Radical and superoxide scavenging activities of matairesinol and oxidized matairesinol

Article abstract of DOI:10.1271/bbb.60096

The radical and superoxide scavenging activities of oxidized matairesinols were examined. It could be assumed that the free benzylic position was important for higher radical scavenging activity. The different level of activity was observed between 7′-oxomatairesinol (Mat 2) and 7-oxomatairesinol (Mat 3). The activity of 8-hydroxymatairesinol was lower than that of matairesinol (Mat 1). The superoxide scavenging activity of the oxidized matairesinols was also demonstrated for the first time. It is assumed that the pKa value of phenol in the oxidized matairesinols affected this activity.

Full text of DOI:10.1271/bbb.60096

Biosci. Biotechnol. Biochem., 70 (8), 1934–1940, 2006
Radical and Superoxide Scavenging Activities
of Matairesinol and Oxidized Matairesinol
y
Satoshi YAMAUCHI,1; Takuya SUGAHARA,1 Yuki NAKASHIMA,1 Akihiro OKADA,1
3
Koichi AKIYAMA,2 Taro KISHIDA,1 Masashi MARUYAMA,1 and Toshiya MASUDA
1Faculty of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama, Ehime 790-8566, Japan
2Integrated Center for Sciences, Ehime University, 3-5-7 Tarumi, Matsuyama, Ehime 790-8566, Japan
3Faculty of Integrated Arts and Sciences, University of Tokushima, Tokushima 770-8502, Japan
Received February 20, 2006; Accepted April 6, 2006; Online Publication, August 23, 2006
[doi:10.1271/bbb.60096]
The radical and superoxide scavenging activities of
oxidized matairesinols were examined. It could be
assumed that the free benzylic position was important
for higher radical scavenging activity. The different
level of activity was observed between 70-oxomatairesi-
nol (Mat 2) and 7-oxomatairesinol (Mat 3). The activity
of 8-hydroxymatairesinol was lower than that of mat-
airesinol (Mat 1). The superoxide scavenging activity of
the oxidized matairesinols was also demonstrated for
the first time. It is assumed that the pKa value of phenol
in the oxidized matairesinols affected this activity.
Mat 3 from the reaction of matairesinol (Mat 1) with
2,2-diphenyl-1-picrylhydrazyl (DPPH).5) It could be
assumed from this result that the 7 position of
matairesinol played a more important role in the
antioxidant activity than the 70 position. Many oxidized
lignans are biosynthesized by plants and the preparation
of oxidized lignan has been reported.6) Evaluating
the antioxidant activity of oxidized lignan is important
for the utilization of lignans. Our experiment compared
the radical scavenging activity of oxidized matairesinol.
This is the first report about the relationship between
the lignan structure and superoxide scavenging activity
by employing Mat 1–Mat 6 and Lio 74) (Fig. 1).
Key words: matairesinol; radical scavenging activity;
superoxide scavenging activity
Results and Discussion
Lignans are widely distributed among plant biore-
sources containing dietary plants and in the waste of the
forestry and pulp industry. Our efforts are continuing to
clarify the effect of dietary plants containing many kinds
of lignan and to discover new worth in waste from plant
bioresources. Research into antioxidant activity is
important to evaluate the effect of plant materials
containing lignans on health.1) Except for the aromatic
portion, the structure of lignans had not been paid
attention to in respect of antioxidant activity. The
relationship between the structure of lignans and their
antioxidant activities has recently been reported.2–4) The
fact that the structure, apart from the aromatic portion,
of lignans affected the degree of antioxidant activity was
clarified by these reports. In particular, in the case of the
3,4-dibenzyltetrahydrofuran type of lignan, a higher
oxidative degree of the benzylic position decreased the
antioxidant activity.4) This present report describes the
radical and superoxide scavenging activity by employ-
ing matairesinol and oxidized matairesinol. Mat 1–Mat
6, and Lio 7 were prepared to examine the effect of
the chemically active 7, 70, and 80 positions on the
antioxidant activity (Fig. 1). Mat 2 and Mat 4 were first
prepared. P. C. Eklund and co-workers have isolated
Preparation of the compounds
Matairesinol (Mat 1) and Lio 7 were prepared from
L-glutamic acid by employing previously reported
methods.4) As shown in scheme 1, Mat 2, Mat 3, and
Mat 4 were respectively prepared from 6, 8, and 11 and
Mat 5 and Mat 6 were prepared from 13. Compounds 6,
8, 11, and 13 were synthesized from L-glutamic acid.4)
Aldol product 6 was oxidized by pyridinium chloro-
chromate (PCC) to ketone 7 in 95% yield. Hydro-
genolysis of 7 gave Mat 2 in 86% yield. To obtain Mat
3, compound 8 was selected as the starting material.
After ꢀ-benzylation of 8 by using potassium hexameth-
yldisilazane (KHMDS) and 4-benzyloxy-3-methoxyben-
zyl bromide (58%), cleavage of the silyl ether by using
n-Bu4NF and PCC oxidation gave 10 in 42% yield (2
steps). Hydrogenolysis of 10 gave Mat 3 in 45% yield.
The value of specific rotation was higher than that in
literature,6) in which Mat 3 had been prepared from
naturally occurring material. There is a possibility that
the naturally occurring lignan was a mixture of enan-
tiomers.7) Mat 4 was prepared from 114) via 12 by
desilylation and PCC oxidation (58% yield in 2 steps)
followed by hydrogenolysis (90% yield). To prepare 80-
y
To whom correspondence should be addressed. Fax: +81-89-977-4364; E-mail: syamauch@agr.ehime-u.ac.jp

Antioxidant Activity of Lignan
1935
O
O
O
O
O
O
7'
O
O
O
8'
Ar1
Ar1
Ar1
Ar1
O
O
O
7
Ar1
Ar1
Ar1
Ar1
Matairesinol (Mat 1)
Mat 2
Mat 3
Mat 4
O
O
O
OH
OH
Ar1
Ar1
Ar1
O
O
OH
Ar1
Ar1
Ar1
Ar1 =
OCH3
Mat 5
Mat 6
Lio 7
Fig. 1. Matairesinol and Oxidized Matairesinols.
hydroxymatairesinol (Mat 5 and Mat 6), benzyl lactone
13 was treated with KHMDS and 4-benzyloxy-3-
methoxybenzyl bromide, giving dibenzyllactone 14 in
47% yield. ꢀ-Hydroxylation of 14 was achieved by
using oxodiperoxymolybdenum(pyridine)(hexamethyl-
phosphoric triamide) (MoOPH) and KHMDS8) to give
15 (13%) and 16 (10%). Lactone 14 was recovered
(43%). Finally, hydrogenolysis of 15 and 16 respective-
ly gave Mat 5 (86%) and Mat 6 (73%). The stereo-
chemistry of Mat 5 and Mat 6 was confirmed by
comparing with 1H-NMR reference data.8) Although
optically active compounds are not necessary to achieve
this result, all compounds were obtained from L-
glutamic acid. Oxidized matairesinols Mat 2 and Mat
4 were synthesized for the first time.
45
40
35
30
25
20
15
10
5
a
b
c
d
e
f
Antioxidant activity of Mat 1–Mat 6
The radical scavenging activity was measured by
employing DPPH in methanol (Fig. 2). In this study,
matairesinol (Mat 1) showed the strongest activity, this
activity being stronger than that of sesamol which is a
well known natural antioxidant. The activities of 70- and
7-oxomatairesinol Mat 2 and Mat 3 were weaker than
that of matairesinol (Mat 1). Oxidation at the benzylic
position of matairesinol decreased the activity, the
tendency also being observed in previous research about
tetrahydrofuran-type lignans.4) These results mean that a
free benzylic position was necessary for higher activity.
To show antioxidant activity, the presence of a benzyl
radical is necessary to give oxidized marairesinol.5) In
the case of the symmetric lactone type of lignan, a
different level of activity was observed between Mat 2
and Mat 3, the activity of Mat 3 being a little weaker
than that of Mat 2. One reason for this difference would
be due to the production of an ꢀ,ꢁ-unsaturated com-
pound between the 80-position and 70-benzylic position
0
Mat 1
Mat 2
Mat 3
Mat 5
Mat 6 Sesamol
Fig. 2. Radical Scavenging Activity of Compounds Mat 1–Mat 3,
Mat 5, and Mat 6.
Conditions: final concentration of a test sample, 20 mM; DPPH,
0.1 mM; detected at 517 nm. Each data value shows the mean ꢀ SD
(n ¼ 3). Different letters denote significant difference at p < 0:01
(Student’s t-test).
of Mat 35) resulting from the presence of a 70-benzyl
radical and ꢀ-radical. Since the 70-benzyl radical formed
an ꢀ,ꢁ-unsaturated compound without any reaction with
oxygen, the conversion of Mat 3 to an ꢀ,ꢁ-unsaturated
compound reduced the activity. Another reason could
have been the production of the enol form from the ꢁ-
diketone of Mat 2. The antioxidant activity of the ꢁ-

1936
S. YAMAUCHI et al.
O
OH
O
O
O
O
b
Ar2
a
Ar2
Mat 2
Ar2
Ar2
7
6
O
O
O
O
O
Ar2
Ar2
O
c
d, a
b
O
TIPSO
H
Mat 3
TIPSO
H
H
H
Ar2
Ar2
Ar2
10
8
9
O
H
O
O
HO
Ar2
O
O
H
H
b
Ar2
O
d, a
Mat 4
TIPSO
H
Ar2
12
Ar2
11
O
O
OH
Ar2
b
Mat 5
O
O
O
c
Ar2
e
O
Ar2
15
O
OH
Ar2
Ar2
14
b
Ar2
13
Mat 6
O
Ar2
16
Scheme 1. Preparation of Mat 2–Mat 6.
(a) PCC, MS 4A, CH2Cl2, r.t., 16 h (7, 95%; 10, 42%, 2 steps; 12, 58%, 2 steps); (b) H2, 5% Pd/C, 2 h (Mat 2, 86%; Mat 3, 67%; Mat 4, 90%;
Mat 5, 86%; Mat 6, 73%); (c) KHMDS, 4-BnO-3-MeOC6H3CH2Br, THF, ꢁ70 ꢂC, 30 min (9, 58%; 14, 47%); (d) n-Bu4NF, THF, 0 ꢂC;
(e) KHMDS, MoOPH, THF, ꢁ70 ꢂC, 30 min (15, 13%; 16, 10%; recovered 14, 43%).
diketone has been reported.9) The enol form from Mat 2
increased the activity. Considering the resonance effect
of the phenoxy radical with the benzylic radical, the
former factor would be important. The activities of 80-
OH derivatives Mat 5 and Mat 6 were higher than those
of Mat 2 and Mat 3. These results indicate that two free
benzylic positions were more important for higher
activity than the ꢀ-position. The activities of Mat 5
and Mat 6 were a little weaker than that of Mat 1. It
could be assumed that the 80-position of matairesinol
also contributed to the radical scavenging activity due
to the presence of the ꢀ-radical. Unexpectedly, Mat 6
showed higher activity than that of diastereomer Mat 5.
It could be assumed that the cis-benzylic radical was a
little more stable than the trans-benzylic radical.
and Lio 7 are shown in Fig. 3. It can be assumed that
the pKa value for phenol affected the activity. In the
presence of a benzyl ketone, which is an electron-
withdrawing group, the pKa value of phenol was
decreased. When this benzyl ketone was one of the
carbonyl group of the ꢁ-diketone, the pKa value for
phenol was increased because of formation of the enol
form which decreased the electron-withdrawing effect of
the benzyl ketone. The activity of Mat 3 was highest due
to the electron-withdrawing effect by the 7-benzyl
ketone group. On the other hand, due to the formation
of the enol form, the activity of Mat 2 was weaker than
that of Mat 3, showing the same level of activity as that
of matairesinol (Mat 1). In the case of Mat 4, double
enol forms of two benzyl ketones which is a conjugated
diene caused the lowest activity. The activity of one
The superoxide scavenging activities of Mat 1–Mat 6

Antioxidant Activity of Lignan
1937
(0.18 g, 0.84 mmol), and MS 4A (0.1 g) in CH2Cl2 (20
ml) was stirred at room temperature for 16 h before
addition of ether. After the mixture was filtered, the
filtrate was concentrated. The residue was applied to
silica gel column chromatography (EtOAc/hexane =
80
70
60
50
40
30
20
10
0
a
ab
1/2) to give ꢀ-benzoyl lactone 7 (0.40 g, 0.72 mmol,
20
b
ac
95%) as a colorless oil, ½ꢀꢃ
¼ þ31 (c 1.1, CHCl3).
D
b
NMR ꢂH (CDCl3): 2.75 (1H, dd, J 13.7, 8.3 Hz,
ArCHH), 2.83 (1H, dd, J 13.7, 8.3 Hz, ArCHH), 3.40
(1H, m, 3-H), 3.79 (3H, s, OCH3), 3.92 (3H, s, OCH3),
4.13 (1H, dd, J 8.8, 5.9 Hz, 4-HH), 4.26 (1H, d, J 6.3 Hz,
2-H), 4.52 (1H, dd, J 8.8, 7.3 Hz, 4-HH), 5.11 (2H, s,
OCH2Ar), 5.23 (2H, s, OCH2Ar), 6.62 (1H, d, J 7.8 Hz,
ArH), 6.65 (1H, s, ArH), 6.79 (1H, d, J 7.8 Hz, ArH),
6.86 (1H, d, J 8.8 Hz, ArH), 7.31–7.50 (12H, m, ArH).
NMR ꢂC (CDCl3): 37.9, 41.3, 53.5, 55.9, 56.0, 70.8,
71.1, 71.9, 111.2, 112.0, 112.5, 114.3, 121.0, 124.3,
127.20, 127.24, 127.9, 128.2, 128.5, 128.7, 128.9, 130.6,
136.0, 137.0, 147.2, 149.6, 149.9, 153.3, 172.9, 191.5.
FABMS m=z (%): 552 (Mþ, 31), 461 (Mþ–C6H5CH2,
10), 325 (Mþ–(C6H5CH2O)(CH3O)PhCH2, 28), 253
((C6H5CH2O)(CH3O)C6H3C(=O)C, 35), 241 ((C6H5-
CH2O)(CH3O)C6H3C(=O), 34), 163 ((CH3O)(HO)-
C6H3CH2(CH)2, 29), 137 ((CH3O)(HO)C6H3CH2, 12),
91 (C6H5CH2, 100). HRFABMS (Mþ): Calcd. for
C34H32O7, 552.2148. Found, 552.2148.
c
d
Mat 1 Mat 2 Mat 3 Mat 4 Mat 5 Mat 6
Lio 7
Fig. 3. Relative Superoxide Scavenging Activity of Mat 1–Mat 6
and Lio 7.
Final concentration of a test sample, 160 mM. Conditions: final
concentration of a test sample, 20 mM; DPPH, 0.1 mM; detected at
517 nm. Each data value is the mean ꢀ SD (n ¼ 3). Different letters
denote significant difference at p < 0:01 (Student’s t-test).
(2S,3R)-2-(4-Hydroxy-3-methoxybenzoyl)-3-(4-hydroxy-
3-methoxybenzyl)-4-butanolide (Mat 2). A reaction
mixture of benzyl ether 7 (0.40 g, 0.72 mmol) and 5%
Pd–C (0.32 g) in EtOAc (10 ml) was stirred at the
ambient temperature for 2 h under H2 gas before
filtration. The filtrate was concentrated, and then the
residue was applied to silica gel column chromatography
of the diastereomers Mat 5 was weaker than that of
matairesinol (Mat 1). It was found that a proton from the
ꢀ position would have little effect on the activity. The
pKa value of the ꢀ-hydroxy group is low because of a
hydrogen bond with the carbonyl group of lactone. The
activity of Lio 7 was almost the same as that of
matairesinol (Mat 1), Mat 2, and Mat 3.
This study compared the radical scavenging activity
of 70-oxomatairesinol, 7-oxomatairesinol, and 80-hy-
droxymatairesinol for the first time. The relationship
between the structure of oxidized matairesinol and the
superoxide scavenging activity was also examined for
the first time. These results will contribute to the
practical utilization of lignans.
(EtOAc/hexane = 1/1) to give Mat 2 (0.23 g, 0.62
20
mmol, 86%) as a colorless oil, ½ꢀꢃ
¼ þ58 (c 1.1,
D
CHCl3). NMR ꢂH (CDCl3): 2.75 (1H, dd, J 13.9, 8.1 Hz,
ArCHH), 2.84 (1H, dd, J 13.9, 7.6 Hz, ArCHH), 3.36
(1H, m, 3-H), 3.80 (3H, s, OCH3), 3.91 (3H, s, OCH3),
4.15 (1H, dd, J 8.8, 5.9 Hz, 4-HH), 4.27 (1H, d, J 6.4 Hz,
2-H), 4.53 (1H, dd, J 8.8, 6.8 Hz, 4-HH), 5.61 (1H, s,
ArOH), 6.28 (1H, s, ArOH), 6.62 (1H, s, ArH), 6.64 (1H,
d, J 7.8 Hz, ArH), 6.83 (1H, d, J 7.8 Hz, ArH), 6.89 (1H,
d, J 8.3 Hz, ArH), 7.30 (1H, dd, J 8.3, 2.0 Hz, ArH), 7.44
(1H, d, J 2.0 Hz, ArH). NMR ꢂC (CDCl3): 38.0, 41.6,
53.5, 55.8, 56.1, 71.9, 110.6, 111.3, 114.0, 114.6, 121.7,
125.0, 128.4, 129.3, 144.6, 146.7, 151.3, 173.1, 191.5.
EIMS m=z (%): 372 (Mþ, 20), 163 ((HO)(CH3O)C6H3-
CH2(CH)2, 100), 151 ((HO)(CH3O)C6H3C(=O), 98),
137 ((HO)(CH3O)C6H3CH2, 33). HREIMS (Mþ):
Calcd. for C20H20O7, 372.1209. Found, 372.1208.
Experimental
Melting point (mp) data are uncorrected. NMR data
were measured by a JNM-EX400 spectrometer using
TMS as standard (0 ppm), and optical rotation values
were evaluated with a HORIBA SEPA-200 instrument.
The silica gel used was Wakogel C-300 (Wako, 200–
300 mesh). The numbering of compounds follows
IUPAC nomenclatural rules.
(2R,3R)-2-(4-Benzyloxy-3-methoxybenzyl)-3-[(R)-(4-
benzyloxy-3-methoxyphenyl)(triisopropylsilyloxy)meth-
yl]-4-butanolide (9). To a solution of KHMDS (7.42 ml,
0.5 M in toluene, 3.71 mmol) in THF (60 ml) was added
a solution of lactone 8 (1.50 g, 3.09 mmol) in THF (10
ml) at ꢁ70 ꢂC. After 15 min at ꢁ70 ꢂC, a solution of
(2S,3R)-2-(4-Benzyloxy-3-methoxybenzoyl)-3-(4-ben-
zyloxy-3-methoxybenzyl)-4-butanolide (7). A reaction
mixture of benzyl alcohol 6 (0.42 g, 0.76 mmol), PCC

1938
S. YAMAUCHI et al.
4-benzyloxy-3-methoxybenzyl bromide (0.95 g, 3.09
mmol) in THF (10 ml) was added. After stirring at
ꢁ70 ꢂC for 30 min, sat. aq. NH4Cl solution was added.
The organic solution was separated, washed with brine,
and dried (Na2SO4). Concentration followed by silica
gel column chromatography (EtOAc/toluene = 1/4)
3.76 (3H, s, OCH3), 3.94 (3H, s, OCH3), 4.06–4.18 (2H,
m, 3-H, 4-HH), 4.39 (1H, dd, J 7.3, 7.3 Hz, 4-HH), 5.52
(1H, br. s, ArOH), 6.21 (1H, br. s, ArOH), 6.57 (1H, d,
J 7.8 Hz, ArH), 6.62 (1H, s, ArH), 6.74 (1H, d, J 7.8 Hz,
ArH), 6.90 (1H, d, J 8.3 Hz, ArH), 7.21 (1H, d, J 8.3 Hz,
ArH), 7.35 (1H, s, ArH). NMR ꢂC (CDCl3): 34.3, 44.8,
46.3, 55.6, 56.0, 68.2, 109.9, 111.6, 113.7, 114.2, 119.7,
122.1, 123.4, 128.5, 128.8, 144.5, 146.5, 146.9, 151.3,
177.2, 194.8. EIMS m=z (%): 372 (Mþ, 63), 194 (Mþ–
(HO)(CH3O)C6H3CH2CHC=O, 100), 151 ((HO)(CH3O)-
C6H3C=O, 74), 137 ((HO)(CH3O)C6H3CH2, 63).
HREIMS (Mþ): Calcd. for C20H20O7, 372.1209. Found,
372.1210.
gave benzyl lactone 9 (1.27 g, 1.79 mmol, 58%) as a
20
colorless oil, ½ꢀꢃ
¼ þ37 (c 1.8, CHCl3). NMR ꢂH
D
(CDCl3): 0.92–0.98 (21H, m, iso-Pr), 2.49 (1H, m, 2-H),
2.61–2.68 (2H, m, ArCH2), 2.80 (1H, m, 3-H), 3.79 (3H,
s, OCH3), 3.80 (3H, s, OCH3), 4.12 (1H, dd, J 9.3,
7.8 Hz, 4-HH), 4.43 (1H, dd, J 9.3, 6.4 Hz, 4-HH), 4.52
(1H, d, J 6.8 Hz, ArCHOTIPS), 5.12 (2H, s, OCH2Ar),
5.14 (2H, s, OCH2Ar), 6.42 (1H, dd, J 8.3, 2.0 Hz, ArH),
6.56–6.59 (2H, m, ArH), 6.67 (1H, d, J 2.0 Hz, ArH),
6.73 (1H, d, J 8.3 Hz, ArH), 6.77 (1H, d, J 8.3 Hz, ArH),
7.18 (1H, m, ArH), 7.24–7.37 (7H, m, ArH), 7.42–7.44
(2H, m, ArH). NMR ꢂC (CDCl3): 12.5, 17.9, 18.0, 34.7,
43.6, 47.7, 55.8, 55.9, 68.2, 71.0, 75.6, 109.9, 112.9,
113.6, 113.9, 118.7, 121.4, 127.3, 127.4, 127.8, 127.9,
128.2, 128.5, 129.0, 130.6, 135.4, 136.9, 137.2, 147.1,
147.8, 149.7, 179.0. Anal. Found: C, 73.02; H, 7.61.
Calcd. for C43H54O7Si: C, 72.64; H, 7.66%.
(2S,3R)-2,3-Bis(4-benzyloxy-3-methoxybenzoyl)-4-bu-
tanolide (12). The same synthetic method as that
described for compound 10 gave 12 from 11 as colorless
crystals, mp 147–149 ꢂC (EtOAc), in 58% yield (2
20
steps), ½ꢀꢃ
þ6 (c 1.1, CHCl3). NMR ꢂH (CDCl3):
D
3.90 (3H, s, OCH3), 3.91 (3H, s, OCH3), 4.39 (1H, dd, J
8.8, 7.3 Hz, 4-HH), 4.72 (1H, dd, J 8.8, 8.8 Hz, 4-HH),
5.03 (1H, ddd, J 8.8, 7.3, 7.3 Hz, 3-H), 5.11 (1H, d, J
7.3 Hz, 2-H), 5.21 (4H, s, OCH2Ar), 6.90 (1H, d, J
7.3 Hz, ArH), 6.93 (1H, d, J 8.8 Hz, ArH), 7.30–7.42
(10H, m, ArH), 7.48 (1H, d, J 8.3 Hz, ArH), 7.51 (1H, s,
ArH), 7.61 (1H, d, J 1.5 Hz, ArH), 7.71 (1H, dd, J 8.3,
2.0 Hz, ArH). NMR ꢂC (CDCl3): 45.2, 50.4, 56.0, 68.0,
70.7, 70.8, 110.9, 111.6, 112.1, 112.2, 123.3, 125.2,
127.1, 125.2, 127.1, 128.0, 128.1, 128.2, 128.4, 128.61,
128.63, 135.8, 135.9, 149.5, 149.9, 153.5, 153.6, 171.3,
190.2, 194.4. Anal. Found: C 71.81%, H 5.33%. Calcd
for C34H30O8: C 72.07%, H 5.34%.
(2R,3R)-3-(4-Benzyloxy-3-methoxybenzoyl)-2-(4-ben-
zyloxy-3-methoxybenzyl)-4-butanolide (10). A reaction
solution of silyl ether 9 (0.88 g, 1.24 mmol) and n-
Bu4NF (1.36 ml, 1 M in THF, 1.36 mmol) in THF (15 ml)
was stirred at 0 ꢂC for 1 h before addition of sat. aq.
NH4Cl solution. The organic solution was separated,
washed with brine, and dried (Na2SO4). Concentration
gave crude benzyl alcohol. This crude benzyl alcohol
was converted to 10 by PCC oxidation by the same
method as that described for compound 7. Colorless
(2S,3R)-2.3-Bis(4-hydroxy-3-methoxybenzoyl)-4-buta-
nolide (Mat 4). The same synthetic method as that
crystals, mp 105–107 ꢂC (EtOAc), 42% yield (2 steps),
20
½ꢀꢃ
¼ þ13 (c 1.2, CHCl3). NMR ꢂH (CDCl3): 2.97
described for Mat 2 gave Mat 4 from compound 12 as a
20
D
(1H, dd, J 14.2, 7.3 Hz, ArCHH), 3.03 (1H, dd, J 14.2,
5.4 Hz, ArCHH), 3.53 (1H, m, 2-H), 3.70 (3H, s, OCH3),
3.90 (3H, s, OCH3), 4.01–4.14 (2H, m, 3-H, 4-HH), 4.35
(1H, dd, J 7.8, 7.3 Hz, 4-HH), 5.03 (2H, s, OCH2Ar),
5.21 (2H, s, OCH2Ar), 6.54 (1H, dd, J 8.3, 2.0 Hz, ArH),
6.63 (1H, d, J 2.0 Hz, ArH), 6.66 (1H, d, J 8.3 Hz, ArH),
6.81 (1H, d, J 8.3 Hz, ArH), 7.15 (1H, dd, J 8.3, 2.0 Hz,
ArH), 7.29–7.41 (11H, m, ArH). NMR ꢂC (CDCl3):
34.2, 44.5, 46.4, 55.6, 55.9, 68.2, 70.7, 70.9, 110.6,
111.8, 112.8, 113.9, 121.4, 122.6, 127.1, 127.2, 127.7,
128.1, 128.4, 128.6, 128.9, 130.1, 135.8, 137.0, 147.1,
149.6, 149.8, 153.2, 177.2, 194.9. Anal. Found: C,
73.54; H, 5.81. Calcd. for C34H32O7: C, 73.89; H,
5.84%.
colorless oil in 90% yield, ½ꢀꢃ
þ43 (c 1.1, CHCl3).
D
NMR ꢂH (CDCl3): 3.83 (3H, s, OCH3), 3.84 (3H, s,
OCH3), 4.37 (1H, dd, J 8.8, 6.8 Hz, 4-HH), 4.73 (1H, dd,
J 8.8, 8.8 Hz, 4-HH), 5.02 (1H, ddd, J 8.8, 6.8, 6.8 Hz, 3-
H), 5.14 (1H, d, J 6.8 Hz, 2-H), 6.56 (2H, s, ArOH), 6.90
(2H, d, J 8.8 Hz, ArH), 7.46–7.48 (2H, m, ArH), 7.54
(1H, d, J 2.0 Hz, ArH), 7.68 (1H, dd, J 8.3, 2.0 Hz,
ArH). NMR ꢂC (CDCl3): 45.3, 50.3, 55.9, 56.0, 68.8,
110.4, 111.1, 114.2, 114.3, 124.1, 125.8, 127.4, 127.9,
146.6, 147.1, 151.6, 151.7, 171.7, 190.3, 194.4; EIMS
m=z (%): 386 (Mþ, 13), 342 (Mþ–CO2, 14), 235 ([Mþ–
C(=O)C6H3(OH)(OCH3), 100], 151 [Mþ–C(=O)-
C6H3(OCH3)(OH), 99], 123 [(C6H3(OH)(OCH3), 47);
HREIMS (Mþ): calcd. for C20H18O8, 386.1002. Found:
386.1002.
(2R,3R)-3-(4-Hydroxy-3-methoxybenzoyl)-2-(4-hydroxy-
3-methoxybenzyl)-4-butanolide (Mat 3). The same syn-
(2R,3R)-2,3-Bis(4-benzyloxy-3-methoxybenzyl)-4-bu-
tanolide (14). The same synthetic method as that
thetic method as that described for Mat 2 gave Mat 3 from
20
10 as a colorless oil in 67% yield, ½ꢀꢃ
¼ þ75 (c 0.2,
described for 9 gave 14 from 13 as a colorless oil in
20
D
20
THF), ½ꢀꢃ
¼ þ44:3 (c 1.0, THF) in the literature.6)
47% yield, ½ꢀꢃ
¼ 11 (c 5.7, CHCl3). NMR ꢂH
D
D
NMR ꢂH (CDCl3): 2.99 (1H, dd, J 13.6, 6.8 Hz, ArCHH),
3.05 (1H, dd, J 13.6, 6.8 Hz, ArCHH), 3.52 (1H, m, 2-H),
(CDCl3): 2.41–2.51 (2H, m, 2-H, 2-ArCHH), 2.51–
2.62 (2H, m, 3-H, 2-ArCHH), 2.88 (1H, dd, J 9.8, 5.9

Antioxidant Activity of Lignan
1939
Hz, 3-ArCHH), 2.92 (1H, dd, J 9.8, 5.9 Hz, 3-ArCHH),
3.79 (3H, s, OCH3), 3.82 (3H, s, OCH3), 3.83–3.88 (1H,
m, 4-HH), 4.10 (1H, dd, J 8.3, 7.3 Hz, 4-HH), 5.10 (4H,
s, OCH2Ar), 6.45 (1H, d, J 8.3 Hz, ArH), 6.48 (1H, s,
ArH), 6.56 (1H, d, J 7.8 Hz, ArH), 6.69 (1H, s, ArH),
6.73–6.77 (2H, m, ArH), 7.24–7.36 (6H, m, ArH), 7.40–
7.42 (4H, m, ArH). NMR ꢂC (CDCl3): 34.5, 38.2, 41.0,
46.5, 56.0, 71.1, 71.2, 112.5, 112.9, 114.1, 114.3, 120.5,
121.3, 127.2, 127.3, 127.7, 127.80, 127.83, 128.49,
128.52, 130.8, 131.1, 137.1, 147.0, 147.1, 149.7, 149.8,
178.7. Anal. Found: C, 75.74; H, 6.50. Calcd. for
C34H34O6: C, 75.82; H, 6.36%.
(2S,3S)-2-Hydroxy-2,3-bis(4-hydroxy-3-methoxyben-
zyl)-4-butanolide (Mat 5). The same synthetic method as
that described for Mat 2 gave Mat 5 from compound 15
20
as a colorless oil in 86% yield, ½ꢀꢃ
¼ ꢁ30 (c 1.1,
D
CHCl3). NMR ꢂH (CDCl3): 2.46–2.53 (2H, m, 3-H, 3-
ArCHH), 2.73 (1H, s, OH), 2.88–2.96 (1H, m, 3-
ArCHH), 2.91 (1H, d, J 13.7 Hz, 2-ArCHH), 3.08 (1H,
d, J 13.7 Hz, 2-ArCHH), 3.82 (3H, s, OCH3), 3.84 (3H,
s, OCH3), 3.97–4.04 (2H, m, 4-H2), 5.56 (1H, s, ArOH),
5.58 (1H, s, ArOH), 6.59–6.62 (3H, m, ArH), 6.69 (1H,
s, ArH), 6.81–6.83 (2H, m, ArH). NMR ꢂC (CDCl3):
31.6, 42.1, 43.8, 55.89, 55.94, 70.1, 76.4, 111.5, 112.7,
114.3, 114.5, 121.3, 121.4, 126.1, 130.3, 144.3, 145.0,
146.56, 146.60, 178.5. EIMS m=z (%): 374 (Mþ, 55),
358 (Mþ–OH + 1, 23), 137 ((HO)(CH3O)C6H3CH2,
100). HREIMS (Mþ): Calcd. for C20H22O7, 374.1366.
Found, 374.1365.
(2S,3S)-2,3-Bis(4-benzyloxy-3-methoxybenzyl)-2-hy-
droxy-4-butanolide (15) and (2R,3S)-2,3-bis(4-benzyl-
oxy-3-methoxybenzyl)-2-hydroxy-4-butanolide (16). To
a solution of KHMDS (4.24 ml, 0.5 M toluene solution,
2.12 mmol) in THF (8.5 ml) was added a solution of
lactone 14 (0.57 g, 1.06 mmol) in THF (8.5 ml) at
ꢁ70 ꢂC. After the mixture was stirred at ꢁ70 ꢂC for
30 min, MoOPH (0.56 g, 1.29 mmol) was added. The
resulting reaction mixture was stirred at ꢁ70 ꢂC for 2 h
before additions of sat. aq. Na2SO3 solution and EtOAc.
The organic solution was separated, washed with 2 M aq.
HCl solution, sat. aq. NaHCO3 solution, and brine, and
dried (Na2SO4). Concentration followed by silica gel
column chromatography (EtOAc/hexane = 1/4) gave
(2S)-15 (80 mg, 0.14 mmol, 13%) as a colorless oil and
(2R)-16 (60 mg, 0.11 mmol, 10%) as a colorless oil.
(2R,3S)-2-Hydroxy-2,3-bis(4-hydroxy-3-methoxyben-
zyl)-4-butanolide (Mat 6). The same synthetic method
as that described for Mat 2 gave Mat 6 from com-
pound 16 as colorless powder, mp 170–172 ꢂC, in 73%
20
yield, ½ꢀꢃ
¼ ꢁ4:9 (c 1.2, CHCl3). NMR ꢂH (CDCl3):
D
2.64 (1H, dd, J 13.4, 11.7 Hz, 3-ArCHH), 2.88–2.99
(2H, m, OH, 3-H), 2.95 (2H, s, 2-ArCH2), 3.12 (1H,
dd, J 13.4, 1.3 Hz, 3-ArCHH), 3.83–3.88 (1H, m, 4-
HH), 3.86 (3H, s,OCH3), 3.87 (3H, s, OCH3), 4.17 (1H,
dd, J 8.8, 8.6 Hz, 4-HH), 5.62 (1H, s, ArOH), 5.70 (1H,
s, ArOH), 6.63–6.68 (2H, m, ArH), 6.72–6.74 (2H, m,
ArH), 6.83–6.86 (2H, m, ArH). NMR ꢂC (CDCl3): 32.0,
38.2, 48.1, 55.9, 56.0, 69.4, 75.8, 110.8, 112.9, 114.4,
114.6, 121.1, 123.3, 124.7, 129.6, 144.5, 145.2, 146.5,
146.8, 178.0. EIMS m=z (%): 374 (Mþ, 38), 137
((HO)(CH3O)C6H3CH2, 100). HREIMS (Mþ): Calcd.
for C20H22O7, 374.1366. Found, 374.1366.
Lactone 14 (0.25 g, 0.46 mmol, 43%) was recovered.
20
(2S)-15, ½ꢀꢃ
¼ ꢁ25 (c 0.5, CHCl3). NMR ꢂH
D
(CDCl3): 2.49–2.53 (2H, m, 3-H, 3-ArCHH), 2.88–
2.95 (1H, m, 3-ArCHH), 2.91 (1H, d, J 13.7 Hz, 2-
ArCHH), 3.08 (1H, d, J 13.7 Hz, 2-ArCHH), 3.83 (3H, s,
OCH3), 3.84 (3H, s, OCH3), 3.96–4.05 (2H, m, 4-H2),
5.119 (2H, s, ArCH2O), 5.124 (2H, s, ArCH2O), 6.58–
6.60 (2H, m, ArH), 6.64 (1H, s, ArH), 6.74 (1H, s, ArH),
6.78–6.80 (2H, m, ArH), 7.27–7.29 (2H, m, ArH), 7.31–
7.37 (4H, m, ArH), 7.41–7.43 (4H, m, ArH). NMR ꢂC
(CDCl3): 31.6, 42.1, 43.8, 56.0, 70.0, 71.0, 71.1, 76.3,
112.8, 113.9, 114.0, 114.4, 120.8, 122.4, 127.21, 127.22,
127.82, 127.84, 128.5, 131.6, 137.0, 137.1, 146.9, 147.6,
149.6, 149.8, 178.3. Anal. Found: C, 73.46; H, 6.36.
Measurement of the radical scavenging activities of
Mat 1–Mat 3, Mat 5 and Mat 6. To an appropriate
amount of a sample in a methanol solution (4.9 ml) was
added 100 ml of 5 mM DPPH in a methanol solution.
After the solution had stood at 25 ꢂC for 0.5 h, the
absorbance at 517 nm was measured. The antiradical
activity was evaluated from the decreased value of the
51-nm absorption, which was calculated by the follow-
ing equation: %scavenging effect = [(A0–A1)/A0] ꢄ
100, where A0 is the absorption of the control (without a
sample) and A1 is the absorption of the mixture
containing a sample.
Calcd. for C34H34O7: C, 73.63; H, 6.18%. (2R)-16,
20
½ꢀꢃ
¼ ꢁ11 (c 0.9, CHCl3). NMR ꢂH (CDCl3): 2.64
D
(1H, dd, J 13.7, 11.7 Hz, 3-ArCHH), 2.82 (1H, s, OH),
2.88–2.98 (1H, m, 3-H), 2.96 (2H, s, 2-ArCH2), 3.12
(1H, dd, J 13.7, 3.9 Hz, 3-ArCHH), 3.82–3.84 (1H, m, 4-
HH), 3.88 (6H, s, OCH3), 4.17 (1H, dd, J 8.8, 7.3 Hz, 4-
HH), 5.14 (4H, s, OCH2Ar), 6.62 (1H, d, J 8.3 Hz, ArH),
6.71–6.72 (2H, m, ArH), 6.77 (1H, d, J 2.0 Hz, ArH),
6.81–6.85 (2H, m, ArH), 7.28–7.30 (2H, m, ArH), 7.32–
7.39 (4H, m, ArH), 7.43–7.45 (4H, m, ArH). NMR ꢂC
(CDCl3): 32.0, 38.2, 48.0, 56.0, 69.3, 71.0, 71.1, 75.8,
112.1, 113.9, 114.1, 114.4, 120.4, 122.6, 126.0, 127.2,
127.3, 127.8, 128.5, 130.9, 137.0, 137.1, 147.1, 147.8,
149.5, 150.0, 177.8. Anal. Found: C, 73.23; H, 6.47.
Calcd. for C34H34O7: C, 73.63; H, 6.18%.
Measurement of the superoxide scavenging activities
of Mat 1–Mat 6 and Lio 7. The superoxide scavenging
activity was determined by using a WST superoxide
dismutase assay kit (Dojindo Molecular Technology,
Kumamoto, Japan). Briefly, superoxide radicals were
generated by the xanthine/xanthine oxidase system, and
reduced WST-8 to water-soluble formazane which
exhibited an absorption maximum at 450 nm. Decreased
absorption of the reaction mixture indicated increased

1940
S. YAMAUCHI et al.
Synthesis and antioxidant activity of oxygenated furo-
furan lignans. Biosci. Biotechnol. Biochem., 68, 183–192
(2004).
superoxide scavenging activity. The reaction mixture
was incubated at 37 ꢂC for 20 min, and the absorption
was read at 450 nm by a Bio-Rad 680 microplate eader.
The capability of scavenging the superoxide radicals
was calculated by using the following equation: %
scavenging effect = [(A0–A1)/A0] ꢄ 100, where A0 is
the absorption of the control (without a sample) and A1
is the absorption of the mixture containing a sample.
3) Yamauchi, S., Hayashi, Y., Kirikihira, T., and Masuda,
T., Synthesis and antioxidant activity of olivil-type
lignans. Biosci. Biotechnol. Biochem., 69, 113–122
(2004).
4) Yamauchi, S., Hayashi, Y., Nakashima, Y., Kirikihira, T.,
Yamada, K., and Masuda, T., Effect of benzylic oxygen
on the antioxidant activity of phenolic lignans. J. Nat.
Prod., 68, 1459–1470 (2005).
Statistical analysis. Each results is expressed as the
means ꢀ standard deviation (SD: n ¼ 3). Student’s t-
test was used to assess the statistical significance of a
difference which was considered to be significant at
p < 0:01.
˚
5) Eklund, P. C., Langvik, O. K., Warna, J. P., Salmi, T. O.,
Willfor, S. M., and Sjoholm, R. E., Chemical studies on
˚
¨
¨
¨
antioxidant mechanism and free radical scavenging
properties of lignans. Org. Biomol. Chem., 3, 3336–
3347 (2005).
6) Eklund, P. C., and Sjoholm, R. E., Oxidative trans-
¨
Acknowledgments
formation of the natural lignan hydroxymatairesinol with
2,3-dichloro-5,6-dicyano-1,4-benzoquinone. Tetrahedron,
59, 4515–4523 (2003).
We measured the 400 MHz NMR data at INCS of
Ehime University. Our thanks to the president of Ehime
University for supporting this project. We are also
grateful to Marutomo Co. for financial support.
7) Ralph, J., Peng, J., Lu, F., Hatfield, R. D., and Helm, R.
F., Are lignans optically active? J. Agric. Food Chem., 47,
2991–2996 (1999).
8) Moritani, Y., Fukushima, C., Ukita, T., Miyagishima, T.,
Ohmizu, H., and Iwasaki, T., Stereoselective syntheses
of cis- and trans-isomers of ꢀ-hydroxy-ꢀ,ꢁ-dibenzyl-ꢃ-
butyrolactone lignans: new syntheses of (+)-tracheloge-
nin and (+)-guayadequiol. J. Org. Chem., 61, 6922–6930
(1996).
9) Osawa, T., Sugiyama, Y., Inayoshi, M., and Kawakishi,
S., Antioxidative activity of tetrahydrocurcuminoids.
Biosci. Biotechnol. Biochem., 59, 1609–1612 (1995).
References
1) Osawa, T., Phenolic antioxidants in dietary plants as
antimutagens. In ‘‘Phenolic Compounds in Food and
Their Effects on Health II,’’ eds. Huang, M.-T., Ho, C.-T.,
and Lee, C. Y., American Chemical Society, Washington,
DC, pp. 135–149 (1992).
2) Yamauchi, S., Ina, T., Kirikihira, T., and Masuda, T.,