R.; Taveira
(18) Ishida, M.; Serit, M.; Nakata, K.; Juneja, L. R.; Kim, M.; Takahashi, S.
Several antifeedants from Neem oil, Azadirachta indica A. Juss., against
Reticulitermes speratus Kolbe (Isoptera: Rhinotermitidae). Biosci.,
Biotechnol., Biochem. 1992, 56, 1835–1838.
~
Magalhaes, M.; Jackman, L. M. The neoflavonoids group of natural
products-I. Tetrahedron 1965, 21, 2683–2696.
(8) Eyton, W. B.; Ollis, W. D.; Fineberg, M.; Gottlieb, O. R.; Salignac de
(19) Ohmura, W.; Doi, S.; Aoyama, M. Antifeedant activity of fla-
vonoids and related compounds against the subterranean
termite Coptotermes formosanus Shiraki. J. Wood Sci. 2000, 46,
149–153.
(20) Hattori, S. Syntrophic acetate-oxidizing microbes in methanogenic
environments. Microbes Environ. 2008, 23, 118–127.
~
~
Souza Guimaraes, I.; Taveira Magalhaes, M. The neoflavonoids
group of natural products-II. Tetrahedron 1965, 21, 2697–2705.
(9) Barnes, M. F.; Ollis, W. D.; Sutherland, I. O.; Gottlieb, O. R.;
~
Taveira Magalhaes, M. The neoflavonoids group of natural pro-
ducts-III. Tetrahedron 1965, 21, 2707–2715.
(10) Seshadri, T. R. Polyphenols of Pterocarpus and Dalbergia woods.
Phytochemistry 1972, 11, 881–898.
(21) Kane, M. D.; Breznak, J. A. Acetonema longum gen. nov. sp. nov., an
H2/CO2 acetogenic bacterium from the termite, Pterotermes occi-
dentis. Arch. Microbiol. 1991, 156, 91–98.
(11) Syafii, W. Antitermitic compounds from the heartwood of Sonokeling
wood (Dalbergia latifolia Roxb.). Indon. J. Trop. Agric. 2000, 9, 55–58.
(12) Sekine, N.; Ashitani, T.; Murayama, T.; Ogiyama, K.; Takahashi, K.
The bioactive extracts of heartwood of Dalbergia latifolia. Mokuzai
Gakkaishi 2009, 55, 29–36.
(13) Donnelly, D. M. X.; Nangle, B. J.; Prendergast, J. P.; O’Sullivan,
A. M. Dalbergia species;V. Isolation of R-5-O-methyllatifolin from
Dalbergia cochinchinensis, Pierre. Phytochemistry 1968, 7, 647–649.
(14) Tellez, M. R.; Ikhalas, A. K.; Kobaisy, M.; Schrader, K. K.; Dayan,
F. E.; Osbrink, W. L. A. Composition of the essential oil of Lepidium
meyenii (Walp.). Phytochemistry 2002, 61, 149–155.
(15) Seru, G.; Pannakal, S. T.; Serge, F.; Hartmut, L. Antitermitic quini-
nones from Diospyros sylvatica. Phytochemistry 2004, 65, 1265–1271.
(16) Kofujita, H.; Ota, M.; Takahashi, K.; Kawai, Y.; Hayashi, Y. A
diterpene quinone from the bark of Cryptomeria japonica. Phyto-
chemistry 2002, 61, 895–898.
(22) Brune, A.; Miambi, E.; Breznak, J. A. Roles of oxygen and the
intestinal microflora in the metabolism of lignin-derived phenylpro-
panoids and other monoaromatic compounds by termites. Appl.
Environ. Microbiol. 1995, 61, 2688–2695.
(23) Schultz, T. P.; Hubbard, T. F.; Jin, L.Jr.; Fisher, T. H.; Nicholas, D. D.
Role of stilbenes in the natural durability of wood: Fungi-
cidal structure-activity relationships. Phytochemistry 1990, 29, 1501–
1507.
(24) Schultz, T. P.; Cheng, Q.; Boldin, W. D.; Hubbard, T. F.; Jin, L. Jr.;
Fisher, T. H.; Nicholas, D. D. Comparison of the fungicidal
activities of (E)-4-hydroxylated stilbenes and related bibenzyls.
Phytochemistry 1991, 30, 2939–2945.
Received November 20, 2008. Revised manuscript received May 14,
2009. Accepted May 18, 2009.