Cancer Letters 153 (2000) 5156

www.elsevier.com/locate/canlet

Preferential cytotoxicity of caffeic acid phenethyl ester analogues

on oral cancer cells
Yean-Jang Lee a,*, Pel-Hu Liao b, Wen-Kang Chen c, Chi-Chiang Yang d
a
Department of Chemistry, National Changhua University of Education, No. 1, Ching Der Road, Changhua 50058, Taiwan
b
Institute of Toxicology, Chung Shan Medical and Dental College, Taichung, Taiwan
c
Institute of Biochemistry, Chung Shan Medical and Dental College, Taichung, Taiwan
d
School of Medical Technology, Chung Shan Medical and Dental College, Taichung, Taiwan
Received 4 October 1999; received in revised form 3 January 2000; accepted 4 January 2000

Abstract
As part of our previous search for new compounds with improved biological activities including antibiotic, antiviral, anti-
inammatory, and tumor growth inhibition activities, we synthesized some caffeic acid phenethyl ester (CAPE)-like
compounds from commercially available caffeic acid. Nine chemicals were tested by the 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide assay on the growth of buccal mucosal broblast (BF), oral submucosus broblast (OSF), neck
metastasis of Gingiva carcinoma (GNM), and tongue squamous cell carcinoma (TSCCa) cells. CAPE and its ethyl analogue
show signicant cytotoxicity on OSF, GNM, and TSCCa cells, but not on BF cells. The results suggest that CAPE-like
compounds may be potential chemotherapy agents against oral cancer. q 2000 Published by Elsevier Science Ireland Ltd.
All rights reserved.
Keywords: Caffeic acid phenethyl ester; Gingiva carcinoma; Tongue squamous cell carcinoma

1. Introduction or MC. CAPE-like analogues exhibited a broad spec-

Recently we synthesized [1] and studied [2] the
constituents of natural honey-propolis, which is a
folk medicine employed for treating various ailments.
We assayed them in peripheral blood mononuclear
cells (PBMC) infected by M-tropic (JRCSF), T-tropic
(NL-4-3), and Dual tropic (89.6) of HIV-1 isolates.
Caffeic acid phenethyl ester (CAPE) and caffeic acid
methyl ester (MC) showed signicant inhibition on
the growth of HIV-replication in PBMC cells. 3-
(3,4-Dimethoxy-phenyl)-acrylic phenethyl ester
(PEDMC) showed less inhibition effect than CAPE
* Corresponding author. Fax: 1 886-4-7211190.
E-mail address: leeyj@cc.ncue.edu.tw (Y.J. Lee)
trum of medical applications including antibiotic,
antiviral, anti-inammatory, and tumor growth inhibi-
tion [35]. Furthermore, Rao et al. have shown that
CAPE has chemoprevention [610] and antitumor
properties [11,12]. Since the biological properties of
propolis and its constituent compounds have become
a point of particular interest, and extraction of these
compounds required large quantities of propolis, we
have designed and undertaken the synthesis of nine
CAPE-like analogues for investigating oral cancer.
The commercial availability of caffeic acid
prompted us to prepare some caffeic acid esters
from this compound. However, selective esterication
of the a, b-unsaturated carboxyl group presents a
problem because of the solubility of caffeic acid.

0304-3835/00/$ - see front matter q 2000 Published by Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0304-383 5(00)00389-X
52 Y.J. Lee et al. / Cancer Letters 153 (2000) 5156

Although CAPE (1) has been synthesized by several cated. CHCl3 (7.26 ppm) was used an internal stan-
groups [7,13,14], their preparations gave only poor dard for silyl compounds. TMS was employed as an
yields of the target compound and the synthetic internal standard for other compounds. 13C NMR
approaches are very time consuming and complicated. spectra were recorded in CDCl3 with CHCl3 (77.0
To overcome these technical difculties, we have ppm) as an internal standard. Solvents were freshly
found that CAPE and its derivatives can be easily distilled from phosphorus pentoxide or CaH2, and
prepared in a one pot reaction in high yield (86%) THF from sodium diphenyl ketyl, prior to use. All
[15]. reactions were carried out under nitrogen atmosphere
We report herein the effect of nine agents on the otherwise stated. Silica gel (silica gel 60, 230400
growth of normal human buccal mucosal broblast mesh, Merck) was used for chromatography. Organic
(BF), oral submucosus broblast (OSF), neck metas- extracts were dried over anhydrous MgSO4.
tasis of Gingiva carcinoma (GNM), tongue squamous General procedure for esters from cinnamic acid.
cell carcinoma (TSCCa) cells using the MTT assay. Caffeic acid (1.02 g, 5.6 mmol) was dissolved in 25 ml
Differential cytotoxicity was observed in BF versus of dioxane (Dried), followed by SOCl2 (0.6 ml,
GNM and TSCCa cell lines in the presence of nine 8.2 mmol). The mixture was heated to reux (1008C
synthetic CAPE-like analogues. oil bath) for 1 h, and then phenethyl alcohol (1.0 ml,
8.4 mmol) was introduced dropwise. The heating was
continued for 1 h, then the resulting solution was
2. Material and methods concentrated in vacuo. The crude product was
subjected to column chromatography (Silica EA:hex-
2.1. Chemicals
ane 1:1) to give the desired CAPE 1 (1.37 g, 86 %).
Compounds 27 were synthesized according to the
Phenethyl 3-(3,4-dihydroxyphenyl)acrylate (CAP-
procedure described above (Table 1). These
E), phenethyl 3-(3,4-dimethoxyphenyl)acrylate (PE-
compounds were analyzed by 1H NMR, 13C NMR,
DMC), methyl 3-(3,4-dihydroxyphenyl)acrylate (M-
and mass spectrum.
C), ethyl 3-(3,4-dihydroxyphenyl)acrylate (EC), phe-
Analytical data for selected compounds. 3: mp 158-
nethyl 3-(4-bromophenyl)acrylic (BrCAPE), ethyl 3-
1598C (lit. 1591608C); 1H NMR (400 MHz, DMSO-
(3,4-dimethoxyphenyl)acrylate (6), ethyl 3-(4-meth-
d6): d 3.75 (s, 3 H), 6.32 (d, J 8:2 Hz, 1 H), 6.90 (d,
oxyphenyl)acrylate (7) were synthesized by our de-
J 4:0 Hz, 1 H), 7.09 (d, J 4:0 Hz, 1 H), 7.20 (s, 1
veloped procedure. Ethyl 2,3-dibromo-3-(3,4-dim-
H), 7.57 (d, J 8:2 Hz, 1 H), 8.34 (s, 1 H), 8.53 (s, 1
ethoxyphenyl) propiolate (8) and ethyl 2,3-dibromo-
H); 4: m.p. 1421438C (lit. 149.58C); 1H NMR (200
3-(4-methoxyphenyl) propiolate (9) were prepared by
MHz, DMSO-d6 and CDCl3): d 1.24 (t, J 7:1 Hz, 3
a literature procedure [15]. DMEM, RPMI, penicillin-
H), 4.13 (q, J 7:1 Hz, 2 H), 6.13 (d, J 15:9 Hz, 1
streptomycin, trypsin-EDTA, dulbeccols phosphate
H), 6.73 (d, J 8:1 Hz, 1 H), 6.84 (dd, J 8:1, 1.9
buttered saline, MTT (3-[4,5dimethylthiazol2yl]-
2,5-diphenyl tetrazolium bromide) were purchased
Table 1
from Sigma Chemical Company. BF, OSF, GNM,
CAPE-like analogues prepared.
and TSCCA cells were obtained from Yi-Chao
Chang, Chung Shan Medical and Dental College at Product Method Reaction time % Yield
Taichung in Taiwan. Solvents (E. Merck Co., Darm-
1 TsOH Cata 34 days 40
stadt, Germany) and other reagents or dishes (Nunc, 1 DCC/DMAP 2 days 46
Denmark) were obtained from the indicated suppliers. 1 Nitrobenzene 8h 50
1 One pot 2h 86
2.2. Synthesis and identication of the CAPE-like 2 One pot 2h 92
analogues (19) 3 One pot 2h 95
4 One pot 2h 98
Melting points were determined on a Mel Temp II 5 One pot 2h 88
6 One pot 2h 98
melting point apparatus and are uncorrected. 1H NMR
7 One pot 2h 96
spectra were taken in CDCl3 unless otherwise indi-
 Y.J. Lee et al. / Cancer Letters 153 (2000) 5156
Hz, 1 H), 6.97 (d, J 1:9 Hz, 1 H), 7.41 (d, J 15:9
Hz, 1 H); 13C NMR (200 MHz, DMSO-d6 and CDCl3)
d 14.0, 59.5, 113.9, 114.2, 115.5, 121.0, 125.5, 144.6,
145.3, 148.0, 166.5; 6: m.p. 48498C (lit. 568C); 1H
NMR (200 MHz, CDCl3): d 1.32 (t, J 7:0 Hz, 3 H),
3.73 (s, 6 H), 4.19 (q, J 7:0 Hz, 2 H), 6.39 (d, J
16 Hz, 1 H), 6.61 (d, J 9:0 Hz, 1 H), 7.05 (s, 1 H),
7.10 (d, J 9:0 Hz, 1 H), 7.64 (d, J 16 Hz, 1 H); 7:
m.p. 49508C (lit. 48498C); 1H NMR (400 MHz,
CDCl3): d 1.32 (t, J 7:0 Hz, 3 H), 3.81 (s, 3 H),
4.24 (q, J 7:0 Hz, 2 H), 6.29 (d, J 16 Hz, 1 H),
6.88 (d, J 9:0 Hz, 2 H), 7.46 (d, J 9:0 Hz, 2 H),
7.63 (d, J 16 Hz, 1 H). Satisfactory combustion
analysis were obtained for 1, 2, 4, and 5.
Ethyl 2,3-dibromo-3-(3,4-methoxyphenyl)propio-
nate (8). The literature procedure was followed [15].
To a solution of 6 (2.0 g, 8.4 mmol) in 30 ml of CHCl3
at 08C was added Br2 (0.4 ml, 8.4 mmol) dropwise.
The mixture was warmed and stirred at room tempera-
ture for 1 h. The solvent was evaporated and the resi-
due was subjected to column chromatography (SiO2,
ether/hexane 1:1) to gave 8 (3.3 g, 99%) as a white
solid: m.p. 1101118C (lit. 1071088C); 1H NMR
(400 MHz, CDCl3): d 1.37 (t, J 7:2 Hz, 3 H),
3.88 (s, 3 H), 3.90 (s, 3H), 4.34 (q, J 7:2 Hz, 2
H), 4.81 (d, J 11:6 Hz, 1 H), 5.33 (d, J 11:6
Hz, 1 H), 6.84 (d, J 8:4 Hz, 1 H), 6.88 (d, J 2:4
Hz, 1 H), 6.95 (dd, J 8:4, 2.4 Hz, 1 H); 13C NMR d
14.0, 47.4, 51.6, 55.9, 56.0, 62.6, 110.5, 110.8, 120.8,
129.7, 149.0, 149.6, 167.5.
Ethyl 2,3-dibromo-3-(4-methoxyphenyl)propionate
(9). Compound 9 (4.0 g, 99 %) was obtained from 7
(2.3 g, 11 mmol) as a white solid according to the
procedure described above: m.p. 1071088C (lit.
1111128C); 1H NMR (200 MHz, CDCl3): d 1.33
(t, J 7:1 Hz, 3 H), 3.77 (s, 3 H), 4.31 (q, J 7:10
Hz, 2 H), 4.78 (d, J 11:9 Hz, 1 H), 5.32 (d, J 11:9
Hz, 1 H), 6.86 (d, J 6:7 Hz, 2 H), 7.29 (d, J 6:7
Hz, 2 H); 13C NMR (200 MHz,CDCl3): d 13.8, 47.4,
51.1, 55.3, 62.5, 114.2, 129.3, 129.6, 160.2, 167.8.
2.3. Cell growth
BF, OSF, and TSCCa cells were grown containing
DMEM medium. GNM cells were cultured
using RPMI medium in 75 cm 2 ask. The cells
were incubated at 378C in 5% CO2 humidied atmo-
sphere. BF, OSF, GNM and TSCCa cells had a
53
doubling time of about 31 34 h under these condi-
tions. Cells were inoculated every 3 days to maintain
their normal growth.
2.4. MTT assay
The inhibitory effect of CAPE-like agents on the
growth of oral cells was assessed with a MIT assay
according to method by Alley et al. [16]. Twenty
microliters DMSO or different concentrations of
CAPE-like agents in DMSO were added in a series
of 96-well culture plates, and cell cultures were incu-
bated at 378C in 5 % CO2 humidied atmosphere for
48 h. Cell line growth and growth inhibition were
measured at 550 nm with a UV scanning spectrophot-
ometer (Shimadzu UV-260).
In brief, cells in triplicate (3 wells) were plated into
a series of 96-well culture plates (100 ml culture/well)
at a cell density of 2 104 cells/ml. Following a 24-h
incubation at 378C, 5 % CO2, 100 % relative humidity,
100 ml of culture medium, culture medium containing
drug was dispensed within appropriate wells.
2.5. Statistical analysis
The results are reported as means ^ standard devia-
tions from individual determinations. Statistical
differences were analyzed according to Student's t-
test; signicant differences were established at
P , 0:05.
3. Results and discussion
Our previous synthetic method was performed with
nitrobenzene to provide CAPE in 50% yield. It was
worth noting that although reactions described could
be regarded, as straightforward, manipulations were
sometimes difcult due to the synthesis and solubility
of the acid chloride. In addition, there are health risks
from toxic solvents, such as benzene and nitroben-
zene. Furthermore the methods required substantial
amounts of alcohol, reagents, and solvents for ester-
ication. To overcome these technical difculties, we
have developed a concise synthetic strategy for the
synthesis of CAPE-like analogues. Our results gave
a practical route in comparison with others (Table 1).
CAPE-like analogues have drawn our attention
because of their potent chemoprevention and anti-
54 Y.J. Lee et al. / Cancer Letters 153 (2000) 5156
Fig. 1. MTT assay of CAPE at various concentration in BF, OSF,
GNM, TSCCa cells.
tumor properties. Recently we have investigated the
antioxidant activity of CAPE, and the result indicated
that CAPE could function as a radical inhibitor,
bleaching 93.4% of DPPH (1,1-diphenyl-2-picrylhy-
drazyl) at 2 mM [1]. In addition, Su et al. [17] reported
that CAPE suppressed oncogene expression and
inhibited tumor cell growth. A number of studies
have shown that CAPE and its analogues exhibited
preferential cytotoxicity toward tumor cells [11] and
inhibited HIV-1 integrase [18]. In our investigation,
the cytotoxicities of nine agents on the growth of BF,
OSF, GNM, and TSCCa were studied by the MTT
assay. Treatment with 25200 mM CAPE and 50
400 mM EC shown signicant cytotoxicity on OSF,
GNM, and TSCCa cells, but not on BF cells (Fig. 1,2).
Signicant cytotoxicity was exhibited at 100 mM EC
Fig. 2. MTT assay of EC at various concentration in BF, OSF,
GNM, TSCCa cells.
Fig. 3. MTT assay of MC at various concentration in BF, OSF,
GNM, TSCCa cells.
and CAPE, OSF, GNM, and TSCCa cell growth was
inhibited compared with normal cell BF when
concentration of EC and CAPE was 100 mM. The
cell growth of BF, OSF, GNM, and TSCCa were
inhibited by 0, 57, 73, 54% at 100 mM EC, respec-
tively. In 100 mM CAPE, cell survival of BF, OSF,
GNM, and TSCCa was 82, 36, 44, and 55%, respec-
tively.
As shown in Fig. 3, MC cytotoxicity was similar to
CAPE at 200 mM MC. However, normal BF cells had
65% survival. It is possible that methanol, a metabolic
product of MC, showed dose-dependent inhibition
activity on BF cell. BrCAPE expressed specically
against GNM carcinoma cells (Fig. 4). At 400 mM
BrCAPE, the cell survival growth of BF, OSF,
GNM, and TSCCa was 90, 100, 35, and 76%, respec-
Fig. 4. MTT assay of BrCAPE at various concentration in BF, OSF,
GNM, TSCCa cells.
 Y.J. Lee et al. / Cancer Letters 153 (2000) 5156
Fig. 5. MTT assay of PEDMC at various concentration in BF, OSF,
GNM, TSCCa cells.
tively. The reasons why BrCAPE exerts preferential
cytotoxicity on GNM carcinoma cell are still not
clear. The cytotoxicities of PEDMC and agents 69
on the growth of BF, OSF, GNM, and TSCCa cells
treated with 25200 mM and 50400 mM, respec-
tively, are shown in Figs. 59. Based on the MTT
assay results none of the analogs of CAPE are better
than the CAPE or EC in inducing the cytotoxicity on
the oral cancer cells. This implies that ortho bis-
hydroxylation and a tethered conjugated double
bond are required for signicant inhibitory potency.
Grunberger et al. [11] have shown that CAPE could
efciently inhibit incorporation of [ 3H]thymidine into
the DNA of Fisher rat embryo broblast (CREF),
Fig. 6. MTT assay of 3-(3,4-dimethoxyphenyl) acrylic acid
phenethyl ester in BF, OSF, GNM, TSCCa cells.
55
Fig. 7. MTT assay of 3-(4-methoxyphenyl) acrylic acid ethyl ester
in BF, OSF, GNM, TSCCa cells.
human breast carcinoma (MCF-7), and melanoma
(SK-MEL-28 and SK-MEL-170) cell lines in culture.
Further investigations should address how CAPE, EC,
and BrCAPE affect cytodical activities in tumor and
normal cells.
In conclusion, a rapid route to CAPE 1 and its
derivatives has been achieved. Moreover, the ready
accessibility of CAPE-like analogues will simplify
further investigations into its mode of action, and
may lead to an understanding of the observed differ-
ential effects on a molecular level. Further in vivo
assays are currently underway to evaluate the efcacy
of these compounds as chemopreventive agents
against oral cancer.
Fig. 8. MTT assay of 2,3-dibromo-3(3,4-dimethoxyphenyl) propio-
nic acid ethyl ester in BF, OSF, GNM, TSCCa cells.
56 Y.J. Lee et al. / Cancer Letters 153 (2000) 5156
Fig. 9. MTT assay of 2,3-dibromo-3-(4-methoxyphenyl) propoinic
acid ethyl ester in BF, OSF, GNM, TSCCa cells.
Acknowledgements
The authors are indebted to Dentist Yi-Chao Chang
for the generous gift of BF, OSF, GNM and TSSCa
cells. For nancial support, we thank the National
Science Council of the Republic of China (Research
Grant NSC 87-2113-M-040-001).
References
[1] W.K. Chen, C.F. Tsai, P.H. Liao, S.C. Kuo, Y.J. Lee, Synth-
esis of caffeic acid esters as antioxidants by esterication via
acyl chlorides, Chin. Pharmacol. J. 51 (1999) 271278.
2 C.C. Yang, New target for anti-retroviral therapy, Abstracts of
AIDS Symposium at Chung Shan, Taiwan, February 1999.
[3] A. Kujumgiev, V.B. Bankova, A. Ignatova, S. Popov, Anti-
bacterial activity of propolis some of is compounds and their
analogs, Pharmazie 48 (1993) 785786.
[4] J. Focht, S.H. Hansen, J.V. Nielsen, A. van den Berg-Segers,
R. Riezler, Bactericidal effect of propolis in vitro against
agents causing upper respiratory tract infections, Arzneim.-
Forsch. 43 (1993) 921923.
[5] M.T. Khayyal, M.A. El-Ghazaly, A.S. El-Khatib, Mechanisms
involved in the antiinammatory effect of propolis extract,
Drugs Exp. Clin. Res. 19 (1993) 197203.
[6] C.V. Rao, D. Desai, B. Simi, N. Kulkarni, A. Amin, B.S.
Reddy, Inhibitory effect of caffeic acid esters on azoxy-
methane-induced biochemical changes and aberrant crypt
foci formation in rat colon, Cancer Res. 53 (1993) 41824188.
[7] O.K. Mirzoeva, G.F. Sud'ina, M.A. Pushkareva, G.A. Korshu-
nova, N.V. Sumbatian, S.D. Varfolomeev, Lipophilic deriva-
tives of caffeic acid as lipoxygenase inhibitors with
antioxidant properties, Bioorg. Khim. 21 (1995) 143151.
[8] C.V. Rao, D. Desai, A. Rivenson, B. Simi, S. Amin, B.S.
Reddy, Chemoprevention of colon carcinogenesis by pheny-
lethy1-3-methylcaffeate, Cancer Res. 55 (1995) 23102315.
[9] Z.Z. Su, J. Lin, D. Grunberger, P.B. Fisher, Growth suppres-
sion and toxicity induced by caffeic acid phenethyl ester
(CAPE) in type 5 adenovirus-transformed rat embryo cells
correlate directly with transformation progression, Cancer
Res. 54 (1994) 18651870.
[10] K. Frenkel, H. Wei, R. Bhimani, J. Ye, J.A. Zadunaiasky,
M.T. Huang, T. Ferraro, A.H. Conny, D. Grunberger, Inhibi-
tion of tumor promoter-mediated processes in mouse skin and
bovine lens by caffeic acid phenethyl ester, Cancer Res. 53
(1993) 12551261.
[11] D. Grunberger, R. Banerjee, K. Eisinger, E.M. Oltz, L. Efros,
M. Caldwell, V. Estevez, K. Nakanishi, Preferential cytotoxi-
city on tumor cells by caffeic acid phenethyl ester isolated
from propolis, Experientia 44 (1988) 230232.
[12] C.V. Rao, C. Desai, B. Kaul, S. Amin, B.S. Reddy, Effect of
caffeic acid esters on carcinogen-induced mutagenicity and
human colon adenocarcinoma cell growth, Chem.-Biol.
Interac. 84 (1992) 277290.
[13] T.R. Burke, M.R. Fesen, A. Mazumder, J. Wang, A.M.
Carothers, Hydroxylated aromatic inhibitors of HIV-1 inte-
grase, J. Med. Chem. 38 (1995) 41714178.
[14] J.H. Chen, Y. Shao, M.T. Huang, C.K. Chin, C.T. Ho, Inhibi-
tory effect of caffeic acid phenethyl ester on human leukemia,
Cancer Lett. 108 (1996) 211214.
[15] E.N. Jacobsen, L. Deng, Y. Furukawa, L.E. Martinez, Enan-
tioselective catalytic epoxidation of cinnamate esters, Tetra-
hedron 50 (1994) 43234334.
[16] M.C. Alley, D.A. Scudiero, A. Monks, M.L. Hursey, M.J.
Czerwinski, D.L. Fine, B.J. Abbott, J.G. Mayo, R.H. Shoe-
maker, M.R. Boyd, Feasibility of drug screening with panels
of human tumor cell lines using a microculture tetrezolium
assay, Cancer Res. 48 (1988) 589601.
[17] Z.Z. Su, J. Lin, M. Prewett, N.Z. Goldstein, P.B. Fisher, Apop-
tosis mediates the selective toxicity of caffeic acid phenethyl
ester (CAPE) toward oncogene-transformed rat embryo bro-
blast cells, Anticancer Res. 15 (1995) 18411848.
[18] M.R. Fesen, Y. Pommier, F. Leteurtre, S. Hiroguchi, J. Yung,
K. Kohn, Inhibition of HIV-1 integrase by avones, caffeic
acid phenethyl ester (CAPE) and related compounds,
Biochem Pharmacol. 48 (1994) 595608.