Anti-inflammatory effect of caffeic acid methyl ester and its mode of action through the inhibition of prostaglandin E2, nitric oxide and tumor necrosis factor-alpha production.

Article abstract of DOI:10.1016/j.bcp.2004.08.002

The anti-inflammatory effects of caffeic acid (CA), caffeic acid methyl ester (CM) and di-O-acetylcaffeic acid (DAC) were investigated in rats using the carrageenin-induced edema model and the antinociceptive effects of these compounds were also assessed in mice by means of the acetic acid-induced abdominal constriction test and hot plate test. CM (10mg/kg, p.o.) showed the most potent anti-inflammatory and antinociceptive effects in these animal models. To investigate the mechanism of the anti-inflammatory action, we examined the effects of these compounds on the lipopolysaccharide (LPS)-induced NO and PGE2 responses in the murine macrophage cell line, RAW 264.7. Our data indicate that CM is the most potent inhibitor of NO and PGE2 production and it also significantly decreased tumor necrosis factor-alpha (TNF-alpha) release. Consistent with these observations, the protein and mRNA expression levels of iNOS and COX-2 were found to be inhibited by CM in a dose-dependent manner. Furthermore, CM inhibited the nuclear factor-kappaB (NF-kappaB) activation induced by LPS, which was associated with the prevention of the degradation of the inhibitor kappaB, and subsequently with decreased p65 protein levels in the nucleus. Taken together, our data indicate that the anti-inflammatory properties of CM might result from the inhibition of iNOS, COX-2 and TNF-alpha expression through the down-regulation of NF-kappaB binding activity.

Full text of DOI:10.1016/j.bcp.2004.08.002

Biochemical Pharmacology 68 (2004) 2327–2336
www.elsevier.com/locate/biochempharm
Anti-inflammatory effect of caffeic acid methyl ester and its
mode of action through the inhibition of prostaglandin E2,
nitric oxide and tumor necrosis factor-a production
a
Kyung-Min Shin , In-Tae Kim , Young-Mi Park , Joohun Ha , Jong-Won Choi ,
a
a
b
c
d
Hee-Juhn Park , Yong Sup Lee , Kyung-Tae Lee
e
a,
*
a
College of Pharmacy, Kyung-Hee University, Dongdaemun-ku, Hoegi-Dong, Seoul 130-701, South Korea
b
College of Medicine, Kyung-Hee University, Dongdaemun-ku, Hoegi-Dong, Seoul 130-701, South Korea
c
College of Pharmacy, Kyung-Sung University, Daeyeon-Dong, Pusan 608-736, South Korea
d
Division of Applied Plant Sciences, Sang-Ji University, Woosan-Dong, Wonju 220-702, South Korea
e
Life Sciences Division, Korea Institute of Science and Technology, P.O. Box 131, Cheongryang, Seoul 130-650, South Korea
Received 18 June 2004; accepted 4 August 2004
Abstract
The anti-inflammatory effects of caffeic acid (CA), caffeic acid methyl ester (CM) and di-O-acetylcaffeic acid (DAC) were investigated
in rats using the carrageenin-induced edema model and the antinociceptive effects of these compounds were also assessed in mice by
means of the acetic acid-induced abdominal constriction test and hot plate test. CM (10 mg/kg, p.o.) showed the most potent anti-
inflammatory and antinociceptive effects in these animal models. To investigate the mechanism of the anti-inflammatory action, we
examined the effects of these compounds on the lipopolysaccaride (LPS)-induced NO and PGE responses in the murine macrophage cell
2
line, RAW 264.7. Our data indicate that CM is the most potent inhibitor of NO and PGE production and it also significantly decreased
2
tumor necrosis factor-alpha (TNF-a) release. Consistent with these observations, the protein and mRNA expression levels of iNOS and
COX-2 were found to be inhibited by CM in a dose-dependent manner. Furthermore, CM inhibited the nuclear factor-kB (NF-kB)
activation induced by LPS, which was associated with the prevention of the degradation of the inhibitor kB, and subsequently with
decreased p65 protein levels in the nucleus. Taken together, our data indicate that the anti-inflammatory properties of CM might result
from the inhibition of iNOS, COX-2 and TNF-a expression through the down-regulation of NF-kB binding activity.
#
2004 Elsevier Inc. All rights reserved.
2
Keywords: Caffeic acid methyl ester; Anti-inflammation; Nitric oxide; Prostaglandin E ; TNF-a; NF-kB
1
. Introduction
respectively, are known to be involved in the pathogenesis
of many chronic diseases including multiple sclerosis,
Parkinson’s and Alzheimer’s diseases, and colon cancer
[1]. NO is produced by iNOS in macrophages, hepatocytes
and renal cells, under the stimulation of lipopolysaccharide
(LPS), tumor necrosis factor-alpha (TNF-a), interleukin-1
or interferon-g [2]. It can also react with superoxide anion
radicals to form the even stronger oxidant, peroxinitrite
Chronic inflammations and infections lead to the up-
regulation of a series of enzymes and signaling proteins in
affected tissues and cells. Among these pro-inflammatory
enzymes, the inducible forms of nitric oxide synthase
(NOS) and cyclooxygenase (COX), which are responsible
for increasing the levels of NO and prostaglandins (PGs),
(
ONOO) [3,4]. Many studies have shown that the chronic
Abbreviations: LPS, lipopolysacchride; EMSA, electrophoretic mobi-
lity shift assay; iNOS, inducible nitric oxide synthase; FBS, fetal bovine
serum; TNF, tumor necrosis factor; NO, nitric oxide; MTT, 3-(4,5-
dimethylthiazoyl-2-yl) 2,5 diphenyl tetrazolium bromide; ECL, enhanced
chemical luminoscence
phase of inflammation is closely associated with an
increase in iNOS activity [5]. The most conclusive evi-
dence for NO acting as a mediator of tissue injury has been
obtained in the case of arthritis, which is based on studies
in an animal model, human osteoarthritis and rheumatoid
arthritis [6]. However, the constitutive epithelial and
*
Corresponding author. Tel.: +82 2 9610860; fax: +82 2 9663885.
E-mail address: ktlee@khu.ac.kr (K.-T. Lee).
0006-2952/$ – see front matter # 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.bcp.2004.08.002

2328
K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
neuronal forms of NOS contribute relatively little to in-
flammation and carcinogenesis. COX is the enzyme that
converts arachidonic acid to PGs. Like NOS, COX has
been found to exist in two isoforms and one of these, COX-
effects of caffeic acid derivatives on the lipopolysaccharide
(LPS)-induced NO, PGE and TNF-a release via NF-kB
binding were also investigated, in order to gain a better
insight into the mechanism of the observed anti-inflam-
matory action.
2
2
, is an inducible form which is responsible for the pro-
duction of large amounts of pro-inflammatory PGs at the
inflammatory site [7]. Furthermore, TNF-a is one of the
most important pro-inflammatory cytokines and is mainly
produced by monocytes and macrophages [8]. It is secreted
during the early phase of acute and chronic inflammatory
diseases such as asthma, rheumatoid arthritis, septic shock
and other allergic diseases, as well as the activation of T
cells [8].
2. Materials and methods
2.1. Materials
Caffeic acid methyl ester and di-O-acetylcaffeic acid
were prepared from caffeic acid as follows: caffeic acid
methyl ester (CM): a solution of caffeic acid (2 g,
11.1 mmol) in methanol (50 mL) was treated with a cat-
alytic amount of c-H SO and heated at reflux for 10 h. The
The expression of these pro-inflammatory enzymes and
cytokines is regulated at multiple levels, including during
the transcriptional, post-transcriptional, translational and
post-translational steps [9,10]. One of the most ubiquitous
transcription factors that regulate the gene expressions
involved in cellular proliferation, inflammatory responses
and cell adhesion is NF-kB. The functionally active NF-kB
exists mainly as a heterodimer, consisting of subunits of the
Rel family, which is normally sequestered in the cytosol as
an inactive complex, by binding to IkBs in unstimulated
cells [11]. The activation mechanism of NF-kB involves
the phosphorylation of IkBs in two critical serine residues
2
4
reaction mixture was cooled at room temperature and
concentrated. The residue was dissolved in EtOAc and
washed successively with water and brine. The ethylace-
tate (EtOAc) layer was dried over anhydrous MgSO and
4
purified by flash column chromatography on silica gel
(EtOAc:n-hexane = 1:2) to give caffeic acid methyl ester
1
(1.53 g, 71%) as a white solid. H NMR (500 MHz,
pyridin-d ) d: 3.71 (3H, s, COOCH ), 6.56 (1H, d, J =
5
3
15.9 Hz, H-8), 7.16 (1H, dd, J = 1.9 and 8.1 Hz, H-6), 7.19
(1H, d, J = 8.1 Hz, H-5), 7.57 (1H, d, J = 1.9 Hz, H-2), 7.92
3
2
36
(
Ser and Ser ) via the IkB kinase (IKK) signalosome
13
complex. Once the IkBs are phosphorylated, they are
targeted for ubiquitination and subsequent degradation
by 26S proteosome [12–14]. The resulting free NF-kB
is translocated to the nucleus, where it binds to the kB
binding sites in the promoter regions of the target genes,
thereby controlling their expression [15].
(1H, d, J = 15.9 Hz, H-7); C NMR (125.5 MHz, pyridine-
d ) d: 51.2 (COOCH ), 114.6 (C-8), 115.7 (C-2), 116.7 (C-
5
3
5), 122.0 (C-6), 126.8 (C-1), 146.0 (C-7), 147.7 (C-3),
150.5 (C-4), 167.8 (C-9). Di-O-acetylcaffeic acid (DAC): a
solution of caffeic acid (2 g, 11.10 mmol) in pyridine
(5.0 mL) was treated with acetic anhydride (3 mL) and
stirred overnight. The reaction mixture was poured into
ice-containing water and the resulting solid was filtered
and washed with water. The solid was further purified by
flash column chromatography on silica gel (CH Cl :
Caffeic acid and its derivatives are nonflavonoid cate-
cholic compounds, which are distributed in many medic-
inal plants [16]. It has been reported that caffeic acid has
anti-inflammatory, anti-mutagenic and anti-carcinogenic
activities [17–20], and some of its derivatives show anti-
oxidant and anti-inflammatory activities [21–23].
Although the anti-inflammatory activities of caffeic acid
derivatives have already been studied, their mechanisms of
action are not fully understood. In this study, we investi-
gated the anti-inflammatory and antinociceptive activities
of caffeic acid, caffeic acid methyl ester and di-O-acet-
ylcaffeic acid in animal models (Fig. 1). Furthermore, the
2
2
MeOH = 10:1) to give di-O-acetylcaffeic acid (1.23 g,
1
42%) as a white solid. H NMR (500 MHz, pyridine-d )
5
d: 2.24, 2.27 (each 3H, s, acetyl), 6.84 (1H, d, J = 16.0 Hz,
H-8), 7.35 (1H, d, J = 8.3 Hz, H-5), 7.47 (1H, dd, J = 1.9
and 8.3 Hz), 7.67 (1H, d, J = 1.9 Hz, H-2), 7.93 (1H, d, J =
13
16.0 Hz, H-7); C NMR (125.5 MHz, pyridine-d ) d:
5
20.79 (CH CO) and 20.81 (CH CO), 122.4 (C-8), 123.8
3
3
(C-2), 124.9 (C-5), 127.1 (C-6), 134.6 (C-1), 142.7 (C-7),
43.8 ( C-3), 144.6 (C-4), 168.7, 168.8 (CH CO), 169.3
1
3
(
C-9). The RAW 264.7 mouse macrophage-like cell line
was obtained from the Korean Cell Line Bank. DMEM,
fetal bovine serum (FBS), penicillin and streptomycin were
purchased from Gibco Life Technologies. COX-2, iNOS,
p65 and IkB-a monoclonal antibodies and the peroxidase-
conjugated secondary antibody were purchased from Santa
Cruz Biotechnology. The enzyme immunoassay (EIA) kit
used for the determination of prostaglandin E and TNF-a
2
was obtained from R&D Systems. NS-398, a COX-2
enzyme inhibitor was obtained from Calbiochem.
The RNA extraction kit was purchased from Intron
Fig. 1. Chemical structures of caffeic acid, caffeic acid methyl ester and di-
O-acetylcaffeic acid.

K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
2329
Biotechnology. The iNOS, COX-2, TNF-a and b-actin
oligonucleotide primers were purchased from Bioneer. 3-
of the hind paws or the jumping movements before and
after drug administration. The cut-off time was 10 s and
10.0 mg/kg of morphine sulphate (Kuju Pharmaceutical,
CO), administered intraperitoneally, was used as the refer-
ence drug [28].
(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-tertazolium bro-
mide (MTT), sulfanilamide, aprotinin, leupeptin, phenyl-
methylsulfonylfluoride (PMSF), dithiothreitol (DTT), L-
6
N -(1-iminoethyl) lysine (L-NIL), Escherichia coli LPS
and all other chemicals were purchased from Sigma.
2.5. Cell culture and sample treatment
2
.2. Animals
The RAW 264.7 cells were cultured in DMEM contain-
ing 10% heat-inactivated FBS, penicillin (100 units/mL)
and streptomycin sulfate (100 mg/mL), and maintained in a
ICR male mice weighing 20–25 g and Sprague–Dawley
male rats weighing 100–120 g were purchased from the
Daehan Biolink and maintained under constant conditions
humidified atmosphere of 5% CO . The cells were incu-
2
bated with the compounds being tested at various concen-
trations or with the positive chemicals and stimulated with
LPS (1 mg/mL) for the period indicated.
(
temperature: 20 ꢀ 2 8C, humidity: 40–60%, 12 h light/
dark cycle) and acclimatized for 1 week or more. For 24 h
period before the experiment, only water was offered to the
animals. To minimize the effect of their daily enzyme
activity variations, the animals were all sacrificed at 10:00–
2.6. MTT assay for cell viability
1
2:00 a.m. All animal experiments were approved by the
RAW 264.7 viability after 24 h of continuous exposure
to the tested compounds were measured with a colorimetric
assay based on the ability of mitochondria in viable cells to
reduce MTT as described previously [21].
University of Kyungsung Animal Care and Use Commit-
tee, and all procedures were conducted in accordance with
the ‘‘Guide for the Care and Use of Laboratory Animals’’
published by the Korea National Institute of Health.
2.7. Nitrite determination
2
.3. Carrageenin-induced paw edema
The nitrite, which accumulated in the culture medium,
The initial hind paw volume of the Sprague–Dawley
was measured as an indicator of NO production by means
of the Griess reaction. Briefly, 100 mL of cell culture
medium (without phenol red) was mixed with an equal
volume of Griess reagent (equal volumes of 1% (w/v)
sulfanilamide in 5% (v/v) phosphoric acid and 0.1% (w/v)
naphthylethylenediamine–HCl), incubated at room tem-
perature for 10 min, and then the absorbance was measured
at 550 nm using a microplate reader. Fresh culture medium
was used as the blank in all experiments. The amount of
nitrite in the samples was obtained by means of the NaNO2
serial dilution standard curve and the nitrite production was
measured.
strain rats was determined volumetrically. A 1% solution of
carrageenin in saline (0.1 mL/rat) was injected subcuta-
neously into the right hind paw 1 h after the test substance
had been administered orally. The test samples were dis-
solved in 10% Tween 80 and diluted with saline. The same
volume of pure solvent was used for the control group. The
test solution (10 mg/kg) was administered orally on the
seven consecutive days prior to the injection of the carra-
geenin. The paw volumes were measured upto 5 h after the
injection at intervals of 60 min and the volume of the
edema was measured with a plethysmometer [24]. Ibupro-
fen, an anti-inflammatory drug, was used as a standard [25].
2.8. PGE and TNF-a assay
2
2.4. Acetic acid-induced abdominal constriction
and hot plate method
The PGE and TNF-a levels in the macrophage culture
2
medium were quantified using EIA kits according to the
manufacturer’s instructions.
The acetic acid-induced abdominal constriction test was
performed as described by Whittle [26]. The vehicle,
aminopyrine (100 mg/kg) and test solution (10 mg/kg)
were administered orally 30 min before the experiment
and 0.1 mL/10 g of 0.7% acetic acid–saline was then
injected i.p. 10 min after the injection. Subsequently, the
frequency of abdominal constriction in the mice was
counted for the next 10 min. The hot plate test was used
to measure the response latencies according to the method
described previously by Eddy and Leimback [27], with
minor modifications. In these experiments, the hot plate
2.9. Western blot analysis
Cellular proteins were extracted from both the control
and CM-treated RAW 264.7 cells. The cells were collected
by centrifugation and washed once with phosphate-buf-
fered saline (PBS). The washed cell pellets were resus-
pended in extraction lysis buffer (50 mM HEPES, pH 7.0,
250 mM NaCl, 5 mM EDTA, 0.1% Nonidet P-40, 1 mM
PMSF, 0.5 mM DTT, 5 mM NaF, 0.5 mM Na VO ), con-
3
4
(
Ugo Basile, model-DS 37) was maintained at 56 ꢀ 1 8C.
taining 5 mg/mL each of leupeptin and aprotinin, and
incubated for 30 min at 4 8C. Fifty micrograms of cellular
The reaction time was noted by observing either the licking

2330
K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
protein from both the treated and untreated cell extracts
was electroblotted onto a nitrocellulose membrane follow-
ing separation using 8–12% SDS-polyacrylamide gel elec-
trophoresis. The immunoblots were blocked overnight in
Tween 20/Tris-buffered saline (TTBS) containing 5% (w/
v) non-fat milk at 4 8C, followed by incubation for 4 h with
a 1:1000 dilution of monoclonal anti-iNOS, COX-2, NF-
kB (p65), IkB-a and b-actin antibodies. The blots were
washed twice with TTBS and incubated with a 1:1000
dilution of horseradish peroxidase-conjugated goat anti-
mouse IgG secondary antibody for 1 h at room tempera-
ture. The blots were washed again three times with TTBS
and then developed by enhanced chemiluminescence
were prepared according to a previously described proce-
dure with a slight modification [28]. The cell pellet was
resuspended in hypotonic buffer (10 mM HEPES, pH 7.9,
1.5 mM MgCl , 10 mM KCl, 0.2 mM PMSF, 0.5 mM
2
DTT, 10 mg/mL aprotinin) and incubated on ice for
15 min. Then the cells were lysed by the addition of
0.1% Nonidet P-40 and vigorous vortexing for 10 s. The
nuclei were pelleted by centrifuging them at 12,000 ꢁ g for
1 min at 4 8C and resuspended in high-salt buffer (20 mM
HEPES, pH 7.9, 25% glycerol, 400 mM KCl, 1.5 mM
MgCl , 0.2 mM EDTA, 0.5 mM DTT, 1 mM NaF, 1 mM
2
Na VO ). A 10 mg sample of the nuclear extract was mixed
3
4
0
with the double-stranded NF-kB oligonucleotide (5 -
0
(ECL).
AGTTGAGGGGACTTTCCCAGGC-3 ) end-labeled by
ꢂ32
[
g
P] dATP (underlining indicates a kB consensus
2
.10. RNA preparation and polymerase chain reaction
sequence or a binding site for the NF-kB/cRel homodi-
meric and heterodimeric complexes). The binding reac-
tions were performed at 37 8C for 30 min in 30 mL of
reaction buffer at pH 7.5 containing 10 mM Tris–HCl,
100 mM NaCl, 1 mM EDTA, 4% glycerol, 1 mg of
poly(dI-dC) and 1 mM DTT. The specificity of binding
was examined by competition with the 80-fold unlabeled
oligonucleotide. DNA–protein complexes were separated
from the unbound DNA probe on native 5% polyacryla-
mide gels at 100 V in 0.5ꢁ TBE buffer. The gels were
vacuum-dried for 1 h at 80 8C and exposed to X-ray film at
ꢂ70 8C for 24 h.
1
Total cellular RNAwas isolated using an Easy Blue kit
according to the manufacturer’s instructions. From each
sample, 1 mg of RNA was reverse-transcribed (RT) using
MuLV reverse transcriptase, 1 mM dNTP and 0.5 mg/mL
of oligo (dT12–18). Then, PCR analysis was performed on
aliquots of the cDNA preparations, in order to detect iNOS,
COX-2, TNF-a and b-actin (as an internal standard) gene
expression using a thermal cycler. The reactions were
carried out in a volume of 25 mL containing 1 unit of
Taq DNA polymerase, 0.2 mM dNTP, 10ꢁ reaction buffer
0
0
and 100 pmol of the 5 and 3 primers (final concentration).
After initial denaturation for 2 min at 95 8C, 30 amplifica-
tion cycles were performed for iNOS, COX-2 and TNF-a
2.12. Statistical analysis
(
1 min of 95 8C denaturation, 1 min of 55 8C annealing and
min of 72 8C extension). The PCR primers used in this
Statistical analysis was performed using one-way ana-
lysis of variance (ANOVA) followed by Duncan’s test for
multiple comparisons, and the Student’s t-test for single
comparisons. The data are reported as means ꢀ S.E.M.
The numbers of independent experiments assessed are
given in the figure legends.
1
study are listed below and were purchased from Bioneer;
0
sense strand iNOS primer, 5 -AATGGCAACATCAGG-
0
TCGGCCATCACT-3 ; anti-sense strand iNOS primer,
0
0
5
strand COX-2 primer, 5 -GGAGAGACTATCAAGATA-
-GCTGTGTGTCACAGAAGTCTCGAACTC-3 ; sense
0
0
0
GT-3 ; anti-sense strand COX-2 primer, 5 -ATGGTCAG-
0
0
TAGA CTTTTACA-3 ; sense strand TNF-a primer, 5 -
0
3. Results
ATGAGCACAGAAAGCATGATC-3 ; anti-sense strand
0
0
TNF-a primer, 5 -TACAGGCTTGTCACTCGAATT-3 ;
0
3.1. In vivo anti-inflammatory and antinociceptive
activity of caffeic acid, caffeic acid methyl ester and
di-O-acetyl caffeic acid
sense strand b-actin primer, 5 -TCATGAAGTGTGAC-
0
GTTGACATCCGT-3 ; anti-sense strand b-actin primer,
0
0
5
-CCTAGAAGCATTTGCGGTGCACGATG-3 .
After
amplification, portions of the PCR reactions were electro-
phoresed on 2% agarose gel and visualized by ethidium
bromide staining and UV irradiation.
We examined the anti-inflammatory effect of CA, CM
and DAC using the carrageenin-induced edema model and
maximal edema inhibition was observed at 3 h after the
edema induction. In particular, treatment with CM (10 mg/
kg, p.o.) reduced the edema rate by 23.6% at 3 h and
ibuprofen (100 mg/kg, p.o.) treatment decreased the edema
rate by 42.2% at 3 h (Table 1). This reduction remained
statistically significant for more than 5 h after the edema
induction. In addition, the antinociceptive effect of the test
samples was assayed using two different models, i.e., by
the acetic acid-induced abdominal constriction test and the
hot plate test in mice. In the case of the former test, CM and
2
.11. Preparation of nuclear extraction and EMSA
RAW 264.7 macrophages were plated in 100-mm dishes
6
(
5 ꢁ 10 cells). The cells were treated with various CM
concentrations (25, 50 and 100 mM) or CM 100 mM for the
periods indicated, stimulated with LPS for the periods
indicated, washed twice with PBS, scraped into 1 mL of
cold PBS and pelleted by centrifugation. Nuclear extracts

K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
2331
Table 1
Effect of caffeic acid, caffeic acid methyl ester and di-O-acetylcaffeic acid on carrageenan-induced paw edema in rat
Treatment
Dose (mg/kg)
Swelling volume (mL)
1
h
2 h
3 h
4 h
5 h
m
m
m
m
n
d,e,f,g
e,f,g
d,e,f
f,g
a
b,c
c,d,e
b,c
g
h
Control
CA
1.27 ꢀ 0.02
1.25 ꢀ 0.03
1.22 ꢀ 0.02
1.25 ꢀ 0.20
0.86 ꢀ 0.02
2.18 ꢀ 0.02
2.16 ꢀ 0.01
2.18 ꢀ 0.02
2.11 ꢀ 0.02
1.47 ꢀ 0.03
3.01 ꢀ 0.02
2.65 ꢀ 0.01
2.77 ꢀ 0.02
2.30 ꢀ 0.02
1.74 ꢀ 0.02
2.52 ꢀ 0.02
2.33 ꢀ 0.02
2.50 ꢀ 0.02
2.08 ꢀ 0.02
1.59 ꢀ 0.05
1.86 ꢀ 0.02
1.91 ꢀ 0.01
1.91 ꢀ 0.02
1.57 ꢀ 0.03
1.20 ꢀ 0.01
b
h
10
a,b
d,e,f
h,I,j
h
DAC
10
j,k
m
CM
10
k,l
i,j,k
Ibuprofen
100
Samples were orally administered for 1 week after carrageenin-induced paw edema. The rats were tested 24 h for last treated materials; values represent means
ꢀ
S.E.M. (n = 6); values sharing the same superscript letter are not significantly different from each other (p < 0.05) by Duncan’s multiple range test.
CA were found to significantly reduce the number of
abdominal constrictions induced by 0.7% acetic acid solu-
tion. The percentages of protection afforded by CM and
CA after their oral administration were 67.8 and 55.6%,
respectively (Table 2). Although this test constitutes a non-
specific model (e.g., anti-cholinergic, anti-histaminic and
other agents also show activity in this test), it is widely used
for analgesic screening and involves local peritoneal recep-
tors (cholinergic and histamine receptors) and the media-
tors of acetylcholine and histamine. The results of the hot
plate test in mice (Table 2) shows that CM and CA at a dose
of 10 mg/kg significantly increased the latency of the
jumping response without affecting the animal’s ability
to detect pain. Morphine (10 mg/kg, i.p.) was used as a
positive control in the hot plate test. Among the tested
compounds, caffeic acid methyl ester showed the most
significant anti-inflammatory activity in the carrageenin-
induced paw edema assay and it also exhibited the most
significant antinociceptive activities in the abdominal con-
striction and hot plate assays. The potencies of these three
compounds were found to be in the following order: CM >
CA > DAC.
with IC50 values of 330, 45.8 and 294.5 mM, respectively
(Fig. 2A). L-NIL (10 mM) was used as a positive inhibitor.
To examine whether the tested compounds could inhibit
PGE production in the same manipulation, the cells were
2
pre-incubated with the CA, CM and DAC for 1 h and then
activated with 1 mg/mL LPS for 24 h. As shown in Fig. 2B,
CM significantly inhibited the production of PGE in a
2
dose-dependent manner. The cytotoxic effects of CA, CM
and DAC were evaluated in the presence or absence of LPS
using the MTT assay and these compounds did not affect
the cell viability of RAW 264.7 cells in either the presence
or absence of LPS even at a dose of 400 mM after a period
of 24 h (data not shown).
3.3. Inhibition of LPS-induced TNF-a production
by CM
Since CM was the most potent inhibitor of the pro-
inflammatory mediators among the tested compounds, we
further investigated the effect of CM on the LPS-induced
TNF-a release using an enzyme immunoassay and RT-
PCR. Pre-treatment of the cells with CM for 1 h decreased
both the TNF-a production and mRNA expression in a
dose-dependent manner (Fig. 3A and B).
3.2. Effect of caffeic acid, caffeic acid methyl ester and
di-O-acetyl caffeic acid on NO and PGE production
2
3.4. Effects of CM on LPS-induced iNOS and COX-2
protein and mRNA expressions
To assess the effects of CA, CM and DAC on the LPS-
induced NO production in RAW 264.7 cells, cell culture
medium was harvested and the production of nitrite was
measured using the Griess reaction. CA, CM and DAC
reduced the NO production in a dose-dependent manner
Western blot and RT-PCR analyses were performed to
determine whether the inhibitory effect of CM on these
pro-inflammatory mediators (NO and PGE ) was related to
2
Table 2
Antinociceptive effect of caffeic acid, caffeic acid methyl ester and di-O-acetylcaffeic acid by acetic acid-induced abdominal constriction and hot plate method
in mice
Treatment
Dose (mg/kg)
Frequency (count/10 min)
Inhibition (%)
Action time (s)
Prolongation (%)
a
d
Control
CA
40.91 ꢀ 1.43
0
26.70 ꢀ 0.72
0
b
c
10
10
18.13 ꢀ 0.46
55.6
4.1
62.95 ꢀ 1.09
140
a
d
DAC
38.95 ꢀ 1.03
31.35 ꢀ 1.32
24
b,c
b
CM
10
13.23 ꢀ 0.73
67.8
82.79 ꢀ 2.26
210.5
495.9
a
Morphine
Aminopyrine
10
159.22 ꢀ 2.87
c
100
9.06 ꢀ 0.45
78
Values represent means ꢀ S.E.M. (n = 10); values sharing the same superscript letter are not significantly different each other (p < 0.05) by Duncan’s multiple
range test.

2332
K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
Fig. 3. The effect of caffeic acid methyl ester on LPS-induced TNF-a
release in RAW 264.7 cells. (A) The cells were pretreated for 1 h with
different concentrations (25, 50 and 100 mM) of caffeic acid methyl ester
(
CM), and then LPS (1 mg/mL) was added and the cells were incubated for
4 h. Control (Con) values were obtained in the absence of LPS or caffeic
2
acid methyl ester (CM). (B) Total RNA was prepared for the RT-PCR
analysis of TNF-a gene expression from RAW 264.7 macrophages that
were stimulated with LPS (1 mg/mL) in the absence or presence of various
concentrations (25, 50 and 100 mM) of CM for 4 h. The TNF-a-specific
sequences (351 bp) were detected by staining the gel (2% agarose) with
ethidium bromide. PCR of b-actin was performed to verify that the initial
cDNA contents of the samples were similar. The values represent the means
Fig. 2. The effects of caffeic acid, caffeic acid methyl ester and di-O-
acetylcaffeic acid on NO (A) and PGE (B) production in RAW 264.7 cells.
A) The cells were pretreated for 1 h with different concentrations (25, 50
2
*
**
***
ꢀ
S.E.M. from three independent experiments. p < 0.05, p < 0.01,
0.001 vs. LPS-treated group; the significance of the difference between
p
(
<
the treated groups was evaluated using the Student’s t-test.
and 100 mM) of caffeic acid and its derivatives, and then LPS (1 mg/mL)
was added and the cells were incubated for 24 h. Control (Con) values were
obtained in the absence of LPS or caffeic acid and its derivatives. Ten
6
micromolar of L-N -(1-iminoethyl)lysine (L-NIL) was present in the assay
inhibited by 68.9% in cells treated with CM (100 mM)
(Fig. 4A). Under the same conditions, the COX-2 mRNA
levels were also significantly decreased in a similar way
as a positive control. The values are the means ꢀ S.E.M. from three
*
**
***
independent experiments. p < 0.05, p < 0.01, p < 0.001 vs. LPS-
treated group; the significance of the difference between the treated groups
was evaluated using the Student’s t-test. (B) The conditions of sample
treatment were identical to those described for Fig. 2(A). Ten micromolar of
NS-398 was used as a positive control in the assay. The values represent the
(
Fig. 4B). CM did not affect the expression of the house-
keeping gene, b-actin. In general, these results are con-
sistent with the profile of the inhibitory effect of CM on NO
and PGE release (Fig. 2).
**
***
means ꢀ S.E.M. from three independent experiments. p < 0.01, p <
0.001 vs. LPS-treated group; the significance of the difference between the
2
treated groups was evaluated using the Student’s t-test.
3.5. Inhibition of LPS-induced NF-kB activation by CM
the modulation of the expression of the iNOS and COX-2
enzymes. In unstimulated RAW 264.7 cells, neither iNOS
nor COX-2 protein or mRNA were detectable. In response
to LPS, the expression level of iNOS was markedly
augmented and CM significantly inhibited iNOS protein
induction in a dose-dependent manner (Fig. 4A). RT-PCR
analysis showed that the amount of iNOS mRNA was
correlated with its protein level (Fig. 4B). A similar pattern
was observed when the effect of CM on the LPS-induced
COX-2 expression level was examined; the densitometric
analysis of three independent experiments demonstrated
that the COX-2 protein expression induced by LPS was
To further investigate the mechanism of the CM-
mediated inhibition of iNOS, COX-2 and TNF-a transcrip-
tion, we focused on NF-kB, which is known to transacti-
vate iNOS, COX-2, TNF-a and other genes [11].
Electrophoretic mobility shift assay (EMSA) analyses
demonstrated that the LPS-induced NF-kB DNA binding
activity in RAW 264.7 macrophages was significantly
reduced by treatment with CM in a dose-dependent manner
(Fig. 5). The extent of this reduction was in a range similar
to that of the iNOS and COX-2 proteins and iNOS, COX-2
and TNF-a mRNA expressions.

K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
2333
Fig. 5. The inhibition of NF-kB DNA binding by caffeic acid methyl ester
CM) was found to occur in a dose-dependent manner. RAW 264.7 cells
(
were incubated for 60 min with different concentrations of CM. The
specificity of binding was examined by competition with a 80-fold cold
probe (C.P.). The arrow indicates the position of the NF-kB band. The data
shown are representative of three independent experiments.
3.6. Effect of CM on degradation of IkB-a and nuclear
translocation of p65
In unstimulated cells, NF-kB is sequestered in the
cytosol by its inhibitor, IkB, which upon LPS stimulation
is phosphorylated by its inhibitor IkB kinases, ubiquiti-
nated and rapidly degraded via the 26S proteosome, thus
releasing NF-kB [11]. We further investigated whether CM
(100 mM) could inhibit the LPS-stimulated degradation of
IkB-a in the cells, by conducting a Western blot assay with
the anti-IkB-a antibody. Fig. 6 shows that the LPS induced
the degradation of IkB-a after 10–20 min, and that this
degradation was significantly blocked by pre-treatment
with CM (100 mM). It was investigated as to whether
CM prevented the translocation of the subunit of NF-
kB, p65 from the cytosol to the nucleus after its release
from IkBs. In the case of the treatment with CM, a decrease
in the level of p65 in the nuclear fraction was detected by
Western blot analysis until 60 min (Fig. 6A). This cascade
Fig. 4. The effects of caffeic acid methyl ester on LPS-induced iNOS and
COX-2 proteins (A) and mRNAs (B) expression in RAW 264.7 cells. (A)
Lysates were prepared from control or 24 h LPS (1 mg/mL)-stimulated cells
alone or in combination with increasing concentrations (25, 50 and 100 mM)
of CM. All lanes contained 50 mg of total proteins. A representative
immunoblot from three separate experiments is shown. (B) Total RNA
was prepared under conditions identical to those described for Fig. 3B and
RT-PCR was performed to compare the mRNA level of iNOS and COX-2.
PCR of b-actin was performed to verify that the initial cDNA contents of the
samples were similar. The experiment was repeated twice with similar
results being obtained in both cases.

2334
K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
Fig. 6. The inhibition of IkB-a degradation and the nuclear translocation of p65 (A) and NF-kB DNA binding (B) by caffeic acid methyl ester (CM) were found
to occur in a time-dependent manner. (A) RAW 264.7 macrophages were treated with LPS (1 mg/mL) and 100 mM of caffeic acid methyl ester (CM) for the
indicated periods of time. For each condition, the protein level of IkB-a in the cytosolic fraction and the p65 level in the nuclear proteins were determined by
Western blot analysis. (B) The nuclear protein from the LPS-stimulated cells was incubated for the indicated time with 100 mM of CM and the electrophoretic
mobility shift assay was performed. The specificity of binding was examined by competition with a 80-fold cold probe (C.P.). The arrow indicates the position of
the NF-kB band. The data shown are representative of three independent experiments.
was also confirmed by the inhibition of the NF-kB DNA
binding activity between 30 and 60 min after the treatment
with CM (100 mM) (Fig. 6B).
plate test in mice was used to examine the antinociceptive
effect, since it is well known that centrally acting analgesic
drugs elevate the pain threshold of mice towards heat. It
was found that CM significantly increased the hot plate
reaction time in mice. The antinociceptive activities shown
by CM in these models indicate that CM may possess
peripherally and centrally mediated antinociceptive prop-
erties.
4
. Discussion
While investigating the anti-inflammatory and antino-
ciceptive effects of CA, CM and DAC in vivo, we found
that CM significantly decreased the edema induced by
carrageenin, in which the peak edema is characterized by
the presence of PGs [29]. The antinociceptive effects of the
tested samples were assayed using two different models,
i.e., the acetic acid-induced abdominal constriction test and
the hot plate test in mice. CM was found to significantly
inhibit the acetic acid-induced abdominal constriction
response at a dose of 10 mg/kg. In order to distinguish
between the peripheral and central analgesic action, the hot
In the present study, we screened the effect of caffeic
acid and its derivatives on the LPS-induced pro-inflam-
matory molecules, including NO and PGE . Thus, CM was
2
found to be the most potent inhibitor of these mediators
among the tested compounds. To further explore the
possible mechanism of these inhibitions by CM, the
expression levels of the iNOS and COX-2 proteins and
the iNOS, COX-2 and TNF-a mRNA levels were exam-
ined. The inhibition by CM of the LPS-stimulated expres-
sion of these molecules in RAW 264.7 cells was not

K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
2335
attributable to cytotoxicity as assessed by the MTT assay
and the expression of the housekeeping gene b-actin. The
inhibition of the iNOS and COX-2 gene expression was
evidenced by reductions in the mRNA levels in a parallel
concentration-dependent manner with a similar pattern.
production via gene expression and this inhibition was
found to be caused by the blocking of NF-kB activation in
RAW 264.7 macrophages. This anti-inflammatory effect
was also evident in vivo. These findings enhance the
importance of caffeic acid methyl ester as a compound
for treating LPS-induced sepsis syndrome and rheumatoid
arthritis.
Thus, the inhibition of NO and PGE release may be
2
attributed to the suppression of iNOS and COX-2 mRNA
transcription followed by protein expression.
The expression of iNOS and COX-2 in murine macro-
phages has been shown to be dependent on NF-kB activa-
tion [11]. The possibility that CM might inhibit the activity
of NF-kB was examined. The results indicate that the
inhibition by CM on the expression of the iNOS and
COX-2 proteins and iNOS, COX-2 and TNF-a mRNA
was most likely due to the suppression of NF-kB. This is
consistent with previous reports that NF-kB response
elements are present on the promoters for the iNOS,
COX-2 and TNF-a genes [30–34]. NF-kB is primarily
composed of two proteins, p50 and p65, which are also
referred to as RelA and cRel, respectively [11]. In the
resting states, NF-kB is present in the cytosol and is bound
to the inhibitory protein, IkB. Following its induction by a
variety of agents such as LPS, TNF-a and tissue plasmino-
gen activator, IkB is phosphorylated to trigger proteolytic
degradation via the 26S proteosome. NF-kB is then
released from IkB and is translocated into the nucleus,
where it binds to the kB binding sites in the promoter
regions of the target genes. In the present study, we found
that CM blocked the LPS-induced activation of NF-kB via
the inhibition of the degradation of IkB. These results
suggest that the inhibition by CM of the LPS-induced
expression of the iNOS and COX-2 genes occurs through
the blocking of NF-kB activation, although the inhibition
of other factors such as AP-1, the interferon response
element and the g-activated site may also be involved.
Although several studies have shown the anti-oxidative
effects of caffeic acid and some of its derivatives [22–24],
we go so far as to propose the suppression of the transcrip-
tional activity of NF-kB as a possible mechanism by which
CM prevents the inhibition of kB degradation. Caffeic acid
phenethyl ester (CAPE), a structural derivative of caffeic
acid, has been shown as a potent and specific inhibitor of
NF-kB activation [35] and this may provide the molecular
basis for its multiple immunomodulatory and anti-inflam-
Acknowledgments
This research was supported by a Grant (PF002104-07)
from the Plant Diversity Research Center of the 21st
Frontier Research Program funded by the Ministry of
Science and Technology of the Korean government and
by the Korea Science & Engineering Foundation (No. R13-
2002-020-01002-0).
References
[
[
1] Heiss E, Herhaus C, Klimo K, Bartsch H, Gerhauser C. Nuclear factor
Kappa B is a molecular target for sulforaphane-mediated anti-inflam-
matory mechanisms. J Biol Chem 2001;276:32008–15.
2] Kuo PC, Schroeder RA. The emerging multifaceted roles of nitric
oxide. Ann Surg 1995;221:220–35.
[3] ChoCMzynski P, Sacchi N. Single-step method of RNA isolation by
acid guanidinium thiocyanate–phenol–chloroform extraction. Ann
Biochem 1987;162:156–9.
[
4] Szab o´ C, Zingarelli B, O’Connor M, Salzman AL. DNA strand
breakage, activation of poly-ADP ribosyl synthetase and cellular
energy depletion are involved in the cytotoxicity in macrophages
and smooth muscle cells exposed to peroxynitrite. Proc Natl Acad
Sci USA 1996;93:1753–8.
[
5] Miller MJ, Grisham MB. Nitric oxide as a mediator of inflammation?
You had better believe it Mediators Inflamm 1995;4:387–96.
6] Cochran FR, Selph J, Sherman P. Insights into the role of nitric oxide
in inflammatory arthritis. Med Res Rev 1996;16:547–63.
[
[7] Weisz A, Cicatiello I, Esumi H. Regulation of the mouse inducible-
type nitric oxide synthase gene promoter by interferon-gamma, bac-
G
terial lipopolysaccharide and N -monomethyl-L-arginine. Biochem J
1
996;316:209–15.
[
8] Palladino MA, Bahjat FR, Theodorakis EA, Moldawer LL. Anti-TNF-
alpha therapies: the next generation. Nat Rev Drug Discov
2003;2:736–46.
[9] Vodovotz Y, Bogdn C, Paik J, Xie OW, Nathan C. Mechanisms of
suppression of macrophage nitric oxide release by transforming
growth factor beta. J Exp Med 1993;178:605–13.
[
[
[
10] Rao KM. Molecular mechanisms regulating iNOS expression in
various cell types. J Toxicol Environ Health B Crit Rev 2000;3:27–58.
11] Baeuerle PA. I Kappa B-NF-Kappa B structures: at the interface of
inflammation control. Cell 1998;95:729–31.
2
+
matory activities. Moreover, CAPE stimulated Ca -acti-
+
vated K -current in pituitary GH cells [36] and induced
3
12] Brown K, Gerberger S, Carlson L, Franzoso G, Siebenlist U. Control
of I Kappa B-alpha proteolysis by site-specific, signal-induced phos-
phorylation. Science 1995;267:1485–8.
apoptosis by activation of Fas in human breast cancer cells
[37] via NF-kB inhibition. These diverse effects of CAPE
might be attributed to its ability to pass through the cell
membrane more easily than caffeic acid [38]. We also
suggest that the more hydrophobic structure of caffeic acid
methyl ester compared to the other two compounds may be
responsible for its characteristic physicochemical proper-
ties for its anti-inflammatory action.
[
[
[
13] DiDonato J, Mercurio F, Rosette C, Wu-Li J, Suyang H, Ghosh S, et al.
Mapping of the inducible IkB phosphorylation sites that signal its
ubiquitination and degradation. Mol Cell Biol 1996;4:1295–304.
14] O’Conell MA, Benett BL, Mercurio F, Manning AM, Makman N.
Role of IKK1 and IKK2 in lipopolysaccharide signaling in human
monocytic cells. J Biol Chem 1998;46:30410–4.
15] Sanchez-Perez I, Benitah SA, Martinez-Gomariz M, Lacal JC, Perona
R. Cell stress and MEKK1-mediated c-Jun activation modulate NF-kB
activity and cell viability. Mol Biol Cell 2002;13:2933–45.
The results of the present study indicate that CM is a
potent inhibitor of the LPS-induced NO, PGE and TNF-a
2

2336
K.M. Shin et al. / Biochemical Pharmacology 68 (2004) 2327–2336
[
16] Moridani MJ, Scobie H, Jamshidzadeh A, Salehi P, Olrie PJ. Caffeic
Acid, chlorogenic acid and dihydrocaffeic acid metabolism: glu-
tathione conjugate formation. Drug Metab Dispos 2001;29:1432–9.
17] Challi BC, Bartlett CD. Possible cocarcinogenic effects of coffee
constituents. Nature 1975;254:522–33.
2
E production from murine macrophage RAW 264.7 cells. Biol Pharm
Bull 2003;26:481–6.
[29] Spector W. The suppression by antiesterases of increased capillary
permeability in acute inflammation. J Pathol Bacterol 1960;84:
391–403.
[
[
18] Koshihara Y, Neichi T, Murota S, Lao A, Fujimoto Y, Tapamo T.
Caffeic acid is a selective inhibitor for leukotriene biosynthesis.
Biochim Biophys Acta 1984;792:92–7.
[30] Xie W, Merrill JR, Bradshaw WS, Simmons DL. Structural determi-
nation and promoter analysis of the chicken mitogen-inducible pros-
taglandin G/H synthase gene and genetic mapping of the murine
homolog. Arch Biochem Biophys 1993;300:247–52.
[31] Xie QW, Kashiwabara Y, Nathan C. Role of transcription factor NF-
Kappa B/Rel in induction of nitric oxide synthase. J Biol Chem
1994;269:4705–8.
[
[
[
19] Tanaka T, Kojima T, Kawamori T, Wang A, Suzuki M, Okamoto K, et
al. Inhibition of 4-nitroquinoline-1-oxide-induced rat tongue carcino-
genesis by the naturally occurring plant phenolics caffeic, ellagic,
chlorogenic and ferulic acids. Carcinogenesis 1993;14:1321–5.
20] Tanaka T, Kojima T, Kawamori T, Yoshimi N, Mori M. Chemopre-
vention of diethylnitrosamine-induced hepatocarcinogenesis by a
simple phenolic acid protocatechuic acid in rats. Cancer Res
[32] Chen F, Kuhn DC, Sun SC, Gaydos LJ, Demers LM. Dependence and
reversal of nitric oxide production on NF-kB in silica and lipopoly-
saccharide induced macrophages. Biochem Biophys Res Commun
1995;214:839–46.
1993;53:2775–9.
21] Hsiao G, Shen MT, Chang WC, Cheng WC, Cheng YW, Pan SL, et al.
A novel antioxidant, octyl caffeate, suppression of LPS/IFN-gamma-
induced inducible nitric oxide synthase gene expression in rat aortic
smooth muscle cells. Biochem Pharmacol 2002;65:1383–92.
[33] Schmedtje JFJr. Ji YS, Liu WL, DuBois RN, Runge MS. Hypoxia
induces cyclooxygenase-2 via the NF-Kappa B p65 transcription
factor in human vascular endothelial cells. J Biol Chem 1997;
272:601–8.
[
[
22] Wells PG, Kim PM, Laposa RR, Nicol CJ, Pharman T, Winn LM.
Oxidative damage in chemical teratogenesis. Mutat Res 1997;396:
[34] Roshak AK, Jackson JR, CMGough K, Chabot-Fletcher M, Mochan E,
Marshall LA. Manipulation of distinct NF-kB proteins alters inter-
leukin-1b-induced human rheumatoid synovial fibroblast prostaglan-
65–78.
23] Kikujak H, Hisamoto M, Hirose K, Akiyama K, Tamiguchi H.
Antioxidant properties of ferulic acid and its related compounds. J
Agri Food Chem 2002;50:2161–8.
2
din E formation. J Biol Chem 1996;271:31496–501.
[35] Natarajan K, Singh S, Burke Jr TR, Grunberger D, Aggarwal BB.
Caffeic acid phenethyl ester is a potent and specific inhibitor of
activation of nuclear transcription factor NF-Kappa B.. Proc Natl
Acad Sci USA 1996;93:9090–5.
[
[
24] Matsuda H, Samunawa K, Kubo M. Anti-inflammatory activity of
ginsenoside Ro. Planta Med 1990;56:19–23.
25] Li DW, Lee EB, Kang SS, Hyun JE, Whang WK. Activity-guided
isolation of saponins from Kalopanax pictus with anti-inflammatory
activity. Chem Pharm Bull 2002;50:900–3.
[36] Lin MW, Yang SR, Huang MH, Wu SN. Stimulatory actions of caffeic
acid phenethyl ester, a known inhibitor of NF-Kappa B activation, on
+
+
Ca2 -activated K -current in pituitary GH3 cells. J Biol Chem
2004;279:26885–92.
[
[
[
26] Whittle BA. The use of change in capillary permeability to distin-
guish between narcotic and anagesic. Br J Pharmacol 1949;22:
[37] Watabe M, Hishikawa K, Takayanagi A, Shimizu N, Nakaki T. Caffeic
acid phenethyl ester induces apoptosis by inhibition of NF-Kappa B
and activation of Fas in human breast cancer MCF-7 cells. J Biol Chem
2004;279:6017–26.
2
46–460.
27] Eddy NB, Leimback D. Synthetic analgesics. II. Dithienbutenyl-
and dithienbutenylamines. Pharmacol Exp Ther 1953;107:
85–93.
J
3
[38] Frenkel K, Wei H, Bhimani R, Ye J, Zadunaisky JA, Huang MT, et al.
Inhibition of tumor promoter-mediated processes in mouse skin
and bovine lens by caffeic acid phenethyl ester. Cancer Res 1993;
53:1255–61.
28] Kim RG, Shin KM, Kim YG, Jeong HJ, Ha JH, Choi JW, et al.
Inhibition of methanol extract from the aerial parts of Saururus
chinensis on lipopolysaccharide-induced nitric oxide and prostagladin