 |
INTRODUCTION |
Phorbol esters exert a variety of effects in cellular systems that
include proliferation, malignant transformation, differentiation, and
cell death (1-5). The multiplicity of effects of phorbol esters on
biological systems is associated with the existence of various phorbol
ester receptors, which include several
PKC1 isozymes and novel
non-kinase receptors (
- and 
-chimaerins, unc-13, and Ras-GRP).
The PKC family comprises at least 10 serine-threonine-kinases subject
to different biochemical regulation (5, 6). PKC isozymes can be
classified into three groups: calcium-dependent or
"classical" (PKCs , 1, 2, and 
), calcium-independent or "novel" (PKCs 
, 
, 
, and 
), and "atypical" (PKCs 
and 
/
). Only the classical and novel PKC isozymes are receptors
for phorbol esters and the lipid second messenger diacylglycerol.
The existence of several PKC isozymes with unique cofactor
requirements, intracellular localization, and cellular/tissue
distribution suggests specialized roles for each isozyme in the control
of cellular functions. An issue of relevance is that individual PKC isozymes may exhibit either similar or opposite biological effects. In
NIH 3T3 fibroblasts, for example, PKC and PKC promote opposite effects on proliferation: whereas PKC induces cell proliferation and
malignant transformation, PKC inhibits cell growth (7). A second
level of complexity involves the host cell: overexpression of PKC
induces cell growth arrest in G2/M in Chinese hamster ovary
cells and HL60 cells (8-10), but it markedly enhances
anchorage-independent growth and metastatic potential in mammary
adenocarcinoma cells (11). Interestingly, and of particular relevance
to the present study, PKC induces apoptosis after DNA damage in
hemopoietic cells and keratinocytes, a process that involves its
proteolytic cleavage and generation of an active catalytic fragment
(12, 13).
The aim of this study was to explore the role of PKC as a mediator
of phorbol ester-mediated responses in LNCaP cells, a widely used model
of androgen-dependent prostate cancer (14). Notably,
phorbol esters induce apoptosis in LNCaP cells (15, 16), an effect that
is also observed in thymocytes and breast cancer cells (17-19).
Phorbol ester-induced apoptosis in LNCaP cells may involve the lipid
second messenger ceramide (20). Powell et al. (16) reported
a persistent translocation of PKC in phorbol ester-induced
apoptosis. The contribution of other PKC isozymes present in LNCaP
cells as mediators of phorbol ester-mediated apoptosis, however, has
not been examined to date.
To evaluate the involvement of PKC in phorbol ester-induced
apoptosis, we used a replication-deficient adenovirus for this novel
PKC isozyme (PKCAdV). The evidence presented in this study implicates PKC as a mediator of phorbol ester-induced apoptosis in
LNCaP cells. Interestingly, and contrary to observations in other cell
types, PKC-mediated apoptosis in LNCaP cells does not involve its
proteolytic activation by caspase-3, suggesting that allosteric
activation of the enzyme is sufficient to trigger an apoptotic response
in these cells.
| |
EXPERIMENTAL PROCEDURES |
Cell Culture--
The human prostate cancer cell line LNCaP and
U-937 human promonocytic leukemia cells were obtained from the American
Type Culture Collection (Manassas, VA). LNCaP and U-937 cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum
and penicillin (100 units/ml)-streptomycin (100 µg/ml) at 37 °C in
a humidified 5% CO2 atmosphere. LNCaP cells stably
overexpressing Bcl-2 and the corresponding vector-transfected cells
(21) were a kind gift of Dr. L. Lothstein (University of Tennessee,
Memphis, TN).
PKC and Dominant Negative PKC Adenoviruses--
A
XhoI-MluI fragment comprising the full-length
PKC cDNA (7) was inserted into pCA4-FLAG, a modified version of
the vector pCA4 (Microbix Biosystems Inc., Toronto, Ontario, Canada).
pCA4-FLAG includes XhoI-MluI sites for
subcloning, followed by a C-terminal FLAG tag and a stop codon in frame
(22). A replication-deficient adenovirus for PKC (PKCAdV) was
generated by standard techniques using 293 packaging cells (22, 23).
Recombinant adenoviruses were isolated from single plaques and
amplified in 293 cells. Titers of viral stocks were normally higher
than 1 × 10 9 pfu/cell. The absence of wild type
adenovirus was confirmed by polymerase chain reaction using primers for
the E1 region. An adenovirus for the LacZ gene was used as a control
(22, 24). Generation of a dominant negative PKC adenovirus
(DN-PKCAdV) is described elsewhere (25).
Infection of LNCaP Cells with PKCAdV--
Subconfluent LNCaP
cells in six-well plates were infected with PKCAdV for 14 h at
different multiplicities of infection (MOIs) (1-300 pfu/cell) in RPMI
1640 medium supplemented with 2% fetal bovine serum. After removal of
the virus, cells were incubated for an additional 24 h in RPMI
1640 medium supplemented with 10% fetal bovine serum. In several
experiments, PMA at different concentrations (0.3-100 nM)
or vehicle (ethanol) was added for 1 h following the adenoviral
infection, and cells were then grown in RPMI 1640 medium supplemented
with 10% fetal bovine serum for different times. For determination of
cell number, cells were harvested by trypsinization (0.25% trypsin and
1 mM EDTA in Hanks' balanced salt solution) and counted in
a hemocytometer.
PKC Activity--
PKC activity in total cellular lysates was
determined by phosphorylation of the -pseudosubstrate peptide, as
described previously by Kazanietz et al. (26). Briefly,
cells in six-well plates were lysed in 150 µl of lysis buffer (50 mM Tris-HCl, pH 7.4, and CompleteTM protease
inhibitor mixture tablets; Roche Molecular Biochemicals). The reaction
was carried out in a total volume of 50 µl containing 50 mM Tris-HCl, pH 7.4, 250 µg/ml bovine serum albumin, 1 mM EGTA, 100 µg/ml phosphatidylserine, 100 nM
PMA, 10 mM -pseudosubstrate peptide, 25 mM
ATP, and 7.5 mM magnesium acetate. After incubation at
30 °C for 5 min, 25 µl of each reaction was spotted onto Whatman PE-81 paper. The paper was washed three times with 0.1 M
phosphoric acid and once with acetone and air-dried, and radioactivity
was counted in a scintillation counter.
[3H]PDBu Binding--
[3H]PDBu
binding in cellular lysates was measured using the polyethylene glycol
precipitation assay (27), as described previously (26, 28). The assay
was performed in a total volume of 250 µl, using 100 µg/ml
phosphatydylserine, 1 mM EGTA, 15 nM
[3H]PDBu, and 50 µl of cell lysate.
Western Blot Analysis--
Cells were harvested into lysis
buffer containing 50 mM Tris-HCl, pH 6.8, 10% glycerol,
2% SDS, 0.00125% bromphenol blue, and 5% -mercaptoethanol and
then lysed by sonication. Equal amounts of protein (10 µg) were
subjected to SDS-polyacrylamide gel electrophoresis and transferred to
nitrocellulose membranes. Membranes were blocked with 5% milk, 0.1%
Tween 20 in phosphate-buffered saline and then incubated with one of
the following antibodies: anti-PKC (1:3000, Upstate Biotechnology,
Inc., Lake Placid, NY), anti-PKC (1:1000, Transduction Laboratories,
Lexington, KY), anti-PKC (1:1000, Transduction Laboratories),
anti-PKC (1:1000, Life Technologies Inc.), anti-PKC (1:3000,
Santa Cruz Biotechnology, Santa Cruz. CA), anti-PKCµ (1:1000,
Transduction Laboratories), anti-FLAG (1:500, Sigma), anti-Bcl-2
(1:1000, Zymed Laboratories Inc., San Francisco, CA), or
anti-caspase-3 (1:1000, Transduction Laboratories). Membranes were
washed three times with 0.1% Tween 20/phosphate-buffered saline and
incubated with secondary antibody conjugated to anti-mouse or
anti-rabbit horseradish peroxidase (1:3000, Bio-Rad). Bands were
visualized by the enhanced chemiluminescence (ECL) Western blotting
detection system (Amersham Pharmacia Biotech).
Subcellular Fractionation--
LNCaP cells were harvested into a
lysis buffer containing 20 mM Tris-HCl, pH 7.4, 5 mM EGTA, 5 µg/ml 4-(2-aminoethyl)-benzenesulfonylfluoride (AEBSF), 1 µg/ml pepstatin A, 5 µg/ml aprotinin, and 5 µg/ml
leupeptin and lysed by sonication. Separation of soluble and particular fractions was performed by ultracentrifugation as described previously (28). Briefly, the cytosolic (soluble) fraction was obtained by
collection of the supernatant after centrifugation of the total lysate
(1 h at 100,000 × g at 4 °C). The remaining pellet
represents the particulate fraction. Protein concentration of the total
lysate and fractions was determined using the Bio-Rad protein assay
kit. Equal amounts of protein for each fraction (10 µg) were
subjected to SDS-polyacrylamide gel electrophoresis and transferred to
nitrocellulose membranes. Membranes were blocked with 5% milk in
phosphate-buffered saline and subsequently immunostained with anti-PKC
antibodies, as described above.
Apoptosis Assays--
DNA laddering was measured using the
Apoptotic DNA Ladder Kit from Roche Molecular Biochemicals.
To assess morphological changes in chromatin structure of LNCaP cells
undergoing apoptosis, cells were stained with
4',6-diamidino-2-phenylindole (DAPI, Sigma). Cells were trypsinized as
described above, mounted on glass slides, and fixed in 70% ethanol.
Cells were then stained for 20 min with 1 mg/ml DAPI and examined by
fluorescence microscopy. Apoptosis was characterized by chromatin
condensation and fragmentation. The incidence of apoptosis in each
preparation was analyzed by counting 500 cells and determining the
percentage of apoptotic cells.
For flow cytometry analysis, cells were fixed in 70% ethanol and
stained with propidium iodide (1 mg/ml). Cell cycle progression and
apoptosis was analyzed in an EPICS XL flow cytometer (Coulter Corp.,
Hialeah, FL). For each treatment 7500 events were recorded.
Caspase-3 activity was measured with the colorimetric assay kit from
BioVision (Palo Alto, CA) that uses as a substrate the chromophore
p-nitroanilide-DEVD.
| |
RESULTS |
Overexpression of PKC in LNCaP Cells after Infection with
PKCAdV--
A recombinant adenovirus for PKC, PKCAdV, was
generated as described under "Experimental Procedures." The
recombinant PKC was engineered to have a FLAG epitope-tag in frame
at the C-terminal end. Infection of LNCaP cells with PKCAdV at
different MOIs resulted in a concentration-dependent
expression of PKC, as determined by Western blotting with either
anti-FLAG (Fig. 1A, top panel) or anti-PKC antibodies (Fig. 1A, bottom panel).
Immunofluorescence using the anti-FLAG antibody reveals an efficiency
of infection higher than 90% at a MOI of 10 pfu/cell (data not shown).
Expression of FLAG-tagged PKC was detected 6 h after infection.
Maximum levels of expression were obtained after 24 h and lasted
for at least 7 days (data not shown).
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 1.
Expression of PKC in
PKCAdV-infected LNCaP cells. Cells were
infected with PKCAdV for 14 h at different MOIs, as indicated
in the figure. A, cells lysates were analyzed by Western
blot using anti-FLAG or anti-PKC antibody. Similar results were
obtained in two additional experiments. B, PKC activity was
measured in lysates of LNCaP cells, 24 h after infection with
PKCAdV or LacZAdV. Kinase activity was measured by phosphorylation
of -pseudosubstrate peptide as described under "Experimental
Procedures." Almost identical results were observed in two additional
experiments. C, [3H]PDBu binding was
determined in lysates of LNCaP cells infected with PKCAdV or LacZAdV
for 24 h, using the MOIs indicated in the figure.
[3H]PDBu binding was measured by the polyethylene glycol
precipitation assay described by Sharkey and Blumberg (27). Results in
B and C are expressed as mean ± S.E. of
three independent experiments.
|
|
Infection of LNCaP cells with PKCAdV (MOI = 1-100 pfu/cell)
resulted in a concentration-dependent increase in PKC
activity in cell extracts (Fig. 1B), as determined by
phosphorylation of -peptide, a specific PKC substrate. The minimum
MOI necessary to achieve a significant increase in total PKC activity
in these cells was 10 pfu/cell. No significant changes in PKC activity were observed when LNCaP cells were infected with a LacZAdV (MOI = 100 pfu/cell). Next, we measured whether infection of LNCaP cells with
PKCAdV increases the level of phorbol ester binding, using
[3H]PDBu as a radioligand. Elevated levels of phorbol
ester binding were observed in cells infected with PKCAdV but not in
those infected with LacZAdV (Fig. 1C). LNCaP cells infected
with LacZAdV did not show any changes in endogenous PKC
immunoreactivity but did show high levels of LacZ expression as
detected with X-Gal staining (data not shown). For most of the
experiments presented here, we have used MOIs in the range of 10-30
pfu/cell to avoid any potential nonspecific effects that might occur
when PKC is overexpressed at high levels. At MOI = 10-30 pfu/cell,
total cellular [3H]PDBu binding activity was elevated
3-8-fold over basal levels, respectively (see Fig. 1C).
It is well established that phorbol esters induce the redistribution of
PKC isozymes from cytosolic to particulate fractions, a process known
as PKC translocation. To confirm that the FLAG-tagged PKC was
responsive to phorbol esters, we infected LNCaP cells with PKCAdV
(MOI = 10 pfu/cell) and subsequently incubated the cells with PMA
for different times. As shown in Fig. 2,
PMA induces translocation of FLAG-tagged PKC from the cytosolyc
(soluble) to the particulate fraction. 4-PMA, the inactive isomer of
PMA, was completely inefficient at inducing translocation.
Translocation of FLAG-tagged PKC followed a similar pattern to that
observed for the endogenous phorbol ester-responsive isozymes PKC
and PKC. As expected, the phorbol ester-unresponsive PKC is not translocated by PMA in LNCaP cells. From these data, we conclude that
infection of LNCaP prostate cancer cells with PKCAdV results in an
efficient expression of a catalytically active, phorbol ester-responsive PKC.
View larger version (42K):
[in this window]
[in a new window]
|
Fig. 2.
Translocation of PKC isozymes in
PKCAdV-infected LNCaP cells. LNCaP cells
were infected with PKCAdV (MOI = 30 pfu/cell, 14 h) and
24 h later incubated with PMA or 4-PMA (100 nM) for
different times, as indicated. Soluble and particulate fractions were
then separated by ultracentrifugation and subjected to Western blot
analysis using the antibodies indicated in the figure.
|
|
Activation of PKC Induces Apoptosis in LNCaP
Cells--
Incubation of LNCaP cells with PMA (0.3-100
nM, 1 h) results in a
concentration-dependent decrease in cell number compared with control (vehicle-treated) cells, as judged by cell counting at 24, 48, and 72 h after PMA treatment. The corresponding inactive isomer, 4-PMA (100 nM), was completely ineffective (Fig.
3A). We then infected LNCaP
cells with PKCAdV at MOIs ranging from 0.3 to 100 pfu/cell and
determined the cell number every 24 h for a total period of
72 h. A concentration-dependent reduction in cell
number was observed after infection with PKCAdV but not with LacZAdV
(MOI = 100 pfu/cell) (Fig. 3B). The inhibitory effect of PMA was markedly enhanced in cells infected with PKCAdV,
suggesting that activation of PKC results in a decrease in cell
count (Fig. 3C).
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 3.
Effect of PMA and
PKCAdV on LNCaP cell number.
A, LNCaP cells were treated with PMA or its inactive isomer
4-PMA for 1 h. After extensive washing to remove the phorbol
esters, cells were cultured for 24-72 h and then counted. Results are
the mean ± S.E. of three independent experiments. B,
LNCaP cells were infected with PKCAdV (MOI = 0.3-100 pfu/cell)
or LacZAdV (MOI = 100 pfu/cell) for 14 h. Cell number was
determined 24-72 h later. Results are the mean ± S.E. of three
independent experiments. C, LNCaP cells were infected with
PKCAdV or LacZAdV (MOI = 10 pfu/cell) for 14 h, and
24 h later, they were incubated with PMA (0.3-3 nM,
1 h). Control cells received vehicle (ethanol). Cell number was
determined 72 h later. Results are expressed as the percentage of
cells relative to vehicle-treated cells and are expressed as the
mean ± S.E. of three independent experiments.
|
|
Morphological examination of LNCaP cells infected with PKCAdV after
PMA treatment revealed a large number of cells with characteristics distinctive of apoptosis, including cellular shrinkage and nuclear fragmentation. No signs of apoptosis were observed in either
PKCAdV- or LacZAdV-infected cells in the absence of PMA treatment
(Fig. 4A). Treatment of LNCaP
cells with PMA induces a pattern of DNA fragmentation characteristic of
apoptosis, which can be visualized as a DNA ladder. DNA fragmentation
was significantly higher in cells infected with PKCAdV (MOI = 30 pfu/cell) after PMA treatment (Fig. 4B). PMA treatment of
PKCAdV-infected cells also results in a large number of
TUNEL-positive cells (data not shown). Flow cytometry analysis of
PKCAdV-infected cells after PMA treatment revealed a large number of
cells with sub-G0/G1 DNA content, consistent with the presence of apoptotic cells. The PMA effect was lower in
LNCaP cells infected with LacZAdV compared with PKCAdV-infected cells (Fig. 4C) or noninfected cells (data not shown). A
dose-response analysis for the apoptotic effect of PMA is shown in Fig.
5. Cells were infected with either
PKCAdV or LacZAdV (MOI = 30 pfu/cell) and treated with
increasing concentrations of PMA (0.1-10 nM, 1 h).
The results shown in Fig. 5 revealed a markedly higher apoptotic effect
of PMA in LNCaP cells infected with PKCAdV than in noninfected cells
or cells infected with LacZAdV. These findings collectively suggest
that activation of PKC induces apoptosis in LNCaP cells.
View larger version (55K):
[in this window]
[in a new window]
|
Fig. 4.
Overexpression of PKC
potentiates PMA-induced apoptosis. LNCaP cells were infected
with either PKCAdV or LacZAdV (MOI = 30 pfu/cell) for 14 h. Twenty-four hours later, cells were treated with either PMA (10 nM) or vehicle (ethanol) for 1 h. Apoptosis was
evaluated 48 h later. A, LNCaP cells were stained with
DAPI and assessed for nuclear morphology by fluorescence microscope.
Apoptotic cells are indicated with arrows. Similar results
were observed in three experiments. B, DNA fragmentation was
monitored by electrophoresis in 2% agarose gels after staining with
ethidium bromide. Lane 1, LacZAdV; lane 2,
PKCAdV; lane 3, LacZAdV + PMA; lane 4,
PKCAdV + PMA. Ms, molecular size. C, after
staining with propidium iodide, DNA content was analyzed by flow
cytometry. The DNA histograms show a marked accumulation of
PKCAdV-infected cells in sub-G0/G1 after PMA
treatment. A representative experiment is shown. Similar results were
observed in two additional experiments.
|
|
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 5.
Concentration dependence analysis of
PMA-induced apoptosis in LNCaP cells. Cells were infected with
PKCAdV or LacZAdV for 14 h (MOI = 30 pfu/cell), and
24 h later, PMA was added for 1 h at different
concentrations. Cells were collected 48 h later and stained with
DAPI. The incidence of apoptosis in each preparation was analyzed by
counting 500 cells and determining the percentage of apoptotic cells.
Results are the mean ± S.E. of three independent
experiments.
|
|
Expression of PKC Isozymes in LNCaP Cells infected with
PKCAdV--
LNCaP cells express the classical PKC, the novel
PKC and PKC, the atypical PKC and PKC, and PKCµ (Fig.
6). Low levels of PKC were also
observed (data not shown). PKC, PKC, and PKC were not detected
in LNCaP cells. These results agree with those previously reported by
Powell et al. (16). Infection of LNCaP cells with PKCAdV
did not result in any significant change in the levels of PKC,
PKC, PKC, PKC, or PKCµ. Only a reduction in the levels of
PKC was observed after PMA treatment, both in LacZAdV and
PKCAdV-infected cells.
View larger version (49K):
[in this window]
[in a new window]
|
Fig. 6.
Expression of PKC isozymes in LNCaP cells
infected with PKCAdV. Cells were infected
for 14 h with PKCAdV or LacZAdV at the MOIs indicated in the
figure. Twenty-four hours later, cells were treated with either 10 nM PMA (PMA +) or ethanol as vehicle (PMA
-) for 1 h. Expression of PKC isozymes in cell extracts was
evaluated using specific PKC antibodies as described under
"Experimental Procedures."
|
|
Effect of PKC Inhibitors, z-VAD, and Bcl-2
Overexpression--
Pretreatment of LNCaP cells with the PKC inhibitor
GF 109203X (bisindolylmaleimide I, 5 µM) completely
blocked PMA-induced apoptosis in both noninfected and
PKCAdV-infected cells. We also used an unrelated antagonist of PKC,
bryostatin I. Bryostatin I is an atypical inhibitor of PKC function:
although it activates PKC isozymes in vitro and promotes few
PKC-dependent effects, it fails to activate PKC-mediated
responses in most cases. In those instances in which bryostatin 1 is
unable to activate PKC, it blocks those responses mediated by PMA (1,
29). In LNCaP cells, bryostatin 1 (10 nM) was ineffective
in inducing apoptosis. However, bryostatin 1 inhibited PMA-induced
apoptosis in both noninfected and in PKCAdV-infected cells (Fig.
7A).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 7.
Effects of PKC inhibitors, z-VAD, and Bcl-2
overexpression. A, cells were infected with PKCAdV
for 14 h (MOI = 30 pfu/cell) and 24 h later 10 nM PMA (PMA +) or ethanol (PMA -) was added for 1 h.
PKC inhibitors (5 µM GF 109202X or 10 nM
bryostatin 1) were added before and during PMA or vehicle treatment.
Results are the mean ± S.E. of three independent experiments.
B, cells were infected with PKCAdV and treated with PMA
as described in A. The pan-caspase inhibitor z-VAD (50 µM) was added 30 min before and during PMA or vehicle
treatment. C, Western blot analysis of Bcl-2 expression
after infection with PKCAdV and PMA treatment, using an anti-Bcl-2
antibody. D, LNCaP cells overexpressing Bcl-2 (solid
bars) or control cells (open bars) were infected for
14 h with PKCAdV (MOI = 10 pfu/cell) and 24 h later
treated for 1 h with 100 nM PMA, as indicated in the
figure. In A, B, and D, cells were
collected 48 h later and stained with DAPI. The incidence of
apoptosis in each preparation was analyzed by counting 500 cells in
each case and determining the percentage of apoptotic cells. Results
are the mean ± S.E. of three independent experiments.
|
|
It is well established that caspases, a family of Asp-directed
cysteine-proteases, play a pivotal role in transducing the apoptotic signal. Caspases can be specifically inhibited in cells by cell-permeable peptides (30). Treatment of LNCaP cells with the
pan-caspase inhibitory peptide z-VAD markedly reduced PMA-induced apoptosis in PKCAdV-infected cells (Fig. 7B).
Bcl-2 plays a central role in cell death signaling by forming
heterodimers with pro-apoptotic proteins Bax and Bad and inhibiting their function (31). In LNCaP cells, the apoptotic effect mediated by
activation of PKC did not result in any noticeable changes in Bcl-2
levels (Fig. 7C). The relationship between PKC-mediated apoptosis and Bcl-2 was further explored in LNCaP cells stably transfected with a Bcl-2 mammalian expression vector (LNCaP-Bcl-2), which express higher levels of Bcl-2 compared with control
(vector-transfected) cells (21). Cells overexpressing Bcl-2 showed a
marked resistance to PMA induced apoptosis when infected with PKCAdV
compared with control cells (Fig. 7D). To rule out the
possibility that this effect is a consequence of differential
sensitivity to PKCAdV-infection, we determined PKC activity in
control and Bcl-2-overexpressing cells. Similar levels in PKC activity
were observed in both cases after infection with PKCAdV infection
(14.6 ± 0.2-fold versus 14.1 ± 0.3-fold increase
over basal in cellular PKC activity after infection with PKCAdV at
100 pfu/cell), suggesting a similar sensitivity to adenoviral infection
in both cell lines.
PKC-mediated Apoptosis Does Not Involve Its Proteolytic
Cleavage--
An emerging model postulates that activation of caspases
leads to the proteolytic cleavage of kinases, which are either
activated or inactivated during the apoptotic process. Recent studies
have shown that PKC is proteolytically activated at the onset of
apoptosis induced by DNA-damaging agents in hemopoietic cells and
keratinocytes (12, 13). Proteolytic cleavage of PKC by caspase-3 at
the V3 (hinge) domain of the enzyme releases a catalytically active fragment of approximately 40 kDa. In order to evaluate whether such
cleavage occurs in LNCaP cells, cells were infected with PKCAdV and
then treated with PMA or vehicle. The presence of a catalytic fragment
was evaluated by Western blot using either an anti-FLAG or an
anti-PKC antibody in samples collected at different times, ranging
from 15 min to 24 h. Either of these antibodies recognizes the
C-terminal kinase domain of FLAG-tagged PKC. Interestingly, we could
not detect the presence of PKC catalytic fragment either at short
times or after several hours of PMA treatment (Fig.
8A). Very low levels of
cleavage were detected 24 h after PMA treatment. Fig.
8B revealed, however, a significant apoptotic response after
6 and 12 h of PMA treatment that reached a maximum at 24 h.
Thus, PKC-mediated apoptosis after PMA treatment does not correlate
with the generation of a catalytically active fragment of the enzyme.
These results prompted us to evaluate whether caspase-3 is activated
during PMA-induced apoptosis in PKCAdV-infected cells. Fig.
8C shows that no significant changes in caspase-3 activity
were detected after PMA treatment. Likewise, we did not detect any
cleavage of caspase-3 under similar experimental conditions (Fig.
8D). The unexpected lack of involvement of caspase-3 was
confirmed by the inefficacy of a specific caspase-3 inhibitor, DEVD, to
block PMA-induced apoptosis in LNCaP cells infected with PKCAdV
(Table I). U-937 cells treated with the
DNA-damaging agent camptothecin (4 µM) were used as
positive controls for these experiments. In this case, a marked
increase in caspase-3 activity (Fig. 8C) and caspase-3
cleavage (Fig. 8D) was observed, and DEVD markedly
inhibited apoptosis (Table I).
View larger version (42K):
[in this window]
[in a new window]
|
Fig. 8.
PKC-mediated
apoptosis in LNCaP cells does not involve the generation of a catalytic
fragment. Cells were infected with PKCAdV (MOI = 30 pfu/cell) for 14 h; 24 h later, they were treated with
vehicle or 100 nM PMA (1 h), and samples were collected at
the times indicated in the figure. Mw, molecular weight.
A, cell lysates were subjected to Western blot analysis
using an anti-FLAG or an anti-PKC antibody. B, incidence
of apoptosis in cells infected with PKCAdV and treated with PMA or
vehicle. C, caspase-3 activity in cell lysates, evaluated
with a colorimetric assay that uses as a substrate the chromophore
p-nitroanilide-DEVD (see under "Experimental
Procedures"). D, cell lysates were subjected to Western
blot analysis with an anti-caspase-3 antibody. In C and
D, U-937 cells treated with camptothecin (Cm)
were used as positive controls.
|
|
View this table:
[in this window]
[in a new window]
|
Table I
Effect of the caspase-3 inhibitor DEVD
Apoptosis in LNCaP cells was induced by infection with PKCAdV
(MOI = 30 pfu/cell, 14 h) and subsequent treatment with PMA
(100 nM, 1 h). Apoptosis in U-937 cells was induced by
treatment with camptothecin (4 µM). Caspase inhibitors (3 µM) were added 1 h before and during PMA or vehicle
treatment. The incidence of apoptosis in each preparation was analyzed
by counting 500 cells and determining the percentage of apoptotic
cells. Results are the mean ± S.E. of three independent
experiments. In all cases, the percentage of apoptotic cells was <2%
in the absence of apoptotic stimuli. Numbers in parentheses show the
percentage of apoptosis relative to control without caspase inhibitor;
ND, not determined.
|
|
Inhibition of PMA-induced Apoptosis by a PKC Inhibitor and a
Dominant Negative PKC Mutant--
The data presented above strongly
implicate PKC as a mediator of PMA induced apoptosis in LNCaP cells.
To further explore the role of PKC, we decided to evaluate the
effect of rottlerin, a selective PKC inhibitor, on PMA-induced
apoptosis. Rottlerin preferentially inhibits PKC over PKC or
PKC (32). Pretreatment of LNCaP cells with rottlerin partially
inhibited PMA-induced apoptosis in noninfected cells LNCaP cells (Table
II).
View this table:
[in this window]
[in a new window]
|
Table II
Effect of the PKC inhibitor rottlerin
LNCaP cells were treated for 1 h with PMA in the presence of
rottlerin, added at the concentrations indicated in the table 1 h
before and during PMA treatment. The incidence of apoptosis in each
preparation was analyzed by counting 500 cells and determining the
percentage of apoptotic cells. Results are the mean ± S.E. of
three independent experiments. Numbers in parentheses show the
percentage of apoptotic cells relative to cells not treated with the
inhibitor.
|
|
The involvement of PKC was further confirmed by introducing a
kinase-deficient mutant of PKC in LNCaP cells. Kinase-deficient mutants of PKC were shown to act as dominant negatives that
interfere with PKC function. We used an adenovirus for a PKC
mutant in which an Arg to Lys mutation was introduced in the
ATP-binding site (DN-PKCAdV). Infection of LNCaP cells with
DN-PKCAdV results in the expression of a catalytically inactive
PKC (Fig. 9A). Interestingly, expression of the kinase-deficient PKC blocks the
apoptotic response of PMA (100 nM) in LNCaP cells.
Infection with DN-PKCAdV at MOIs of 30 and 100 pfu/cell resulted in
50 and 61% inhibition, respectively, of PMA-induced apoptosis (Fig. 9B).
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 9.
Inhibition of PMA-induced apoptosis by a
dominant negative PKCAdV. A,
LNCaP cells were infected with dominant negative PKCAdV at the MOIs
indicated in the figure for 14 h and lysed 24 h later. PKC
activity in cell lysates was measured as described under
"Experimental Procedures." Inset, representative Western
blot using an anti-PKC antibody. Lane 1, control cells;
lane 2, DN-PKCAdV-infected cells (MOI = 100 pfu/cell). B, LNCaP cells were infected with either
DN-PKCAdV or LacZAdV for 14 h at the MOIs indicated in the
figure. After 24 h, PMA 100 nM was added for 1 h.
Cells were collected 48 h later and stained with DAPI. The
incidence of apoptosis in each preparation was analyzed by counting 500 cells and determining the percentage of apoptotic cells. Results are
the mean ± S.E. of three independent experiments.
|
|
| |
DISCUSSION |
To study the role of PKC as mediator of apoptotic responses in
LNCaP cells, we overexpressed this novel PKC isozyme by means of an
adenoviral expression system. Adenovirus-mediated delivery of proteins
has been successfully used in LNCaP cells (33, 34) and has proven to be
an efficient approach at introducing PKC isozymes into cells (25).
Infection of LNCaP cells with PKCAdV allowed for a controlled
expression of a catalytically active, phorbol ester-responsive PKC.
Our results provide strong evidence that activation of PKC induces
apoptosis in LNCaP cells.
The complexity in the regulation of PKC activity, which includes not
only regulation by diacylglycerol but also by tyrosine phosphorylation
and association to tyrosine kinases, suggests an important role for
this PKC isozyme in proliferation and cell death (35-38). Recent
reports have implicated PKC as a pro-apoptotic kinase. Emoto
et al. (12) have shown that exposure of U-937 cells to
DNA-damaging agents results in proteolytic cleavage of PKC with the
release of an active 40-kDa fragment corresponding to its C-terminal
kinase domain. Denning et al. (13) have recently reported
that in keratinocytes exposed to UV radiation, the generation of an
active catalytic fragment correlates with the apoptotic event. A
similar effect has been recently observed in etoposide-treated salivary
gland acinar cells (39). A proteolytic cleavage site for caspase-3 has
been identified at the V3 (hinge) region of PKC (12). Inhibition of
caspases using peptide inhibitors blocks the proteolytic cleavage of
PKC and apoptosis in U-937 cells and keratinocytes after DNA damage
(12, 13). To our surprise, we found no evidence for PKC proteolytic
cleavage after PMA activation in LNCaP cells. The unexpected lack of
effect of a caspase-3 inhibitor (DEVD) to block PKC-mediated
apoptosis and the absence of caspase-3 activation explains the lack of
proteolytic cleavage for this PKC isozyme. In agreement with our
results, caspase-3-independent apoptosis has also been described in
other systems, including human ovarian cancer cell lines treated with
cisplatin (40), leukemic cells treated with arsenic trioxide (41) or
tumor necrosis factor (42), MCF-7 mammary carcinoma cells treated
with tributyrin (43), and macrophages exposed to NO donors (44).
Caspase-3 is important for the typical morphologies associated with
apoptosis and for the formation of apoptotic DNA ladders. In LNCaP
cells, caspase-3 is a critical mediator of apoptosis induced by sodium phenylacetate (45). The present experiments have not addressed the
mechanisms by which PKC-induced apoptosis bypasses caspase-3 activation. However, the inhibitory effect of a pan-caspase inhibitor strongly suggests the involvement of other caspases. Other alternative mechanisms, such as the recently reported association and
phosphorylation of DNA-dependent kinases by PKC, may be
implicated in PKC-dependent apoptosis (46). It was
also shown that PKC-mediated apoptosis in keratinocytes involves the
alteration of mitochondria function (22). Although studies in
hemopoietic cells and keratinocytes have clearly shown that PKC is
proteolytically activated at the onset of (12, 13), the issue of
whether cleavage is required or not for apoptosis was still unanswered.
Our results in LNCaP cells unquestionably show that cleavage of
PKC is not required for apoptosis.
Activation of PKC by phorbol esters and diacylglycerol is associated
with the translocation of the enzyme from the cytosolic to particulate
fractions and stimulation of activity through an allosteric mechanism
(6). As expected, in LNCaP cells PKC is redistributed to the
particulate fraction after PMA treatment. Although prolonged
translocation of PKC isozymes can be associated with their proteolytic
cleavage by calpains and other proteases, removal of PMA in our
experimental conditions results in a fast dissociation of PKC from
the particulate fraction and return of the enzyme to the cytosol (data
not shown), as previously shown by Blumberg and co-workers (47). The
fact that apoptosis in LNCaP cells infected with PKCAdV was only
observed after phorbol ester treatment strongly suggests that
allosteric activation of PKC is sufficient to induce apoptosis. Our
results are consistent with recent studies presented by Chen et
al. (48) that show that UVB radiation-induced apoptosis in JB6
epidermal cells requires translocation of PKC to the membrane.
Interestingly, overexpression of a PKC catalytic fragment in HeLa
and NIH 3T3 cells induced apoptosis, whereas overexpression of
full-length PKC was ineffective (49). In the latter study, however,
cells were not challenged with phorbol esters or any other stimuli.
Therefore, it is likely that PKC-mediated apoptosis may proceed
through two distinct mechanisms, namely proteolytic cleavage (after DNA
damage) and allosteric activation (after direct activation). Each
mechanism may involve different signaling pathways. Our results suggest that PKC may act as a primary effector or is involved in a pathway that signals for apoptosis.
Expression of a dominant negative PKC mutant inhibits PMA-induced
apoptosis in LNCaP cells. It is tempting to speculate that the partial
inhibition observed by expression of this dominant negative PKC
mutant, as well as the partial inhibition observed by the PKC
inhibitor rottlerin, reflects the involvement of other phorbol
ester-responsive PKC isozymes in addition to PKC, namely PKC and
PKC. In preliminary experiments using a PKC adenovirus, we have
found that this PKC isozyme also mediates apoptosis in LNCaP
cells.2 Therefore, more than
one PKC isozyme may signal to apoptosis in LNCaP cells. It will be
important to use dominant negative forms of each PKC isozyme to address
this issue. Whether individual PKC isozymes promote apoptosis in LNCaP
cells by similar or different mechanisms is not known. Powell et
al. (16) observed a persistent membrane translocation of PKC
during PMA-induced apoptosis in LNCaP cells. Notably, a caspase
cleavage site is not present in PKC (12), which also supports a
model of allosteric activation of PKC in phorbol ester-induced
apoptosis. Evidence obtained from several cellular models suggests that
PKC-mediated apoptosis is not restricted to PKC. PKC, an isoform
highly homologous to PKC, mediates programmed cell death in U-937
myeloid leukemia cells in response to DNA damaging agents, an effect
that involves its proteolytic cleavage (50). PKCI is also
proteolytically activated after treatment of HL60 human promyelocytic
leukemia cells with the anticancer agents UCN-01, camptothecin, and
etoposide. Interestingly, similar treatment results in activation of
PKC without proteolytic cleavage (51).
PKC isozymes may also signal to inhibit apoptosis (4, 53, 54). It is
conceivable that the balance between expression and/or activation of
different isozymes may result in either pro-apoptotic or
anti-apoptotic signaling. The anti-apoptotic effect of PKC isozymes may
involve the up-regulation of Bcl-2 levels (52, 55). However, we did not
observe any significant changes in Bcl-2 levels after PKC
overexpression in LNCaP cells. Our observations that overexpression of
Bcl-2 in LNCaP cells prevents PKC-mediated apoptosis in LNCaP cells
emphasizes a role for Bcl-2 as a pro-survival signal in prostate cancer
cells, as described by Marcelli et al. (45).
In summary, our results provide the first evidence that PKC is a
mediator of phorbol ester-induced apoptosis in LNCaP cells. Clarifying
specific biological functions of individual PKC isozymes in prostate
cancer cells may underscore signaling pathways controlling cell growth
or death and provide us with novel therapeutic targets for studying the
progression of the disease.
(PKC
,
,
TOP
ABSTRACT
INTRODUCTION
