In vivo anti-inflammatory and antinociceptive effects of liriodendrin isolated from the stem bark of Acanthopanax senticosus

Article abstract of DOI:10.1055/s-2003-41127

In the present study, liriodendrin isolated by activity-guided fractionation from the ethyl acetate (EtOAc) extracts of the stem bark of Acanthopanax senticosus, was evaluated for anti-inflammatory and antinociceptive activities. Liriodendrin (5, 10 mg/kg/day, p.o.) significantly inhibited the increase of vascular permeability induced by acetic acid in mice and reduced an acute paw edema induced by carrageenan in rats. When the analgesic activity was measured by the acetic acid-induced writhing test and hot plate test, liriodendrin showed a dose-dependent inhibition in animal models. In addition, syringaresinol, the hydrolysate of liriodendrin, more potently inhibited the LPS-induced production of NO, PGE₂ and TNF-α production of macrophages than liriodendrin. Consistent with these observations, the expression level of iNOS and COX-2 enzyme was decreased by syringaresinol in a concentration-dependent manner. These results suggest that the anti-inflammatory and antinociceptive effects of liriodendrin after oral administration were attributable to the in vivo transformation to syringaresinol, which may function as the active constituent.

Full text of DOI:10.1055/s-2003-41127

Hyun-Ju Jung1
Hee-Juhn Park1
Ryung-Gue Kim2
Kyoung-Min Shin2
Joohun Ha3
In vivo Anti-Inflammatory and Antinociceptive Effects
of Liriodendrin Isolated from the Stem Bark of
Acanthopanax senticosus
Jong-Won Choi4
Hyoung Ja Kim5
Yong Sup Lee5
Kyung-Tae Lee2
Abstract
tion of NO, PGE2 and TNF-a production of macrophages than lir-
iodendrin. Consistent with these observations, the expression
In the present study, liriodendrin isolated by activity-guided level of iNOS and COX-2 enzyme was decreased by syringaresi-
fractionation from the ethyl acetate (EtOAc) extracts of the stem nol in a concentration-dependent manner. These results suggest
bark of Acanthopanax senticosus, was evaluated for anti-inflam- that the anti-inflammatory and antinociceptive effects of lirio-
matory and antinociceptive activities. Liriodendrin (5, 10 mg/kg/ dendrin after oral administration were attributable to the in
day, p.o.) significantly inhibited the increase of vascular perme- vivo transformation to syringaresinol, which may function as
ability induced by acetic acid in mice and reduced an acute paw the active constituent.
edema induced by carrageenan in rats. When the analgesic activi-
ty was measured by the acetic acid-induced writhing test and Key words
hot plate test, liriodendrin showed a dose-dependent inhibition Acanthopanax senticosus ´ Araliaceae ´ liriodendrin ´ syringaresi-
in animal models. In addition, syringaresinol, the hydrolysate of nol ´ anti-inflammation ´ antinociceptive
liriodendrin, more potently inhibited the LPS-induced produc-
610
Introduction
have been isolated [3]. However, no triterpene derivative has
been isolated from the stem bark. Fujikawa et al. [4] have reported
The stem bark of Acanthopanax senticosus Harms (Eleutherococcus that chlorogenic acid and syringaresinol di-O-b-D-glucoside (lirio-
senticosus Maxim, Araliaceae), very often called Siberian Ginseng, dendrin) isolated from the stem bark of A. senticosus could repre-
has been used as a tonic for chronic bronchitis, hypertension, is- sent the anti-gastric ulcer activity of the extract. Polysaccharide
chemic heart disease and anti-stress. From the leaves of A. and sesamin isolated from A. senticosus were reported to have
senticosus, many saponins named ciwujianosides have been re- anti-tumor activities via immunostimulation and induction of
ported [1], [2]. From the stem bark, many phenolic compounds in- apoptosis [5], [6]. However, this plant has not been investigated
cluding lignan glycosides (the glycosides of pinoresinol, mediore- previously for systematic isolation of its anti-inflammatory com-
sinol and syringaresinol), isofraxidin, isofraxidin 7-O-glucoside, ponents. We report herein the bioassay-guided chromatographic
chlorogenic acid, 2,6-dimethoxy-p-benzoquinone and syringin fractionation of an ethyl acetate soluble extract of A. senticosus,
Affiliation
1 Division of Applied Plant Sciences, Sangji University, Woosan-Dong, Wonju, Korea
2 College of Pharmacy, Kyung Hee University, Hoegi-Dong, Seoul, Korea
3 College of Medicine, Kyung Hee University, Hoegi-Dong, Seoul, Korea
4 College of Pharmacy, Kyungsung University, Dayeon-Dong, Pusan, Korea
5 Division of Life Sciences, Korea Institute of Science and Technology, Cheongryang, Seoul, Korea
Correspondence
Dr. Kyung-Tae Lee ´ Department of Biochemistry ´ College of Pharmacy ´ Kyung-Hee University ´
Dongdaemun-Ku ´ Hoegi-Dong 130±701 ´ Seoul ´ Korea ´ Phone: +82-2-9610860 ´ Fax: +82-2-9663885 ´
E-mail: ktlee@khu.ac.kr
Funding
This research was supported by a grant (PF002104±07) from Plant Diversity Research Center of 21st Frontier
Research Program funded by Ministry of Science and Technology of Korean government
Received September 20, 2002 ´ Accepted April 26, 2003
Bibliography
Planta Med 2003; 69: 610±616 ´ ꢀ Georg Thieme Verlag Stuttgart ´ New York ´ ISSN 0032-0943

which has led to the isolation of the anti-inflammatory lirioden- 5.2) and the solution was left at 378C for 2 days. The precipitate
drin (Fig.1).
was collected by filtration and extracted with CH2Cl2. The com-
bined organic solvent was evaporated to dryness and residue
was purified by preparative TLC on RP-18 (0.25 mm, 20”20 cm,
Merck), developed with MeOH-H2O (64:36), to afford 13.2 mg
of syringaresinol (2a). The spectral data and optical rotation
value {[a]D25: + 62.28 (c 0.1, CHCl3)} were identical with those re-
Materials and Methods
Plant material
A. senticosus was collected in August 1998 in Kangwon Province, ported in the literature [8].
Korea, and the plant was identified by Prof. S. Y. Yun (Division of
Applied Plant Sciences, Sang-Ji University, Wonju, Korea). A vou- Kalopanaxsaponin A {3; hederagenin 3-O-[a-L-rhamnopyrano-
cher specimen (#NATCHEM-20) was deposited in the herbarium syl-(1®2)-a-L-arabinopyranoside]}: colorless needles from
of Life Science and Natural Resources, Sang-Ji University, Wonju, MeOH, m.p. 265±2688C (dec.), [a]D25: + 188 (c 0.3, MeOH) [9].
Korea.
Hederoside D2 {4; hederagenin 3-O-[b-D-glucopyranosyl-(1®2)-
Extraction, fractionation and isolation
a-L-arabinopyranoside]}: Colorless needles, m.p. 248±2508C,
Dried stem bark (4.8 kg) of A. senticosus was cut and extracted [a]D25: + 408 (c 1.66, pyridine) [10].
with MeOH three times under reflux. The extract was filtered
and evaporated on a rotary evaporator under reduced pressure Copies of the original spectra are obtainable from the author of
to give a viscous mass (630 g) of MeOH extract. This extract was correspondence.
successively fractionated into CHCl3, EtOAc and n-BuOH, and
dried under vacuum to give CHCl3 extract (180 g), EtOAc extract Chemicals
(12 g), and n-BuOH extract (87 g). The most active EtOAc extract DMEM medium, fetal bovine serum (FBS), penicillin, and strepto-
was subjected to silica gel (600 g, 7”70 cm, Merck, Art 7734, mycin were obtained from Life Technologies Inc. (Grand Island,
Germany) column chromatography using the eluent CHCl3- NY), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-
MeOH-H2O (73:27:10, lower phase) and fractions A (300± mide (MTT), sulfanilamide, aprotinin, leupeptin, phenylmethyl-
700 mL, tR; 60±140 min), B (800±1100 mL, tR; 160±220 min), sulfonyl flouride (PMSF), dithiothreitol, N-mono-methylarginine
C (1200±1400 mL), and D (1500±1800 mL) were obtained. Re- (NMA), indomethacin and E. coli lipopolysaccharide (LPS) were
crystallization of the fractions A and B in MeOH gave caffeic purchased from Sigma Chemical Co. (St. Louis, MO). Trypsin was
acid (1) as pale yellow crystals and liriodendrin (2) as colorless supplied from Kawayam (Japan). COX-2 and iNOS monoclonal
needles, respectively. Purification of fraction C was performed antibodies and the peroxidase conjugated secondary antibody
by silica gel column chromatography with the eluting solvent were purchased from Santa Cruz Biotechnology, Inc. (Santa
of CHCl3-MeOH-H2O (75:25:10, lower phase). Compound 3 Cruz, CA). EIA kits for prostaglandin E2 and TNF-a were obtained
(kalopanaxsaponin A) was detected at the elution volume of from Amersham Pharmacia Biotech. (Piscataway, NJ) and R&D
350±450 mL, recrystallized in MeOH and obtained as a white systems (Minneapolis, MN), respectively.
powder. Purification of fraction D gave hederoside D2 (4, elution
611
volume; 500±750 mL) as a white powder by the similar proce- Animals
dure as fraction C.
ICR male mice weighing 20±25 g and Sprague-Dawley male rats
weighing 100±120 g were purchased from Korean Experimental
Caffeic acid (1): pale yellow crystals from MeOH, m.p. 223± Animal Co. and maintained in a constant condition (tempera-
1
2258C. This compound was identified by comparison of the H- ture: 20  28C, humidity: 40±60%, light/dark cycle: 12 h) for
NMR and 13C-NMR data with those of an authentic specimen.
two weeks or more. Twenty-four hours before the experiment,
only water was offered to the animals. Considering the variation
Liriodendrin (2): colorless needles from MeOH, m.p. 257±2588C, of enzyme activity during one day, the animals were sacrificed at
[a]D25: ±18.58 (c 0.54, 50% EtOH). FAB-MS: m/z = 765 [C34H46O18
Na]+ [7].
+
fixed time (10:00 a.m.±12:00 a.m.). These experiments were
approved by the University of Kyungsung Animal Care and Use
Committee. All procedures were conducted in accordance with
Syringaresinol (2a): b-Glycosidase (4.6 mg) was added to a solu- the ªGuide for Care and Use of Laboratory Animalsº published
tion of liriodendrin (2, 39.0 mg) in 10 mL of NaOAc buffer (pH by the National Institutes of Health.
Acetic acid-induced vascular permeability in mice
The method is based on that of Whittle [11]. The mice were dosed
Fig. 1 Chemical structures of
liriodendrin (2) and syringare-
orally with the test substances suspended in 0.2% carboxymethyl
cellulose (CMC) solution 30 min before the injection of acetic
acid-saline solutions. Thirty min after the injection (i.p., 0.7%
acetic acid-saline 0.1 mL/10 g), 4% pontamine sky blue (10 mL/
kg) was also injected to the tail vein. After 20 min, the mice
were sacrificed and then the pigment exuded to abdominal cav-
ity was washed with 10 mL of distilled water. This washed solu-
tion was centrifuged (3,000 rpm, 10 min) and then the absor-
bance of supernatants was measured at 580 nm wavelength
sinol (2a).
Jung H-J et al. In vivo Anti-inflammatory¼ Planta Med 2003; 69: 610±616

using a UV-VIS spectrophotometer. The vascular permeability ef- dium nitrite standard curve freshly prepared in culture medi-
fects were expressed in terms of dye amount per 30 g weight of um. PGE2 and TNF-a levels in macrophage culture medium
mouse, which leaked into the intraperitoneal cavity. Aminopyr- were quantified by EIA kits according to the manufacture's in-
ine was used as a standard drug [12].
structions.
Carrageenan-induced edema in rats
Western blot assay
The initial hind paw volume of the Sprague-Dawley strain rats Cellular proteins were extracted from control and liriodendrin/
was determined volumetrically. A 1% solution of carrageenan in syringaresinol-treated RAW 264.7 cells. The washed cell pellets
saline (0.1 mL/rat) was injected subcutaneously into the right were resuspended in extraction lysis buffer (50 mM HEPES pH
hind paw 1 h after the test substances had been administered 7.0, 250 mM NaCl, 5 mM EDTA, 0.1% Nonidet P-40, 1 mM phenyl-
orally. The control group received the vehicle. Paw volumes methylsulfonyl fluoride, 0.5 mM dithiothreitol, 5 mM NaF, 0.5
were measured up to 2 h at intervals of 30 min, and the volume mM Na orthovanadate) containing 5 mg/mL each of leupeptin
of the edema was measured with a plethysmometer. Indometha- and aprotinin and incubated for 15 min at 48C. Cell debris was
cin was used as a standard drug [12].
removed by microcentrifugation, followed by quick freezing of
the supernatants. Protein concentration was determined by Bio-
Rad protein assay reagent according to the manufacture's in-
Measurement of trypsin activity
The trypsin inhibitor activity was measured according to the struction. 40±50 mg of cellular proteins from treated and un-
method of Bieth et al. [13] and Fritz et al. [14]. The absorbance treated cell extracts were electroblotted onto a nitrocellulose
of N-benzoylarginine-p-nitroanilide (BAPNA) was measured at membrane following separation on a 10% SDS-polyacrylamide
405 nm indicating the conversion of BAPNA to N-benzoylargi- gel electrophoresis. The immunoblot was incubated overnight
nine and p-nitroaniline by trypsin activity. In brief, trypsin solu- with blocking solution (5% skim milk) at 48C, followed by incu-
tion (50 mU/mL) and triethanolamine buffer solution (0.2 M tri- bation for 4 h with a 1:500 dilution of monoclonal anti-iNOS and
ethanolamine, 20 mM CaCl2, pH 7.8) were added to the test tubes COX-2 antibody (Santacruz, CA, U.S.A.). Blots were washed two
containing inhibitor assay (I) and trypsin reference assay (Tr), times with PBS and incubated with a 1:1000 dilution of horse-
respectively, and preincubated at 378C for 5 min. Then, the sub- radish peroxidase-conjugated goat anti-mouse IgG secondary
strate of BAPNA was added and its reaction rate was determined antibody (Santacruz, CA, U.S.A.) for 1 h at room temperature.
every 1 min for 5 min by the absorbances taken at 405 nm by a Blots were again washed three times in tween 20/Tris-buffered
UV-VIS spectrophotometer. The enzyme activity was expressed saline (TTBS) and then developed by enhanced chemilumines-
as IU/mL.
cence (Amersham Life Science, Arlington Heights, IL, U.S.A.).
Writhing test in mice
Data analysis
The acetic acid abdominal constriction test was performed as de- Statistical analysis was performed using one-way analysis of var-
scribed by Whittle [11]. Vehicle, aminopyrine (100 mg/kg) and iance (ANOVA) followed by Duncan's test for multiple compari-
test solution (100 and 250 mg/kg) were orally administered 30 sons, and Student's t test for single comparisons. The data are re-
min before the experiment, and 0.1 mL/10 g of 0.7% acetic acid- ported as mean  S.E.M. The number of independent experi-
saline was then injected i.p. 10 min after the injection, the fre- ments assessed are given in the legends for the figures.
quency of writhing in mice was counted for the succeeding 10
612
min.
Results
Hot plate test
The hot plate test was used to measure the response latencies ac- Activity-guided fractionation and isolation of active
cording to the method described previously by Eddy and Leim- compounds
back [15], with minor modifications. In these experiments, the The MeOH extract of the stem bark of A. senticosus exhibited
hot plate (Ugo Basile, model-DS 37) was maintained at 56  18C. anti-inflammatory and antinociceptive effects as demonstrated
The reaction time was noted by observing either the licking of by the acetic acid-induced vascular permeability, carrageenan-
the hind paws or the jumping movements before and after drug induced anti-edema, writhing and hot plate tests (Tables 1±3).
administration. The cut-off time was 20 sec and morphine sul- This extract was also found to inhibit carrageenan-induced
fate 4.0 mg/kg (Kuju Pharmaceutical CO), administered intraperi- trypsin activity (Table 1). This extract was successfully fractio-
toneally, was used as reference drug [16].
nated into CHCl3, EtOAc and BuOH extracts for the activity-
guided fractionation. The potency of the fractions after 100
and 250 mg/kg oral administration was shown to be as follows:
Nitrite, PGE2 and TNF-a assay
Nitrite accumulation, an indicator of NO synthesis, was meas- EtOAc extract > CHCl3 extract > BuOH extract (data are not
ured in the culture medium by the Griess reaction [17]. Briefly, shown). Silica gel column chromatography of the EtOAc extract
100 mL of cell culture medium were mixed with 100 mL of Griess led to the isolation of caffeic acid (1), liriodendrin (2), kalopa-
reagent [equal volumes of 1% (w/v) sulfanilamide in 5% (v/v) naxsaponin A (3) and hederoside D2 (4), which were identified
phosphoric acid and 0.1% (w/v) naphthylethylenediamine- on the basis of physical and spectral data. Compounds 1, 3, and
HCl] and incubated at room temperature for 10 min, and then 4 have not previously been isolated from the stem bark of A.
the absorbance at 550 nm was measured in a microplate reader. senticosus. Compounds 1±4 (5 and 10 mg/kg) were tested for
Fresh culture medium was used as the blank in all experiments. anti-inflammatory and antinociceptive activity in the bioassay
The amount of nitrite in the samples was calculated from a so- used. The activity at those assays was consistently shown as
Jung H-J et al. In vivo Anti-inflammatory¼ Planta Med 2003; 69: 610±616

Table 1 Inhibitory effect of A. senticosus MeOH extract and its active component, liriodendrin on dye leakage into the peritoneal cavity and tryp-
sin activity
Treatment
Dose
(mg/kg)
Route
No. of
mice
Amount of P.S.B.*
(mg/30 g b. wt)
Inhibition (%)
Trypsin activity
(IU/mL)
Inhibition (%)
Normal
p.o.
p.o.
p.o.
p.o.
p.o.
p.o.
p.o.
10
10
10
10
10
10
10
9.6  6.541,e
247.6  15.8a
187.6  9.26b
161.9  10.7c
139.5  6.47d
107.2  7.27e
106.6  4.53e
±
149
89
63
40
8
0.142  0.015e
0.347  0.030a
0.323  0.017a
0.274  0.012b
0.233  0.014c
0.192  0.020d
0.137  0.025e
±
Control
144
128
93
MeOH ext.
100
250
5
Liriodendrin
Aminopyrine
64
10
35
100
7
±4
* 4% pontamine sky blue. The assay procedure was described in the experimental methods.
1 Values are expressed mean  S.E.M. The number of animal used for each group was 10.
a,b.c.d.e Values sharing the same superscript letter are not significantly different (P < 0.05) each other by Duncan's multiple range test.
Table 2 Inhibitory effect of MeOH extract of A. senticosus and its active component, liriodendrin on carrageenan-induced edema of the hind paw
in rats
Treatment
Dose (mg/kg)
30
60
90
120 (min)
Paw Volume Inhibiton
Paw Volume Inhibiton
Paw Volume Inhibiton
Paw Volume Inhibiton
(ml)
(%)
(ml)
(%)
(ml)
(%)
(ml)
(%)
Control
±
20.8  2.541,a,b,
22.6  1.98a
21.7  2.67a
18.5  2.31b,c + 10
17.9  2.15c
17.6  1.54c
±
39.2  3.43a
36.7  2.24a,b
34.7  1.46b
30.7  2.52c
25.5  2.17d
14.8  1.25e
0
44.7  2.97a
39.1  2.67b
35.7  2.17c
26.6  1.49d
23.6  1.83e
10.3  1.20f
±
55.8  4.35a
47.8  3.28b
38.2  2.53c
34.2  3.11d
26.8  2.29e
12.9  1.37f
±
MeOH ext.
100
250
5
±10
+ 5
+ 10
+ 21
+ 33
+ 62
+ 13
+ 20
+ 40
+ 47
+ 78
+ 14
+ 32
+ 39
+ 52
+ 77
±5
Liriodendrin
10
+ 14
+ 14
Indomethacin 10
The assay procedure was described in the experimental methods.
1 Values are expressed mean  S.E.M. The number of animal used for each group was 10.
a,b,c,d,e,f Values sharing the same superscript letter are not significantly different (P < 0.05) each other by Duncan's multiple range test.
613
Table 3 Antinociceptive effect of A. senticosus MeOH extract and its active component, liriodendrin by acetic acid-induced writhing and hot-
plate method in mice
Treatment
Dose
(mg/kg)
Frequency
(Count/10 min)
Inhibition
(%)
Action time
(sec)
Increase
(%)
Control
47.0  2.011,d
35.0  2.51c
33.0  3.21c
31.6  2.22b,c
17.7  1.27b,c
15.3  1.00b
9.0  1.01a
±
±
25.6  2.18a
48.9  1.47b
53.7  2.36b
58.8  3.01b
60.3  3.39c
71.2  3.24c
±
±
MeOH ext.
100
250
500
5
91
24
30
33
62
65
±
110
130
136
178
558
±
Liriodendrin
10
Aminopyrine
Morphine
100
4
168.4  20.7d
81
The assay procedure was described in the experimental methods.
1 Values are expressed mean  S.E.M. The number of animal used for each group was 10.
a,b,c,d Values sharing the same superscript letter are not significantly different (P < 0 05) each other by Duncan's multiple range test.
follows: liriodendrin > caffeic acid > kalopanaxsaponin A > he- In vivo anti-inflammatory and antinociceptive effects
deroside D2. The activities of liriodendrin, caffeic acid, kalopa- The total dye amount which leaked into the peritoneal cavity
naxsaponin A were shown to be statistically significant, was 247  15.8 mg/30 g body weight in the vehicle control group.
whereas hederoside D2 was inactive.
Liriodendrin (5 and 10 mg/kg, p.o.) significantly reduced the dye
Jung H-J et al. In vivo Anti-inflammatory¼ Planta Med 2003; 69: 610±616

leakage as shown in Table 1. A standard drug, aminopyrine
(100 mg/kg, p.o.), strongly reduced the leakage. Moreover, lirio-
dendrin (5 and 10 mg/kg, p.o.) also exhibited a significant inhibi-
tory effect on carrageenan-induced edema throughout the ex-
perimental period as shown in Table 2. A standard drug, indome-
thacin (10 mg/kg, p.o.), showed higher inhibition than that of lir-
iodendrin. As shown in Table 1, the control value for carragee-
nan-induced trypsin activity was 0.347  0.03 IU/mL. Treatment
with either liriodendrin (5 and 10 mg/kg, p.o.) or aminopyrine
(100 mg/kg, p.o.) significantly suppressed trypsin activity.
The antinociceptive effect of test samples was assayed in two dif-
ferent models: the acetic acid-induced writhing test and hot-
plate test in mice. Liriodendrin showed high antinociceptive ac-
tivity by oral administration (5 and 10 mg/kg). It was found that
liriodendrin significantly increased the hot-plate reaction time
and inhibited the acetic acid-induced writhing response
(Table 3). In these experiments, aminopyrine was used as a posi-
tive control, and oral administration of aminopyrine (100 mg/kg)
exhibited a strong antinociceptive effect.
Inhibition by liriodendrin and syringaresinol of LPS-induced
NO2±, PGE2 and TNF-a production
To determine the effects of liriodendrin and syringaresinol on NO
production in RAW 264.7 cells, the cells were treated with LPS
(1 mg/mL) in the presence or absence of liriodendrin or syringar-
esinol for 24 h. 50 mM NMA was used as positive inhibitor (from
34.8  1.7 mM to 19.8  0.6 mM). Syringaresinol showed a higher
inhibitory effect on LPS-induced NO production (Fig. 2A) than lir-
iodendrin, and this effect was more evident at high concentra-
tion (120 mM). Additionally, both liriodendrin and syringaresinol
had no quenching effect on the Griess reagent at the concentra-
tion used. In addition, LPS-induced production of PGE2 and TNF-
a was also significantly inhibited by syringaresinol (Fig. 2B, 2C),
whereas liriodendrin showed mild activity. The cytotoxic effects
of liriodendrin and its hydrolysate, syringaresinol were evaluat-
ed in the absence or presence of LPS by the MTT assay. Lirioden-
drin and syringaresinol did not affect the cell viability at the con-
centrations used (30, 60, 120 mM) regardless of the presence of
LPS.
614
Effects of liriodendrin and syringaresinol on LPS-induced
iNOS and COX-2 expression
To find the relationship between the inhibitory effect of syringar-
esinol on these inflammatory mediators (NO and PGE2) and a
modulation of iNOS and COX-2 induction, we examined their ex-
pression levels by Western blot analysis. In response to LPS, the
expression level of iNOS was markedly induced, and syringaresi-
nol significantly inhibited the iNOS induction in a concentration-
dependent manner (Fig. 3). A similar pattern was observed when
Fig. 2 Effect of 2 and 2a on nitrite (A), PGE2 (B) and TNF-a (C) produc-
tion by LPS-induced RAW 264.7 macrophage for 24 h. The values are
means of 3 determinations  S.E.M. *p < 0.05, **p < 0.01, ***p <
0.001 with respect to the control LPS-treated group.
the effect of syringaresinol was examined on the LPS-induced drug. In this communication, we attempted to find the anti-in-
COX-2 expression level. These results are consistent with the flammatory and antinociceptive principles including liriodendrin,
±
profile of the inhibitory effect of syringaresinol on NO2 and and then compared the in vitro anti-inflammatory activities of lir-
PGE2 release (Fig. 4).
iodendrin and syringaresinol. We found that the EtOAc extract of
A. senticosus has significant anti-inflammatory and antinocicep-
tive effects in animals. The activity-guided fractionation led to
identification of active components such as caffeic acid, lirioden-
drin, kalopanaxsaponin A and hederoside D2. We have already re-
Discussion
A variety of biological activities of A. senticosus has been reported ported the antinociceptive and anti-inflammatory activity of kalo-
since this crude drug has been used as a tonic and anti-rheumatic panaxsaponin A isolated from Kalopanax pictus (Araliaceae) [18].
Jung H-J et al. In vivo Anti-inflammatory¼ Planta Med 2003; 69: 610±616

Fig. 3 Modulation by 2 and 2a of LPS-induced iNOS
expression in RAW 264.7 cells. Lysates were prepar-
ed from control or 24 h-LPS (1 mg/mL) stimulated
cells alone or in combination with increasing con-
centrations (30-60-120 mM) of liriodendrin and
syringaresinol. All lanes contained 50 mg of total
proteins. A representative immunoblot from three
separate experiments is shown.
Fig. 4 Modulation by 2 and 2a of LPS-induced COX-
2 expression in RAW 264.7 cells. Lysates were pre-
pared from control or 24 h-LPS (1 mg/mL) stimulated
cells alone or in combination with increasing con-
centrations (30-60-120 mM) of liriodendrin and
syringaresinol. All lanes contained 50 mg of total
proteins. A representative immunoblot from three
separate experiments is shown.
Liriodendrin was found to significantly increase hot-plate reac- by syringaresinol occurs mainly through the regulation of iNOS
tion time in mice. It is known that centrally acting analgesic and COX-2 gene expression. However, we cannot rule out the
drugs elevate the pain threshold of mice towards heat. In order possibility that syringaresinol directly interferes with iNOS and
to distinguish between the central and peripheral analgesic ac- PGE2 activity. Moreover, the development of hyperalgesic states
tion, the acetic acid-induced writhing response in mice was during inflammation is thought to be mediated by pro-inflam-
used to examine the effect. It was found that liriodendrin signifi- matory cytokines such as TNF-a, IL-1 and IL-6 [20]. Therefore,
cantly inhibited the acetic acid-induced writhing response in a we investigated whether syringaresinol and liriodendrin could
dose-dependent manner. The abdominal constriction is related possibly influence the formation of TNF-a in an in vitro model.
to the sensitization of nociceptive receptors to prostaglandins. It We demonstrated that liriodendrin and syringaresinol signifi-
is therefore possible that liriodendrin exerts an analgesic effect cantly inhibit TNF-a production in a concentration-dependent
probably by inhibiting the synthesis or action of prostaglandins.
manner and syringaresinol has higher activity than liriodendrin.
615
In many cases, it appears that the active moiety in many glyco- Based on these results, we concluded that oral administration of
sides existing in plant resources should be attributed to the agly- liriodendrin results in a potent anti-inflammatory and antinoci-
cone. This suggestion was supported by our previous results in- ceptive activity. In addition, the true active moiety of lirioden-
dicating that liriodendrin is transformed to the more cytotoxic drin could be syringaresinol producible from the glycoside in
syringaresinol by human intestinal bacteria [19]. Syringaresinol the living system rather than the gross structure.
is much less frequently found in Araliaceae plants than its glyco-
sides, suggesting that the former is more unstable than the latter.
This is presumed from the production of a variety of artifacts on References
acid hydrolysis of liriodendrin.
1
Shao CJ, Kasai R, Xu J, Tanaka O. Saponins from leaves of Acanthopanax
senticosus Harms., Ciwujia: Structures of ciwujianosides B, C1, C2, C3,
C4, D1, D2 and E. Chem Pharm Bull 1988; 36: 601±8
It is well known that macrophages play a crucial role in both
non-specific and acquired immune responses, and macrophage
activation by LPS leads to a functionally diverse series of respon-
ses including the production of pro-inflammatory cytokines (IL-1,
TNF-a and IL-6), the activation of phospholipase A2 which pro-
duces lipid metabolites of arachidonic acid such as PGs, and NO
production. In this study, we demonstrated that LPS-induced NO
production was inhibited by liriodendrin and syringaresinol in a
concentration-dependent manner without notable cytotoxicity.
The inhibitory effect of syringaresinol was higher than that of lir-
iodendrin. Results from Western blot analysis further indicated
that LPS-induced iNOS expression in RAW 264.7 macrophages
was significantly blocked by syringaresinol. In addition to inhibi-
tion of NO release and iNOS induction, syringaresinol also poten-
tially inhibited PGE2 production and COX-2 gene expression in
LPS-treated RAW 264.7 macrophages. Therefore, these results
suggest that inhibition of LPS-induced NO and PGE2 production
2 Shao CJ, Kasai R, Xu JD, Tanaka O. Saponins from leaves of
Acanthopanax senticosus Harms., Ciwujia: II. Structures of ciwujiano-
sides A1, A2, A3, A4 and D3. Chem Pharm Bull 1989; 37: 42±5
3 Nishibe S, Kinoshita H, Takeda H, Okano G. Phenolic compounds from
stem bark of Acanthopanax senticosus and their pharmaceutical effect
in chronic swimming stressed rats. Chem Pharm Bull 1990; 38: 1763±5
4 Fujikawa T, Yamaguchi A, Morita I, Takeda H, Nishibe S. Protective ef-
fects of Acanthopanax senticosus Harms from Hokkaido and its com-
ponents on gastric ulcer in restrained cold water stressed rats. Biol
Pharm Bull 1996; 19: 1227±30
5 Fang JN, Proksch A, Wagner H. Immunologically active polysacchar-
ides of Acanthopanax senticosus. Phytochemistry 1985; 24: 2619±22
6 Hibasami H, Fujikawa T, Takeda H, Nishibe S, Satoh T, Fujisawa T, Na-
kashima K. Induction of apoptosis by Acanthopanax senticosus Harms
and its component, sesamin in human stomach cancer KATO III cells.
Oncol Rep 2000; 7: 1213±6
7 Tsukamoto H, Hisada S, Nishibe S. Coumarins from Fraxunus japonicus
and F. mandshurica var. japonica, Chem Pharm Bull 1985; 33: 4069±
73
Jung H-J et al. In vivo Anti-inflammatory¼ Planta Med 2003; 69: 610±616

15
16
8 Deyama T, Ikawa T, Nishibe S. The constituents of Eucommia ulmoides
Oliv. II. Isolation and structures of three new lignan glycosides. Chem
Pharm Bull 1985; 33: 3651±7
Eddy NB, Leimback D. Synthetic analgesics. II. Dithienbutenyl-and di-
thienbutenylamines. J Pharmacol Exp Ther 1953; 107: 385±93
Carvalho JC, Silva MF, Maciel MA, Pinto AC, Nunes DS, Lima RM, Bastos
JK, Sarti SJ. Investigation of anti-inflammatory and antinociceptive ac-
tivities of trans-dehydrocrotonin, a 19-nor-clerodane diterpene from
Croton cajucara. Part 1. Planta Medica 1996; 62: 402±4
Green LC, Wagner DA, Glogowsky J, Skipper PL, Wishnok JS, Tannen-
baum SR. Analysis of nitrate, nitrite, and (15N)nitrate in biological
fluids. Anal Biochem 1982; 126: 131±8
Choi JW, Huh K, Kim SH, Lee KT, Park HJ, Han YN. Antinociceptive and
anti-rheumatoidal effects of Kalopanax pictus extract and its saponin
components in experimental animals. J Ethnopharmacol 2002; 79:
199±204
Kim DH, Lee KT, Bae EA, Han MJ, Park HJ. Metabolism of liriodendrin
and syringin by human intestinal bacteria and their relation to in
vitro cytotoxicity. Arch Pharm Res 1999; 22: 30±4
9 Park HJ, Kwon SH, Lee JH, Lee KH, Miyamoto K, Lee KT. Kalopanaxsapo-
nin A is a basic saponin structure for the anti-tumor activity of heder-
agenin monodesmosides. Planta Medica 2001; 67: 118±21
10
17
18
Higuchi R, Miyahara K, Kawasaki T. Seed saponins of Akebia quinata
Decne I. Hederagenin 3-O-glycosides. Chem Pharm Bull 1972; 20:
1935±9
Whittle BA. The use of change in capillary permeability to distinguish
between narcotic and anagesic. Br J Pharmacol 1949; 22: 246±460
Matsuda H, Samukawa K, Kubo M. Anti-inflammatory activity of gin-
senoside Ro. Planta Medica 1990; 56: 19±23
Bieth J, Metais P, Warter J. Activation, inhibition and protection of
tryptic and a-chymotryptic activity by normal human serum. Clin Chi-
mica Acta 1969; 20: 69±80
11
12
13
19
14
Fritz H, Trautschold I, Werle E. Determination of the molecular weight
of new trypsin inhibitors with the aid of sephadex gel filtration. Hoppe
Seyler Z Physiol Chem 1965; 342: 253±63
20 Watkins LR, Maier SF, Goehler LE. Immune activation: the role of
proinflammatory cytokines in inflammation, illness response and
pathological pain states. Pain 1995; 63: 289±302
616
Jung H-J et al. In vivo Anti-inflammatory¼ Planta Med 2003; 69: 610±616