Oncogene (2005) 24, 6877–6889
& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00
www.nature.com/onc
Sodium butyrate sensitizes human glioma cells to TRAIL-mediated
apoptosis through inhibition of Cdc2 and the subsequent downregulation
of survivin and XIAP
Eun Hee Kim1, Hee Sue Kim1, Seung U Kim2,3, Eun Joo Noh3, Jong-Soo Lee3,4
and Kyeong Sook Choi*,1,2
1
Institute for Medical Sciences, Ajou University School of Medicine, Suwon, South Korea; 2Brain Disease Research Center, Ajou
University School of Medicine, Suwon, South Korea; 3Division of Neurology, UBC Hospital, University of British Columbia,
Vancouver, Canada; 4Research Institute, National Cancer Center, South Korea
In TNF-related apoptosis-inducing ligand (TRAIL)-resistant
glioma cells, co-treatment with nontoxic doses of sodium
butyrate and TRAIL resulted in a marked increase of
TRAIL-induced apoptosis. This combined treatment was
also cytotoxic to glioma cells overexpressing Bcl-2 or
Bcl-xL, but not to normal human astrocytes, thus offering
an attractive strategy for safely treating resistant gliomas.
Cotreatment with sodium butyrate facilitated completion
of proteolytic processing of procaspase-3 that was partially
blocked by treatment with TRAIL alone. We also found that
treatment with sodium butyrate significantly decreased the
protein levels of survivin and X-linked inhibitor of apoptosis
protein (XIAP), two major caspase inhibitors. Overexpression of survivin and XIAP attenuated sodium butyratestimulated TRAIL-induced apoptosis, suggesting its involvement in conferring TRAIL resistance to glioma cells.
Furthermore, the kinase activities of Cdc2 and Cdk2 were
significantly decreased following sodium butyrate treatment,
accompanying downregulation of cyclin A and cyclin B,
as well as upregulation of p21. Forced expression of Cdc2
plus cyclin B, but not Cdk2 plus cyclin A, attenuated
sodium butyrate/TRAIL-induced apoptosis, overriding
sodium butyrate-mediated downregulation of survivin and
XIAP. Therefore, Cdc2-mediated downregulation of survivin
and XIAP by sodium butyrate may contribute to the
recovery of TRAIL sensitivity in glioma cells.
Oncogene (2005) 24, 6877–6889. doi:10.1038/sj.onc.1208851;
published online 27 June 2005
Keywords: TRAIL; sodium butyrate; apoptosis; glioma
Introduction
TNF-related apoptosis-inducing ligand (TRAIL) is
considered to be a promising candidate for anti-cancer
*Correspondence: KS Choi, Institute for Medical Sciences, Ajou
University School of Medicine, San 5, Wonchon-dong, Youngtong-gu,
Suwon 442-749, South Korea; E-mail: [email protected]
Received 19 October 2004; revised 22 March 2005; accepted 21 April
2005; published online 27 June 2005
therapy, since the cytotoxic activity of TRAIL is
relatively selective for cancer cells compared to normal
cells both in vitro and in vivo (Ashkenazi et al., 1999;
Walczak et al., 1999). However, recent studies have
shown that many types of cancer cells are resistant to the
apoptotic effects of TRAIL (Wang and El-Deiry, 2003;
Shankar and Srivastava, 2004), suggesting that TRAIL
alone is ineffective for the treatment of certain cancers.
Therefore, to establish effective TRAIL-based cancer
therapy, an understanding of the molecular mechanisms
that confer TRAIL resistance to cancer cells and
identification of the sensitizers capable of overcoming
TRAIL resistance is required.
Malignant gliomas, the most common primary brain
tumors, are known to invade the surrounding normal
brain tissue. This often results in incomplete surgical
resection, local recurrence, and poor responses to
multimodal therapeutic interventions, such as radiotherapy and chemotherapy (Legler et al., 1999). Many
glioma cells are known to be resistant to TRAIL, despite
their expression of the TRAIL receptor (Knight et al.,
2001). Therefore, a new therapeutic strategy should be
developed to restore TRAIL-induced apoptotic potency
to glioma cells.
The short-chain fatty acid sodium butyrate is capable
of inducing cell cycle arrest, differentiation, and
apoptosis in a variety of cancer cells (Chen et al.,
2003). In human glioma cells, sodium butyrate has been
shown to inhibit proliferation through modulation of
the protein levels of various cell cycle regulators (Ito
et al., 2001; Kamitani et al., 2002). Sodium butyrate
most likely acts through inhibition of deacetylases, thus
leading to hyperacetylation of chromatin components
such as histones and nonhistone proteins, and ultimately
to alterations in gene expression (Kruh, 1982). Several
histone deacetylase (HDAC) inhibitors, administered
intravenously or intraperitoneally, inhibit tumor growth
in animal cancer models of breast, prostate, lung and
stomach cancers, neuroblastoma, and leukemia, with
little or no toxicity (Shankar and Srivastava, 2004).
Derivatives of butyrate, SAHA, MS-275, and FK228,
all of which are capable of acting as HDAC inhibitors,
are currently being investigated in clinical trials (Piekarz
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6878
and Bates, 2004); preliminary results suggest their
potential applicability in cancer treatment. Recent
studies have shown that leukemia, melanoma, mesothelioma, colon cancer, and breast cancer cells can be
sensitized with HDAC inhibitors to undergo TRAILmediated apoptosis (Rosato et al., 2003; Zhang et al.,
2003; Chopin et al., 2004; Guo et al., 2004; Kim et al.,
2004a; Nakata et al., 2004; Neuzil et al., 2004).
However, the underlying mechanisms by which HDAC
inhibitors sensitize cancer cells to TRAIL-induced
apoptosis are still poorly understood. Recently, upregulation of DR5, a death receptor of TRAIL (Guo
et al., 2004; Kim et al., 2004a; Nakata et al., 2004),
downregulation of FLIP (Hernandez et al., 2001; Guo
et al., 2004), or p21 induction (Chopin et al., 2004) has
been proposed as the potential mechanisms through
which HDAC inhibitors enhance TRAIL-induced
apoptosis. In this study, we show that combined
treatment with subtoxic doses of sodium butyrate and
TRAIL dramatically induces apoptosis in TRAILresistant glioma cells. However, this treatment is not
cytotoxic to normal human astrocytes. Furthermore, the
present study provides the first evidence that inhibition
of Cdc2 activity by sodium butyrate contributes to the
recovery of TRAIL sensitivity in glioma cells through
downregulation of two major inhibitor of apoptosis
proteins (IAPs), survivin and X-linked IAP (XIAP).
Results
Sodium butyrate sensitizes TRAIL-induced apoptosis in
glioma cells but not in normal astrocytes
Many human glioma cell lines are resistant to TRAIL,
as we have reported previously (Kim et al., 2004b).
First, we investigated whether sodium butyrate could
sensitize these glioma cells to TRAIL-induced apoptosis. Four glioma cell lines were treated with sodium
butyrate alone, TRAIL alone, or in combination with
sodium butyrate and TRAIL for 24 h. While cellular
viabilities of glioma cells, assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide
(MTT) assay, were not significantly reduced by treatment with sodium butyrate or TRAIL alone, cell
viability was significantly decreased by the combined
treatment; both when holding the concentration of
TRAIL fixed and varying the concentrations of sodium
butyrate, and, conversely, when holding the concentration of sodium butyrate fixed and varying TRAIL
(Figure 1a). These results demonstrate that the combined treatment with sodium butyrate and TRAIL
effectively induces cell death in these TRAIL-resistant
glioma cells. Next, we investigated whether this combined treatment affects the viability of normal human
astrocytes. The astrocytes were also resistant to either
Figure 1 Subtoxic doses of sodium butyrate sensitize human glioma cells but not normal human astrocytes to TRAIL-induced
apoptosis. (a) Effect of sodium butyrate and/or TRAIL on the viability of glioma cell lines. Four glioma cell lines were treated with
sodium butyrate for 30 min, and further treated with TRAIL for 24 h at the indicated concentrations. Cellular viability was assessed by
MTT assay. Graphs represent the average and standard deviation of three individual experiments. (b) Effect of sodium butyrate and/or
TRAIL on the cellular viabilities of astrocytes. Human primary astrocytes were treated with TRAIL in the absence or presence of
sodium butyrate for 24 h at the indicated concentrations
Oncogene
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6879
TRAIL or sodium butyrate alone (Figure 1b). Furthermore, combined treatment with sodium butyrate and
TRAIL did not induce any significant cell death in
human astrocytes. These results suggest that the sensitizing regimens using sodium butyrate and TRAIL may be
preferentially toxic for glioma cells over normal astrocytes. Next, we investigated whether the apoptosis by this
combined treatment could be blocked by overexpression
of Bcl-2 or Bcl-xL. Interestingly, cellular viabilities were
also significantly reduced in stable cell lines overexpressing Bcl-2 or Bcl-xL (Figure 2a and b). These results
suggest that combined treatment with sodium butyrate
and TRAIL may have a therapeutic effect on glioma cells
overexpressing Bcl-2 or Bcl-xL, which are resistant to
many other chemotherapeutic drugs.
Cotreatment with sodium butyrate and TRAIL recovers
incomplete activation of caspases in TRAIL-resistant
glioma cells
We examined whether the sensitizing effect of sodium
butyrate on TRAIL-mediated apoptosis was associated
with the activation of caspases. U87MG or A172 cells
were treated with 5 mM sodium butyrate for 16 h, or
100 ng/ml TRAIL alone, or pretreated with sodium
butyrate (30 min), followed by TRAIL for the indicated
times. Treatment with 5 mM sodium butyrate alone for
16 h did not induce any proteolytic processing of
caspases (Figure 3a). In response to TRAIL, the 32kDa procaspase-3 was partially cleaved to a 20-kDa
intermediate form after 4 h, but further cleavage into the
active p17 subunit was not detected, nor were other
caspase processing events in both cell lines. However,
treatment with sodium butyrate plus TRAIL induced
the cleavage of caspase-3 into the p20 intermediate form
at 4 h, and its subsequent cleavage into the active p17
subunit after 8 h in A172 cells. Caspase-7, -8, and -9
were also progressively processed into their respective
active forms after combined treatment. We further
assessed the cleavage of several key death substrates
that indicate activation of caspases, including FAK
(caspase-3), PARP (substrate for csapase-3, -7), and Bid
(caspase-8) (Earnshaw et al., 1999). They were degraded
from 8 h after the combined treatment, whereas they
were not degraded following treatment with TRAIL or
sodium butyrate alone (Figure 3b). Similar activation
patterns of caspases in U87MG cells were also observed
in response to sodium butyrate plus TRAIL (Figure 3a
and b). These results suggest that TRAIL resistance in
these glioma cell lines may be associated with a
proteolytic processing blockade of procaspase-3, and a
failure in the subsequent caspase amplification cascade,
and co-treatment with sodium butyrate may contribute
to the relief of proteolytic processing blockade in glioma
cells exposed to TRAIL.
Caspase-8 has been reported to play a critical role as
an initiator caspase in TRAIL-induced apoptosis (Seol
et al., 2001) and the fundamental activation event of
caspase-8 has been reported to be the dimerization of
procaspase-8 monomers rather than their cleavage (Boatright and Salvesen, 2003). Therefore, we first investigated
whether caspase-8 is also important for apoptosis
induced by the combined treatment with sodium butyrate
and TRAIL. Overexpression of CrmA, a viral protein
inhibitor of caspase-8 (Zhou et al., 1997), almost
completely blocked the cell death by this combined
treatment (Figure 4a), demonstrating the critical role of
caspase-8 in this process. We next examined whether
inhibition of caspase-8 activity affects the proteolytic
cleavage of procaspase-3 following TRAIL treatment.
CrmA overexpression completely inhibited the TRAILinduced cleavage of procaspase-3 into the p20 intermediate form (Figure 4b). These results suggest that
caspase-8 is involved at the initial step of TRAIL-induced
proteolytic processing of procaspase-3 in these glioma
cells. Combined treatment-induced proteolytic processing
of caspase-3 into its active forms was also completely
blocked by CrmA overexpression (Figure 4c).
Combined treatment with sodium butyrate and TRAIL
reduces the expression of survivin and XIAP
Figure 2 Overexpression of Bcl-2 or Bcl-xL does not attenuate
sodium butyrate-sensitized TRAIL-induced apoptosis. (a) Analysis
of Bcl-2 or Bcl-xL expression in the stably transfected cell lines.
Western blotting using an anti-Flag Ab was performed to confirm
the overexpression of Flag-tagged Bcl-2 or Flag-tagged Bcl-xL in
selected cell lines. (b) Effect of the combined treatment with sodium
butyrate and TRAIL in the cells overexpressing Bcl-2 or Bcl-xL.
U87MG glioma cells overexpressing Bcl-2 or Bcl-xL were treated
with 5 mM sodium butyrate plus 100 ng/ml TRAIL for 24 h, and
cellular viability was measured by MTT assay
We investigated the possible underlying mechanisms by
which sodium butyrate sensitizes glioma cells to
TRAIL-induced apoptosis. Recently, upregulation
of the TRAIL receptor DR5 by HDAC inhibitors
including sodium butyrate has been proposed to be
involved in the sensitization of Jurkat and HCT116
colon cancer cells to TRAIL-induced apoptosis (Guo
et al., 2004; Kim et al., 2004a; Nakata et al., 2004).
Therefore, we first examined whether sodium butyrate
also enhanced TRAIL-induced apoptosis of glioma cells
through inhibition of HDAC activity and DR5 upregulation. Acetylation levels of histone H3 and H4 were
Oncogene
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6880
Figure 3 Cotreated sodium butyrate effectively restores the proteolytic processing of caspase-3 incompleted by TRAIL alone.
(a) Effect of sodium butyrate and/or TRAIL on the proteolytic processing of caspases. U87MG or A172 cells were treated with 5 mM
sodium butyrate alone, 100 ng/ml TRAIL alone, or a combination of both for the indicated time points. Cell extracts were prepared for
Western blotting to detect the changes in the expression of caspases. Equal loading of the protein samples was confirmed by Western
blotting of a-tubulin. (b) Effect of sodium butyrate and/or TRAIL on the degradation of the substrate proteins of caspases. To confirm
the activation of caspases, Western blotting of the substrate proteins was performed
significantly increased in U87MG and A172 glioma cells
in a dose- and time-dependent manner in response to
sodium butyrate (Figure 5a), demonstrating that their
intracellular HDAC activity was inhibited by sodium
butyrate treatment. However, basal DR5 protein levels
were fairly high in both U87MG cells and A172 cells,
and DR5 upregulation following exposure to sodium
butyrate was not dramatic. Recently, reduction in the
protein levels of the caspase inhibitor FLIP by sodium
butyrate has been implicated in the sensitization of
colon cancer cells and Jurkat cells to TRAIL-induced
apoptosis (Hernandez et al., 2001; Guo et al., 2004).
However, FLIP protein levels were not changed following sodium butyrate treatment in either U87MG or
A172 cells (Figure 5a). These results suggest that other
mechanisms may be responsible for the dramatic
sensitizing effect of sodium butyrate on TRAIL-induced
apoptosis in these glioma cells.
We next examined whether apoptosis induced by
combined treatment with sodium butyrate and TRAIL
was associated with changes in the levels of apoptosisrelated proteins. Bcl-2 protein levels were not altered in
either U87MG cells or A172 cells following treatment
with 5 mM sodium butyrate and 100 ng/ml TRAIL,
whereas this treatment resulted in slightly decreased BclxL protein levels in A172 cells but not in U87MG cells
Oncogene
(Figure 5b). Both the active form and total protein levels
of Akt gradually decreased during apoptosis induced by
the combined butyrate/TRAIL treatment. Protein levels
of two major IAP proteins, survivin and XIAP, were
also significantly reduced during sodium butyratestimulated TRAIL-induced apoptosis, while c-IAP2
protein levels were unchanged. Since many antiapoptotic proteins, including XIAP and Akt, have been
recently reported to be substrates of caspases (Deveraux
et al., 1999; Kirsch et al., 1999; Rokudai et al., 2000),
we examined whether the decrease in the levels of
these proteins during apoptosis induced by combined
butyrate/TRAIL treatment was correlated with enhanced
caspase activity. CrmA overexpression almost completely
blocked the downregulation of Akt by combined
butyrate/TRAIL treatment (Figure 5c), suggesting that
the decrease in the activity and total protein levels of
Akt by the combined treatment resulted from the
activation of caspases. In contrast, CrmA overexpression did not completely block downregulation of XIAP.
Interestingly, survivin still remained downregulated in
CrmA-overexpressing cells treated with sodium butyrate
and TRAIL. Together, these results suggest that another
non-caspase control mechanism may exist for the
downregulation of survivin and XIAP protein levels by
sodium butyrate.
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6881
Figure 4 Caspase-8 plays a critical role not only in TRAILinduced proteolytic processing of procaspase-3 into the p20
intermediate form but also in sodium butyrate-sensitized TRAILinduced apoptosis. (a) Overexpression of CrmA blocks sodium
butyrate-sensitized TRAIL-induced apoptosis. The sublines overexpressing CrmA were treated with 5 mM sodium butyrate plus
100 ng/ml TRAIL for 24 h and the cellular viability was measured
using calcein-AM and EthD-1. Data represent means of triplicate
determinations. Protein levels of CrmA in the respectively selected
stable cell lines are shown in the left panel. (b) CrmA overexpression blocks TRAIL-induced proteolytic processing of procaspase-3. The sublines overexpressing CrmA (clone no. 7) were
treated with 100 ng/ml TRAIL alone for the indicated time points
and cell extracts were prepared for Western blotting of caspase-3.
Cell extracts of U87MG cells treated with 100 ng/ml TRAIL for
16 h were used as a control for Western blotting of caspase-3.
(c) CrmA overexpression blocks the proteolytic processing of
caspase-3 induced by the combined treatment with sodium butyrate
and TRAIL. The sublines overexpressing CrmA (clone no. 7) were
treated with 5 mM sodium butyrate and 100 ng/ml TRAIL in
combination for the indicated time points and cell extracts were
prepared for Western blotting of caspae-3. Cell extracts of U87MG
cells treated with sodium butyrate plus TRAIL for 16 h were used
as a control for Western blotting
Sodium butyrate enhances TRAIL-induced apoptosis
by downregulation of survivin and XIAP
Next, we investigated whether sodium butyrate itself
affects the expression of survivin and XIAP. Treatment
with sodium butyrate alone led to a significant dose- and
Figure 5 Effect of the combined treatment with sodium butyrate
and TRAIL on the expression of the proteins associated with
intracellular regulators of apoptosis. (a) Changes in the acetylation
levels of histone H3, H4 and expression of DR5 or FLIP following
treatment with sodium butyrate. U87MG cells or A172 cells were
treated with sodium butyrate at the indicated concentrations for
24 h or at 5 mM for the indicated times. Western blotting was
performed to detect the changes in the acetylation levels of histone
H3 or H4, and protein levels of DR5 or FLIP. a-Tubulin protein
levels served as the control for protein loading. (b) Effect of the
combined treatment with sodium butyrate and TRAIL on the
expression of antiapoptotic proteins. U87MG or A172 cells were
treated with 5 mM sodium butyrate and 100 ng/ml TRAIL in
combination for the indicated time points, and cell extracts were
prepared for Western blotting of the indicated proteins. (c) Effect
of CrmA overexpression on the expression of Akt, survivin, and
XIAP following the combined treatment. The sublines overexpressing CrmA (clone no. 7) were treated with 5 mM sodium
butyrate and 100 ng/ml TRAIL in combination, and cell extracts
were prepared for Western blotting
time-dependent decrease in survivin protein levels, and,
to a lesser extent, a dose- and time-dependent decrease
in XIAP protein levels (Figure 6a). In contrast, both the
activity and total protein levels of Akt were not altered
by sodium butyrate treatment. To investigate whether
this downregulation of survivin and/or XIAP by sodium
butyrate is critical to trigger TRAIL-induced apoptosis,
we analysed the cytotoxic effect of sodium butyrate plus
TRAIL on stable cell lines overexpressing survivin or
XIAP. In both types of cells, apoptosis induced by
combined butyrate/TRAIL treatment was considerably,
but not completely, attenuated (Figure 6b). These results
suggest that high expression of survivin and/or XIAP
may be involved in TRAIL resistance in glioma cells.
However, sodium butyrate-mediated downregulation of
survivin and XIAP may trigger the completion of
Oncogene
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6882
procaspase-3 processing initiated by TRAIL, resulting
in irreversible cell death.
Sodium butyrate inhibits Cdc2 and Cdk2 activity through
modulation of the expressions of various cell cycle
regulators
We next investigated the possible upstream signaling
pathways involved in sodium butyrate-mediated downregulation of survivin and XIAP. We have recently
demonstrated that treatment with roscovitine, an
inhibitor of Cdc2 and Cdk2 kinase, results in downregulation of survivin and XIAP in glioma cells (Kim
et al., 2004b). Furthermore, sodium butyrate has
previously been reported to increase the expression of
p21, a representative CDK inhibitor, in a variety of
cancer cells (Ito et al., 2001; Pellizzaro et al., 2001).
Therefore, we examined expression of various cell cycle
regulators in response to sodium butyrate. Consistent
with the results of other researchers (Ito et al., 2001;
Pellizzaro et al., 2001), p21 protein levels were significantly increased by treatment with 5 mM sodium
butyrate in both U87MG and A172 cells (Figure 7a). In
contrast, cyclin A and cyclin B protein levels decreased
dramatically, whereas Cdk2 and cyclin E protein levels
were not significantly altered. While Cdc2 protein levels
Figure 6 Critical role of survivin and XIAP in conferring glioma
cells resistance to TRAIL-induced apoptosis. (a) Effect of sodium
butyrate on survivin, XIAP, and Akt protein levels. U87MG and
A172 cells were treated with sodium butyrate at the indicated
concentrations for 24 h or at 5 mM for the indicated times. Cell
extracts were prepared for Western blotting. (b) Effect of survivin
or XIAP overexpression on apoptosis induced by the combined
treatment with sodium butyrate and TRAIL. Control U87MG cells
and U87MG sublines overexpressing survivin or XIAP were
treated with 5 mM sodium butyrate and 100 ng/ml TRAIL for
24 h, and cellular viability was measured using calcein-AM and
EthD-1. Values from each treatment group are expressed as a
percentage relative to the untreated control U87MG cells (100%).
Data represent means of triplicate determinations
Oncogene
were slightly decreased only in U87MG cells, Cyclin D
protein levels were slightly increased only in A172 cells
in response to sodium butyrate treatment. Phosphorylation levels of pRB gradually decreased following sodium
butyrate treatment in these glioma cells. RT–PCR
analysis demonstrated that mRNA levels of both cyclin
A and cyclin B were significantly downregulated by
sodium butyrate (Figure 7b), suggesting that sodium
butyrate modulates the expressions of these cyclins at
Figure 7 Effect of sodium butyrate on the expressions of cell cycle
regulators and activities of Cdc2 and Cdk2. (a) Changes in the
protein levels of the cell cycle regulators following treatment with
sodium butyrate. U87MG or A172 cells were treated with 5 mM
sodium butyrate at the indicated time points, and cell extracts were
prepared for Western blotting. (b) Changes in mRNA levels of
cyclin A and cyclin B following treatment with sodium butyrate.
U87MG cells were treated with 5 mM sodium butyrate at the
indicated time points and RT–PCR was performed to detect the
mRNAs of cyclin A and cyclin B. (c) Inhibition of Cdc2 and Cdk2
kinase by sodium butyrate. U87MG cells were treated with 5 mM
sodium butyrate alone, 100 ng/ml TRAIL alone or 5 mM sodium
butyrate plus 100 ng/ml TRAIL for 24 h. Following treatments, cell
extracts were prepared and Cdc2 or Cdk2 immune complex kinase
assays on histone H1 were performed. Representative results of
three independent experiments are shown. Fold change in the
respective kinase activity was calculated from liquid scintillation
counting of each gel slice
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6883
the transcriptional level. We next examined whether the
kinase activities of Cdc2 and Cdk2 were affected by
sodium butyrate treatment. Treatment of U87MG cells
with 5 mM sodium butyrate alone or combined with
100 ng/ml TRAIL dramatically inhibited histone H1associated Cdc2 and Cdk2 activity, whereas treatment
with 100 ng/ml TRAIL alone did not (Figure 7c). Taken
together, these results demonstrate that inhibition of
Cdc2 and Cdk2 activity by sodium butyrate may result
from the combined effects of downregulation of cyclin A
and cyclin B plus p21 upregulation.
Forced expression of Cdc2 plus cyclin B attenuates
sodium butyrate-stimulated TRAIL-induced apoptosis
Next, we investigated whether sodium butyrate-mediated
downregulation of Cdc2 and/or Cdk2 activity is critical
for its sensitizing effect on TRAIL-induced apoptosis.
U87MG cells were co-transfected with expression vectors
encoding Cdc2 plus cyclin B or Cdk2 plus cyclin A, and
cellular viability was assessed following combined
treatment with sodium butyrate and TRAIL. Interestingly, co-expression of Cdc2 plus cyclin B, but not Cdk2
plus cyclin A, considerably attenuated apoptosis induced
by the combined treatment (Figure 8a). We further
examined whether this attenuating effect on apoptosis by
co-expression of Cdc2 and cyclin was directly associated
with the modulation of survivin and XIAP protein levels.
U87MG cells were transfected with expression vectors
encoding Cdc2 plus cyclin B or Cdk2 plus cyclin A, after
which cells were treated with sodium butyrate. Forced
expression of Cdc2 with cyclin B, but not Cdk2 with
cyclin A, led to a considerable recovery of survivin
protein levels, which were significantly downregulated by
sodium butyrate treatment (Figure 8b). Sodium butyrate-mediated downregulation of XIAP protein levels
was similarly modulated by expression of these cell cycle
regulators. These results suggest that inhibition of Cdc2
activity is critical for downregulation of survivin and
XIAP, providing one mechanism by which sodium
butyrate overcomes barriers against TRAIL-induced
apoptosis of glioma cells.
Other HDAC inhibitors also sensitize glioma cells
to TRAIL-induced apoptosis through Cdc2-mediated
downregulation of survivin and XIAP
We investigated whether other HDAC inhibitors could
sensitize glioma cells to TRAIL-induced apoptosis via
the same mechanism that was observed with sodium
butyrate. Cotreatment with trichostatin A (TSA) or
suberic bishydroxamate (SBHA) significantly enhanced
TRAIL-induced apoptosis in U87MG cells (Figure 9a).
Neither DR5 nor FLIP protein levels were significantly
altered by TSA or SBHA treatment (Figure 9b). However, both cyclin A and cyclin B protein levels dramatically decreased in response to either TSA or SBHA
treatment in a dose-dependent manner. Experiments
using the promoter-luciferase reporter of cyclin A or
cyclin B clearly demonstrated that the respective treat-
Figure 8 Forced expression of Cdc2 plus cyclin B but not Cdk2
plus cyclin A attenuates sodium butyrate-sensitized TRAILinduced apoptosis by recovering the protein levels of survivin and
XIAP. (a) Effect of the expression of cyclin A plus Cdk2 or cyclin B
plus Cdc2 on the apoptosis induced by sodium butyrate and
TRAIL. U87MG cells were co-transfected with the expression
vectors encoding Cdc2 and cyclin B or co-transfected with the
expression vectors encoding Cdk2 and cyclin A at the indicated
concentrations. Transfected cells were then treated with 5 mM
sodium butyrate and 100 ng/ml TRAIL in combination for 24 h.
Cellular viability was measured using calcein-AM and Etd-1. The
values from each treatment group are expressed as a percentage
relative to the untreated control U87MG cells (set at 100%).
(b) Effect of the expression of Cdc2 plus cyclin B or Cdk2 plus
cyclin A on the expression of survivin and XIAP following
treatment with sodium butyrate. U87MG cells were transfected
with plasmids encoding Cdc2 plus cyclin B or Cdk2 plus cyclin A at
the indicated concentrations; transfected cells were then treated
with 5 mM sodium butyrate for 12 h. Cell extracts were prepared for
Western blotting of survivin and XIAP. Expression of transfected
Cdk2, cyclin A, Cdc2, or cyclin B was confirmed by Western
blotting of these proteins. a-Tubulin protein levels served as the
control for protein loading
ment with sodium butyrate, TSA, or SBHA downregulated the promoter activities of cyclin A and cyclin B
in a dose-dependent manner, suggesting that these
HDAC inhibitors commonly downregulate cyclin A and
cyclin B at the transcriptional level (Figure 9c). Interestingly, p21 protein levels were significantly upregulated by
SBHA treatment, but not by TSA treatment (Figure 9b).
Cdc2 protein levels were slightly decreased only by
Oncogene
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6884
SBHA. However, pRB phosphorylation levels were
dramatically downregulated by treatment with either
SBHA or TSA. Furthermore, Cdc2 kinase activity on
histone H1 was also inhibited by treatment with TSA or
SBHA (Figure 10a), as seen with sodium butyrate
treatment (Figure 7c). Survivin and XIAP protein levels
were also downregulated following TSA or SBHA
treatment (Figure 10b), as was observed following sodium
butyrate treatment (Figure 6a). These results therefore
suggest that HDAC inhibitors may sensitize glioma cells
to TRAIL-induced apoptosis through Cdc2-mediated
downregulation of survivin and XIAP in common.
Figure 10 Both TSA and SBHA downregulate Cdc2 activity and
protein levels of survivin and XIAP. (a) Inhibition of Cdc2 activity
following treatment with HDAC inhibitors. U87MG cells were
treated with 100 ng/ml TRAIL, 5 mM sodium butyrate, 100 ng/ml
TSA, or 80 mM SBHA for 24 h. Following treatments, cell extracts
were prepared and Cdc2 immune complex kinase assays on histone
H1 were performed. Representative results of three independent
experiments are shown. Fold changes in the respective kinase
activity were calculated from liquid scintillation counting of each
gel slice. (b) Effect of TSA or SBHA on the expression of survivin
and XIAP. U87MG cells were treated with 100 ng/ml TSA or
80 mM SBHA for 24 h at the indicated concentrations. Cell extracts
were prepared for Western blotting of survivin and XIAP
Discussion
Figure 9 Both TSA and SBHA sensitize glioma cells to TRAILinduced apoptosis, accompanying the significant downregulation of
cyclin A and cyclin B. (a) Cotreated TSA or SBHA sensitizes
glioma cells to TRAIL-induced apoptosis. U87MG cells were
treated with TSA or SBHA for 30 min, and were further treated
with TRAIL for 24 h at the indicated concentrations. Cellular
viability was assessed by MTT assay. Graphs represent the average
and standard deviation of three individual experiments. (b) Effect
of TSA or SBHA on the protein levels of DR5, FLIP, cyclin A,
cyclin B, p21, Cdc2, and pRB. U87MG cells were treated with TSA
or SBHA for 24 h at the indicated concentrations. Western blotting
to detect changes in the expression of indicated proteins was
performed. (c) Treatment with HDAC inhibitors downregulates the
promoter activities of cyclin A and cyclin B. U87MG cells were
transiently transfected with the promoter-reporter constructs,
pCyclin A-Luc, and pCyclin B-Luc, respectively. Luciferase activity
was measured as described in Materials and methods. Similar
results were obtained from three independent experiments
Oncogene
Malignant gliomas, the most common primary brain
tumors, are known to invade surrounding normal brain
tissue, which results in incomplete surgical resection and
a high frequency of local recurrence. Moreover,
malignant gliomas remain poorly responsive to multimodal therapeutic interventions, including radiotherapy
and chemotherapy (Legler et al., 1999). The resistance of
glioblastoma multiforme to treatments is related to its
cellular heterogeneity (Misra et al., 2000) and the
exceptional migratory nature of the tumor cells, which
are able to diffusely infiltrate normal brain tissue
(Lipinski et al., 2003). Due to this poor prognosis and
the absence of successful therapeutics for glioma
patients, new therapeutic paradigms for the treatment
of malignant gliomas need to be explored.
For effective cancer therapeutics, it is very important
to induce apoptosis specifically in malignant tumor cells,
but not in normal cells. TRAIL is a recently identified
member of the TNF family that is capable of inducing
apoptosis in various tumor cells with minimal toxicity to
normal cells (Wiley et al., 1995). In this regard, TRAIL
is an attractive candidate for cancer treatment. However, various types of cancer cells, including breast,
prostate, ovarian and lung carcinoma, multiple myeloma, leukemia, and glioma cells, are resistant to
apoptosis induced by TRAIL (Shankar and Srivastava,
2004). Therefore, use of TRAIL alone is not an
appropriate option for treating these types of cancer.
Combination therapy may sensitize TRAIL-resistant
cancer cells to TRAIL-induced apoptosis. Resistance to
TRAIL appears to occur through the modulation of
various molecular targets, including differential expression of death receptors, constitutively active Akt and
NF-kB, overexpression of c-FLIP, mutations in Bax
and Bak, and defects in the release of mitochondrial
proteins, in resistant cells (Shankar and Srivastava,
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6885
2004). Moreover, high expression of IAPs may also
contribute to TRAIL resistance, since IAPs such as
survivin, XIAP, c-IAP1, and c-IAP2 block apoptosis at
the effector phase, a point at which multiple signaling
pathways converge (Deveraux and Reed, 1999).
HDAC inhibitors are potent regulators of gene
expression through their effect on the acetylation of
core histones (Marks et al., 2001; Vigushin and
Coombes, 2002). HDAC inhibitors have recently
emerged as promising chemotherapeutic agents. Inhibition of HDACs may contribute to the induction of cell
cycle arrest during the G1 and/or G2 phase, differentiation, and/or apoptosis in various cancer cells (Marks
et al., 2001). Several HDAC inhibitors can halt tumor
growth in animal models with little toxicity, while the
inhibition of HDAC is normally not sufficient to cause
cell death in nontumor cells (Vigushin et al., 2001;
Atadja et al., 2004). This selectivity for tumor cells
makes these compounds particularly attractive for
cancer therapy, and several newly developed HDAC
inhibitors have been tested on cancer patients in clinical
trials (Piekarz and Bates, 2004). Among the short-chain
fatty acids, butyrate appears to have the most profound
effect on growth inhibition and differentiation in various
kinds of cancer cells (Heerdt et al., 1994, 1997). It is
often suggested that the antitumorigenic effects of
sodium butyrate are exerted by inhibition of HDACs
(Bernhard et al., 1999; Siavoshian et al., 2000). In
glioma cells, sodium butyrate has been reported to
inhibit not only cell proliferation (Ito et al., 2001;
Kamitani et al., 2002) but also invasion (Ito et al., 2001)
as well as VEGF secretion (Sawa et al., 2002). These
results suggest that sodium butyrate may be a useful
therapeutic agent for treating malignant gliomas. However, the molecular mechanisms underlying these effects
have not been fully understood.
Recent studies have shown that leukemia, melanoma,
mesothelioma, colon cancer, and breast cancer cells
can be sensitized with HDAC inhibitors to undergo
TRAIL-mediated apoptosis (Rosato et al., 2003;
Zhang et al., 2003; Chopin et al., 2004; Guo et al.,
2004; Kim et al., 2004a; Nakata et al., 2004; Neuzil
et al., 2004), suggesting that combining HDAC inhibitor with TRAIL may be a promising therapeutic
strategy. Sodium butyrate has been proposed to disrupt
the association of HDAC with Sp1, which in turn
may lead to the decondensation of local chromatin,
activate the transcription of DR5, and activate TRAILinduced apoptotic signaling pathways in colon
cancer cells (Kim et al., 2004a). Decreased expression
of FLIP, which functions as a dominant negative
inhibitor of caspase-8 due to its structural resemblance
to caspase-8 without proteolytic activity, has also
been proposed to be involved in this sensitizing effect
of sodium butyrate (Hernandez et al., 2001). In our
study, combined treatment with sodium butyrate and
TRAIL, but not either agent alone, significantly induced
apoptosis in various glioma cell lines. However, no
significant upregulation of DR5 occurred, and FLIP
protein levels were not altered by sodium butyrate
treatment, suggesting that other mechanisms may exist
for sodium butyrate-enhanced TRAIL-induced apoptosis
in glioma cells.
Previous reports have indicated that survivin and
XIAP are overexpressed in gliomas and that these high
expression levels correlate with abbreviated patient
survival, unfavorable prognosis, and resistance to
therapy (Chakravarti et al., 2002). In our study, survivin
and XIAP protein levels were significantly suppressed
following treatment with sodium butyrate, and the
extent of survivin downregulation was greater than that
of XIAP. However, apoptosis was not induced in glioma
cells by treatment with 5 mM sodium butyrate alone.
These results suggest that downregulation of survivin
and XIAP by sodium butyrate may not be sufficient to
trigger activation of caspases in these glioma cells.
Interestingly, the 32-kDa procaspase-3 was partially
cleaved to a 20-kDa intermediate form at 4 h following
treatment with TRAIL, but further cleavage into the
active p17 subunit did not occur in glioma cells,
indicating that the proteolytic processing of procaspase-3 can be primed, but not completed by TRAIL.
Reportedly, caspase-3 is initially synthesized as an
inactive 32-kDa precursor, which is subsequently
processed by proteolytic cleavage at ESMDkS (amino
acids 25–29) and IETDkS (amino acids 172–176) to
produce the mature caspase-3 (Nicholson et al., 1995).
Cleavage at the IETDkS site seems to occur first,
yielding a mature p12 subunit and a p20 intermediate
product (Han et al., 1997). Subsequently, a cleavage
occurs at the ESMDkS site in the p20 intermediate,
yielding the mature p17 subunit. Together, the p12 and
p17 subunits then form the mature caspase-3. In
TRAIL-induced apoptosis, caspase-8, rather than caspase-10, has been reported to play a critical role as an
initiator caspase (Seol et al., 2001; Suliman et al., 2001).
The fundamental activation event of caspase-8 has been
reported to be the dimerization of procaspase-8 monomers rather than their cleavage (Boatright and Salvesen,
2003). Instead, the cleavage event is thought to provide
stability to the dimer generated during death-inducing
signaling complex (DISC) formation. To examine the
role of caspase-8 in our system, we tested the effect of
caspase-8 inhibition on the proteolytic cleavage of
procaspase-3 following TRAIL treatment. As shown in
Figure 4b, overexpression of CrmA, a viral protein
inhibitor of caspase-8, completely inhibited the TRAILinduced cleavage of procaspase-3 into the p20 intermediate form, suggesting that caspase-8 activity may be
required for this initial proteolytic processing step of
procaspase-3 in our system. Therefore, treatment with
TRAIL alone may induce dimerization of procaspase-8,
leading to partial activation that is sufficient for initial
proteolytic processing of procaspase-3 into p20, but not
for the autoproteolytic processing of procaspase-8 into
its active forms. Sun et al. (2002) have reported that
XIAP does not inhibit the initial processing of
procaspase-3 into its p20 intermediate form by caspase-8, but does inhibit the caspase activity of p20/p12
and autocatalytic processing of p20 into p17. Moreover,
ubiquitin-protein ligase activity of XIAP has been
reported to promote proteasomal degradation of active
Oncogene
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6886
form caspase-3, but not procaspase-3 (Suzuki et al.,
2001). Many studies have also established the role
of survivin in apoptosis, and suggested that its
antiapoptotic function is related to its ability to
directly or indirectly inhibit caspases (Zaffaroni and
Daidone, 2002). Therefore, it is very possible that
XIAP and survivin overexpressed in glioma cells play
a critical role in blocking of the complete proteolytic
processing of the p20 intermediate form into the p17
subunit of caspase-3 following exposure to TRAIL.
However, after combined treatment with sodium
butyrate and TRAIL, reduction of survivin and
XIAP protein levels by sodium butyrate may allow
the complete autoprocessing cascade of the p20
intermediate form that is generated by TRAIL into
qthe p17 active subunit and subsequently amplify
the activation of other caspases. Therefore, downregulation of survivin and XIAP may be one mechanism
by which sodium butyrate overcomes TRAIL resistance
in glioma cells. Overexpression of either survivin
or XIAP reduced the degree of apoptosis induced
by sodium butyrate/TRAIL treatment, thereby highlighting the functional significance of sodium butyratemediated downregulation of survivin and/or XIAP
during apoptosis.
Recently, we have shown that inhibition of Cdc2
activity by roscovitine downregulates survivin and
XIAP, thereby contributing to enhancement of
TRAIL-induced apoptosis (Kim et al., 2004b). Moreover, Chopin et al. (2004) have reported that butyrate
and TRAIL synergistically induce apoptosis in breast
cancer cells and that this process requires p21. p21 is a
representative inhibitor of cyclin-dependent kinases
(Cdks) (Brugarolas et al., 1999). Therefore, in the
present study, we further examined the underlying
mechanisms involved in sodium butyrate-mediated
downregulation of survivin and XIAP by investigating
whether it was also associated with the inhibition of Cdk
activity. In our study, p21 protein levels were significantly enhanced in U87MG and A172 cells following
sodium butyrate treatment. In contrast, cyclin A and
cyclin B protein levels were significantly reduced by the
same treatment. Not only histone H1-associated Cdc2
and Cdk2 kinase activity but also endogenous phosphorylation levels of pRB were also significantly
decreased following sodium butyrate treatment. Very
interestingly, exogenously expressed cyclin B plus Cdc2,
but not cyclin A plus Cdc2, suppressed sodium butyrateenhanced TRAIL-induced apoptosis. Downregulation
of survivin and XIAP by sodium butyrate was also
alleviated by this forced co-expression of Cdc2 plus
cyclin B, but not by that of Cdk2 plus cyclin A. Taken
together, these results suggest that suppression of Cdc2
activity by sodium butyrate plays a critical role in the
reduction of survivin and XIAP protein levels. In the
present study, we also present evidence that TSA and
SBHA, HDAC inhibitors classified as hydroxamic acids
(Vigushin and Coombes, 2002), also sensitize glioma
cells to TRAIL-induced apoptosis through inhibition of
Cdc2 activity and the subsequent downregulation of
survivin and XIAP. Recently, various HDAC inhibitors
Oncogene
have been shown to upregulate p21 through the
transcriptional control (Gui et al., 2004; Yokota et al.,
2004). However, p21 was not upregulated by treatment
with TSA, although the sensitizing effect of TSA
on TRAIL-mediated apoptosis was similar to that
of sodium butyrate or SBHA (Figures 1a and 9a).
These results suggest that p21 upregulation itself
may not be essential for the reduction of survivin and
XIAP protein levels, as well as for the sensitizing
effect of these HDCA inhibitors on TRAIL-induced
apoptosis. We further investigated the role of p21
in sodium butyrate-facilitated TRAIL-induced apoptosis using isogenic colon cancer cell lines differing only
in p21 status. HCT116 p21 þ / þ cells were relatively
sensitive to TRAIL and TRAIL-mediated apoptosis
was significantly potentiated by cotreatment with
sodium butyrate. However, HCT116 p21/ cells
showed similar responses (Supplementary Figure 1),
indicating that p21 may not play a critical role
in apoptosis induced by sodium butyrate plus
TRAIL in these cells, and further suggesting that
the requirement of p21 in HDAC inhibitor-facilitated
TRAIL-mediated apoptosis may differ according to the
cell type. In contrast, both cyclin A and cyclin B protein
levels were significantly downregulated by either TSA or
SBHA. Our RT–PCR analysis and experiments using
promoter-luciferase reporter clearly demonstrated that
treatment with HDAC inhibitors transcriptionally
downregulates cyclin A and cyclin B (Figures 7b and
9c). Although downregulation of cyclin A and cyclin B
mRNA by TSA has been previously reported (Nair
et al., 2001), we show here for the first time that the
transcriptional downregulation of cyclin A and cyclin B
mRNA may be a common effect of HDAC inhibitors,
including sodium butyrate, TSA, and SBHA, on glioma
cells. Future work will be needed to assess whether
NF-Y, a transcriptional factor complex that binds to
CCAAT boxes located in the cyclin A (Kramer et al.,
1997) and cyclin B (Katula et al., 1997) gene promoters,
may be targeted by HDAC inhibitors for downregulation of cyclin A and cyclin B. Taken together, the
inhibition of Cdc2 activity by HDAC inhibitors may
result from the combined effects of transcriptional
repression of cyclin A and cyclin B plus transcriptional
upregulation of p21. Subsequently, this decreased Cdc2
activity by HDAC inhibitors leads to the reduction of
survivin and XIAP protein levels, contributing to the
recovery of TRAIL sensitivity in glioma cells (summarized in Figure 11).
In our study, treatment with sodium butyrate and
TRAIL did not enhance TRAIL-induced apoptosis in
normal human astrocytes, whereas the same treatment
significantly induced apoptosis in various human glioma
cells. Moreover, overexpression of either Bcl-2 or Bcl-xL
did not block apoptosis induced by combined butyrate/
TRAIL treatment. Therefore, co-treatment with sodium
butyrate and TRAIL is a potentially safe and attractive
treatment strategy against intractable human malignant
gliomas, particularly in cases where resistance to
apoptosis induced by anticancer drugs is due to overexpression of either Bcl-2 or Bcl-xL.
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6887
number less than 5 were used in the present study. Immunofluorescence study indicated that better than 99% of cells
expressed glial fibrillary acidic protein (GFAP)-positive
immunoreactivity, a cell-type-specific marker for astrocytes.
Permission to use human brain tissues for research was
granted by the Ethics Committee of the University.
Measurement of cellular viability
Cellular viability was assessed by MTT assay. Cells
(5 103 cells/well) were plated in a 96-well plate. Following
treatment, the MTT stock solution (5 mg/ml in PBS) was
added to the medium, and incubated with cells at 371C for 3 h
to allow cell-mediated reduction of MTT. The absorbance was
measured at 540 nm in a microtiter plate reader. Alternatively,
the cell viability was assessed by double labeling of cells with
2 mM calcein-AM and 4 mM Etd-1. The calcein-positive live cells
and Etd-1-positive dead cells were visualized using a fluorescence microscope (Nikon Diaphot 300, Japan).
Figure 11 Schematic diagram of the apoptotic pathway induced
by the combined treatment with HDAC inhibitor and TRAIL. In
TRAIL-resistant glioma cells, the 32-kDa procaspase-3 is partially
cleaved to a 20-kDa intermediate form in response to TRAIL, but
further cleavage into the active p17 subunit and other caspase
processing events do not occur. HDAC inhibitors suppress Cdc2
activity by the combined effects of downregulation of cyclin A and
cyclin B plus p21 upregulation. Cotreated HDAC inhibitors
facilitate completion of proteolytic processing of procaspase-3,
which is partially blocked by treatment with TRAIL alone, through
inhibition of the Cdc2 activity and the subsequent downregulation
of XIAP and survivin
Materials and methods
Chemicals and antibodies (Abs)
Recombinant human TRAIL/Apo2 ligand was from KOMA
Biotech Inc. (Korea). Calcein-AM and ethidium homodimer-1
(Etd-1) were from Molecular Probe (Eugene, OR, USA).
Sodium butyrate, TSA, and SBHA were obtained from Sigma
(St Louis, MO, USA). The following Abs were used: anticaspase-8, caspase-3, caspase-7, survivin, FLIP, and XIAP
(Stressgen, British Columbia, Canada); anti-caspase-9, Rb,
FAK, Bcl-2, Bcl-xL, c-IAP2, cyclin A, cyclin B, cyclin D,
cyclin E, Cdc2, Cdk2, and Myc (Santa Cruz Biotechnologies,
Santa Cruz, CA, USA); anti-phospho-Akt, Akt, and Bid (Cell
Signaling, Beverly, MA, USA); anti-Flag M2 (Sigma, St Louis,
MO, USA); anti-CrmA (BD Pharmingen, San Diego, CA,
USA); anti-a-tubulin and DR5 (Calbiochem., San Diego, CA,
USA); anti-acetylated histone H3, acetylated histone H4 and
anti-PARP (Upstate biotechnology, Lake Placid, NY, USA);
horseradish peroxidase-conjugated anti-rabbit IgG and horseradish peroxidase-conjugated anti-mouse IgG HRP (Zymed
Laboratories, Inc., South San Francisco, CA, USA).
Culture of glioma cell lines and normal human astrocytes
The human malignant glioma cell lines U87MG, U251MG,
A172, and T98 were cultured in Dulbecco’s modified Eagle’s
medium (DMEM) (GIBCO-BRL, Life Technologies, NY,
USA) supplemented with 10% fetal bovine serum and
antibiotics (GIBCO-BRL, Life Technologies, NY, USA).
The primary cultures of normal human astrocytes were
prepared from 14-week-gestation fetal cerebrum tissues as
described previously (Kim, 1985; Kim et al., 1986). Human
astrocyte cultures were subcultured in DMEM containing 10%
fetal bovine serum every 2 weeks and cell cultures passage
Immunoblotting
Cells were washed in PBS and lysed in boiling sodium dodecyl
sulfate/polyacrylamide gel electrophoresis (SDS–PAGE) sample buffer (62.5 mmol/l, Tris (pH 6.8), 1% SDS, 10% glycerol,
and 5% b-mercaptoethanol). The lysates were boiled for 5 min,
separated by SDS–PAGE, and transferred to an Immobilon
membrane (Millipore, Bedford, MA, USA). After blocking
nonspecific binding sites for 1 h by 5% skim milk, membranes
were incubated for 2 h with specific Abs. Membranes were then
washed three times with TBST and incubated further for 1 h
with horseradish peroxidase-conjugated anti-rabbit, -mouse or
-goat Ab. Visualization of protein bands was accomplished
using ECL (Amersham Life Science, Buckinghamshire, UK).
RT–PCR
Total RNA was isolated from U87MG cells treated with 5 mM
sodium butyrate for the indicated time points using RNAZolB
(Tel-Test, Friendswood, TX, USA). Total RNA (2 mg) from
each cell culture was reverse transcribed using oligo-dT
primers and AMV reverse transcriptase (TaKaRa, Otsu,
Shiga). The cDNAs were amplified by PCR (941C for 30 s,
601C for 30 s and 721C for 1 min) with Taq DNA polymerase.
To ensure exponential amplification, four aliquots were
removed from each PCR assay at cycles 20, 25, 30, or 35
cycles (which were determined in preliminary experiments to
produce the weakest detectable PCR product for each gene)
and every 2–4 cycles thereafter. Amplified products were
analysed by agarose gel electrophoresis at cycles within the
linear range of mRNA amplification. The sequences of
oligonucleotide primers used for RT–PCR and the expected
transcript sizes are as follows: cyclin A, 50 -TCCAAGAG
GACCAGGAGAATATCA-30 and 50 -TCCTCATGGTAG
TCTGGTACTTCA-30 ; cyclin B1, 50 -AAGAGCTTTAAA
CTTTGGTCTGGG-30 and 50 -CTTTGTAAGTCCTTGATTT
ACCATG-30 ; GAPDH, 50 -CGTCTTCACCACCATGGA
GA-30 and 50 -CGGCCATCACGCCACAGTTT-30 .
Cdc2 and Cdk2 immune complex kinase assay
After treatment of U87MG cells with HDAC inhibitor alone
or sodium butyrate plus TRAIL as indicated, cells were lysed
in buffer A (1% Nonidet P-40, 50 mM Tris-HCl (pH 8.0),
150 mM NaCl, 10 mg/ml aprotinin, 10 mg/ml leupeptin, 20 mM
NaF, 1 mM Na3VO4, 10 mg/ml PMSF) at 41C for 15 min. Cell
lysates were cleared by centrifugation at 13 000 rpm for 15 min.
Protein concentrations were quantitated by Bio-Rad protein
Oncogene
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6888
assay. A total of 500 mg of protein was used for each
immunoprecipitation. Cdc2 or Cdk2 in cell extracts was
incubated with anti-Cdc2 or anti-Cdk2 Ab (1 mg/reaction) for
3 h at 41C, respectively. 15 ml of protein A/G-agarose
(Oncogene Research Products, Cambridge, MA, USA) was
added into the mixture, which was then further incubated for
1 h. Immune complexes were centrifuged at 2500 r.p.m for
5 min and the precipitates were washed three times with buffer
A and twice with kinase buffer (50 mM Tris-HCl (pH 7.5),
10 mM MgCl2, 1 mM DTT). Cdc2 or Cdk2 kinase assay on
histone H1 was performed by mixing the respective immune
complexes with 5 mg of histone H1 and 10 mCi of [g-32P]ATP in
30 ml of kinase buffer. The kinase reaction was performed at
301C for 30 min and then terminated with 2 SDS–PAGE
sample buffer. The reaction mixtures were resolved by SDS–
PAGE analysis. Gels were stained with Coomassie Blue
staining solution and dried. The extent of phosphorylation
was measured by liquid scintillation counting of the gel slices
of each substrate.
Establishment of the stable U87MG cell lines overexpressing
Bcl-2, Bcl-xL, CrmA, XIAP, or survivin
U87MG cells were transfected with the following: a mammalian expression vector containing Flag-tagged bcl-2 cDNA and
bcl-xL (Huang et al., 1997) (kindly provided by Dr A Strasser,
The Walter and Eliza Hall Institute of Medical Research,
Australia); a vector containing CrmA cDNA (Tewari et al.,
1995), kindly provided by Professor VM Dixit (University of
Michigan Medical School); a vector containing Myc-tagged
XIAP (Nomura et al., 2003) (kindly provided by Professor T
Nomura, Oita Medical University, Japan). Stable cell lines
overexpressing Bcl-2 or Bcl-xL were selected in fresh media
containing puromycin (4 mg/ml). Overexpression of Bcl-2 or
Bcl-xL in the stable cell lines was analysed by Western blotting
using anti-Flag Ab. Stable cell lines overexpressing CrmA or
XIAP were selected in fresh media containing G418 (500
mg/ml). Overexpression of CrmA in the stable cell lines was
analysed by Western blotting using anti-CrmA Ab, while
overexpression of XIAP in the respective stable cell lines
was analysed using anti-c-Myc Ab (9E10). To establish
stable cell lines overexpressing survivin, U87MG cells were
co-transfected with a flag/survivin fusion protein expression
vector (Kobayashi et al., 1999), kindly provided by Professor
T Tokuhisa (Chiba University Graduate School of Medicine,
Japan), and pcDNA3. Transfected cells were selected
with fresh media containing G418 (500 mg/ml) and overexpression of survivin was analysed by Western blotting
using anti-Flag Ab.
Expression of Cdc2, Cdk2, cyclin A or cyclin B by transient
transfection
U87MG cells were plated into 24-well plates at 5 104
cells/well. After 24 h, cells were transfected with the
plasmid encoding Cdc2, Cdk2 (kindly provided by
Dr Sander van den Heuvel, Massachusetts General
Hospital), cyclin A, or cyclin B (kindly provided by Dr Paul
Robbins, University of Pittsburgh) at the indicated concentrations using Lipofectamine Plus reagent (Gibco BRL,
Grand Island, NY, USA) following the manufacturer’s
instructions. Transfected cells were incubated for 24 h and
then further treated with 5 mM sodium butyrate and/or 100 ng/
ml TRAIL for 24 h. Transfection efficiency, assessed using a
green fluorescence protein-encoding plasmid (pEGFP-C1;
from Promega), reached about 70% in these experiments.
Cellular viability was assessed using calcein-AM and Etd-1 as
described above. To further confirm the expression of
transgenes, Western blotting of the transfected proteins was
performed.
Transfection and luciferase assays
U87MG cells were seeded at 5 105 cells in 60 mm culture
dishes. The following day, 60–80% confluent cells were
transiently transfected in duplicate with 1 mg of pCyclin
A-Luc or pCyclin B-Luc, the reporter construct containing
the promoter of human cyclin A or cyclin B1 (kindly provided
by Dr DY Shin, Dankook University College of Medicine),
together with 0.2 mg of pCMV-b-gal reporter plasmid
using Lipofectamine Plus reagent (Gibco BRL, Grand
Island, NY, USA) following the manufacturer’s instructions.
In all transfection assays, pSV-luc reporter plasmid was
transfected as negative control, while the transfection of
pCMV-b-gal reporter plasmid was carried out to assess
transfection efficiency. Transfected cells were incubated in
serum-containing media for 24 h and luciferase activity was
measured in 20 ml of cell lysates using a luciferase assay system
(Promega, Madison, WI, USA) on a TD-20/20 Luminometer
(Turner designs, Sunnyvale, CA, USA) for 12 s. Luciferase
activity was normalized for b-galactosidase activity in cell
lysates.
Acknowledgements
We thank Professor A Strasser (The Walter and Eliza Hall
Institute of Medical Research) for providing us with Bcl-2 and
Bcl-xL expression vector; Professor VM Dixit (University of
Michigan Medical School) for CrmA expression vector; and
Professor T Tokuhisa (Chiba University) for a flag/survivin
fusion protein expression vector; Professor Y Nomura (Oita
Medical University) for XIAP expression vector; Dr S van den
Heuvel (Massachusetts General Hospital) for plasmids expressing Cdc2 and Cdk2; Dr P Robbins (University of Pittsburgh)
for plasmids expressing cyclin A and cyclin B; Dr DY Shin
(Dankook University college of Medicine) for the reporter
constructs containing the promoters of human cyclin A and
cyclin B1; Dr B Vogelstein for human colon cancer cell lines,
HCT116 wild-type cells and HCT116 p21/ cells. This study
was supported by grants from the National R & D Program
for Cancer Control (2003), Ministry of Health & Welfare,
Korea and the KOSEF/BDRC Ajou University (R11-1998052-08009).
References
Ashkenazi A, Pai RC, Fong S, Leung S, Lawrence DA,
Marsters SA, Blackie C, Chang L, McMurtrey AE, Hebert
A, DeForge L, Koumenis IL, Lewis D, Harris L, Bussiere J,
Koeppen H, Shahrokh Z and Schwall RH. (1999). J. Clin.
Invest., 104, 155–162.
Atadja P, Gao L, Kwon P, Trogani N, Walker H, Hsu M,
Yeleswarapu L, Chandramouli N, Perez L, Versace R, Wu
Oncogene
A, Sambucetti L, Lassota P, Cohen D, Bair K, Wood A and
Remiszewski S. (2004). Cancer Res., 64, 689–695.
Bernhard D, Ausserlechner MJ, Tonko M, Loffler M,
Hartmann BL, Csordas A and Kofler R. (1999). FASEB
J., 13, 1991–2001.
Boatright KM and Salvesen GS. (2003). Curr. Opin. Cell Biol.,
15, 725–731.
Sensitization of TRAIL-induced apoptosis by sodium butyrate
EH Kim et al
6889
Brugarolas J, Moberg K, Boyd SD, Taya Y, Jacks T and
Lees JA. (1999). Proc. Natl. Acad. Sci. USA, 96,
1002–1007.
Chakravarti A, Noll E, Black PM, Finkelstein DF, Finkelstein
DM, Dyson NJ and Loeffler JS. (2002). J. Clin. Oncol., 20,
1063–1068.
Chen JS, Faller DV and Spanjaard RA. (2003). Curr. Cancer
Drug Targets, 3, 219–236.
Chopin V, Slomianny C, Hondermarck H and Le Bourhis X.
(2004). Exp. Cell. Res., 298, 560–573.
Deveraux QL and Reed JC. (1999). Genes Dev., 13, 239–252.
Deveraux QL, Leo E, Stennicke HR, Welsh K, Salvesen GS
and Reed JC. (1999). EMBO J., 18, 5241–5251.
Earnshaw WC, Martins LM and Kaufmann SH. (1999). Annu.
Rev. Biochem., 68, 383–424.
Gui CY, Ngo L, Xu WS, Richon VM and Marks PA. (2004).
Proc. Natl. Acad. Sci. USA, 101, 1241–1246.
Guo F, Sigua C, Tao J, Bali P, George P, Li Y, Wittmann S,
Moscinski L, Atadja P and Bhalla K. (2004). Cancer Res.,
64, 2580–2589.
Han Z, Hendrickson EA, Bremner TA and Wyche JH. (1997).
J. Biol. Chem., 272, 13432–13436.
Heerdt BG, Houston MA and Augenlicht LH. (1994). Cancer
Res., 54, 3288–3293.
Heerdt BG, Houston MA and Augenlicht LH. (1997). Cell
Growth Differ., 8, 523–532.
Hernandez A, Thomas R, Smith F, Sandberg J, Kim S, Chung
DH and Evers BM. (2001). Surgery, 130, 265–272.
Huang DC, Cory S and Strasser A. (1997). Oncogene, 14,
405–414.
Ito N, Sawa H, Nagane M, Noguchi A, Hara M and Saito I.
(2001). Neurosurgery, 49, 430–436.
Kamitani H, Taniura S, Watanabe K, Sakamoto M,
Watanabe T and Eling T. (2002). Neuro-oncology, 4, 95–101.
Katula KS, Wright KL, Paul H, Surman DR, Nuckolls FJ,
Smith JW, Ting JP, Yates J and Cogswell JP. (1997). Cell
Growth Differ., 8, 811–820.
Kim EH, Kim SU, Shin DY and Choi KS. (2004b). Oncogene,
23, 446–456.
Kim SU, Moretto G, Lee V and Yu RK. (1986). J. Neurosci.
Res., 15, 303–321.
Kim SU. (1985). J. Neuroimmunol., 8, 255–282.
Kim YH, Park JW, Lee JY and Kwon TK. (2004a).
Carcinogenesis, 25, 1813–1820.
Kirsch DG, Doseff A, Chau BN, Lim DS, de Souza-Pinto NC,
Hansford R, Kastan MB, Lazebnik YA and Hardwick JM.
(1999). J. Biol. Chem., 274, 21155–21161.
Knight MJ, Riffkin CD, Muscat AM, Ashley DM and
Hawkins CJ. (2001). Oncogene, 20, 5789–5798.
Kobayashi K, Hatano M, Otaki M, Ogasawara T and Tokuhisa
T. (1999). Proc. Natl. Acad. Sci. USA, 96, 1457–1462.
Kramer A, Carstens CP, Wasserman WW and Fahl WE.
(1997). Cancer Res., 57, 5117–5121.
Kruh J. (1982). Mol. Cell Biochem., 42, 65–82.
Legler JM, Ries LA, Smith MA, Warren JL, Heinmann EF,
Kaplan RS and Linnet MS. (1999). J. Natl. Cancer Inst., 91,
1382–1390.
Lipinski CA, Tran NL, Bay C, Kloss J, McDonough WS,
Beaudry C, Berens ME and Loftus JC. (2003). Mol. Cancer
Res., 1, 323–332.
Marks PA, Richon VM, Breslow R and Rifkind RA. (2001).
Curr. Opin. Oncol., 13, 477–483.
Misra A, Chattopadhyay P, Dinda AK, Sarkar C, Mahapatra
AK, Hasnain SE and Sinha S. (2000). J. Neurooncol., 48,
1–12.
Nair AR, Boersma LJ, Schiltz L, Chaudhry MA and Muschel
RJ. (2001). Cancer Lett., 166, 55–64.
Nakata S, Yoshida T, Horinaka M, Shiraishi T, Wakada M
and Sakai T. (2004). Oncogene, 23, 6261–6271.
Neuzil J, Swettenham E and Gellert N. (2004). Biochem.
Biophys. Res. Commun., 314, 186–191.
Nicholson DW, Ali A, Thornberry NA, Vaillancourt JP, Ding
CK, Gallant M, Gareau Y, Griffin PR, Labelle M, Lazebnik
YA, Munday NA, Raju SM, Smulson ME, Yamin TT and
Miller DK. (1995). Nature, 376, 37–43.
Nomura T, Mimata H, Takeuchi Y, Yamamoto H, Miyamoto
E and Nomura Y. (2003). Urol. Res., 31, 37–44.
Pellizzaro C, Coradini D, Daniotti A, Abolafio G and
Daidone MG. (2001). Int. J. Cancer, 91, 654–657.
Piekarz R and Bates S. (2004). Curr. Pharm. Des., 10,
2289–2298.
Rokudai S, Fujita N, Hashimoto Y and Tsuruo T. (2000). J.
Cell Physiol., 182, 290–296.
Rosato RR, Almenara JA, Dai Y and Grant S. (2003). Mol.
Cancer Ther., 2, 1273–1284.
Sawa H, Murakami H, Ohshima Y, Murakami M,
Yamazaki I, Tamura Y, Mima T, Satone A, Ide W,
Hashimoto I and Kamada H. (2002). Brain Tumor Pathol.,
19, 77–81.
Seol DW, Li J, Seol MH, Park SY, Talanian RV and Billiar
TR. (2001). Cancer Res., 61, 1138–1143.
Shankar S and Srivastava RK. (2004). Drug Resist. Update, 7,
139–156.
Siavoshian S, Segain JP, Kornprobst M, Bonnet C,
Cherbut C, Galmiche JP and Blottiere HM. (2000). Gut,
46, 507–514.
Suliman A, Lam A, Datta R and Srivastava RK. (2001).
Oncogene, 20, 2122–2133.
Sun XM, Bratton SB, Butterworth M, MacFarlane M and
Cohen GM. (2002). J. Biol. Chem., 277, 11345–11351.
Suzuki Y, Nakabayashi Y and Takahashi R. (2001). Proc.
Natl. Acad. Sci. USA, 98, 8662–8667.
Tewari M, Beidler DR and Dixit VM. (1995). J. Biol. Chem.,
270, 18738–18741.
Vigushin DM and Coombes RC. (2002). Anticancer Drugs, 13,
1–13.
Vigushin DM, Ali S, Pace PE, Mirsaidi N, Ito K, Adcock I
and Coombes RC. (2001). Clin. Cancer Res., 7, 971–976.
Walczak H, Miller RE, Ariail K, Gliniak B, Griffith TS,
Kubin M, Chin W, Jones J, Woodward A, Le T, Smith C,
Smolak P, Goodwin RG, Rauch CT, Schuh JC and Lynch
DH. (1999). Nat. Med., 5, 157–163.
Wang S and El-Deiry WS. (2003). Oncogene, 22, 8628–8633.
Wiley SR, Schooley K, Smolak PJ, Din WS, Huang CP,
Nicholl JK, Sutherland GR, Smith TD, Rauch C, Smith CA
and Goodwin RG. (1995). Immunity, 3, 673–682.
Yokota T, Matuszaki Y, Miyazawa K, Zindy F, Roussel MF
and Sakai T. (2004). Oncogene, 23, 5340–5349.
Zaffaroni N and Daidone MG. (2002). Drug Resist. Update,
5, 65–72.
Zhang XD, Gillespie SK, Borrow JM and Hersey P. (2003).
Biochem. Pharmacol., 66, 1537–1545.
Zhou Q, Snipas S, Orth K, Muzio M, Dixit VM and Salvesen
GS. (1997). J. Biol. Chem., 272, 7797–8000.
Supplementary Information accompanies the paper on Oncogene website (http://www.nature.com/onc)
Oncogene