[CANCER RESEARCH 55, 3576-3583,
August 15, ¡995]
Apoptosis and Altered Redox State Induced by Caffeic Acid Phenethyl Ester
(CAPE) in Transformed Rat Fibroblast Cells1
Chia Chiao, Adelaide M. Carothers, Dezider Grunberger, Gregory Solomon, Gloria A. Preston, and J. Carl Barrett2
Laboratory of Molecular Carciiiogenesis, National Institute of Environmental Health Sciewes, Research Triangle Park, North Carolina 27709 1C. C., G. S., G. A. P., J. C. B.j; Institute of Cancer
Research, College of Physicians and Surgeons, Columbia-Presbyterian Cancer Center and School of Public Health, Columbia University, New York, New York 10032 ¡A.M. C., D. G.]
ABSTRACT
Caffeic acid phenethyl ester (CAPE), which is derived from the propolis
of bee hives, was shown previously to block tumor promoter- and carcin
ogen-generated oxidative processes in several assays and to engender
differential toxicity to some transformed cells. To study the mechanisms of
CAPE-induced differential cytotoxicity, nontumorigenic rat embryo fibroblasts (CREF) and adenovirus (type 5)-transformed CREF cells (Wt3A)
were used. As shown by nucleosomal-length DNA degradation, morpho
logical alterations by electron microscopy, in situ labeling of .V-OII ends,
and the appearance of a hypodiploid cell population by bivariant flow
cytometry, cell death induced by CAPE in the transformed Wt3A cells was
apoptosis. Under the same CAPE treatment condition, CREF cells tran
siently growth arrested. Both CREF and Wt3A cells were radioresistant,
suggesting deficiencies in the proteins controlling the G, checkpoint. To
explore possible mechanisms of CAPE-induced apoptosis, it was deter
mined whether CAPE-induced toxicity was influenced by the redox state
of the cells. Depletion of cellular glutathione (GSH) with buthionine
sulfoximine before CAPE treatment caused CREF sensitive to CAPEinduced cell death. GSH levels were also determined in CAPE-treated
CREF and Wt3A cells. The GSH level in the CREF cells was unaffected
by CAPE, whereas the Wt3A cells showed a significant reduction. When
the GSH levels were increased in Wt3A cells by treatment with the
reducing agent, JV-acetyl-cysteine before CAPE treatment, the Wt3A cells
were partially rescued. Furthermore, Bcl2, which protects cells from
oxidative stress, had a protective effect against CAPE-induced apoptosis
in Wt3A cells. Finally, the sensitivity of Wt3A cells to a known oxidant,
hydrogen peroxide (H2O2), was examined. Wt3A cells were killed by
H2O2-induced apoptosis, whereas CREF cells remained resistant. When
Wt3A cells were treated with catalase, a cellular enzyme that inactivates
H2O2, CAPE-induced apoptosis in Wt3A cells was reduced, further prov
ing that Wt3A cells were more sensitive than CREF cells to oxidative
stress. These results suggest that CAPE can modulate the redox state of
cells. Sensitivity of cells to CAPE-induced cell death may be determined by
the loss of normal redox state regulation in transformed cells.
INTRODUCTION
Propolis is exuded from the bark of conifer trees and carried by
honeybees to the hives. Known for the variety of its beneficial effects,
it has been a popular folk medicine through the ages. Included among
the medicinal properties of propolis are anti-inflammatory, antiviral,
immunostimulatory, and carcinostatic activities (1). CAPE3 is a bio
logically active ingredient of honeybee propolis; its structure is de
picted in Fig. 1. In some cases, CAPE exhibits differential toxicity to
cancer cells versus normal cells. For example, the growth of type 5
Received 3/20/95; accepted 6/19/95.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
1This investigation was partially supported by the Lucille P. Markey Charitable Trust
(A. M. C. and D. G.)
2 Laboratory of Molecular Carcinogenesis,
National Institute of Environmental Health
Sciences, P. O. Box 12233, Research Triangle Park, NC 27709.
3 The abbreviations used are: CAPE, caffeic acid phenethyl ester; BrdUrd, bromodeoxyuridine; BSO, buthionine sulfoximine; CREF, cloned rat embryonic fibroblasts; ddUTP, digoxigenin-conjugated dUTP; GSH, glutathione; H,O,, hydrogen peroxide; NAC,
W-acetyl-cysteine; REF, rat embryo fibroblasts; ROS, reactive oxygen species; TPA,
12-O-teiradecanoylphorbol-13-acetate;
TdT, terminal deoxynucleotidyl transferase; EM,
electron microscopy.
adenovirus-transformed
rat embryo fibroblasts (Wt3A) is differen
tially inhibited by CAPE compared to the nontumorigenic, diploid
parental cells (CREF; Ref. l). Subsequent studies have shown that
CAPE-mediated growth response depends on the transformed phenotype per se (2-4). In addition, the growth of numerous human tumor
cell lines are also suppressed by CAPE treatment. Screening of cancer
cell lines by the National Cancer Institute for IC50 values of CAPEinduced killing showed that effective molar concentration
1.4 X 1(T5 to 5 X 10~7.4
are from
CAPE-associated growth inhibition may relate to effects on oxida
tive processes induced by mitogenic stimuli. Control of cell prolifer
ation in a variety of mammalian cell types is mediated by the binding
of cytokines, growth factors, and hormones to specific cell-surface
receptors, which in turn leads to the generation of O2~ and H2O2
(reviewed in Ref. 5). Tumor promoting agents may stimulate cell
proliferation by increasing the intracellular production of ROS (re
viewed in Ref. 6). The mitigation of tumor promoter- or carcinogenmediated oxidative processes by CAPE was observed by measuring
myeloperoxidase activity of TPA-treated polymorphonuclear leuko
cytes in mouse skin, catalase activity, and intracellular 2'7'-dichlorofluorescin (DCFH) fluorescence to quantitate H2O2 production in
neutrophils and HeLa cells, formation of oxidized bases in DNA of
HeLa cells by HPLC, and nucleoside postlabeling analyses (7, 8), and
production of azoxymethane-induced lipoxygenase metabolites 8(S)and 12(5)-hydroxyeicosatetraenoic
acid (9). CAPE was observed to
inhibit oxidative processes to an extent better than or comparable to
other chemopreventive agents such as tamoxifen, (—)epigallocatechin
gallate, sarcophytol A, and pento-O-galloyl-ß-o-glucose (8). Further
more, lens opacification resulting from oxidative stress during cataract
induction, was reduced by CAPE with the use of both bovine lenses
(7) and whole animals.5
In the present study, we investigated whether CAPE-induced dif
ferential growth effects correlated with a selective killing of Wt3A
cells by apoptosis. A variety of factors can sensitize cells to die by
apoptosis, including expression of oncogenes that deregulate growth,
presence of functionally active p53 protein, suppression of negative
regulators of apoptosis (e.g., Bcl-2), sustained loss of calcium homeostasis, signal transduction inhibition, trophic factor withdrawal, and
oxidative stress and/or redox imbalance (reviewed in Ref. 10). Like
wise, an array of exogenous agents that cause DNA damage can also
induce cell death by apoptosis such as chemicals (11, 12) and ionizing
radiation (13). Regarding the former mediators, many agents that
either induce or rescue cells from apoptosis affect redox changes in
cells through direct or indirect means (reviewed in Ref. 14). The
proliferative capacity of cells depends on the integration of multiple
growth signaling pathways that collectively establish the cellular
redox status. In the present study, we observed that CAPE altered the
redox state of treated CREF and Wt3A cells. Differential CAPEtriggered apoptosis in the Wt3A cells was associated with reduced
oxidant defenses in the viral transformed Wt3A cells and predicts a
similar phenotype in drug-sensitive human tumor cells.
4 NCI, unpublished data.
5 Grunberger et al., unpublished data.
3576
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.
DIFFERENTIAL
CAPE-INDUCED
APOPTOSIS
IN TRANSFORMED
CELLS
EM. Electron microscopic analysis was performed as described previously
(17) on collected cells that were washed in a 4°C0.1 M PIPES buffer and fixed
in a mixture of 2% paraformaldehyde and 2% glutaraldehyde. The fixed cells
were embedded in Epon (Polysciences Inc., Warrington, PA) and stained with
5% uranyl acetate and 2.1% lead citrate. Sections were examined by using a
Phillips 400 transmission electron microscope.
TdT Assay for Apoptosis. Wt3A cells treated with or without CAPE
were incubated for 8 h, collected, and fixed in 1% paraformaldehyde.
The
ApopTag kit of Oncor (Gaithersburg,
MD) was used according to the
manufacturer's protocol for labeling with d-dUTP by TdT and reaction with
Fig. 1. Structure of CAPE.
CREF -CAPE (o)
fluorescein-conjugated
anti-digoxigenin
antibody. Cytocentrifuged
cells
(5-7 X IO5) were examined at a magnification of X400 with a Nikon
Optiphot microscope.
DNA Synthesis and Cell Cycle Analysis. Bivariant flow cytometry was
performed on cells grown in the presence or absence of CAPE for various
times and on cells exposed to 4Gy Rad -y-irradiation. After treatments, cells
WtSA
-CAPE (o)
were labeled with 10 mM BrdUrd for 30 min, fixed in 70% ethanol, and
incubated with fluorescein-conjugated anti-BrdUrd antibody (Becton Dickin
son, San Jose, CA) to identify the S-phase fraction (labeled green). Also,
O)
O
CREF+CAPE (•)
propidium iodide was used to stain total cellular DNA (labeled red) before
scanning with a fluorescence-activated
FACScan (Becton Dickinson). This
method was also applied to examine primary REF, CREF, and Wt3A cells 24
h after exposure to 4 Gy y-irradiation.
GSH Assay. Cellular levels of GSH were measured by using an enzymatic
assay kit (GSH-400) from Bioxytech (Paris, France). Pelleted cells were
Wt3A +CAPE (•) processed
resuspended
centrifugation at 3000 rpm for 10 min and assayed for GSH activity by using
the manufacturer's protocol.
Time (h)
Statistical Analysis. All numerical experimental data were expressed as
mean ± SD. Paired two-sample for means of Student t test was used to
Fig. 2. Growth of CREF and Wi3A cells in the presence or absence of CAPE. The
relative cell numbers reflect only the attached cells.
MATERIALS
in a homogenizer after having been rinsed once with PBS and
in 5% meta-phosphoric acid. Supernatant was retained after
determine the significance of difference.
AND METHODS
RESULTS
CAPE Induces Growth Arrest in CREF Cells and Apoptosis in
Transformed Wt3A Cells. It was reported previously that CAPE
treatment induces dose-dependent differential growth inhibition after
a 72-h incubation in transformed murine Wt3A cells compared with
nontumorigenic, diploid CREF cells (1). We evaluated the growth
inhibitory effect in these cells at early time intervals after the additions
of a single relatively low dose of CAPE (1 /xg/ml). As shown in Fig.
Chemicals. CAPE was synthesized as described (1). Catalase was pur
chased from Sigma Chemical Co. (St. Louis, MO).
Cell Lines and Cell Culture Conditions. Fischer rat CREF and Wt3A
cells (15) were maintained in DMEM supplemented with 10% fetal bovine
serum, and cultured at 37°Cin an incubator containing 10% CO2. Growth in
the presence of NAC used medium buffered with 15 mM HEPES (GIBCOBRL, Gaithersburg, MD). Normal primary REF were isolated from 12- to
16-day-old Fischer rat embryo and maintained in the same growth medium.
Experiments used early passage (3—7)REF cells.
BcI2 Expression Vector Construction and Transfection. A human 1.9
kb-fcc/2 cDNA gene (kindly provided by Dr. Stanley Korsmeyer, Washington
CREF
University, St. Louis, MO) containing all bcl2 coding sequences, was inserted
into the Ec'oRI cloning site of a Moloney murine leukemic retroviral vector,
Time (h)
+CAPE
pLXSHD (16). The orientation of the insertion was determined by using
BamHl digestion and was confirmed by direct sequencing. Vectors with either
sense or antisense orientation were electroporated into Wt3A cells, and the
resistant cells were selected in growth medium containing 10 mM histidinol.
Electroporation was performed by using the Gene pulser (BRL) with 300 V at
a capacitance of 960 fi.F.
Cell Growth Analysis. Cells were seeded at 6000 cells/cm2 cell density
and incubated overnight before CAPE was added to one-half of the cultures.
All experiments used a dose of 1 ng/ml CAPE. At various times after
treatment, the attached cells were trypsinized, and cell numbers were deter
mined by a Coulter counter.
DNA Gel Electrophoresis. Cultured cells were scraped off the culture
dishes, combined with detached cells, and sedimented. Washed cell pellets
were resuspended in cell lysis buffer [10 mw EDTA, 50 mM Tris (pH 8.0),
0.5% sodium lauryl sarcosine, and 0.5 mg/ml proteinase K] and incubated at
55°Cfor 2 h. RNase A was added at a concentration of 0.5 mg/ml, followed
by another 2 h of incubation. DNA was phenol extracted and ethanol precip
itated overnight at —
70°C.After centrifugation, the precipitated DNA was
solubilized in H,0, and electrophoresed
ethidium bromide at 7 V/cm electrical
visualized by UV fluorescence.
in 1.6% agarose pre-impregnated
field. After separation,
with
DNA was
Wt3A
8
9
10
Fig. 3. Agarose gel electrophoresis of fragmented and intact DNA from CAPE-treated
CREF and Wt3A cells. Lane 1, DNA of untreated CREF cells; Lanes 2-5, DNA of CREF
cells exposed to 1 ng/ml of CAPE obtained after 2-, 4-, 6-, and 8-h incubation times,
respectively; Lane 6, DNA of untreated Wt3A cells; Lanes 7-10, DNA of similarly treated
Wt3A cells incubated for 2, 4, 6, and 8 h. respectively.
3577
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.
DIFFERENTIAL
CAPE-INDUCED
APOPTOSIS
2, in the absence of CAPE, cell numbers of CREF and Wt3A cells
increased steadily. CAPE treatment reduced the rate of growth of
CREF cells almost completely, but the cells remained viable. In
contrast, Wt3A cells were killed with <10% of cells viable at 8 h after
CAPE addition. Within 24 h, essentially all cells were detached from
culture dishes. This experiment demonstrated that the differential
growth effects in CAPE-exposed cultures were induced rapidly (with
in 4-8 h), and that growth rates of both cell types were affected by the
drug.
Wt3A cells exposed to CAPE for 8 h did not exclude trypan blue
(data not shown) and were not viable. To determine whether CAPEinduced death of Wt3A cells was due to apoptosis or necrosis, DNA
of these cells was examined for the appearance of nucleosomal length
DNA fragments (18). Gel electrophoresis was performed on DNA
samples from CREF and Wt3A cells exposed to CAPE for 2-8 h (Fig.
3). DNA isolated from control cells, and CAPE-treated CREF cells
did not show any DNA degradation. A nucleosomal length DNA
ladder was detected only in Wt3A cells treated with CAPE. Initial
appearance of DNA degradation occurred 4 h after drug treatment and
was extensive by 8 h.
Additional markers of apoptotic cell death were also examined.
Control CREF and Wt3A cells exhibited normal morphology under
EM (Fig. 4, A and C). CREF cells treated with CAPE were also
normal in appearance (Fig. 40). Typical features of apoptosis (19)
were evident in CAPE-treated Wt3A cells at 8 h (Fig. 4D), includ
ing cell rounding, loss of cell processes, chromatin fragmentation
and condensation, and persistence of normal cytoplasmic organelles. The effect of CAPE on Wt3A cells was also analyzed by
in situ terminal deoxynucleotidyl transferase labeling of 3'-hydroxy DNA ends with d-dUTP, followed by immunohistochemical
¡mageanalysis, another specific assay for apoptosis (20). Only the
8-h, CAPE-treated Wt3A cells incorporated the fluorescent label
(Fig. 5, A and B).
CAPE treatment of human tumor cell lines was shown previ
ously to inhibit DNA synthesis as measured by incorporation of
IN TRANSFORMED
CELLS
[3H]thymidine into acid insoluble material (1). To determine the
effect of CAPE on DNA synthesis and to correlate the drug's effect
on cell cycle progression at early time intervals (4 and 8 h),
bivariant flow cytometry was performed by using BrdUrd-labeled
CREF and Wt3A cell populations. Cells exposed to CAPE were
incubated for different times and then pulse labeled with 10 mM
BrdUrd for 30 min. The proportion of cell populations in S phase
(i.e., BrdUrd-positive staining) was compared with the DNA con
tent of cells (i.e., PI staining). Consistent with inhibition of DNA
synthesis, a reduction in the S-phase fraction of CAPE-treated
CREF cells occurred over time. Among control CREF cells in log
phase growth, 35.4 ±1.8% of the cells were in S phase (Fig. 6A).
Exposure to CAPE for 4 and 8 h caused the number of BrdUrdpositive cells to decline to 3.6 ±3.3% and 8.0 ±1.1% of the
population, respectively (Fig. 6, ß
and C). DNA synthesis of CREF
cells resumed 24 h after CAPE treatment (data not shown). A
similar proportion of the Wt3A population in log phase growth
were in S phase of the cell cycle (34.1 ±1.0%) as was observed
with CREF (Fig. 6D). However, the incorporation of BrdUrd in
Wt3A cells stopped entirely after 4- and 8-h incubation in the
presence of CAPE. In addition, at the 4-h interval a fraction of
CAPE-treated Wt3A cells emerged that contained less than Gr
phase DNA content. The appearance of a hypodiploid cell popu
lation by flow cytometry is another indicator of apoptosis (21).
Defective G, Checkpoint Control in Response to DNA Damage
in Both CREF and Wt3A Cells. Using the BrdUrd-labelingand
bivariant flow cytometry approach described above, we further tested
the growth characteristics of the CREF and Wt3A cells after DNA
damage. Primary REF, CREF, and Wt3A cells were y irradiated (4
Gy), and the percentages of the cells in S and G, were measured 24
h after treatment (Table 1). The REF cells growth arrested in G, as
indicated by the decrease in S-phase cells and increase in G, cells.
However, neither CREF nor Wt3A cells displayed a G, arrest after
irradiation, indicating that both cell types have defects in G, check
point control. G, arrest in response to DNA damage induced by
Fig. 4. Electron micrographs of CAPE-treated
and untreated CREF and Wt3A cells. A, untreated
CREF cells; fi, CREF cells treated with CAPE for
8 h; C, untreated Wt3A cells; D, Wt3A cells treated
with CAPE for 8 h, showing chromatin condensa
tion and fragmentation, loss of cell processes, and
cell rounding. Magnification of all the electron
micrographs, X 1400.
3578
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.
DIFFERENTIAL
CAPE-INDUCED
APOPTOSIS
IN TRANSFORMED
CELLS
BSO. The reduction in cell number was dependent on the dose of
BSO. When CREF cells were treated with 25 /J.M BSO for 14 h,
followed by CAPE addition, the majority of the cells detached
from culture dishes within 25 h. When CREF cells that had been
pretreated with BSO, followed by the addition of CAPE, were
examined by EM, the morphological alterations of both apoptosis
necrosis were evident (data not shown). Thus, depletion of GSH by
BSO made CREF cells sensitive to CAPE-induced cell killing.
These results predicted that GSH levels in CREF and Wt3A cells
may be involved in their differential response to CAPE. To test this
hypothesis, GSH levels were quantitated by a colorometric assay in
CREF and Wt3A cells (Fig. 8). In untreated log phase cultures,
GSH levels were identical in both cell types. CAPE treatment had
little effect on the GSH level in CREF cells; however, a significant
reduction (—42%; P < 0.01) in the GSH level was observed in
Wt3A cells exposed to 1 /xg/ml CAPE for 4 h. Because this result
further associated CAPE sensitivity with a redox imbalance in
Wt3A cells, NAC, a glutathione precursor, was used to test if it
could rescue Wt3A cells from CAPE toxicity. As demonstrated by
the cell counts in Fig. 9, exposure of Wt3A cells to NAC for 1 h
before the addition of CAPE (1 ju.g/ml) partially rescued them. The
protective effect was dose dependent; a 50% rescue was achieved
with 5-10 mM NAC, and a 70% rescue resulted at 15 ITIMNAC
following an 8-h incubation with CAPE. DNA degradation was
also inhibited when CAPE-exposed Wt3A cells were pretreated
with NAC (data not shown). Additionally, the ability of CAPEtreated Wt3A cells to synthesize DNA by [3H]thymidine uptake
into TCA-precipitable material was partially rescued by a 1-h NAC
pretreatment in the above dose range, as well as viability by a
clonigenic assay (data not shown).
Bcl2 Protects Wt3A Cells from CAPE-induced Toxicity.
Because overexpression of Bcl2 has been shown to prevent apop
tosis in other cell systems through inhibition of ROS (26, 27), the
effect of Bcl2 expression in Wt3A cells was examined. When the
human ec/2 gene was transfected into Wt3A cells (Wt3A/bcl2-S),
cell killing by CAPE was substantially reduced after 8 h of treat
ment (Fig. 10/4). As has been noted in other systems, although
protection by Bcl2 was temporary and exposure to CAPE was for
a longer time (24 h), it still killed most of the Wt3A cell popula
Fig. 5. Photomicrographs of Wl3A cells labeled with digoxigenin-dUTP in situ by the
tion. Whereas Wt3A/bcl2-S cells escaped CAPE-induced
cell
TdT apoptosis assay. A, untreated Wt3A cells; B, Wt3A cells 8 h after CAPE treatment,
showing strong nuclear fluorescence.
death at the 8 h time, the entire population bearing the antisense
construct (Wt3A/bcl2-AS) was killed at this point. A low level of
spontaneous DNA degradation was evident in Wt3A/bcl2-AS cells
ionizing radiation depends on proteins such as p53 (22-24). These
untreated by CAPE (Fig. 10ß,far left lane). Although extensive
results suggest that the effector proteins in CAPE-induced growth
DNA fragmentation occurred subsequent to CAPE addition in
these transfectants, none was apparent at 4 or 8 h in Wt3A/bcl2-S
arrest and apoptosis may be different from those controlling the G,
checkpoint.
cells. A slight amount of DNA degradation was apparent in these
Redox State of the Cells Determines Sensitivity to CAPE. Pre
latter cells after 24 h (Fig. lOfl, far right lane). Wt3A cells
transfected with only the drug-resistant vector showed the same
vious studies showed that CAPE modulated oxidative stress im
posed by tumor promoter- or carcinogen-mediated
processes in sensitivity to CAPE as the Wt3A and Wt3A/bcl2-AS cells (data not
several systems (e.g., mouse skin, bovine lens, HeLa cells, and rat
shown). Taken together, expression of human Bcl2 partially res
colon; Refs. 7-9). To test the hypothesis that the specific redox
cued Wt3A cells from CAPE-induced cell killing.
H2O2-induced Differential Growth Effects in CREF and Wt3A
status in CREF and Wt3A cells influences CAPE sensitivity,
cellular levels of reduced GSH and free radical scavengers were
Cells Are Analogous to the Effects of CAPE. Because reduction or
enhancement of the defenses against oxidative stress could either
altered to examine whether immortal CREF cells could be sensi
tized to CAPE-induced apoptosis or whether Wt3A cells could be
increase the killing of CREF cells or the survival of Wt3A cells,
protected from CAPE-triggered
apoptosis. Thus, GSH pools in
respectively, we investigated whether treatment of these cells with an
CREF cells were depleted with BSO, an inhibitor of 7-glutamyl
oxidant (H2O2) yielded results similar to those of CAPE. Fig. HA
shows CREF and Wt3A cell counts at timed intervals after growth in
cysteine synthetase (25), for 14 h at doses of 5 or 25 JUM,and then
medium with or without 600 /XMH2O2. Values reflect only viable,
fresh medium with or without CAPE (1 /xg/ml) was added. As
attached cells that excluded trypan blue. As in the case of treatment by
shown in Fig. 7, a negligible change in cell counts resulted after
CAPE (Fig. 2), CREF cells remained viable in the presence H2O2, but
exposure to 25 /AMBSO alone for 14 h, but the cell counts dropped
the rate of CREF cell growth was inhibited. Exposure to H2O2
sharply after CAPE treatment of cells that were pretreated with
3579
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.
DIFFERENTIAL
CAPE-INDUCED
APOPTOSIS
IN TRANSFORMED
CELLS
Wt3A
CREF
(5
Ó"
600
200
800
400
600
Ohour
Fig. 6. Bivariant flow cytometery showing the fraction
of CREF and Wt3A cell populations synthesizing DNA
after treatment with CAPE and showing the population
distributions in the different cell cycle compartments. A,
untreated CREF cells showing normal population distri
bution for log phase growth; B and C, CREF cells ex
posed to CAPE for 4 and 8 h, respectively, showing
reduction of cells in S phase; D untreated log phase Wt3A
cells; E and F, Wt3A cells exposed to CAPE for 4 and 8
h, respectively, showing absence of cells in S phase and
appearance of hypodiploid cells.
«' »•
.
Less than G, DMA content
O
200
800
400
0
800
400
600
4 hours + CAPE
I-
Less lhan G1 DNAcontent
200
400
600
800
O
200
400
600
800
8hours
+CAPE
Anti-BrdU Fluorescence
severely affected Wt3A cells with <5% of these cells remaining after
24 h of treatment. No DNA degradation was evident in CREF cells
even after 24 h of treatment with H202. However, in similarly treated
Wt3A cells, fragmentation was apparent in DNA samples starting at
4 h (Fig. l Iß).
Because Wt3A cells are differentially sensitive to oxidant H2O2, the
possibility that CAPE may function through formation of intermediate
100000
-BSD, -CAPE (D)
2 5uM BSO 0/N, then -CAPE (o)
Anti-BrdU Fluorescence
intracellular oxidants was examined. Catalase scavenges H2O2 and
converts it to a less reactive species (e.g., H2O + O2). If CAPE
treatment of Wt3A cells induces the formation of intracellular oxi
dants such as H2O2, then simultaneous addition of catalase should
rescue these cells. Fig. 12A shows the number of attached Wt3A cells
counted at different times after addition of CAPE alone or in combi
nation with catalase (50 /¿/ml).After 24 h of treatment, CAPEinduced toxicity was substantially reduced in the presence of catalase.
Moreover, as shown in Fig. 12B, cotreatment with catalase inhibited
CAPE-initiated nucleosomal DNA degradation in Wt3A cells. DNA
fragmentation in CAPE-treated cells without catalase was more
extensive at 8 h than in catalase plus CAPE-treated cells at 24 h.
80000
DISCUSSION
=
1
U
«0
60000
40000
-
20000
-
-BSO, +CAPE (•)
5uM BSO 0/N, then +CAPE («)
25uM BSO 0/N, then +CAPE (•)
20
10
Time (h)
Fig. 7. GSH depletion by BSO influences sensitivity to CAPE in treated CREF cells.
Applying the growth conditions specified in "Materials and Methods," cells were allowed
to attach for 4 h before different concentrations of BSO were added into culture medium.
After overnight incubation (OfN), BSO was removed, and fresh medium containing
CAPE was added. Counts of attached cells were determined at timed intervals.
In this study, we showed that a natural plant product contained in
propolis, CAPE, caused time-dependent changes in the treated rat
cells at low doses. In immortal, parental CREF cells, exposure to
CAPE transiently blocked cell proliferation. In adenovirus-transformed Wt3A cells, CAPE treatment elicited an alternative response,
cell death by apoptosis. We documented that the effect of CAPE was
apoptosis by four techniques: DNA laddering, EM, analysis, image
analysis of d-dUTP-labeled cells by using TdT, and bivariant flow
cytometry. An evaluation of CREF and Wt3A cell lines for growth
arrest in response to ionizing radiation indicated a deficient Gj check
point control in comparison to proficient primary rat embryo fibroblasts. Loss of the Gj checkpoint control has been associated with the
loss of wild-type p53 function (22-24). It is possible that CAPEinduced transient growth arrest in CREF cells and apoptosis in Wt3A
cells are p53 independent. We also found that CAPE-sensitive RKO
3580
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.
DIFFERENTIAL
CAPE-INDUCED
APOPTOSIS
Table 1 Evaluation of the Gìcell cycle commi in CREF and Wl3A cells after
Igammai-ray irradiation
>of cells in S phase"
% of cells in G | phase
postirradiation7.1
±0.3*29.0
REFCREFWl3AControl20.7
0.435.4
±
1.834.3
±
0.237.7
±
±3.357.2
5
0.452.9
±
postirradiation72.2
3.5''53.3
±
0.446.6
±
±2.324-h
±0.8Control61.
±1.524-h
±1.1
" Percentage of S-phase populations was determined using BrdUrd incorporation
followed by bivariant flow cytometry.
Significantly reduced compared to control; P < 0.001.
' Significantly increased compared to control; P < 0.005.
70 n
60 -
0
O
O.
50 40 -
at
E
O)
30 -
(O
20 -
0
10-
1234
CREF-
CREF+
Wt3A-
Wt3A +
Fig. 8. Assay of GSH levels in CAPE-treated and untreated CREF and Wt3A cells.
Cell extracts were prepared 4 h after the addition of CAPE. *, the difference in GSH levels
measured in CAPE-treated versus untreated Wt3A cells was significant (P < 0.01).
IN TRANSFORMED
CELLS
catalase treatment induces apoptosis in HeLa cells (31). Adult rat
Leydig cells undergo apoptosis after treatment with an alkylating
chemical, ethane dimethanesulfonate, but are rescued by inhibition
of GSH synthesis with BSO (32). These phenomena may be
rationalized if oxidative processes are vital for survival of these
cells. Thus, exogenous agents that affect a redox change can
produce opposite effects under different circumstances.
It was observed previously that CAPE reduced oxidative stress
induced by the tumor promoter TPA and by the carcinogen AOM
(7, 9). Both TPA and AMO generate mitogenic signals in the
treated cells (9, 33, 34). Stimulation of cell proliferation may be
accomplished by intracellular ROS induced by tumor promoters
and carcinogens (5, 6). Because CAPE inhibits DNA synthesis, it
may antagonize effects of TPA and AMO, possibly through signal
transduction pathways. On the basis of our work and studies by
others, we propose that CAPE modulates the redox state of cells.
Depending on the specific conditions and cell type, the outcome of
CAPE treatment may differ. Like CAPE, the effect of Bcl2 ex
pression may alter the redox state of cells. Both anti- and prooxidant effect have been reported for Bcl2 (26, 27, 35). Because
expression of Bcl2 partially protected Wt3A cells from CAPEinduced apoptosis, it is possible that Bcl2 can elicit only a transient
redox change, which in normal cells would allow other growth
options that are not available in the transformed cells used here and
by others (36, 37).
In conclusion, this study and studies in progress show that CAPE
can induce apoptosis in transformed rodent and human cells in culture.
Sensitivity of transformed cells to antitumor drugs may be determined
by the inability of transformed cells to synthesize GSH in response to
oxidative stress (38, 39). Wt3A cells appear to be more sensitive than
CREF cells to oxidative stress imposed by CAPE. Therefore, cell
types that are predicted to be sensitive to CAPE should demonstrate
signs of chronic oxidative stress and deficiency in response to oxida
tive stress. The apparent benign effects of CAPE on normal cells may
make it a useful adjunct to chemotherapy especially in treatment of
tumors with mutant p53.
cells (a human colon carcinoma cell line with wild-type p53) transfected with the E6 oncogene of human papilloma virus retained their
sensitivity to CAPE.6 Because E6 selectively targets p53 for
degradation (28, 29), this further suggests that CAPE may function
through p53-independent pathways.
Control of cell proliferation versus apoptosis in a variety of
mammalian cell types is mediated by receptor-mediated events that
may lead to the generation of O2~ and H2O2. Therefore, cell
proliferation and apoptosis are alternative responses that depend on
the specific cellular redox balance at a given time (reviewed in
Refs. 5 and 14). In the case of CREF and Wt3A cells, increase of
oxidative stress appeared to make these cells sensitive to CAPEinduced cell death. CREF cells, which do not die after treatment
with CAPE alone, could be made to undergo apoptosis by raising
their level of the oxidative stress (e.g., depletion of GSH by
pretreatment with BSO; Fig. 7). In contrast, the reduction of
oxidative stress by the GSH precursor, NAC, allowed Wt3A cells
to maintain viability and to be partially spared from CAPE-triggered apoptosis (Fig. 9). Catalase, which reduces oxidative stress,
also rescued Wt3A cells from CAPE toxicity (Fig. 12). Reduction
of oxidative stress by reducing agents and rescue from apoptosis
was also observed in other cell systems (30). However, in some
cells reduction of oxidative stress causes apoptosis. For example,
'' Chiao et al., unpublished data.
1.2 -i
Wt3A cells, 8 hours cell count
1.0 -
O
S3
0.8-
E
3
•
U
S
o
NAC
CAPE
0.0
(mM)
(1ug/ml)
0
-
Fig. 9. Protection from CAPE-induced toxicity in Wl3A cells by NAC. After the
standard cell seeding and overnight growth, cultures were incubated with various con
centrations of NAC (0-15 mm) for 1 h. Subsequently, CAPE was added, and attached
cells were counted after an H-h incubation.
3581
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.
DIFFERENTIAL
CAPE-INDUCED
APOPTOSIS
IN TRANSFORMED
CELLS
AS
Time(h)
0
8 24 0
8 24
Fig. 10. Protection from CAPE-induced toxicity
in Wt3A cells by Bcl2. Wt3A transfectants harbor
ing the bcl2-sense (5) or antisense (AS) constructs
were cultured overnight before CAPE treatment. A,
cell numbers of Wt3A cells at various times after
addition of CAPE. B, electrophorctic analysis of
DNA from these transfectants for fragmentation.
Lanes /-.?, Wt3A/bcl2-AS
DNA from untreated
cells, or CAPE-treated ones that were incubated for
8 or 24 h, respectively. Lanes 4-6, Wt3A/bcl2-S
DNA from similarly untreated and treated cells.
123456
10
20
Time (h)
B.
CREF- (o)
CREF
Time(h)
Fig. 11. Differential toxicity of H2O2 in Wt3A
cells compared to CREF cells. A, after plating and
overnight growth, both cell types were exposed to
H2O2 (6(M)¿AM).
The number of attached cells was
determined at various times. B. DNA from treated
and untreated cells was prepared in parallel and
analyzed by gel electrophoresis for fragmentation.
Lanes 1-4, CREF cells untreated or treated with
H2O2 for 4, 8, and 24 h, respectively. Lanes 5-7,
Wt3A cells untreated or treated with H2O2 for 4
and 8 h, respectively.
0
4
8
|t Wt3A
24
o
4
8
Wt3A- «
4
CREF+Hj02(B)
VW3A+HjOj(•)
10
12
34567
20
Time (h)
4000
-CAPE(o)
Fig. 12. Rescue of Wt3A cells from CAPEinduced apoptosis by catalase. A, Wt3A cells after
seeding and overnight growth were treated with
CAPE with or without catalase (50 (¿/ml).Attached
cells were counted at various times after this treat
ment. B, DNA samples from parallel cultures ana
lyzed by electrophoresis for fragmentation. Lanes I
and 2, DNA from control Wt3A cells and CAPEtreated cells collected at 8 h, respectively. Lanes 3
and 4, DNA from CAPE-treated cells simulta
neously incubated with catalase for 8 or 24 h,
respectively.
+CAPE
Time (h) ""o(T
+CAPE
+Catalase
8 24
+CAPE,+50u/ml catalase(•)
+CAPE(l)
10
12
34
20
Time (h)
3582
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.
DIFFERENTIAL
CAPE-INDUCED
APOPTOSIS
CELLS
suppressor gene. Cancer Res.. 54: 4214—1223, 1994.
18. Wyllic, A. Glucocorticoid-induccd thymocyte apoplosis in association with endoge
nous endonuclease activation. Nature (Lond.), 284: 555-556, 1980.
19. Kerr, J., Wyllie, A., and Curde, A. Apoptosis: a basic biological phenomenon with
wide-ranging implication in tissue kinetics. Br. J. Cancer, 26: 239-257.
20. Gorczyca. W., Gong, J.. and Darzynkiewic/. Z. Detection of DNA strand breaks in
individual apoptotic cells by the /// .V//Hterminal deoxynucleotidyl transferasc assays.
Cancer Res., 53: 1945-1951, 1993.
Telford, W.. King, L.. and Franker. P. Rapid quantitation of apoptosis in pure and
heterogeneous cell populations using flow cytomctry. J. Immunol. Methods, 172:
1-16, 1994.
22. Kuerbitz. S.. Plunkett, B., Walsh, W.. and Kastan, M. Wild-type p53 is a cell cycle
checkpoint pathway determinant following irradiation. Prix:. Nail. Acad. Sci. USA,
8V: 7491-7495, 1992.
Brislow, R. G., Jang, A.. Peacock, J.. Chung. S., Benchimol, S., and Hill. R. P. Mutant
p53 increases radioresistance in rat embryo fibroblasts simultaneously transfected
with HPV17-E-7 and/or H-ra.v. Oncogene, u: 1527-1536. 1994.
ACKNOWLEDGMENTS
We lhank Dr. Paul Fisher at Columbia University for kindly providing us
with CREF and Wt3A cells and Jitka Mucha for technical assistance. We are
grateful to John L. Horton at the National Institute of Environmental Health
and Sciences for providing assistance with the EM analysis. Dr. Ha/el B.
Matthews for his helpful suggestions, and Lois A. Annab for providing REF
cells.
REFERENCES
1. Grunbcrgcr, D., Banerjec, R., Eisinger, K., Oltz, E. M.. Efros, L., Caldwell. M,
Estevez. V., and Nakanishi. K. Preferential cytotoxicity on tumor cells by caffeic acid
phenethyl ester isolated from propolis. Experientia, 44: 230-232, 1988.
2. Su, Z-Z., Grunberger, D., and Fisher, P. B. Suppression of adenovirus type 5
El A-mediated transformation and expression of the transformed phenotype by caffeic
acid phenethyl ester (CAPE). Mol. Carcinog., 4: 231-242, 1991.
3. Su, Z-Z., Lin, J., Grunberger, D., and Fisher, P. B. Growth suppression and (oxicity
induced by caffeic acid phenethyl ester (CAPE) in type 5 adenovirus-transformed rat
embryo cells correlated with transformation progression. Cancer Res., 54:
1865-1870, 1994.
4. Lin, J.. Su, Z-Z.. Grunberger, D., Zimmer, S. G., and Fisher, P. B. Expression of the
transformed phenotype induced by diverse acting viral oncogenes mediates sensitivity
to growth suppression induced by caffeic acid phenethyl ester (CAPE). Intl. J. Oncol.,
5: 5-15, 1994.
5. Burton. R. H. Superoxide and hydrogen peroxide in relation to mammalian cell
proliferation. Free Radical Biol. Med., 18: 775-794, 1995.
6. Troll, W., and Weisner, R. The role of oxygen radicals as a possible mechanism of
tumor promotion. Ann. Rev. Pharmacol. Toxicol., 25: 509-528, 1985.
7. Frenkel, K.. Wei. H., Bhimani, R., Ye, J.. Zadunaisky. J. A., Huang. M-T., Fcranro,
T., Conney, A. H., and Grunhcrger. D. Inhibition of tumor promoter-mediated
processes in mouse skin and bovine lens by caffeic acid phenethyl ester. Cancer Res.,
5J: 1255-1261, 1993.
H. Bhimani. R., Troll. W., Grunberger, D.. and Frenkel. K. Inhibition of oxidativc stress
in HeLa cells by chemopreventive agents. Cancer Res., 53: 4528-4533, 1993.
9. Rao, C. V., Desai, D.. Simi, B., Kulkarni, N., Amin, S., and Rcddy, B. S.
Inhibitory effect of caffeic acid esters on azoxymethane-induced
biochemical
changes and aberrant crypt foci formation in rat colon. Cancer Res., 53: 41824188, 1993.
10. McConkey, D., and Orrenius. S. Signal transduction pathways to apoptosis. Trends
Cell Biol., 4: 370-375, 1994.
11. Eastman, A. Activation of programmed cell death by anticancer agents: cisplatin as
a model system. Cancer Cells, 2: 275-280, 1990.
12. Hickman. J. Apoptosis induced by anticancer drugs. Cancer Metastasis Rev., II:
121-139, 1992.
13. Szumiel, I. Ionizing radiation-induced cell death. Int. J. Radial. Biol., 66: 329-341,
1994.
14. Buttke. T., and Sandstrom, P. Oxidative stress as a mediator of apoptosis. Immunol.
Today, /5: 7-10, 1994.
15. Fisher. P.. Babiss, L., Weinstein. I., and Ginsberg, H. Analysis of type 5 adenovirus
transformation with a clone of rat embryo cell line (CREF). Proc. Nati. Acad. Sci.
USA, 79: 3527-3531, 1982.
16. Stockschlaeder, M. A. R., Storb, R., Osborne, W. A. R., and Miller, A. D. L-histidinol
provides effective selection of retrovirus-vector-infected
keratinocytes without im
pairing their proliferative potential. Hum. Gene Ther., 2: 33—39,1991.
17. Preston, G. A., Lang, J. E., Maronpot, R. R., and Barrett, J. C. Regulation ol apoptosis
by low serum in cells of different stages of ncoplastic progression: enhanced suscep
tibility after loss of a senescence gene and decreased susceptibility after loss of tumor
IN TRASSIOKMi:!)
24. Pardo, F. S., Su, M., Borek, C., Preffer, F.. Dombkowski, D., Gerweck. L., and
Schmidt. E. V. Transfection of rat embryo cells with mutant p53 increases the
intrinsic radiation resistance. Radial. Res., 140: 180-185, 1994.
25. Griffith. O. W., and Meister. A. Potent and specific inhibition of glutalhione synthesis
by buthionine sulfoximine (5-n-butyl homocysteine sulfoximine). J. Biol. Chem..
254: 7558-7560. 1979.
Hockcnhery, D. M.. Oltvai, Z. N.. Yin. X-M., Milliman. C. L.. and Korsmcyer, S. J.
Bcl2 functions in an antioxidant pathway to prevent apoptosis. Cell, 75: 241-251.
1993.
27. Kane, D. J., Sarafian, T. A., Anton, R., Hahn, H.. Gralla, E. B., Valentine, J. S., Ord,
T., and Bredesen, D. E. Bcl2 inhibition of neural death: decreased generation of
reactive oxygen species. Science (Washington DC), 262: 1274-1277, 1993.
28. Werness. B. A., Levine, A. L. and Howley. P. M. Association of human papillomavirus type-16 and type-18 E6 proteins with p53. Science (Washington DC), 243:
76-79, 1990.
29. Scheffner. M.. Huibreglse, J. M., Vierstra, R. D., and Howley. P. M. The HPV-16 E6
and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of
p53. Cell, 75: 495-505. 1993.
30. Mayer, M., and Noble, M. /V-acetyl-L-cystcinc is a pluripotent protector against cell
death and enhancer of trophic factor-mediated cell survival in vitro. Proc. Nati. Acad.
Sci. USA, W: 7496-7500, 1994.
31. Burdon. R.. and Gill. V. Cellular generated active oxygen species and HeLa cell
proliferation. Free Radical Res. Commun.. IV: 203-213. 1993.
32. Kelce, W. Buthionine sulfoximine protects the viability of adult rat Leydig cells exposed
to ethane dimethanesulfonatc. Toxicol. Appi. Pharmacol. 125: 237-246, 1994.
33. Nishi/uka. Y. The role of protein kinase C in cell surface signal transduction and
tumor promotion. Nature (Lond.), 308: 639-698, 1984.
34. Blumberg. P. M. Protein kinase C as the receptor for the phorhol ester tumor
promoter. Cancer Res., 48: 1-8, 1988.
Steinman. H. The Bcl2 oncoprotein functions as a pro-oxidant. J. Biol. C'hem., 270:
35.
3487-3490, 1995
36. Vaux. D., Cory. S., and Adams, J. Bcl2 gene promotes hcmopoielic cell survival and
cooperates with c-mvr to immortalize prc-B cells. Nature (Lond.), .W: 440-442.
1988.
37. McDonnell. T.. Deane, N., Platt. F.. Nunez, G.. Jaeger. U., McKcarn J., and Korsmeyer. S. bcl2-immunoglobulin transgenic mice demonstrate extended B cell survival
and follicular lymphoproliferation. Cell. 57: 79-88, 1989.
38. Lee, F. Y. F., Siemann, D. W., and Sutherland. R. M. Changes in cellular glutalhione
content during adriamycin treatment in human ovarian cancer-a possible indicator of
chemosensitivity. Br. J. Cancer, 60: 291-298, 1989.
Meister, M. Glutathione. ascorbate, and cellular protection. Cancer Res., 54:
1969-1975, 1994.
3583
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.
Apoptosis and Altered Redox State Induced by Caffeic Acid
Phenethyl Ester (CAPE) in Transformed Rat Fibroblast Cells
Chia Chiao, Adelaide M. Carothers, Dezider Grunberger, et al.
Cancer Res 1995;55:3576-3583.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/55/16/3576
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on July 31, 2017. © 1995 American Association for Cancer Research.