Cyclooxygenase (COX)-2 is an inflammatory gene catalyzing
prostaglandin E2 (PGE2) production from a substrate
lipid, arachidonic acid.1) Inflammatory signals greatly enhance
the expression level of COX-2, particularly in inflammatory
cells such as monocytes, macrophages, endothelial
cells, and fibroblasts.2) Numerous trials to develop a promising
antiinflammatory drug therefore are still targeted at suppressing
PGE2 production as well as COX-2 expression.3) A
better understanding of the induction mechanism of COX-2
expression and PGE2 production would allow us to consider
various levels of inflammatory signaling pathways as target
molecules. The biochemical pathways include the levels of
transcription factors (nuclear factor (NF)-k B and AP-1) and
upstream signaling cascades such as mitogen-activated protein
kinase (MAPK) containing extracellular signal-regulated
kinase (ERK), p38, and C-Jun N-terminal kinase (JNK),
phosphoinositide-3-kinase (PI3K)/Akt, and inhibitor of kB
(Ik B) kinase (IKK)/IkBa .4,5)
Coffee is a popular beverage worldwide and has been studied
for a long time due to its controversial pharmacology
thought to be either beneficial or adverse. Although coffee
has been reported to increase blood cholesterol level and
blood pressure, and cause hypersecretion of gastric juice, it is
increasingly noted that coffee also has somewhat beneficial
effects such as antiinflammatory and anticarcinogenic properties.
6.8) The major compounds to explain these pharmacologically
beneficial effects are diterpenes such as cafestol
(Fig. 1) and kahweol.6) In spite of numerous papers on the
anticancer activities of coffee and its ingredients, so far, few
studies demonstrated the effect of cafestol on inflammatory
mediator production in macrophages. The Jeong group in
Korea found that cafestol has strong inhibitory activity on
PGE2 production by suppressing the NF-k B activation pathway.
9) They found that the lipopolysaccharide (LPS)-induced
activation of NF-k B is down-regulated by cafestol by preventing
IkBa degradation and inhibiting Ik B kinase (Ik K)
activity.9) Although that study contributed to understanding
the immunopharmacologic roles of cafestol, there is a possibility
that other inflammatory signaling pathways can serve
as target sites for cafestol action. To expand our knowledge
of the beneficial roles of coffee, we consumed to evaluate the
antiinflammatory mechanism of cafestol in terms of other
major inflammatory signaling pathways such as MAPK
(ERK, p38, and JNK) and their relevant transcription factor
MATERIALS AND METHODS
Materials Cafestol (Fig. 1, purity: 99%), phorbol-12-
myristate-13-acetate (PMA), and lipopolysaccharide (LPS,
Escherichia coli 0111:B4) were purchased from Sigma (St.
Louis, MO, U.S.A.). LY294002, wortmannin, U0126,
SP600125, and SB203580 were obtained from Calbiochem
(La Jolla, CA, U.S.A.). Fetal bovine serum (FBS) was obtained
from GIBCO (Grand Island, NY, U.S.A.). The ERK
kinase assay kit was purchased from Millipore (Billerica,
128 Vol. 33, No. 1
Cafestol, a Coffee-Specific Diterpene, Is a Novel Extracellular Signal-
Regulated Kinase Inhibitor with AP-1-Targeted Inhibition of
Prostaglandin E2 Production in Lipopolysaccharide-Activated
Ting SHEN,a,# Jaehwi LEE,b,# Eunji LEE,b Seong Hwan KIM,c Tae Woong KIM,d and Jae Youl CHO*,a
a School of Bioscience and Biotechnology, and Institute of Bioscience and Biotechnology, Kangwon National University;
d Department of Biochemistry, Kangwon National University; Chuncheon 200.701, Republic of Korea: b College of
Pharmacy, Chung-Ang University; Seoul 156.756, Republic of Korea: and c Laboratory of Chemical Genomics, Korea
Research Institute of Chemical Technology; P.O. Box 107, Yuseong-gu, Daejeon 305.600, Republic of Korea.
Received June 11, 2009; accepted September 19, 2009; published online October 14, 2009
Coffee is a popular beverage worldwide with various nutritional benefits. Diterpene cafestol, one of the
major components of coffee, contributes to its beneficial effects through various biological activities such as
chemopreventive, antitumorigenic, hepatoprotective, antioxidative and antiinflammatory effects. In this study,
we examined the precise molecular mechanism of the antiinflammatory activity of cafestol in terms of
prostaglandin E2 (PGE2) production, a critical factor involved in inflammatory responses. Cafestol inhibited both
PGE2 production and the mRNA expression of cyclooxygenase (COX)-2 from lipopolysaccharide (LPS)-treated
RAW264.7 cells. Interestingly, this compound strongly decreased the translocation of c-Jun into the nucleus and
AP-1 mediated luciferase activity. In kinase assays using purified extracellular signal-regulated kinase 2 (ERK2)
or immunoprecipitated ERK prepared from LPS-treated cells in the presence or absence of cafestol, it was found
that this compound can act as an inhibitor of ERK2 but not of ERK1 and mitogen-activated protein kinase kinase
1 (MEK 1). Therefore our data suggest that cafestol may be a novel ERK inhibitor with AP-1-targeted inhibitory
activity against PGE2 production in LPS-activated RAW264.7 cells.
Key words cafestol; prostaglandin E2; cyclooxygenase-2; AP-1; c-Jun; extracellular signal-regulated kinase
Biol. Pharm. Bull. 33(1) 128.132 (2010)
. To whom correspondence should be addressed. e-mail: firstname.lastname@example.org ¨Ï 2010 Pharmaceutical Society of Japan
# These authors contributed equally to this work.
Fig. 1. Structure of Cafestol
MA, U.S.A.). All other chemicals were of reagent grade.
Total or phospho-specific antibodies to c-Jun, c-fos, JNK,
p38, ERK, b -tublin, and lamin A/C were purchased from
Cell Signaling (Beverly, MA, U.S.A.).
Cell Culture RAW264.7 and HEK293 cells (American
Type Culture Collection, Rockville, MD, U.S.A.) were maintained
in complete RPMI-1640 medium (supplemented with
100 U/ml of penicillin, 100m g/ml of streptomycin, and 10%
fetal bovine serum).
PGE2 Production The modulatory effect of cafestol on
PGE2 release was determined as previously described.10)
RAW264.7 cells (2 106 cells/ml) were incubated with LPS
and various concentrations of cafestol for 24 h. Supernatants
were collected and assayed for PGE2 content using the PGE2
enzyme immunoassay kit (Amersham, Little Chalfont, Buckinghamshire,
MTT Assay (Colorimetric Assay) for Measurement of
Cell Proliferation Cell proliferation and the cytotoxicity of
cafestol were determined using the conventional 3-(4,5-dimethylthiazol-
2-yl)-2,5-diphenyltetrazolium bromide (MTT)
assay, as previously reported.10)
Determination of LPS-Inducible COX-2 Gene Expression
For the evaluation of COX-2 mRNA expression levels,
the total RNA from the LPS treated-RAW264.7 cells (5 106
cells/ml) was prepared by adding TRIzol Reagent (Gibco
BRL) according to the manufacturer¡¯s protocol, as reported
previously.11) The primers (Bioneer, Seoul, Korea) used in
this experiment were: COX-2 (F-5 -CACTACATCCTGACCCACTT-
3 and R-5 -ATGCTCCTGCTTGAGTATGT-3 );
and glyceraldehyde-3-phosphate dehydrogenase (GAPDH)
(F-5 -CACTCACGGCAAATTCAACGGCAC-3 and 5 -
Luciferase Reporter Gene Activity Assay Since
RAW264.7 cells are not easy to transfect with interesting
DNA constructs, HEK293 cells (1 106 cells/ml) were transfected
with 1m g of plasmids with AP-1-Luc as well as b -
galactosidase using the calcium phosphate method in a 12-
well plate. The cells were used for experiments 48 h after
transfection. Luciferase assays were performed using the Luciferase
Assay System (Promega, Madison, WI, U.S.A.).12)
Immunoblotting For total protein extraction: RAW
264.7 cells were harvested, washed with cold phosphate
buffered saline (PBS), and lysed in lysis buffer (Tris.HCl
20mM, pH 7.4, EDTA 2mM, EGTA 2mM, b -glycerophosphate
50mM, sodium orthovanadate 1mM, dithiothreitol 1
mM, 1% Triton X-100, 10% glycerol, leupeptin 10m g/ml,
aprotinin 10m g/ml, pepstatin 10m g/ml, benzimidine 1mM
and phenylmethane sulphonylfluoride 2mM) for 30 min with
rotating at 4 ¡ÆC. Lysates were clarified by centrifugation at
16000 g for 10 min at 4 ¡ÆC. For nuclear protein extraction,
nuclear proteins were obtained through three steps. After the
treatment, cells were harvested and lysed in 500m l of lysis
buffer (KCl 50mM, 0.5% Nonidet P-40, HEPES 25mM,
phenylmethylsulfonyl fluoride 1mM, leupeptin 10m g/ml,
aprotinin 20m g/ml, and 100m M 1,4-dithiothreitol) on ice for
4 min. Cells lysates were centrifuged at 14000 rpm for 1 min
at 4 ¡ÆC. In the second step, the pellet was washed with the
washing buffer, which was the same as the lysis buffer excluding
Nonidet P-40. In the final step, the nuclei were incubated
with an extraction buffer (KCl 500mM, 10% glycerol,
HEPES 10mM, NaCl 300mM, 1,4-dithiothreitol 0.1mM,
PMSF 0.1mM, leupeptin 2m g/ml, and aprotinin 2m g/ml) and
centrifuged at 14000 rpm for 5 min. The supernatant was collected
as the nuclear protein extract. Soluble cell lysates were
immunoblotted, and phospho-ERK levels were visualized as
ERK Kinase Assay For evaluating ERK kinase inhibitory
activity with purified enzyme, a kinase profiler service
from Millipore was used. In a final reaction volume of
25m l, ERK1 (human), ERK2 (human), or MAPK kinase
(MEK) 1 (human) (1.5 mU) was incubated with the reaction
buffer. The reaction was initiated by the addition of
MgATP. After incubation for 40 min at room temperature, the
reaction was stopped by the addition of 5 ml of a 3% phosphoric
acid solution. Ten microliters of the reaction product
were then spotted onto a P30 filtermat and washed three
times for 5 min in phosphoric acid 75mM and once in
methanol prior to drying and scintillation counting. To determine
the inhibitory effects of cafestol on LPS-activated ERK
activity, immunoprecipitated ERK prepared from LPStreated
RAW264.7 cells (5 106 cells/ml) in the presence or
absence of cafestol were incubated with myelin basic protein
(MBP), according to the manufacturer¡¯s instructions. The
ERK kinase activity was determined with anti-phospho-MBP
antibody after immunoblotting analysis.
Statistical Analysis Student¡¯s t-test and one-way ANOVA
were used to determine the statistical significance of differences
between values for the various experimental and control
groups. Data expressed as mean S.E.M. were taken
from at least three independent experiments performed in
triplicate (Figs. 2A, B and 3A, D, E). The data (Figs. 2C and
3B, C) are representative of three different experiments with
similar results. p values of 0.05 or less were considered to
represent statistically significant differences.
RESULTS AND DISCUSSION
In this study, we found a novel inhibitory pathway by
which cafestol effectively blocks the AP-1 pathway in LPSstimulated
macrophages. Cafestol clearly suppressed PGE2
production (Fig. 2A) in macrophages after LPS stimulation
in a dose-dependent manner (IC50 value 45.7m M), as did
standard compounds (LY294002 and wortmannin) (Fig. 2A,
right panel). Since there was no effect on normal cell viability
by cafestol after 12-and 24-h incubation (Fig. 2B),
cafestol-derived inhibition does not appear to be due to its
nonspecific action. However, the inhibitory potency was different
from that in a previous report in which the IC50 value
of cafestol was around 5m M.9) Continuous evaluation also indicated
that cafestol-mediated inhibition of PGE2 production
may occur at the transcriptional level, because cafestol decreased
the mRNA synthesis of COX-2 observed under the
same conditions (Fig. 2C). This inhibitory pattern is generally
seen in numerous PGE2 inhibitory compounds with transcriptional
regulatory features,13,14) and accordingly the identification
of the transcription factor(s) and upstream signaling
events were continuously observed.
The transcriptional control of PGE2 production in
macrophages is mediated by redox-sensitive transcription
factors such as NF-k B and AP-1. Because cafestol has been
reported to suppress NF-k B activation,9) we mainly focused
on understanding the involvement of AP-1. To do this, the
January 2010 129
translocational levels of AP-1 (c-Jun and c-fos) in the nuclear
fraction from LPS-treated RAW264.7 cells was first examined
using immunoblotting analysis. As Fig. 3A depicts, the
levels of translocated c-Jun and phospho-c-Jun in the nuclear
fraction were interrupted by cafestol 25 and 100m M, without
any contamination of cytosolic proteins, as confirmed with
cytosolic b -tublin.15) In contrast, c-Jun and c-fos were not
found in the normal nuclear fraction, indicating inducible
activation of AP-1 (c-Jun/c-fos) during LPS treatment. In
agreement with those results, PMA-upregulated luciferase
activity in HEK293 cells transfected with a luciferase reporter
construct containing binding sites for AP-111) was also
dose dependently suppressed by cafestol treatment (Fig. 3C),
suggesting that the molecular target of cafestol appears in a
downstream pathway, activated commonly during both LPS
and PMA treatment for AP-1 activation.
Since the translocation of AP-1 is mediated by the phosphorylation
of MAPK,16) we next investigated the potential
involvement of MAPK in cafestol inhibition. Interestingly,
however, the phosphorylation of MAPK (ERK, JNK, and
p38) was not blocked by various concentrations of cafestol
(Fig. 3B). The negative effects of cafestol on MAPK activation
was also observed even when incubated for various
times (data not shown), implying that the MAPK pathway is
not a target of the pharmacology of cafestol. Nonetheless,
among MAPK inhibitors, U0126, an inhibitor of ERK kinase
(MEK) strongly suppressed the nuclear translocation of AP-1
(c-Jun and c-fos) by LPS in RAW264.7 cells and luciferase
activity mediated by AP-1 activation in HEK293 cells cultured
with PMA (Fig. 3D), suggesting that the ERK (but not
the p38 and JNK) activation pathway could play a central
role in this AP-1 up-regulation event (Fig. 3D). Based on
these results, it was assumed that cafestol may not only block
the ERK pathway but also suppress a biochemical step between
c-Jun translocation and the phosphorylation of ERK.
Because the immunoblotting results shown in Fig. 3B indicated
that cafestol did not block each activational phosphorylation
step of ERK, p38, and JNK, and the inhibitors to p38
and JNK did not suppress or only weakly suppressed AP-1
(c-Jun/c-fos) translocation as well as AP-1-mediated luciferase
activity (Fig. 3D), the only method to determine
cafestol activity was to explore ERK kinase itself. Therefore,
the enzyme kinase assay was carried out to ascertain testing
whether this compound was able to suppress the kinase activity
of ERK (ERK 1/2) and its upstream kinase, MEK1, directly.
17) Interestingly, cafestol only decreased the enzymatic
activity of ERK2, but not that of ERK1 and MEK1, at
100m M (Fig. 3E). These data therefore imply that cafestol
could block PGE2 production, COX-2 expression, and c-
Jun/AP-1 translocation by directly blocking ERK2. Moreover,
the lack of inhibition of MEK1 by cafestol strongly
indicates that ERK2 inhibition by cafestol is not due to a
simple treatment effect. In agreement with the result of the
purified enzyme assay, cafestol also significantly suppressed
immunoprecipitated ERK activity, assessed by measuring
the phosphorylation level of its substrate protein MBP, performed
with LPS-treated cells in the presence or absence of
cafestol (Fig. 3F). Therefore these data suggest that cafestol
could act as an inhibitor of ERK(2).
It is widely reported that the ERK activation pathway is
involved in the pathophysiologic onset of various diseases
such as cancer, stroke, diabetes, cardiovascular disease, au-
130 Vol. 33, No. 1
Fig. 2. Effects of Cafestol on the Production of PGE2 by LPS-Activated RAW264.7 Cells
(A) RAW264.7 cells (1 106 cells/ml) were incubated with various concentrations of cafestol (left panel) or standard drugs (LY294002 and wortmannin) (right panel) in the presence
of LPS (2m g/ml) for 24 h. Culture supernatants were assayed for PGE2 determination using EIA. (B) RAW264.7 cells (1 106 cells/ml) were incubated with various concentrations
of cafestol for 12 or 24 h. Cell viability was determined in the MTT assay. (C) RAW264.7 cells (5 106 cells/ml) were incubated with cafestol in the presence or absence of
LPS (2m g/ml) for 6 h. The mRNA levels of COX-2 and GAPDH were determined with semiquantitative RT-PCR. .. p 0.01 compared with control.
toimmune diseases, atherosclerosis, allergic reactions, inflammation,
and neurologic disorders.18.20) Due to this, numerous
researchers are now developing potent ERK pathway
inhibitors. U0126 and PD98059 are known for MEK-inhibitory
activity. On the other hand, 3-(2-aminoethyl)-5-((4-
ethoxyphenyl)methylene)-2,4-thiazolidinedione has been reported
to inhibit ERK activity (IC50 25m M) by suppressing
ERK binding to its protein substrates, but not to the ATP domain
in HeLa, A549, and SUM-159 tumor cells.21,22) In addition,
pyrazolylpyrrole-type CAY10561 is a potent, ATP-competitive
inhibitor of ERK2 (Ki 2nM) with higher selectivity
for ERK2.23) Whether cafestol binds to ATP-binding sites or
other substrate-binding areas and which functional groups of
cafestol are important for ERK2 inhibition have not yet been
elucidated. Therefore detailed characterization of inhibitory
mechanism of action and modification of chemical structures
to optimize the inhibitory potency should be undertaken.
In conclusion, we found that cafestol inhibits PGE2 production
from LPS-treated RAW264.7 cells at the transcriptional
level. In particular, this compound downregulates the
translocation of c-Jun and AP-1-mediated luciferase activity.
The direct kinase assay showed that cafestol can act as an inhibitor
of ERK2 but not of ERK1 and MEK1. Therefore our
data suggest that cafestol could be a novel ERK inhibitor
with AP-1-targeted inhibitory activity against PGE2 production
in LPS-activated RAW264.7 cells.
Acknowledgments This study was financially supported
January 2010 131
Fig. 3. Effects of Cafestol on AP-1 Activation
(A) RAW264.7 cells (5 106 cells/ml) pretreated with cafestol for 1 h were stimulated in the presence or absence of LPS (2m g/ml) for 30 min. After preparation of the nuclear
fraction, the phospho- or total protein levels of c-Jun, c-fos, b -tublin, and lamin A/C were determined by immunoblotting analysis with their phospho- or total protein antibodies.
(B) RAW264.7 cells (5 106 cells/ml) pretreated with cafestol for 1 h were stimulated in the presence or absence of LPS (2m g/ml) for 1 h. After immunoblotting, the phospho- or
total protein levels of MAPK (ERK, p38, and JNK) and b -actin were determined using phospho-specific or total protein antibodies. (C and D, right panel) HEK293 cells cotransfected
with the plasmid constructs AP-1-Luc (1m g/ml) and b -gal (as a transfection control) were treated with cafestol (C) or MAPK inhibitors (SB203580, SP600125, and U0126)
(D, right panel) in the presence or absence of PMA (0.1m M) for 18 h. Luciferase activity was determined by luminometry. (D, left panel) RAW264.7 cells (5 106 cells/ml) pretreated
with cafestol for 1 h were stimulated in the presence or absence of LPS (2m g/ml) for 30 min. After preparation of the nuclear fraction, the total protein levels of c-Jun, c-fos,
and lamin A/C were determined by immunoblotting analysis with their total protein antibodies. (E) The inhibitory effects of cafestol on kinase activities of ERK1, ERK2, and
MEK1 were determined in a direct kinase assay as described in Materials and Methods. (F) The inhibitory effects of cafestol on immunoprecipitated ERK prepared from LPStreated
RAW264.7 cells were determined by measuring the level of phospho-MBP, as described in Materials and Methods. .. p 0.01 compared with control.
by MEST and KOTEF through the Human Resource Training
Project for Regional Innovation.
1) Cuccurullo C., Fazia M. L., Mezzetti A., Cipollone F., Curr. Med.
Chem., 14, 1595.1605 (2007).
2) Calder P. C., Lipids, 36, 1007.1024 (2001).
3) Leone S., Ottani A., Bertolini A., Curr. Topics Med. Chem., 7, 265.
4) Han E. H., Kim J. Y., Kim H. K., Hwang Y. P., Jeong H. G., Toxicol.
Appl. Pharmacol., 233, 333.342 (2008).
5) Hou D. X., Masuzaki S., Hashimoto F., Uto T., Tanigawa S., Fujii M.,
Sakata Y., Arch. Biochem. Biophys., 460, 67.74 (2007).
6) Cavin C., Holzhaeuser D., Scharf G., Constable A., Huber W. W.,
Schilter B., Food Chem. Toxicol., 40, 1155.1163 (2002).
7) Cornelis M. C., El-Sohemy A., Curr. Opin. Clin. Nutr. Metab. Care,
10, 745.751 (2007).
8) Sudano I., Binggeli C., Spieker L., Luscher T. F., Ruschitzka F., Noll
G., Corti R., Prog. Cardiovasc. Nursing, 20, 65.69 (2005).
9) Kim J. Y., Jung K. S., Jeong H. G., FEBS Lett., 569, 321.326 (2004).
10) Cho J. Y., Baik K. U., Jung J. H., Park M. H., Eur. J. Pharmacol., 398,
11) Lee Y. G., Lee W. M., Kim J. Y., Lee J. Y., Lee I. K., Yun B. S., Rhee
M. H., Cho J. Y., Br. J. Pharmacol., 154, 852.863 (2008).
12) Lee J. Y., Rhee M. H., Cho J. Y., Naunyn Schmiedebergs Arch. Pharmacol.,
377, 111.124 (2008).
13) Yamaguchi K., Ishikawa T., Kondo Y., Fujisawa M., Mol. Cell. Endocrinol.,
283, 68.75 (2008).
14) Krady J. K., Lin H. W., Liberto C. M., Basu A., Kremlev S. G., Levison
S. W., J. Neurosci. Res., 86, 1538.1547 (2008).
15) Lesort M., Attanavanich K., Zhang J., Johnson G. V., J. Biol. Chem.,
273, 11991.11994 (1998).
16) Guha M., Mackman N., Cell Signal., 13, 85.94 (2001).
17) Yoon S., Seger R., Growth Factors, 24, 21.44 (2006).
18) Hilger R. A., Scheulen M. E., Strumberg D., Oncologie, 25, 511.518
19) Jeon S. J., Kwon K. J., Shin S., Lee S. H., Rhee S. Y., Han S. H., Lee
J., Kim H. Y., Cheong J. H., Ryu J. H., Min B. S., Ko K. H., Shin C. Y.,
Biomol. Ther., 17, 70.78 (2009).
20) Kim A. K., Choi H. J., Kim B. G., Park Y. R., Kim J. S., Ryu J. H., Soh
Y., Biomol. Ther., 16, 377.384 (2009).
21) Chen F., Hancock C. N., Macias A. T., Joh J., Still K., Zhong S.,
MacKerell A. D. Jr., Shapiro P., Bioorg. Med. Chem. Lett., 16, 6281.
22) Hancock C. N., Macias A., Lee E. K., Yu S. Y., Mackerell A. D. Jr.,
Shapiro P., J. Med. Chem., 48, 4586.4595 (2005).
23) Aronov A. M., Baker C., Bemis G. W., Cao J., Chen G., Ford P. J., Germann
U. A., Green J., Hale M. R., Jacobs M., Janetka J. W., Maltais F.,
Martinez-Botella G., Namchuk M. N., Straub J., Tang Q., Xie X., J.
Med. Chem., 50, 1280.1287 (2007).
132 Vol. 33, No. 1
- Casos clínicos