Biochem. J. (2011) 438, 349–358 (Printed in Great Britain)
349
doi:10.1042/BJ20101653
Suppression of TG-interacting factor sensitizes arsenic trioxide-induced
apoptosis in human hepatocellular carcinoma cells
Zi-Miao LIU*, Joseph T. TSENG†, Duang-Yang HONG* and Huei-Sheng HUANG*1
*Department of Medical Laboratory Science and Biotechnology, College of Medicine, National Cheng Kung University, Tainan 70101, Taiwan, and †Institute of Bioinformatics and
Biosignal Transduction, College of Bioscience and Biotechnology, National Cheng Kung University, Tainan 70101, Taiwan
HCC (hepatocellular carcinoma) is among the most common and
lethal cancers worldwide with a poor prognosis mainly due to a
high recurrence rate and chemotherapy resistance. ATO (arsenic
trioxide) is a multi-target drug that has been effectively used as
an anticancer drug in acute promyelocytic leukaemia. However,
a Phase II trial involving patients with HCC indicates that the
use of arsenic as a single agent is not effective against HCC.
TGIF (TG-interacting factor) is a transcriptional co-repressor that
interferes with TGF-β (transforming growth factor-β) signalling
which plays a growth-inhibitory role in HCC. In the present study,
we demonstrated that ATO induced hepatocellular apoptosis
via TGF-β/Smad signalling and led to downstream induction
of p21WAF1/CIP1 (p21). However, ATO could also induce TGIF
expression via a post-transcriptional regulation mechanism to
antagonize this effect. Using a biotin-labelled RNA probe pulldown assay and in vivo RNA immunoprecipitation analysis,
we identified that HuR (human antigen R) bound to the TGIF
mRNA 3 -UTR (3 -untranslated region) and prevented it from
degradation. ATO treatment increased the interaction between
HuR and TGIF mRNA, and reduction of HuR expression inhibited
ATO-induced TGIF expression. Moreover, the EGFR (epidermal
growth factor receptor)/PI3K (phosphoinositide 3-kinase)/Akt
pathway was shown to mediate the post-transcriptional regulation
of TGIF in response to ATO. Finally, we also demonstrated that
the down-regulation of TGIF could sensitize ATO-induced HepG2
cell apoptosis. Collectively, we propose that the EGFR/PI3K/Akt
pathway may regulate the post-transcriptional regulation of
TGIF expression to antagonize ATO-induced apoptosis in HCC.
Blockage of the PI3K/Akt pathway or TGIF expression combined
with ATO treatment may be a promising strategy for HCC therapy.
INTRODUCTION
chemotherapy resistance [6]. Therefore development of new drugs
or therapies for the treatment of HCC is urgently needed.
A Phase II trial has been conducted to evaluate the efficacy
and toxicity of single-agent ATO in patients with HCC [7].
However, the trial shows that use of arsenic as a single agent is
not effective against HCC and should be combined with other
chemotherapeutic agents. Results accumulated so far indicate
that ATO can induce some protective signal pathways, which
negatively regulate its antitumoral effects. In combination with
a pharmacological inhibitor of protective signalling, such as
PI3K (phosphoinositide 3-kinase)/Akt and JNK (c-Jun N-terminal
kinase) inhibitors, ATO can enhance ATO-induced growthinhibitory and pro-apoptotic responses [4,8,9]. Therefore the
diverse biological function of ATO should be investigated further
to develop new therapeutic strategies for its promising clinical
uses.
Previously, we have found that two opposing signal pathways
were involved in the ATO-induced apoptosis of A431 cells.
The EGFR [EGF (epidermal growth factor) receptor]/ERK
(extracellular-signal-regulated kinase) pathway can be activated
by ATO to induce p21WAF1/CIP1 (p21) expression and the resultant
apoptosis. In contrast, the JNK pathway acts as a negative
regulator in ATO-induced p21 expression [10,11]. Furthermore,
we elucidated that JNK-induced c-Jun phosphorylation can
recruit the TGIF [TG-interacting factor; also known as TGFβ (transforming growth factor-β)-induced factor homeobox 1;
Arsenic is a metalloid that naturally exists in our environment. It
has been considered as a potent carcinogen, and chronic exposure
to arsenic may lead to various types of tumours, including lung,
renal, prostate, bladder and skin cancers and other malignancies.
In addition, long-term exposure to arsenic is associated with some
chronic diseases, such as blackfoot, diabetes and cardiovascular
diseases [1].
However, one of the major inorganic forms, ATO (arsenic
trioxide), is a chemotherapeutic agent approved by the U.S.
Food and Drug Administration in 2000 for the treatment of APL
(acute promyelocytic leukaemia) patients. ATO (<2 μM) has an
antitumoral effect by inducing cell death or differentiation of APL
cells [2]. Clinical trials with arsenic-based drugs for the treatment
of various forms of cancer are in progress [3]. Several modes of
action for the antitumoral effects of ATO have been proposed,
such as ROS (reactive oxygen species) production, up-regulation
of pro-apoptotic proteins and induction of some death-associated
signalling [1,4]. Therefore we suggest that ATO is a multi-target
drug that may be more effective than single-target agents in cancer
therapy.
HCC (hepatocellular carcinoma) is the fifth most common
cancer worldwide with a poor prognosis, mainly due to a high
recurrence rate and ineffective therapy [5]. Moreover, the common
chemotherapy drug used in HCC may induce side effects and
Key words: Akt, arsenic trioxide, hepatocellular carcinoma,
human antigen R (HuR), TG-interacting factor (TGIF),
transforming growth factor-β (TGF-β).
Abbreviations used: APL, acute promyelocytic leukaemia; ATO, arsenic trioxide; DMEM, Dulbecco’s modified Eagle’s medium; ECL, enhanced
chemiluminescence; EGF, epidermal growth factor; EGFR, EGF receptor; ERK, extracellular-signal-regulated kinase; HCC, hepatocellular carcinoma;
HDAC1, histone deacetylase 1; HGF, hepatocyte growth factor; HuR, human antigen R; JNK, c-Jun N-terminal kinase; MAPK, mitogen-activated protein
kinase; MEK, MAPK/ERK kinase; PARP, poly(ADP-ribose) polymerase; PI, propidium iodide; PI3K, phosphoinositide 3-kinase; RT, reverse transcription;
shRNA, small-hairpin RNA; siRNA, small interfering RNA; Sp1, specificity protein 1; TGF-β, transforming growth factor-β; TGF-βR, TGF-β receptor; TGIF,
TG-interacting factor; 3 -UTR, 3 -untranslated region.
1
To whom correspondence should be addressed (email [email protected]).
c The Authors Journal compilation c 2011 Biochemical Society
350
Z.-M. Liu and others
HUGO gene symbol TGIF1]–HDAC1 (histone deacetylase 1)
complex to Sp1 (specificity protein 1) sites at the p21 promoter
to suppress its activation. We hypothesize that blockage of the
JNK pathway or knockdown of TGIF can enhance ATO-induced
cancer cell apoptosis [9,12].
TGIF is one of the TALE (three-amino-acid loop extension)
subfamily of atypical homeodomain proteins. It functions as
a transcriptional repressor/co-repressor competing with coactivators for Smad interactions, and recruits HDAC1 to repress
TGF-β/Smad-mediated signalling [13]. In humans, the TGIF gene
was mapped to the region at 18p11.3. Mutations in the TGIF gene
have been found in patients with HPE (holoprosencephaly), a
severe brain and craniofacial malformation associated with mental
retardation [14].
TGF-β is a pluripotent cytokine that regulates a broad range of
biological processes, including proliferation, migration, apoptosis
and differentiation. Its signalling pathway shows a dual functional
role in tumour suppression and oncogenic effects in various
cancers [15,16]. Through its serine/threonine kinase receptor
complexes, TGF-β phosphorylates cytoplasmic mediators, the
Smads, which may translocate to the nucleus and co-operate with
transcriptional co-activators or co-repressors to regulate various
TGF-β-responsive genes [16]. In response to TGF-β in HepG2
cells, Smad3/4 associates with c-Jun to transcriptionally activate
p21 at Sp1-binding sites, leading to p21-dependent cellular
senescence [17], and thus eliciting its antitumoral effects [18].
As described above, TGIF can restrict TGF-β signalling
through multiple binding partners in the nucleus; this implies
that TGIF may be involved in some phenomena of cancer.
Overexpression of TGIF was also found in EGF-regulated
hepatocellular carcinogenesis in mice [19]. Therefore the role of
TGIF in HCC should be further elucidated. So far, evidence
of the regulation of TGIF is limited. EGF, as well as HGF
(hepatocyte growth factor), activates the ERK pathway to
phosphorylate C-terminal residues of TGIF to increase its protein
stability [20,21]. Insulin up-regulates TGIF by stabilizing its
protein against degradation via ERK and PI3K activation in 3T3L1 preadipocytes [22].
The results of the present study show that ATO can activate
TGF-β signalling to induce p21 expression, and finally results in
apoptosis in HepG2 cells. However, TGIF expression was also
increased due to the fact that HuR (human antigen R) bound to
TGIF mRNA to enhance its mRNA stability by ATO treatment.
The EGFR/PI3K/Akt pathway was shown to regulate this posttranscriptional regulation mechanism. Furthermore, knockdown
of TGIF could enhance the cytotoxic effects of ATO on HCC
cells.
EXPERIMENTAL
Reagents and antibodies
ATO, actinomycin D, wortmannin, LY294002, Protein A–agarose
beads and streptavidin–agarose were purchased from Sigma–
Aldrich. The Wizard Plus MiniPrep DNA purification system,
Pfu polymerase, luciferase plasmid pGL-3 and the luciferase
assay system were obtained from Promega. Cycloheximide and
PD153035 were purchased from Calbiochem. U0126 and the antiphospho-Smad3 antibody were purchased from Cell Signaling
Technology. LipofectamineTM 2000 reagent, the TRIzol® RNA
extraction kit, SuperScriptTM III, DMEM (Dulbecco’s modified
Eagle’s medium) and Opti-MEM were obtained from Invitrogen.
Antibodies against TGIF (sc-9084), p21 (sc-397), phosphoERK (sc-16982), phospho-Akt (sc-7985), Akt (sc-8312), HuR,
nucleolin, TGF-βR (TGF-β receptor) II and Smad3 were
c The Authors Journal compilation c 2011 Biochemical Society
purchased from Santa Cruz Biotechnology, and an anti-βactin antibody was obtained from Sigma–Aldrich. Antibodies
against caspase 3 (active form) and cleaved PARP [poly(ADPribose) polymerase] were from Upstate Biotechnology. The
shRNA (small-hairpin RNA) of candidate genes and a luciferase
control (pLKO.1-shLuc) plasmid construct were obtained from
the National RNAi Core Facility located at the Institute
of Molecular Biology/Genomic Research Center, Academia
Sinica (Taipei, Taiwan). The shRNA-targeted sequences were
as follows: (i) pLKO.1-TGIF-shRNA targets the human
TGIF gene sequence 5 -GCAAGAGATGAATTGCATTAT-3 ;
(ii) pLKO.1-Smad3-shRNA targets the human Smad3 gene
sequence 5 -GGATTGAGCTGCACCTGAATG-3 ; and (iii)
pLKO.1-Akt1-shRNA targets the human Akt1 gene sequence 5 GGACAAGGACGGGCACATTAA-3 . A chemically synthesized
siRNA (small-interfering RNA) targeted the human HuR gene
sequence 5 -AAGAGGCAAUUACCAGUUUCATT-3 . The p21luc plasmid was provided by Dr Bert Vogelstein (Johns Hopkins
University Medical Institutions, Baltimore, MD, U.S.A.).
Cell culture
The liver cancer cell lines HepG2 and Huh7 were both maintained
in DMEM supplemented with 10 % (v/v) fetal bovine serum,
100 μg/ml streptomycin and 100 units/ml penicillin. Cells were
grown at 37 ◦ C in a humidified atmosphere of air/CO2 (19:1, v/v).
Western blot analysis
Cell lysates (30 μg) were separated by SDS/PAGE (12 %
gels), and then transferred on to a PVDF membrane using a
semi-dry blotting apparatus. After blocking with non-fat dried
skimmed milk, antibodies against TGIF, phospho-ERK, ERK,
phospho-Akt, Akt, TGF-βRII, phospho-Smad3, HuR, nucleolin,
Smad3, cleaved caspase 3 and β-actin were used as primary
antibodies. Rabbit or mouse IgG-specific antibodies conjugated to
horseradish peroxidase were used as secondary antibodies. After
development with an ECL (enhanced chemiluminescence) kit (GE
Healthcare), the density of the immunoblots was determined using
an image analysis system installed with BIO-ID software (Vilber
Lourmat, France).
RT (reverse transcription)–PCR
A sense primer (5 -AGATCTGAATTGTGCCAGTGTTTCTCTTTG-3 ) and an antisense primer (5 -CCATGGCGGCGCTGTCAGAGTGAG-3 ) corresponding to the nucleotide sequence
of human TGIF were used. The RT–PCR method was performed
according to the manufacturer’s instructions with a slight
modification. The mixture contained 3 μg of total RNA, 2.5 μM
oligo(dT), 1.5 mM MgCl2 , 0.01 M DTT (dithiothreitol) and
200 units of SuperScriptTM III reverse transcriptase in a total
volume of 20 μl. The products were amplified on an Eppendorf
MasterCycler under the following conditions. The RT steps
consisted of an RT step (55 ◦ C for 50 min) and a denaturation step
(95 ◦ C for 5 min). The PCR steps consisted of a denaturation
step (95 ◦ C for 30 s), a primer annealing step (58 ◦ C for 30 s)
and an elongation step (72 ◦ C for 1 min) for 30 cycles. For
the final step, the duration of the elongation step was 7 min.
Finally, samples were mixed with loading buffer (50 mM
Tris/HCl, 10 mM EDTA and 0.25 % Bromophenol Blue) and
loaded on to a 1 % agarose gel containing 0.1 μg/ml ethidium
bromide. The agarose gels were run at 135 V for 20 min in a
TAE (Tris/acetate/EDTA) buffer. The gels were observed and
photographed under UV light.
Antagonistic effects of TGIF on ATO-induced apoptosis
Real-time PCR
TRIzol®
Total RNA was isolated from cells using
RNA solution.
DNase I-treated RNA (2 μg) was reverse-transcribed using
SuperScriptTM III and oligo(dT) primers. Real-time PCR was
conducted using the SYBR Green PCR Master Mix (Qiagen)
and was processed on a LightCycler PCR and detection system
(Roche Diagnostics). Each reaction (25 μl) was run in duplicate
and contained 1 μl of cDNA template along with 0.4 μM
TGIF primers (sense, 5 -CGCTACAGGTCTGTAACTGGTTC3 ; antisense, 5 -GATGCCCATCACGGACTC-3 ) or β-actin
primers (sense, 5 -TCCCTGGAGAAGAGCTACGA-3 ; antisense, 5 -ACTCCATGCCCAGGAAGG-3 ). Cycling parameters
were 95 ◦ C for 15 min to activate DNA polymerase, followed
by 40 cycles of 95 ◦ C for 15 s, 60 ◦ C for 20 s, and 72 ◦ C for
30 s. Melting curves were generated at the end of the reaction.
Threshold cycles (Ct ) for each gene tested were normalized to the
housekeeping β-actin gene value (Ct ) and every experimental
sample was referred to its control (Ct ). Fold change values
were expressed as 2 − Ct .
DNA constructs, transient transfection and reporter gene assay
The full-length cDNAs of the human SMAD2/3/4 genes
were amplified by PCR. The PCR products were digested
with EcoRI/XhoI to generate fragments which were then
ligated into the pcDNA3.1( + ) expression vector containing
the CMV (cytomegalovirus) promoter and validated by sequencing.
The human TGIF promoter from − 2393 to + 119 bp
was cloned by PCR using two specific primers (sense, 5 AGATCTGAATTGTGCCAGTGTTTCTCTTTG-3 ; antisense,
5 -CCATGGCGGCGCTGTCAGAGTGAG-3 ) with human genomic DNA from A431 cells as a template. The products of PCR
were cloned into the T&A cloning vector (Yeastern Biotech),
and then confirmed by DNA sequencing. After BglII and NcoI
digestion, the fragments were then subcloned into the pGL3-basic
vector (Promega). The 3 -UTR (3 -untranslated region) of TGIF
(448 bp) was amplified from human HepG2 cells by RT–RCR,
and then cloned into the pGL3 vector (Promega) and designated
pGLTG3UTR.
The transfection method was performed using LipofectamineTM
2000 reagent according to the manufacturer’s instructions with a
slight modification. HepG2 or Huh7 cells were subcultured in a
12-well plate for 24 h before transfection at a density of 8×104
cells in 2 ml of fresh culture medium per well. For use in the
transfection assay, plasmids were mixed with LipofectamineTM
2000 reagent in 1 ml of Opti-MEM, and then incubated at
room temperature (25 ◦ C) for 30 min. Cells were incubated in
the mixture at 37 ◦ C in a humidified atmosphere of air/CO2
(19:1) for a further 48 h, and then lysed for the measurement
of luciferase activity as described previously [23]. The luciferase
activity was determined and normalized to the amount of total
protein.
Flow cytometry analysis
Cell death was evaluated by flow cytometry after staining
with FITC-conjugated Annexin V and 0.5 μg/ml PI (propidium
iodide). An Annexin V–FITC apoptosis detection kit (Biovision)
was used to detect early apoptosis according to the manufacturer’s
instructions with slight modifications. After transfection for 24 h
as described above, HepG2 cells were harvested and washed
twice with ice-cold PBS and resuspended in 100 μl of binding
351
buffer (Biovision). Annexin V (5 μl) and PI (10 μl) were added
and the mixture was incubated for 15 min in the dark. Finally,
400 μl of binding buffer was added to the cells, and the mixture
was analysed with a flow cytometer (FACSCalibur; BD Biosciences).
In vitro RNA synthesis and biotin-labelled RNA probe pull-down
assay
The pGLTG3UTR plasmids were linearized to serve as templates
to generate RNA probes for the pull-down assay. The RNAs
were synthesized using the Riboprobe In Vitro Transcription
System (Promega). Briefly, 1 μg of linearized DNA template was
transcribed by T7 polymerase in the presence of 2.5 mM UTP,
ATP and GTP, 1 mM CTP and 10 mM biotin–CTP for 1.5 h at
37 ◦ C in each 30 μl reaction mixture. The reaction was stopped
by adding 2 units of RQ1 RNase-free DNase I (Promega) for
15 min at 37 ◦ C. The RNA was purified using MicroSpin G-25
columns. For the RNA probe pull-down assay, the biotinylated
TGIF 3 -UTR was incubated with HepG2 cytosolic lysates at
room temperature for 1 h. The complexes were isolated using
streptavidin-conjugated agarose beads, the binding proteins in
the pull-down complex were analysed by Western blotting using
specific antibodies recognizing HuR and nucleolin, and then
visualized by ECL (GE Healthcare).
RNA immunoprecipitation
HepG2 cells were treated with lysis buffer [10 mM Hepes
(pH 8.0), 10 mM KCl, 1.5 mM MgCl2 and 500 units of
RNaseOUT] for 10 min on ice. The cellular extract was
incubated with anti-HuR antibody or mouse control IgG and
Protein A/G–agarose beads at 4 ◦ C overnight followed by
centrifugation (3300 g for 2 min at 4 ◦ C). The immunoprecipitated
protein–RNA complexes were washed three times with lysis
buffer, and RNAs were extracted using TRIzol® reagent. The
TGIF 3 -UTR was detected by RT–PCR (primers used were
sense, 5 -CCCGGGCCCATTTTCAAGCAAAAC-3 ; antisense,
5 -AGATCTATGTCATGAAAAAAAAGGCATCA-3 ).
Statistical analysis
A representative result is presented, and all experiments have
been performed at least three times. The statistical analysis was
determined using Student’s t test. All values are displayed as
means +
− S.D. for three determinations (*P < 0.05, **P < 0.01
and ***P < 0.001 compared with the vehicle).
RESULTS
ATO increased p21 expression and cell apoptosis by activating
the TGF-β signalling pathway
As described previously, TGF-β signalling plays an inhibitory role
in the growth of HCC [18,24]. In hepatoma HepG2 cells, 5 μM
ATO could increase Smad3 phosphorylation, and Smad3 protein
and p21 expression in a time-dependent manner (Figure 1A).
Transient overexpression of Smad3 in HepG2 cells followed by
ATO treatment for 4 h could further enhance p21 promoter activity
(Figure 1B). Conversely, knockdown of Smad3 expression by
shRNA could suppress ATO-induced p21 promoter activation
and protein expression (Figure 1C). In addition, treatment with
5 μM ATO for 20 h increased the active form of caspase 3
expression and cleaved PARP in HepG2 cells, indicating that
c The Authors Journal compilation c 2011 Biochemical Society
352
Figure 1
Z.-M. Liu and others
ATO induced HepG2 cell apoptosis via TGF-β signalling
(A) HepG2 cells were treated with 5 μM ATO for various time periods as indicated. The cell lysates were collected for the determination of phospho-Smad3 (p-Smad3), Smad3 and p21 expression by
using Western blot analysis. β-Actin was used as an internal control. (B) The p21-luc plasmids were co-transfected with the expression vector for Smad3 into cells as described in the Experimental
section. After incubation for 24 h, cells were treated with 5 μM ATO for 4 h, and then harvested for the measurement of luciferase activity and protein expression of Smad3. ***P < 0.001 compared
with control cells (n = 3). (C) The p21-luc plasmids were co-transfected with pLKO.1-Smad3-shRNA or pLKO.1-shLuc plasmids into cells for 24 h. Cells were treated with 5 μM ATO for 4 h, and
then harvested for measurement of luciferase activity of the p21 reporter and protein expression of Smad3, p21 and β-actin. *P < 0.05 compared with control cells (n = 3). (D) HepG2 cells were
transiently transfected with pLKO.1-Smad3-shRNA or pLKO.1-shLuc plasmids for 24 h, followed by treatment with or without 5 μM ATO for 20 h. Cell lysates were collected to analyse Smad3, the
active form of caspase 3, cleaved PARP and β-actin expression by Western blotting. For the Western blots, the molecular mass in kDa is indicated.
cellular apoptosis had occurred. However, knockdown of Smad3,
followed by treatment with ATO, could suppress the apoptotic
events (Figure 1D). Thus we suggest that TGF-β signalling is
involved in ATO-induced p21 expression and apoptosis in HepG2
cells.
ATO induced TGIF expression and led to the attenuation of p21
activation
ATO was shown to induce cell apoptosis; however, the effect
for treating HCC in Phase II trials is not very promising. It will
be interesting to identify the factor which antagonizes the ATOinduced apoptotic effect. Previously, we have demonstrated that
transient overexpression of TGIF could inhibit ATO-induced p21
expression in A431 cells [9,11,12]. In HepG2 cells, transient
overexpression of TGIF could also inhibit ATO-induced p21
promoter activation (Figure 2A, top panel) and protein expression
(Figure 2A, bottom panel) in a dose-dependent manner. This
result prompted us to examine whether ATO can induce TGIF expression in HCC. As shown in Figures 2(B) and 2(C), TGIF
mRNA and protein expression was induced by ATO stimulation in
a time-dependent manner in HepG2 and Huh7 cells respectively.
The multiple bands of TGIF shown in the Western blot analysis
c The Authors Journal compilation c 2011 Biochemical Society
might be isoforms of the protein, as described previously [25].
This result implies that TGIF may be the factor that interferes
with ATO-induced apoptosis.
ATO-induced TGIF expression was dependent on a
post-transcriptional regulation mechanism
To determine whether ATO controls the transcriptional activity of
TGIF, a reporter containing 2.5 kb of the human TGIF promoter
( −2393 to + 119 bp) fused with the firefly luciferase gene was
constructed, as described in the Experimental section. Then,
we transiently transfected the reporter into hepatoma cells and
measured its promoter activity in response to ATO. As shown
in Figure 3(A), ATO could only slightly induce TGIF promoter
activity in HepG2 cells. Compared with the fold increase of
TGIF mRNA after ATO treatment (Figure 2B, approximately 3.8fold), this result highlighted that an mRNA stability regulation
mechanism may be involved in the expression of TGIF. Using
actinomycin D to block the transcriptional event, we measured
the degradation rate of TGIF in response to ATO by real-time
PCR. As shown in Figure 3(B), the degradation rate of TGIF
mRNA slowed down after ATO treatment. A similar result was
also obtained using a traditional RT–PCR assay (Figure 3C). To
Antagonistic effects of TGIF on ATO-induced apoptosis
Figure 2
353
ATO increased TGIF expression to antagonize p21 activation
(A) The p21-luc plasmids were co-transfected with TGIF plasmids into cells as described in the Experimental section. After incubation for 24 h, cells were treated with 5 μM ATO for 4 h, and then
harvested for measurement of luciferase activity and protein expression of TGIF and p21. *P < 0.05 compared with ATO-treated cells (n = 3). For the Western blot, the molecular mass in kDa is
indicated. (B) HepG2 cells or (C) Huh7 cells were treated with ATO for various time periods as indicated. The mRNA and protein expression of TGIF were analysed by RT–PCR and Western blotting
respectively. GAPDH (glyceraldehyde-3-phosphate dehydrogenase) mRNA or β-actin were used as internal controls. WB, Western blot.
Figure 3
ATO increased mRNA stability of TGIF in HepG2 cells
(A) HepG2 cells were transfected with TGIF-luc plasmids (0.5 μg) for 24 h, followed by treatment with 5 μM ATO for various time periods, and then harvested for measurement of luciferase
activity. HepG2 cells were treated with or without 5 μM ATO for 1 h followed by addition of 5 μg/ml actinomycin D (ActD) for another 1 h or 2 h respectively. TGIF mRNA expression was examined
by real-time PCR (B) or RT–PCR (C) as described in the Experimental section. **P < 0.01 compared with control cells (n = 6). (D) HepG2 cells were transfected with pGL-luc (top panel) or
pGLTG3UTR (bottom panel) respectively, followed by treatment with 5 μM ATO for 2 h, and then harvested for measurement of luciferase activity. *P < 0.05 compared with control cells (n = 6).
gapdh, glyceraldehyde-3-phosphate dehydrogenase.
demonstrate further whether the 3 -UTR of TGIF mRNA plays
an important role in regulating the mRNA stability, a reporter
containing the 3 -UTR of TGIF mRNA was constructed. We found
that treatment with ATO for 2 h led to an increase in TGIF 3 -UTR
reporter activity by up to 1.5-fold (Figure 3D, bottom panel).
This indicates that ATO-induced TGIF expression is based on
regulating mRNA stability.
ATO increased the interaction between HuR and the TGIF mRNA
3 -UTR
An increased interaction between RNA-binding proteins,
such as HuR, nucleolin and hnRNPK (heterogeneous
nuclear ribonucleoprotein K), with the 3 -UTR plays an
important role in the regulation of mRNA stability under
c The Authors Journal compilation c 2011 Biochemical Society
354
Z.-M. Liu and others
The EGFR/PI3K signalling pathway mediated ATO-induced TGIF
expression
Figure 4
ATO enhanced HuR binding to the TGIF 3 -UTR
(A) Biotin-labelled TGIF 3 -UTR probes were incubated with or without cytosolic extract from
ATO-treated HepG2 cells. The mixture was analysed by biotin pull-down assay, followed by
Western blot analysis using an anti-HuR antibody. (B) HepG2 cells were treated with or
without 5 μM ATO for 1.5 h, and an RNA immunoprecipitation assay was performed using
anti-HuR or anti-mouse IgG antibodies respectively. The precipitated mRNA was extracted to
analyse TGIF expression by RT–PCR. (C) HepG2 cells were transiently transfected with HuR
siRNA or scrambled control siRNAs. Cell lysates were harvested to detect TGIF, HuR and
β-actin expression by Western blot analysis. The molecular mass in kDa is indicated. IP,
immunoprecipitation.
stimuli [26,27]. Therefore the involvement of these RNAbinding proteins in the regulation of TGIF mRNA in
response to ATO was investigated. A pull-down assay, as
described in the Experimental section, showed increased binding of HuR, but not nucleolin protein, to the biotin-labelled
TGIF 3 -UTR with ATO treatment (Figure 4A); however, protein
expression of HuR and nucleolin was not changed after ATO
treatment. These results indicate that ATO can enhance the binding of HuR to the TGIF 3 -UTR in vitro. Furthermore, the
interaction of HuR with the TGIF 3 -UTR in vivo was explored
using an RNA immunoprecipitation assay. We also found that
more TGIF mRNA could be pulled down by ATO treatment
than by vehicle (Figure 4B). In addition, reduction of the HuR
protein expression by siRNA technology abolished the ATOinduced TGIF expression (Figure 4C). HuR siRNA reduced the
level of HuR protein to approximately 50 %, as well as the level of
ATO-induced TGIF expression being decreased to approximately
50 % after transfection for 24 h.
Collectively, these results indicate that HuR physically interacts
with the TGIF mRNA 3 -UTR to stabilize and increase its mRNA
expression in response to ATO.
c The Authors Journal compilation c 2011 Biochemical Society
EGF, as well as HGF, signalling activates the ERK pathway
leading to phosphorylation of TGIF to enhance its protein stability
[20,21]. We also found that ATO could regulate p21 expression
via EGFR-associated signalling in A431 cells [10,11]. To explore
whether the EGFR-associated signalling pathway was involved in
the ATO-induced TGIF expression in hepatoma cells, the EGFR
inhibitor PD153035 was used to address this question. As shown
in Figure 5, PD153035 completely blocked ATO-induced TGIF
mRNA and protein expression (Figure 5A). In addition, U0126,
a specific MEK [MAPK (mitogen-activated protein kinase)/ERK
kinase] inhibitor, could block ATO-induced sustained ERK1/2
phosphorylation, but it could not change the expression of TGIF
protein (Figure 5B). This indicates that ERK1/2 activation is not
involved in ATO-induced TGIF expression.
The PI3K/Akt signalling pathway is well-known for its effects
in promoting cell survival [28]. Accumulated results suggest
that specific pharmacological inhibitors to the PI3K/Akt survival
signal can improve ATO-induced leukaemia cell apoptosis [8,29].
It has also been shown that insulin increases the TGIF protein level
through the PI3K/Akt signalling pathway in the differentiation
of preadipocytes [22]. In HepG2 cells, we found that ATO was
able to induce Akt (Ser473 ) phosphorylation in a time-dependent
manner (Figure 5C), and pre-treatment with the EGFR inhibitor
PD153035 could block the phosphorylation (results not shown).
Furthermore, pre-treatment with the PI3K inhibitor LY294002
(results not shown) or wortmannin followed by ATO stimulation
could inhibit TGIF expression (Figure 5D). Therefore we suggest
that the EGFR/PI3K-associated signalling pathway is involved in
ATO-induced TGIF expression.
The EGFR/PI3K/Akt signalling pathway regulated TGIF mRNA
stability
We further explored the signalling pathways involved in regulating
TGIF mRNA stability in response to ATO. Pre-treatment with
the EGFR inhibitor PD153035 was able to inhibit ATO-induced
TGIF 3 -UTR reporter activity (Figure 6A). In addition, pretreatment with the PI3K inhibitors wortmannin or LY294002
(results not shown) was able to inhibit ATO-induced TGIF 3 -UTR
reporter activity (Figure 6B). Moreover, silencing of Akt1 by its
shRNA also blocked ATO-induced TGIF 3 -UTR reporter activity
(Figure 6C), as well as TGIF protein expression (Figure 6D).
Therefore we suggest that the EGFR/PI3K/Akt pathway is
involved in ATO-induced post-transcriptional regulation of TGIF.
Knockdown of TGIF enhanced ATO-induced apoptosis
According to the results described above, we suggest that TGIF
may play an antagonistic role in the ATO-induced apoptosis
of HepG2 cells. Therefore knockdown of TGIF followed by
treatment with 5 μM ATO for 20 h in HepG2 cells was performed
and then Annexin V/PI staining followed by flow cytometry was
used to measure the apoptotic cells. As shown in Figure 7(B),
ATO-treated cells were 11.4 % Annexin V + /PI − (early apoptotic)
and 3.9 % Annexin V + /PI + (later apoptotic), whereas the vehicle
cells were 1.8 % Annexin V + /PI − and 1.7 % Annexin V + /PI +
(Figure 7A). In addition, when cells were transfected with
TGIF shRNA (2 μg) followed by treatment with 5 μM ATO for
20 h, early apoptotic cells (Annexin V + /PI − ) were increased
further to 21.4 % (Figure 7C), and cleaved PARP expression
was also increased, as analysed by Western blotting (Figure 7E).
Antagonistic effects of TGIF on ATO-induced apoptosis
Figure 5
355
The EGFR/PI3K signalling pathway, but not the MEK/ERK signalling pathway, was required for ATO-induced TGIF expression
(A) Cells were pre-treated with different amounts of EGFR inhibitor PD153035 for 0.5 h and then treated with 5 μM ATO for 4 h. Total cell lysates were collected for the analysis of expression of TGIF
and β-actin by Western blotting. (B) Pre-treatment with the specific MEK1 inhibitor U0126 for 1 h followed by treatment with 5 μM ATO for various time periods. Cell lysates were collected to detect
TGIF and phospho-ERK1/2 expression by Western blot analysis. (C) Cells were treated with ATO for various time points as indicated. Cell lysates were collected to detect phospho-Akt (Ser473 ), TGIF
and β-actin expression by Western blot analysis. (D) HepG2 cells were pretreated with the PI3K inhibitor wortmannin for 0.5 h, followed by treatment with 5 μM ATO for 4 h. Total cell lysates were
collected for the analysis of phospho-Akt (Ser473 ), Akt, TGIF and β-actin expression by Western blotting. p-, phospho-. For the Western blots, the molecular mass in kDa is indicated.
Figure 6
EGFR/PI3K/Akt signalling was required for ATO-increased mRNA stability of TGIF
HepG2 cells were transfected with the pGLTG3UTR reporter, followed by treatment with 5 μM ATO for various time periods in the absence or presence of (A) the EGFR inhibitor PD153035 or (B) the
PI3K inhibitor wortmannin. (C) The pGLTG3UTR reporter plasmids were co-transfected with pLKO.1-Akt1-shRNA or pLKO.1-shLuc plasmids into cells for 24 h. Cells were treated with 5 μM ATO for
4 h, and then harvested for the measurement of luciferase activity. **P < 0.01 and ***P < 0.001 compared with ATO-treated cells (n = 3). (D) HepG2 cells were transfected with pLKO.1-Akt1-shRNA
or pLKO.1-shLuc plasmids for 24 h. Total cell lysates were collected for the analysis of Akt1/2, TGIF and β-actin expression by Western blotting. The molecular mass in kDa is indicated.
c The Authors Journal compilation c 2011 Biochemical Society
356
Figure 7
Z.-M. Liu and others
Knockdown of TGIF increased ATO-induced apoptosis
HepG2 cells were transiently transfected with empty vectors (A and B) or 2 μg of shRNA for TGIF (C and D) for 24 h, followed by treatment with (B and C) or without (A and D) 5 μM ATO for
20 h. Using FITC-conjugated Annexin V/PI staining followed by analysis of flow cytometry as described in the Experimental section, the dot plot diagrams were shown to represent cell populations
of apoptosis. The lower-left quadrants of the panels represent live cells (Annexin V − /PI − ); the lower-right quadrants represent early apoptotic cells (Annexin V + /PI − ); the upper-left quadrants
represent necrosis cells (Annexin V − /PI + ); and the upper-right quadrants represent late apoptotic cells (Annexin V + /PI + ). (E) The relative protein expression of TGIF and cleaved PARP of each
reaction of flow cytometry was determined by Western blot analysis. β-Actin was used as an internal control. The molecular mass in kDa is indicated.
The knockdown efficiency of TGIF shRNA was measured to
be approximately 80 % (Figure 7E). Therefore we suggest that
silencing of TGIF can enhance ATO-induced apoptosis in HepG2
cells.
Collectively, we suggest that ATO can induce TGF-β signalling
to increase p21 expression and results in HepG2 cell apoptosis.
However, ATO can activate EGFR/PI3K/Akt signalling to
induce TGIF expression via increasing the interaction between
HuR and TGIF mRNA. The increase of TGIF expression can
prevent the apoptosis induced by ATO (Figure 8).
DISCUSSION
Three interesting findings have emerged from the present
study. First, we found that TGF-β signalling was involved
in ATO-induced p21 expression and resultant apoptosis in
HepG2 cells. TGF-β has dual roles in tumour suppression
and tumour promotion. In the early stages of cancer, TGFβ/Smad is considered as a growth inhibitory pathway in many
tumours, including HCC. However, in advanced cancer, TGFβ/Smad can enhance the progression and metastasis of tumours
[24]. In HepG2 cells, TGF-β can induce p21 expression, and
resultant cellular arrest, senescence or apoptosis, to have its
antitumoral effects [17,18,24]. On the other hand, ATO can induce
hepatocellular growth arrest or apoptosis through regulation of
redox homoeostasis [30,31] and PI3K/Akt signalling [32], but
not MAPK signalling [30]. It also has a potential role in cancer
therapy to inhibit angiogenesis and VEGF (vascular endothelial
growth factor) expression [33,34]. Arsenic has a high affinity
for thiol-containing enzymes and proteins, such as phosphatases,
c The Authors Journal compilation c 2011 Biochemical Society
Figure 8 Schematic diagram illustrating the mechanisms of ATO-induced
apoptosis in hepatoma cells
ATO can induce TGF-β signalling to activate p21 expression as well as subsequent cellular
apoptosis. TGIF can also be up-regulated by ATO at its post-transcriptional level via
EGFR/PI3K/Akt signalling to enhance HuR binding to the 3 -UTR, which can antagonize p21
expression. p, phospho-.
and is necessary for their normal cellular activity [35]. Therefore
we suggest that ATO may inhibit phosphatases, specifically
dephosphorylating TGF-βRs, to activate TGF-β signalling.
However, the mechanisms involved should be elucidated
further.
Antagonistic effects of TGIF on ATO-induced apoptosis
Secondly, TGIF was able to be up-regulated by ATO through
a post-transcriptional regulation mechanism, and played an
antagonistic role in ATO-induced hepatocellular apoptosis. TGIF
has been implicated in various types of cancer. The imbalance
of the TGIF-located chromosome (18p11) has been found in
transitional cell carcinoma of the bladder [36], and frequently
overexpression of the region was found in oesophageal carcinoma
by comparative genomic hybridization analysis [37]. In addition,
overexpression of TGIF in EGF-induced HCC in mice was found
[19]. Furthermore, TGIF can physically bind to phosphorylated
c-Jun to recruit HDAC1 to suppress TGF-β/Smad-activated
transcriptional activity [13,38] or to suppress ATO-induced p21
expression in A431 cells [9,12] and in hepatoma cells (Figure 2).
Interestingly, loss of WAF/KIP protein families (p21, p27 and p57)
was reported to correlate with a poorer prognosis in HCC patients
[39,40]. In addition, TGIF also plays a role as a component
of the ubiquitin ligase complex associated with degradation of
Smad2 to promote cellular growth [41]. Sequestration of cPML
(cytoplasmic promyelocytic leukaemia) proteins by TGIF in the
nucleus can negatively regulate TGF-β signalling to promote
cellular transformation [42]. Therefore how to regulate the
WAF/KIP protein families and TGF-β signalling by TGIF should
be addressed further in HCC.
Thirdly, ATO can increase TGIF mRNA stability through
EGFR/PI3K/Akt signalling to regulate HuR binding to the
TGIF mRNA 3 -UTR. EGF phosphorylates C-terminal residues
of TGIF (Thr235 and Thr239 ) to increase its half-life through
the EGFR/Ras/MEK/ERK pathway [20]. However, we found
that pre-treatment with the MEK inhibitor U0126 could not
block the ATO-induced TGIF protein up-regulation. Furthermore,
overexpression of MEK1 did not increase TGIF expression
(results not shown). In addition, the results of treatment with
the protein synthesis inhibitor cycloheximide also indicated
that the post-translational modification of TGIF was not regulated
by ATO in hepatocytes. Using actinomycin D and 3 -UTR
analysis, we found a novel mechanism whereby the ATO-induced
EGFR/PI3K/Akt pathway regulated the stability of TGIF mRNA
in HepG2 cells. Some RNA-binding proteins, such as HuR
and nucleolin, have been reported to be involved in arsenicinduced stabilization of GADD45 (growth arrest and DNA
damage 45) mRNA [43]. We also identified that the transacting factor HuR could bind to the TGIF mRNA 3 -UTR to
increase its mRNA stability in response to ATO (Figure 4).
HuR, originally identified in Drosophila melanogaster to be
essential for neural development [44], belongs to a member of
the embryonic lethal abnormal vision family of RNA-binding
proteins [45]. It can selectively bind to AU-rich elementcontaining regions or uridine-rich stretches to stabilize mRNA
expression of numerous anti-apoptosis-related genes, such as
prothymosin α [46], Bcl-2 and Mcl-1 [47]. In HCC cell lines, DNA
methyltransferase 3B could also be regulated by the PI3K/Akt
pathway to modulate HuR protein at the post-transcriptional level
[48].
In addition, the PI3K/Akt pathway is a well-known signal that
promotes the proliferation and survival of cancer cells [49]. Thus
selective pharmacological inhibitors of the PI3K/Akt pathway
have been used to enhance ATO-induced apoptosis, not only in
leukaemia cells [8,50,51], but also in solid tumour cells [32,52].
FOXO3A (Forkhead box 3A) [32], mTOR (mammalian target
of rapamycin)/p70 S6K (S6 kinase) [50] or p53 [51] might be
regulated by the PI3K/Akt pathway to modulate ATO-induced
apoptosis. We conclude that blockage of the PI3K/Akt pathway
or knockdown of TGIF may be a novel strategy to overcome
drug resistance and enhance the efficacy of ATO treatment in
HCC.
357
AUTHOR CONTRIBUTION
Zi-Miao Liu performed most of the experiments in the paper. Joseph Tseng co-operated
with us to design and carry out the experiments investigating the post-transcriptional
regulation mechanism of TGIF by ATO. Duang-Yang Hong contributed to the construction
of functional Smad plasmids. Huei-Sheng Huang contributed to the experimental design
and paper organization.
ACKNOWLEDGEMENTS
We thank Dr Wen-Chang Chang for his valuable discussions, and Ms Christine Hsieh for
help with English proofreading prior to submission.
FUNDING
This work was supported by the National Science Council of Taiwan [grant numbers
NSC95-2320-B-006-055-MY3, NSC98-2320-B-006-008-MY3].
REFERENCES
1 Miller, Jr, W. H., Schipper, H. M., Lee, J. S., Singer, J. and Waxman, S. (2002)
Mechanisms of action of arsenic trioxide. Cancer Res. 62, 3893–3903
2 Douer, D. and Tallman, M. S. (2005) Arsenic trioxide: new clinical experience with an old
medication in hematologic malignancies. J. Clin. Oncol. 23, 2396–2410
3 Dilda, P. J. and Hogg, P. J. (2007) Arsenical-based cancer drugs. Cancer Treat. Rev. 33,
542–564
4 Platanias, L. C. (2009) Biological responses to arsenic compounds. J. Biol. Chem. 284,
18583–18587
5 Jemal, A., Siegel, R., Ward, E., Hao, Y., Xu, J., Murray, T. and Thun, M. J. (2008) Cancer
statistics, 2008. CA Cancer J. Clin. 58, 71–96
6 Thomas, M. (2009) Molecular targeted therapy for hepatocellular carcinoma. J.
Gastroenterol. 44 (Suppl. 19), 136–141
7 Lin, C. C., Hsu, C., Hsu, C. H., Hsu, W. L., Cheng, A. L. and Yang, C. H. (2007) Arsenic
trioxide in patients with hepatocellular carcinoma: a phase II trial. Invest. New Drugs 25,
77–84
8 Ramos, A. M., Fernandez, C., Amran, D., Sancho, P., de Blas, E. and Aller, P. (2005)
Pharmacologic inhibitors of PI3K/Akt potentiate the apoptotic action of the antileukemic
drug arsenic trioxide via glutathione depletion and increased peroxide accumulation in
myeloid leukemia cells. Blood 105, 4013–4020
9 Huang, H. S., Liu, Z. M. and Hong, D. Y. (2010) Blockage of JNK pathway enhances
arsenic trioxide-induced apoptosis in human keratinocytes. Toxicol. Appl. Pharmacol.
244, 234–241
10 Liu, Z. M. and Huang, H. S. (2006) As2 O3 -induced c-Src/EGFR/ERK signaling is via Sp1
binding sites to stimulate p21WAF1/CIP1 expression in human epidermoid carcinoma A431
cells. Cell. Signalling 18, 244–255
11 Huang, H. S., Liu, Z. M., Ding, L., Chang, W. C., Hsu, P. Y., Wang, S. H., Chi, C. C. and
Chuang, C. H. (2006) Opposite effect of ERK1/2 and JNK on p53-independent
p21WAF1/CIP1 activation involved in the arsenic trioxide-induced human epidermoid
carcinoma A431 cellular cytotoxicity. J. Biomed. Sci. 13, 113–125
12 Liu, Z. M. and Huang, H. S. (2008) Inhibitory role of TGIF in the As2 O3 -regulated
p21WAF1/CIP1 expression. J. Biomed. Sci. 15, 333–342
13 Wotton, D. and Massague, J. (2001) Smad transcriptional corepressors in TGFβ family
signaling. Curr. Top. Microbiol. Immunol. 254, 145–164
14 Gripp, K. W., Wotton, D., Edwards, M. C., Roessler, E., Ades, L., Meinecke, P.,
Richieri-Costa, A., Zackai, E. H., Massague, J., Muenke, M. and Elledge, S. J. (2000)
Mutations in TGIF cause holoprosencephaly and link NODAL signalling to human neural
axis determination. Nat. Genet. 25, 205–208
15 Derynck, R., Akhurst, R. J. and Balmain, A. (2001) TGF-β signaling in tumor suppression
and cancer progression. Nat. Genet. 29, 117–129
16 Pardali, K. and Moustakas, A. (2007) Actions of TGF-β as tumor suppressor and
pro-metastatic factor in human cancer. Biochim. Biophys. Acta 1775, 21–62
17 Kardassis, D., Papakosta, P., Pardali, K. and Moustakas, A. (1999) c-Jun transactivates
the promoter of the human p21WAF1/Cip1 gene by acting as a superactivator of the
ubiquitous transcription factor Sp1. J. Biol. Chem. 274, 29572–29581
18 Senturk, S., Mumcuoglu, M., Gursoy-Yuzugullu, O., Cingoz, B., Akcali, K. C. and Ozturk,
M. (2010) Transforming growth factor-β induces senescence in hepatocellular carcinoma
cells and inhibits tumor growth. Hepatology 52, 966–974
19 Borlak, J., Meier, T., Halter, R., Spanel, R. and Spanel-Borowski, K. (2005) Epidermal
growth factor-induced hepatocellular carcinoma: gene expression profiles in precursor
lesions, early stage and solitary tumours. Oncogene 24, 1809–1819
20 Lo, R. S., Wotton, D. and Massague, J. (2001) Epidermal growth factor signaling via Ras
controls the Smad transcriptional co-repressor TGIF. EMBO J. 20, 128–136
c The Authors Journal compilation c 2011 Biochemical Society
358
Z.-M. Liu and others
21 Yang, S., Nugent, M. A. and Panchenko, M. P. (2008) EGF antagonizes TGF-β-induced
tropoelastin expression in lung fibroblasts via stabilization of Smad corepressor TGIF.
Am. J. Physiol. Lung Cell. Mol. Physiol. 295, L143–151
22 Horie, T., Ono, K., Kinoshita, M., Nishi, H., Nagao, K., Kawamura, T., Abe, Y., Wada, H.,
Shimatsu, A., Kita, T. and Hasegawa, K. (2008) TG-interacting factor is required for the
differentiation of preadipocytes. J. Lipid Res. 49, 1224–1234
23 Huang, H. S., Chang, W. C. and Chen, C. J. (2002) Involvement of reactive oxygen
species in arsenite-induced downregulation of phospholipid hydroperoxide glutathione
peroxidase in human epidermoid carcinoma A431 cells. Free Radical Biol. Med. 33,
864–873
24 Massague, J. (2008) TGFβ in cancer. Cell 134, 215–230
25 Hamid, R., Patterson, J. and Brandt, S. J. (2008) Genomic structure, alternative splicing
and expression of TG-interacting factor, in human myeloid leukemia blasts and cell lines.
Biochim. Biophys. Acta 1779, 347–355
26 Westmark, C. J., Bartleson, V. B. and Malter, J. S. (2005) RhoB mRNA is stabilized by HuR
after UV light. Oncogene 24, 502–511
27 Masuda, K., Abdelmohsen, K. and Gorospe, M. (2009) RNA-binding proteins implicated
in the hypoxic response. J. Cell. Mol. Med. 13, 2759–2769
28 Datta, S. R., Brunet, A. and Greenberg, M. E. (1999) Cellular survival: a play in three Akts.
Genes Dev. 13, 2905–2927
29 Tabellini, G., Cappellini, A., Tazzari, P. L., Fala, F., Billi, A. M., Manzoli, L., Cocco, L. and
Martelli, A. M. (2005) Phosphoinositide 3-kinase/Akt involvement in arsenic trioxide
resistance of human leukemia cells. J. Cell. Physiol. 202, 623–634
30 Kang, S. H., Song, J. H., Kang, H. K., Kang, J. H., Kim, S. J., Kang, H. W., Lee, Y. K. and
Park, D. B. (2003) Arsenic trioxide-induced apoptosis is independent of
stress-responsive signaling pathways but sensitive to inhibition of inducible nitric oxide
synthase in HepG2 cells. Exp. Mol. Med. 35, 83–90
31 Tian, C., Gao, P., Zheng, Y., Yue, W., Wang, X., Jin, H. and Chen, Q. (2008) Redox status
of thioredoxin-1 (TRX1) determines the sensitivity of human liver carcinoma cells
(HepG2) to arsenic trioxide-induced cell death. Cell Res. 18, 458–471
32 Fei, M., Lu, M., Wang, Y., Zhao, Y., He, S., Gao, S., Ke, Q., Liu, Y., Li, P., Cui, X. et al.
(2009) Arsenic trioxide-induced growth arrest of human hepatocellular carcinoma cells
involving FOXO3a expression and localization. Med. Oncol. 26, 178–185
33 Roboz, G. J., Dias, S., Lam, G., Lane, W. J., Soignet, S. L., Warrell, Jr, R. P. and Rafii, S.
(2000) Arsenic trioxide induces dose- and time-dependent apoptosis of endothelium and
may exert an antileukemic effect via inhibition of angiogenesis. Blood 96, 1525–1530
34 Xiao, Y. F., Liu, S. X., Wu, D. D., Chen, X. and Ren, L. F. (2006) Inhibitory effect of arsenic
trioxide on angiogenesis and expression of vascular endothelial growth factor in gastric
cancer. World J. Gastroenterol. 12, 5780–5786
35 Cavigelli, M., Li, W. W., Lin, A., Su, B., Yoshioka, K. and Karin, M. (1996) The tumor
promoter arsenite stimulates AP-1 activity by inhibiting a JNK phosphatase. EMBO J. 15,
6269–6279
36 Richter, J., Beffa, L., Wagner, U., Schraml, P., Gasser, T. C., Moch, H., Mihatsch, M. J. and
Sauter, G. (1998) Patterns of chromosomal imbalances in advanced urinary bladder
cancer detected by comparative genomic hybridization. Am. J. Pathol. 153, 1615–1621
Received 26 October 2010/6 June 2011; accepted 7 June 2011
Published as BJ Immediate Publication 7 June 2011, doi:10.1042/BJ20101653
c The Authors Journal compilation c 2011 Biochemical Society
37 Nakakuki, K., Imoto, I., Pimkhaokham, A., Fukuda, Y., Shimada, Y., Imamura, M.,
Amagasa, T. and Inazawa, J. (2002) Novel targets for the 18p11.3 amplification frequently
observed in esophageal squamous cell carcinomas. Carcinogenesis 23, 19–24
38 Wotton, D., Lo, R. S., Lee, S. and Massague, J. (1999) A Smad transcriptional
corepressor. Cell 97, 29–39
39 Fiorentino, M., Altimari, A., D’Errico, A., Cukor, B., Barozzi, C., Loda, M. and Grigioni,
W. F. (2000) Acquired expression of p27 is a favorable prognostic indicator in patients
with hepatocellular carcinoma. Clin. Cancer Res. 6, 3966–3972
40 Calvisi, D. F., Ladu, S., Pinna, F., Frau, M., Tomasi, M. L., Sini, M., Simile, M. M., Bonelli,
P., Muroni, M. R., Seddaiu, M. A. et al. (2009) SKP2 and CKS1 promote degradation of
cell cycle regulators and are associated with hepatocellular carcinoma prognosis.
Gastroenterology 137, 1816–1826 e1811–1810
41 Seo, S. R., Lallemand, F., Ferrand, N., Pessah, M., L’Hoste, S., Camonis, J. and Atfi, A.
(2004) The novel E3 ubiquitin ligase Tiul1 associates with TGIF to target Smad2 for
degradation. EMBO J. 23, 3780–3792
42 Lin, H. K., Bergmann, S. and Pandolfi, P. P. (2004) Cytoplasmic PML function in TGF-β
signalling. Nature 431, 205–211
43 Zhang, Y., Bhatia, D., Xia, H., Castranova, V., Shi, X. and Chen, F. (2006) Nucleolin
links to arsenic-induced stabilization of GADD45α mRNA. Nucleic Acids Res. 34,
485–495
44 Campos, A. R., Grossman, D. and White, K. (1985) Mutant alleles at the locus elav in
Drosophila melanogaster lead to nervous system defects. A developmental-genetic
analysis. J. Neurogenet. 2, 197–218
45 Good, P. J. (1995) A conserved family of elav-like genes in vertebrates. Proc. Natl. Acad.
Sci. U.S.A. 92, 4557–4561
46 Lal, A., Kawai, T., Yang, X., Mazan-Mamczarz, K. and Gorospe, M. (2005) Antiapoptotic
function of RNA-binding protein HuR effected through prothymosin α. EMBO J. 24,
1852–1862
47 Abdelmohsen, K., Lal, A., Kim, H. H. and Gorospe, M. (2007) Posttranscriptional
orchestration of an anti-apoptotic program by HuR. Cell Cycle 6, 1288–1292
48 Mei, C., Sun, L., Liu, Y., Yang, Y., Cai, X., Liu, M., Yao, W., Wang, C., Li, X., Wang, L.
et al. (2010) Transcriptional and post-transcriptional control of DNA methyltransferase 3B
is regulated by phosphatidylinositol 3 kinase/Akt pathway in human hepatocellular
carcinoma cell lines. J. Cell. Biochem. 111, 158–167
49 Vara, J. A., Casado, E., de Castro, J., Cejas, P., Belda-Iniesta, C. and Gonzalez-Baron, M.
(2004) PI3K/Akt signalling pathway and cancer. Cancer Treat. Rev. 30, 193–204
50 Altman, J. K., Yoon, P., Katsoulidis, E., Kroczynska, B., Sassano, A., Redig, A. J., Glaser,
H., Jordan, A., Tallman, M. S., Hay, N. and Platanias, L. C. (2008) Regulatory effects of
mammalian target of rapamycin-mediated signals in the generation of arsenic trioxide
responses. J. Biol. Chem. 283, 1992–2001
51 Li, Y., Qu, X., Qu, J., Zhang, Y., Liu, J., Teng, Y., Hu, X., Hou, K. and Liu, Y. (2009) Arsenic
trioxide induces apoptosis and G2/M phase arrest by inducing Cbl to inhibit PI3K/Akt
signaling and thereby regulate p53 activation. Cancer Lett. 284, 208–215
52 Frohlich, E., Czarnocka, B., Brossart, P. and Wahl, R. (2008) Antitumor effects of arsenic
trioxide in transformed human thyroid cells. Thyroid 18, 1183–1193