Carcinogenesis vol.31 no.7 pp.1298–1307, 2010
Advance Access publication April 16, 2010
SOX4 overexpression regulates the p53-mediated apoptosis in hepatocellular
carcinoma: clinical implication and functional analysis in vitro
Wonhee Hur1,2, Hyangshuk Rhim3, Chan Kwon Jung4,
Jin Dong Kim1,2, Si Hyun Bae1,2, Jeong Won Jang1,2,
Jin Mo Yang1,2, Seong-Taek Oh5, Dong Goo Kim5,
Hee Jung Wang6, Sean Bong Lee7and Seung Kew Yoon1,2,?
1Department of Internal Medicine and2WHO Collaborating Center of Viral
Hepatitis, The Catholic University of Korea, Seoul 137-701, Korea,3Research
Institute of Molecular Genetics, Catholic Research Institutes of Medical
Science, The Catholic University of Korea, Seoul 137-701, Korea,
4Department of Hospital Pathology and5Department of Surgery, The Catholic
University of Korea, Seoul 137-701, Korea,6Department of Surgery, Ajou
University School of Medicine, Suwon, 443-721, Korea and7Genetics of
Development and Disease Branch, The National Institute of Diabetes and
Digestive and Kidney Diseases/National Institutes of Health, Bethesda,
MD 20892, USA
?To whom correspondence should be addressed. Tel: þ82-2-2258-7534;
Background and aims: The underlying molecular mechanisms of
hepatocellularcarcinoma (HCC)remainpoorlyunderstooddue to
its complex development process. The human T cell-specific tran-
scription factor sex-determining region Y-related high-mobility
group (HMG) box 4 (SOX4) has been linked to development
and tumorigenesis. In this study, we characterized the roles of
SOX4 in regulation of the p53 transcription activity and evaluated
the expression patterns and prognostic value of the transcription
factor SOX4 in HCC. Methods: The expression levels of human
SOX4 were examined in HCC samples obtained from 58 patients
having curative partial hepatectomy. The interaction and effects
of SOX4 on the p53 pathway were assessed in HCC cell lines.
Luciferase reporter assay to examine p53-mediated transcription
of target genes was performed. The association of SOX4 expres-
sion level with tumor recurrence and overall survival was evalu-
ated. Results: We showed that the HMG box domain of SOX4
interacted with p53, resulting in the inhibition of p53-mediated
transcription by the Bax promoter. More importantly, SOX4
overexpression led to a significant repression of p53-induced
Bax expression and subsequent repression of p53-mediated apo-
ptosis induced by g-irradiation. In clinicopathological analysis,
nuclear overexpression of SOX4 was observed in 37 out of 58
(63.8%) HCC samples, and this correlated with diminished risk
of recurrence (P 5 0.014) and improved overall survival time
(P 5 0.045) in HCC patients. Conclusion: These results suggest
that SOX4 contributes to hepatocarcinogenesis by inhibiting
p53-mediated apoptosis and that its overexpression might be
a useful prognostic marker for survival after surgical resection.
Hepatocellular carcinoma(HCC)is the fifth mostcommon cancerin the
world and the third leading cause of cancer-related death globally.
Chronic hepatitis B virus (HBV) and hepatitis C virus infections con-
tribute to HCC development in .80% of the HCC cases worldwide (1).
Other factors associated with HCC include heavy alcohol drinking,
exposure to aflatoxin B1 (AFB1), non-alcoholic fatty liver disease, he-
mochromatosis, diabetes and obesity (2). Although the risk factors for
HCC are well defined, the underlying molecular mechanisms remain
unclear because hepatocarcinogenesis is a complex process associated
with the accumulation of genetic and epigenetic changes that pass
through steps of initiation, promotion and progression. These molecular
events are accompanied by enhanced expression of several factors that
influence cancer cell survival by regulating the cell cycle and apoptosis.
Over the past decade, extensive research has focused on the identifica-
acivated protein kinase, Janus kinases/signal transducers and activators
of transcription, stress response signaling, epidermal growth factor
receptor and transforming growth factor-b pathways (2). Although the
reason why these different signaling pathways are independently
involved in the carcinogenesis of HCC is unclear, it might be due to a
unique manner of each risk factor for HCC in signal transduction.
The SOX gene family (sex-determining region Y-related high-
mobility group [HMG] box) plays a key role in regulating transcrip-
tion in diverse developmental processes (3,4). The SOX family shares
the highly conserved HMG box (5), which binds DNA directly in the
minor helix groove of the DNA helix (6). The SOX proteins show
diverse functions in mammals, because the residues outside the HMG
box domain are variable and may influence the selection of cellular
partner proteins and subsequently lead to DNA-binding stabilities
(3,7,8). Among the many SOX members, SOX4 belongs to the
C subgroup of the family (5,7,9). The SOX4 gene encodes a protein
of 474 amino acids (aa) with three distinguishable domains: an HMG
box (aa 57–135), a glycine-rich region (aa 152–227) and a serine-rich
region (SRR, aa 333–397). The HMG box serves as a DNA-binding
region, whereas the SRR domain serves as a transactivation domain.
Moreover, the central domain (CD) containing the glycine-rich region
located between the HMG box and SRR domains serves as a novel
functional region for promoting apoptotic cell death (10–12).
Previous studies have shown that SOX4 proteins play essential
roles in endocardial ridge development and in the regulation of lym-
phocyte development and differentiation (11,13–16). Recently, the
overexpression of SOX4 has been reported in several tumors, includ-
ing bladder carcinoma, medulloblastoma, prostate cancer, colon can-
cer, breast cancer and HCC cell lines (17–23). Although the precise
mechanisms by which SOX proteins contribute to tumorigenesis are
poorly understood, they regulate their target genes by pairing off with
specific partner factors. This partnering might allow SOX proteins to
act in a cell-specific manner, which would be an essential role in cell
differentiation or tumorigenesis (9). SOX4 does not affect normal
liver functions due to its lack of expression in normal adult liver
(24). To date, little is known about the molecular and clinical involve-
ment of SOX4 in HCC. However, Liao et al. (25) demonstrated that
SOX4 has an important function in liver tumor metastasis, as RNA
interference knock-down reduced HCC cell migration, invasion and
intrahepatic metastasis in an orthotopic liver cancer model.
a number of cellular signaling pathways (26,27) and the loss of its
function is known to be a common feature of many human cancers
has been associated with exposure to AFB1(29). Regarding the rela-
tionship between SOX4 and p53, the induction of SOX4 in response to
DNA damage is critical for p53 stabilization and function (30). To the
best of our knowledge, the expression of SOX4 in human HCC has not
been demonstrated. Hence, in the present study, we evaluated SOX4
expression in human HCC samples. We found SOX4 overexpression in
human HCC, which prompted us to investigate the relationship be-
tween SOX4 overexpression and p53 in the regulation of apoptosis
and in hepatocarcinogenesis. Furthermore, we assessed the clinical
prognostic value of SOX4 protein expression as a biomarker for HCC.
hepatitis B virus; HCC, hepatocellular carcinoma; HEK, human embryonic
kidney; HMG, high-mobility group; IgG, immunoglobulin G; Mdm2, murine
group (HMG) box; SRR, serine-rich region; WT, wild-type.
? The Author 2010. Published by Oxford University Press. All rights reserved. For Permissions, please email: firstname.lastname@example.org 1298
by guest on November 5, 2015
Materials and methods
Fifty-eight paired samples of HCC and their corresponding non-tumorous liver
tissues were obtained from HCC patients who had undergone curative partial
hepatectomy atKangnam StMary’s Hospital atthe CatholicUniversityofKorea
(Seoul, Korea) between January 2000 and December 2002. Informed consent
was obtained for specimen collection in all cases and the study protocol was
approved by the Ethics Committee of the Catholic University of Korea.
The liver specimens were prepared for histopathological evaluation using
conventional paraffin embedding, sectioning and hematoxylin and eosin stain-
ing. Fresh tissue specimens were also frozen in liquid nitrogen and stored until
use. The histopathological grade of tumor differentiation was assessed accord-
ing to the Edmondson–Steiner grade criterion (31).
Tissue microarray generation
Representative tissue areas were marked on standard hematoxylin- and eosin-
stained sections, punched out of the paraffin block with a 2.0 mm punch and
inserted into a recipient paraffin block, resulting in 5 ? 6 arrays for each of the
30 cases. Additionally, three control tissue specimens were inserted near the
5 ? 6 arrays on the same recipient block. Duplicate tissue cores per specimen
were arrayed on a recipient paraffin block to minimize the error introduced by
sampling and to minimize the impact of tissue loss during processing.
Five micrometer-thick sections were cut from the completed array block and
transferred to silanized glass slides. The tissue sections were deparaffinized by
incubation in xylene and rehydrated in a graded series of ethanol–water solu-
tions. Antigen retrieval was performed by heating the sample in 0.01 M citrate
buffer (pH 6.0) using a microwave vacuum histoprocessor (RHS-1, Milestone,
Bergamo, Italy) at a controlled final temperature of 121?C for 15 min. The
endogenous peroxidase activity was blocked by incubating the slides in 3%
polyclonal antibody to SOX4 (Sigma, St Louis, MO), which had been diluted
1:800 in Dako antibody diluent (Dako, Carpinteria, CA) with background-
reducing components, at room temperature for 30 min. After washing, we
detected antibody binding using a Dako EnVision Plus system. The immuno-
reaction was developed with diaminobenzidine (Dako) for 5 min, and hematox-
ylin counterstaining was used. The nuclear and cytoplasmic staining with SOX4
was evaluated for the tumor cells. The immunostaining was considered positive
when .10% of the tumor cells were immunoreactive. The intensity of staining
was graded semiquantitatively as negative, weak, moderate or strong positivity.
The two pathologists independently reviewed all core biopsies.
The clinicopathological characteristics of HCC patients analyzed included age,
sex, tumor size, Edmondson–Steiner grade, pathological tumor-node-metastasis
stage, vascular invasion, intrahepatic metastasis, microscopic involvement of
resection margin, serum alpha-fetoprotein levels, presence of HBV infection
and accompanying liver cirrhosis.
HepG2, Hep3B and human embryonic kidney (HEK) 293 cells (American
Type Culture Collection, Manassas, VA) were cultured in Dulbecco’s modified
Eagle’s medium (Invitrogen, Carlsbad, CA) supplemented with 10% fetal
bovine serum (Invitrogen), 50 U/ml penicillin and 50 lg/ml streptomycin
(Invitrogen) at 37?C in a humidified environment with 5% CO2.
Plasmids and cell transfection
The cDNA encoding the N-terminal region of human SOX4 (aa residues 1–173
encompassing the HMG box domain) was generated by polymerase chain
reaction amplification with Pfu DNA polymerase (Stratagene, La Jolla, CA)
using the pCDM7-SOX4 plasmid as a template (11). The amplified fragment
was inserted into the pGEX-4T-1 plasmid (Amersham Biosciences Corp.,
Cardiff, UK) and designated as pGST-HMG. A plasmid encoding the wild-
type (WT) SOX4 with the C-terminal FLAG epitope tag was generated by
modifying the pGST-HMG and pCDM7-SOX4 plasmids (12), designated as
pFLAG-SOX4 (WT). The following truncated SOX4 constructs of various
sizes were generated by digesting with appropriate unique restriction enzymes
within the SOX4 cDNAs and inserting the appropriate fragment into pFLAG-
CMV5a (Sigma) or modified pFLAG plasmids: HMG (aa 1–173), CD (aa 166–
342) and SRR (aa 343–474) (12). The pDS332-p53 plasmids were kindly
provided by Dr Shin (Dan-kook University, Korea). The verification of
sequence integrity and expression of each plasmid construct was achieved
by DNA sequencing (ABI Prism BigDye Terminator Cycle Sequencing Ready
Reaction Kit; Applied Biosystems, Foster City, CA) and by immunoblot anal-
yses with specific antibodies, respectively.
For all transfections, 1 ? 106cells were plated on a 100 mm culture dish and
1.5 lg of the indicated plasmid was transfected into each cell line using Lip-
ofectamineTMreagent (Invitrogen) according to the manufacturer’s instructions.
For the co-immunoprecipitation experiments, HepG2 and HEK293 cells trans-
fected with expression plasmids were lysed by incubation with lysis buffer
[0.2% digitonin, 20 mM N-2-hydroxyethylpiperazine-N#-2-ethanesulfonic
acid (pH 7.5), 100 mM KCl, 10 mM CaCl2and 50 mM MgCl2] containing
protease inhibitor tablets (Roche, Mannheim, Germany). The total protein
extracts (1 mg) were incubated with anti-p53 antibody (Santa Cruz Biotech-
nology, Santa Cruz, CA) or anti-FLAG beads (Sigma) and then Protein
A agarose (Invitrogen) was added. After 1 h, the beads were washed with
ice-cold phosphate-buffered saline, and the bound immunoprecipitates were
eluted from the beads by boilingin sample buffer [62.5 mM Tris–HCl (pH 6.8),
10% glycerol, 2% sodium dodecyl sulfate (SDS), 144 mM b-mercaptoethanol
and 0.0005% bromophenol blue].
Western blot analysis
Immunoblot analysis was carried out for frozen liver tissues and harvested cells.
The frozen liver tissue samples were pulverized in liquid nitrogen suspended in
RIPA cell lysis buffer [20 mM Tris–HCl (pH 7.5), 150 mM NaCl, 1% Triton
X-100, 1% sodium deoxycholate and 0.1% SDS] containing protease inhibitors
and further dispersed using a dounce homogenizer. The protein extracts and
immunoprecipitated pellets were resolved by 12% SDS–polyacrylamide gel
electrophoresis and transferred to nitrocellulose membranes (Schleicher &
Schuell, Dassel, Germany). The membranes were blocked in 5% skim milk
and then incubated with the following primary antibodies: polyclonal rabbit
anti-SOX4 (1:500; Sigma), monoclonal mouse anti-FLAG (1:1000; Sigma),
polyclonal rabbit anti-p53 (1:1000; Santa Cruz), polyclonal rabbit anti-Bax
(1:1000; Santa Cruz), monoclonal mouse anti-enhanced green fluorescent pro-
tein (1:2000; Santa Cruz) and monoclonal mouse anti-b-actin (1:2500; Sigma).
The blots were then incubated with horseradish peroxidase-conjugated anti-
mouse or anti-rabbit secondary antibodies (1:5000; Amersham) and developed
using the enhanced chemiluminescence system (Amersham). Densitometry val-
ues were determined for the SOX4 and b-actin bands, and the ratio of the SOX4
and b-actin values was calculated for each sample. These values are reported as
the mean ratio from three separate blots. To compare the expression levels of
SOX4 between the tumorous and surrounding non-tumorous liver tissues, the
protein bands were quantified using a Kodak molecular imaging system and
standardized relative to the b-actin expression level.
Cell fractionation and immunofluorescence staining
Transfected HepG2 and Hep3B cells were fractionated into nuclear and cyto-
plasmic fractions using an NE-PER kit (Pierce Biotechnology, Rockford, IL).
Transfected cells were incubated with anti-FLAG (1:500; Sigma) and anti-p53
antibodies (1:500; Santa Cruz), followed by incubation with Cy3-coupled anti-
mouse immunoglobulin G (IgG) (1:500; Jackson Laboratory, Harbor, ME) or
Alexa 488-labeled anti-rabbit IgG (1:500; Jackson Laboratory), respectively.
Cells were stained with 4#,6-diamidino-2-phenylindole (Sigma) to counterstain
the nuclei and were then examined by fluorescence microscopy (Carl Zeiss,
Luciferase reporter assays
andp21waf1werekindlyprovided by A.Fusco(Universita ` degli Studidi Napoli
Federico II, Italy). Cells were transfected with various plasmids, as indicated.
microplate luminometer (Tuner Biosystems, Sunnyvale, CA). The firefly lu-
minescence signal was normalized to the Renilla luminescence signal. The
resultsare reported asthe mean ±SD of at least threeindependentexperiments.
Chromatin immunoprecipitation assays
SOX4 overexpression was shown to suppress p53 transcriptional activity on the
Bax promoter in a dose-dependent manner. To determine if SOX4 is present in
the p53 transcription complex, we performed chromatin immunoprecipitation
assays on the Bax gene promoter with anti-FLAG, anti-p53 or a control IgG.
in 100 mm dish. Cells were then treated with the cross-linking reagent formal-
dehyde (1% final concentration) for 10 min at 37?C, rinsed twice with cold
phosphate-buffered saline and swollen on ice in SDS lysis buffer (1% SDS,
10 mM ethylenediaminetetraacetic acid and 50 mM Tris–HCl, pH 8.1) supple-
mentedwith proteaseinhibitors (Roche).Nuclei were collected and sonicated on
ice. Supernatants were diluted 5-fold in chromatin immunoprecipitation dilution
buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM ethylenediaminetetraacetic
acid, 16.7 mM Tris–HCl, pH 8.1, and 167 mM NaCl) and incubated with 2 ll
Role of SOX4 overexpression in HCC
by guest on November 5, 2015
Immunoprecipitation was performed using protein A/G-agarose for 1 h at 4?C
recovered and reversed according to Upstate’s protocol (Upstate, Chicago, IL).
Final DNA pellets were recovered and analyzed by polymerase chain reaction
using a pair of primers that encompass the Bax promoter region. The primers
for the Bax gene promoter were (5#-TAATCCCAGCGCTTTGGAAG-3# and
5#-GTCCAATCGCAGCTCTAATG-3#) and the glyceraldehyde 3-phosphate
dehydrogenase gene promoter were (5#-AAAAGCGGGGAGAAAGTA-3#
Induction and measurement of apoptosis
To investigate whether SOX4 can modulate the biological activity of p53, we
treated HepG2 and Hep3B cells with c-irradiation to induce p53-mediated apo-
ptosis. The transfected cells with pFLAG-SOX4 (WT) were treated with
7.5 Gy c-irradiation (132Cs, 2.875 Gy/min) using a Gammacell irradiator
(Atomic Energy of Canada Ltd, Ottawa, Canada). To determine the radiation-
induced apoptosis, the cells were washed once with phosphate-buffered saline
iodide (BD Biosciences, San Jose, CA) for 15 min according to the manufac-
turer’s instructions. Stained cells were detected using the FACSCalibur flow
cytometer (BD Biosciences).
Genomic DNA extraction and DNA sequencing analysis
In order to isolate genomic DNA from the paraffin-embedded 12 HCC tissues
for the determination of p53 mutation, tumor foci were identified and five
10 lm-thick sections were cut with standard microtome from every paraffin
wax block and transferred into a microtube. The microtome blade was washed
with xylene and ethanol. After paraffin removal, genomic DNAwas extracted
using QIAamp DNA mini kit (QIAGEN, Hilden, Germany) according to the
manufacturer’s instructions. All DNA samples were stored at ?20?C. Poly-
merase chain reaction amplification of genomic DNA was performed using
primers specific for human p53. Ten different sets of 20mer oligonucleotide
primers were designed using the genomic sequence of p53 (GenBank acces-
sion numbers NM_005228.3). Their sequences have been reported elsewhere
(32–34). The primers were also used for sequencing analysis.
To analyze the correlation intensity of SOX4 expression with clinicopatholog-
ical factors, the unpaired Student’s t-test, Fisher’s exact test or chi-square test,
depending on categorical or numerical date, were applied. Cumulative survival
statistically significant. All statistical analyses were conducted using the SPSS
version 13.0 software package (SPSS Inc., Chicago, IL).
SOX4 expression in HCC
To investigatewhether the SOX4proteinisexpressed inHCC tissuesor
not, western blot analysis was conducted in 20 paired HCCs and cor-
responding non-tumorous liver tissues. As shown in Figure 1A, the
SOX4 was expressed at high levels in 14 of 20 (70%) HCC tissues
and only at low levels in the corresponding non-tumorous liver tissues.
The optical density of each band from the tumorous (T) and non-
tumorous (N) tissues in the same patient was measured using TINA
image analysis software. The ratio between the optical density of pro-
teins of interest and b-actin in the same sample was calculated as the
An immunohistochemical study using anti-SOX4 polyclonal antibod-
ies was performed to determine whether SOX4 is expressed in HCC
samples. Tumor cells showed nuclear and/or cytoplasmic expression of
SOX4 (Figure 1B): (i) weak cytoplasmic, (ii) strong cytoplasmic and
(iii) strong nuclear SOX4 expression. Nuclear SOX4 staining was neg-
ative for 4 (6.9%) patients, weak for 17 (29.3%) patients, moderate for
31 (53.4%) patients and strong (10.3%) for 6 patients. Cytoplasmic
SOX4 staining was negative for 2 (3.4%) patients, weak for 7 (12.1%)
patients, moderate for 29 (50.0%) patients and strong for 20 (34.5%)
patients. In normal liver tissue, hepatocytes were negative but lym-
phocytes were positive for SOX4 (Figure 1B, d). This finding is consis-
tent with prior evidence that B and T lymphocytes express SOX4 (11).
Correlation between SOX4 expression and clinicopathological
parameters in HCC
We analyzed the correlation between the SOX4 expression pattern
and clinicopathological characteristics of the tumors. By statistical
analysis, there were no significant correlation between SOX4 expres-
sion and any clinicopathological characteristics, such as age, gender,
tumor size, tumor grade, vascular invasion, intrahepatic metastasis,
serosal invasion, resection margin,serum alpha-fetoprotein level, hep-
atitis B surface antigen status or cirrhosis (Table I). However, we
found that patients with nuclear SOX4 overexpression (moderate or
strong immunopositivity) had a better prognosis than those with neg-
ative or weak nuclear SOX4 expression (disease-free survival rate,
P 5 0.014; overall survival rates, P 5 0.045; as shown in Figure 2). In
contrast, there was no significant correlation between the cytoplasmic
SOX4 expression and recurrence or survival rates. These results sug-
gest the nuclear location of SOX4 as a prognostic indicator of SOX4
expression-related HCC patients.
SOX4 interacts with p53 in HCC cells
Although recent studies have demonstrated SOX4 to be an important
developmental transcription factor that is often overexpressed in sev-
eral types of cancers, little is known about how SOX4 is regulated in
cancer cells. SOX4 overexpression has been reported in an HCC cell
line (18). More recently, it has been shown that SOX4 regulates p53
stability (30) and p53 target genes such as Bcl 2 binding component 3
(21). In this study, we sought to understand whether human SOX4
binds to p53 in HCC cells. To this end, we conducted co-immunopre-
cipitation studies in HEK293 and HepG2 cells. We used HEK293
cells because it can be efficiently transfected with expression vector.
In brief, HEK293 and HepG2 cells were transiently co-transfected
with pFLAG-SOX4 (WT) and/or pDS332-p53, followed by immuno-
precipitation with anti-FLAG beads and immunoblotting with anti-
p53 polyclonal antibody. The FLAG-immunoprecipitated complex
revealed an interaction between SOX4 and p53 in both the HEK293
and HepG2 cells (Figure 3A). In the reverse experiment, HepG2 cells
were transfected with the pFLAG-SOX4 (WT) expression plasmid
and immunoprecipitated with anti-p53 polyclonal antibody followed
by immunoblotting with anti-FLAG monoclonal antibody. The immu-
noprecipitated complex of SOX4 with endogenous p53 is shown in
Figure 3A. These results demonstrate that SOX4 interacts with p53 in
by cell fractionation and fluorescence immunocytochemistry in HepG2
cells transfected with pFLAG-SOX4 (WT). The results revealed the
nucleus(Figure 3B). This finding indicates that SOX4 and p53 proteins
are co-localized and thus can potentially interact in the nucleus.
HMG domain of SOX4 is required for the interaction with p53
To identify the region of SOX4 required for the nuclear p53 interaction,
SRR domain or CD containing the glycine-rich region motif (12). Each
of these SOX4 mutants was expressed at a level similar to that of the
with p53, we transiently co-transfected HEK293 cells with pDS332-
p53 expression plasmid and the expression plasmid for SOX4 (WT),
the HMG domain, the SRR domain or the CD. The anti-FLAG immu-
noprecipitation from these cells were immunoblotted with anti-p53
antibody. As shown in Figure 3C, SOX4 (WT) and HMG formed
complexes readily with p53, whereas the mutants expressing only the
CD or SRR domain failed to bind p53. These findings indicate that the
SOX HMG domain is necessaryfor the interaction with p53 but the CD
and the extreme C-terminal transactivation region of SOX4 are not
involved in the interaction with p53. These results are consistent with
previous finding reported by Pan et al. (30).
SOX4 suppresses p53-mediated transactivation of p53-responsive
To further define the effects of SOX4 on p53-mediated transcription,
we transfected HepG2 cells harboring WT p53 and Hep3B cells de-
ficient in p53 with pFLAG-SOX4 (WT) and with reporter vectors
W.Hur et al.
by guest on November 5, 2015
Fig. 1. SOX4 expression in HCC. (A) Expression of SOX4 by immunoblot analysis in matchedtumorous (T) and non-tumorous (N) tissues from 20 HCC patients.
The results shown are from one representative experiment among three replicates. The band densities were quantified with TINA image analysis software
using b-actin as a reference. The data are expressed relative to the density of the control. HEK293 cells that were transiently transfected with the pFLAG-SOX4
(WT) expressionvector were used as a positive control. (B) Immunohistochemical staining pattern of SOX4 in HCC. The tumor cells show weak cytoplasmic (a),
strong cytoplasmic (b) or strong nuclear (c) SOX4 expression. SOX4 expression was not observed in normal liver and in non-tumor liver cells (d), whereas
most of the lymphocytic cells are positive for SOX4. Original magnification ?400.
Role of SOX4 overexpression in HCC
by guest on November 5, 2015
carrying the luciferase gene under the control of the p53-responsive
Bax, Mdm2 or p21waf1promoter. The expression of SOX4 reduced
p53 transcriptional activity on the Bax promoter in a dose-dependent
manner in HepG2 cells (Figure 4A) but did not lead to significant
difference between the levels mediated by the Mdm2 nor p21waf1
promoter (data not shown).
Furthermore, ectopically expressed p53 in p53-deficient Hep3B
cells activated the transcription of the Bax-luc vector (Figure 4B).
However, increasing the level of SOX4 protein greatly reduced this
p53-mediated Bax-luc transcription in a dose-dependent manner.
These results suggest that tumor suppressor p53 transcriptional activ-
ity may be inhibited in HCC cells overexpressing SOX4.
To clarify the mechanism by which SOX4 overexpression represses
the p53-mediated Bax-luc transcription, we performed a chromatin
immunoprecipitation analysis. Figure 4C demonstrates that the pre-
cipitation of SOX4 protein cross-linked to DNA under these condi-
tions resulted in successful amplification of a p53-response element
region in the Bax promoter and the precipitation of p53 protein was
specifically recruited to the Bax promoter as a positive control. No
precipitationwas observedwith anti-IgG precipitates,and
Table I. Association of SOX4 expression with clinicopathological characteristics in human HCCs
VariableNo. of cases Nuclear SOX4 expressionP Cytoplasmic SOX4 expressionP
Negative (%) Positive (%)Negative (%) Positive (%)
Mean age ± SD (years)
Grade 1, 2
Grade 3, 4
Hepatitis B surface antigen status
55.1 ± 11.154.3 ± 11.50.723
56.6 ± 11.8 54.2 ± 11.20.927
Fig. 2. Nuclear overexpression of SOX4 correlates with diminished risk of recurrence (A) and improved overall survival time (B) in patients with HCC.
Solid and dotted lines present moderate to strong nuclear expression of SOX4 (n 5 37) and negative or weak nuclear expression of SOX4 (n 5 21) cases,
respectively. Strong nuclear expression of SOX4 was significantly associated with better prognosis (A: P 5 0.014; B: P 5 0.045).
W.Hur et al.
by guest on November 5, 2015
glyceraldehyde 3-phosphate dehydrogenase promoter-specific pri-
mers were used to confirm specificity of recruitment of SOX4/p53
onto the Bax promoter (Figure 4C). These results therefore confirm
that SOX4 is certainly bound to the Bax promoter in HepG2 cells
and may thus act by associating with p53 once it is bound to the
SOX4 suppresses p53-mediated apoptosis induced by c-irradiation
in HCC cells
It is well established that p53 plays a critical role in DNA damage-
inducedapoptosis(35). Based on our previous study that demonstrated
that SOX4 overexpression in HEK293 cells results in increased cell
death (12), we examined p53-mediated apoptosis induced by
Fig. 3. SOX4 protein interacts with p53 in nucleus. (A) HepG2 and HEK293 cells were transfected with the indicated plasmid pairs. Equal amounts of whole-cell
lysates were precleared with the A/G-agarose beads and used for immunoprecipitation (IP) with anti-FLAG beads. SOX4 was co-immunoprecipitated with p53.
The blot was stripped and probed with FLAG antibody to detect immunoprecipitated SOX4 protein. The in vivo interaction between SOX4 and endogenous
p53 was assessed in HepG2 cells. The cell lysates were also immunoprecipitated with anti-FLAG beads followed by immunoblot analysis with anti-p53 antibody.
The reverse experiment was performed with immunoprecipitation using p53 antibody and western blot using anti-FLAG antibody. The blot was stripped and
probed with p53 antibody to detect immunoprecipitated p53 protein. (B) SOX4 is distributed in the nuclear and cytoplasmic compartments of HepG2 cells, and
p53 is localized predominantly in the nucleus. Nuclear and cytoplasmic compartments were prepared as described in the Materials and Methods. Transfected
HepG2 cells were seeded on a slip for dual immunofluorescence analysis. Cells were stained with monoclonal anti-FLAG and polyclonal anti-p53 antibodies.
The detection of SOX4 with Cy3-conjugated secondary antibody (red) shows its distribution in the cytoplasm and nucleus of HepG2 cells; the nuclear localization
of p53 was detected with Alexa 488-labeled secondary antibody (green). An overlay of the images reveals the co-localization of both proteins in the nucleus
(yellow; arrowhead). The nuclei were counterstained with 4#,6-diamidino-2-phenylindole. (C) Mapping of the p53 interaction domain of SOX4. Various
truncated SOX4 constructs were assessed for the ability to interact with p53 by co-immunoprecipitation. HEK293 cells were transiently co-transfected with
pDS332-p53 and the indicated plasmid carrying the pFLAG-SOX4 (WT) vector or a deletion vector. After 24 h, total cell extracts were prepared, and equal
amounts of proteins were immunoprecipitated with anti-FLAG antibody beads, followed by immunoblot analysis of the immunocomplexes using anti-p53
antibody. The relative inputs are the total cell extracts derived from HEK293-transfected cells with the indicated plasmid pairs.
Role of SOX4 overexpression in HCC
by guest on November 5, 2015
c-irradiation in cells expressing SOX4 (WT) in order to evaluate the
effects of SOX4 on the biological activity of p53 and then quantified
by the annexin V–fluorescein isothiocyanate. We observed apoptosis
rates of ?160 and 135% in SOX4 (WT)-transfected HepG2 and
Hep3B cells versus in the cells transfected with mock vector alone
(Figure 5A and B). Ionizing radiation increased the apoptotic popula-
tion to .140% in mock vector-transfected HepG2 and Hep3B cells,
respectively. The c-irradiation-induced apoptotic cell death was mark-
edly lower (84%) in SOX4 (WT)-transfected HepG2 but not in SOX4
(WT)-transfected Hep3B cells (Figure 5A and B). We next examined
whether reduced apoptosis in the SOX4-expressing cell population
was caused by p53 transcription repression, as determined indirectly
by measuring the endogenous Bax expression level, in irradiated
SOX4 (WT)-expressing HepG2 and Hep3B cells. The Bax expression
level was not significantly different between the pFLAG-SOX4 (WT)-
and mock vector-transfected HepG2 and Hep3B cells in the absence of
irradiation (Figure 5C). It decreased significantly in HepG2 cells ex-
pressing SOX4 following irradiation as compared with the control
cells (Figure 5C). However, Bax expression decreased markedly in
SOX4-expressing HepG2cells following irradiation,ascomparedwith
the control cells (Figure 5C). These results suggest that the inhibition
of c-irradiation-induced p53-mediated apoptosis occurs through
a SOX4 (WT)-mediated repression of p53 transactivational activity.
Although various risk factors for HCC are well recognized, the mech-
anisms of hepatocarcinogenesis are not completely understood due to
the complexity of the process. However, in general, HCC develops in
the setting of cirrhosis or chronic hepatitis in which continuous ex-
posure to injurious stimuli such as hepatitis viruses elicits chronic
inflammation and hepatocyte regeneration. However, little is known
about what leads to hepatocarcinogenesis. Numerous studies have
shown that cellular signaling pathways involved in HCC differ in
different settings based on various risk factors: the p53 pathway is
affected by HBV, AFB1 and hemochromatosis; Wnt/b-catenin is
affected by HBV, hepatitis C virus, AFB1and alcohol; the mitogen-
acivated protein kinase pathway is affected by HBV and hepatitis
Mounting evidences suggest that the SOX4 protein is involved in the
development of several tumor types (17,18,20–23,25); however, the
way in which SOX4 exerts its oncogenic effects during the malignant
transformation of normal cells remains unclear. In particular, there has
been no report regarding the role of SOX4 on hepatocarcinogenesis, to
date. In this study, we analyzed SOX4 expression by western blotting
and immunohistochemical staining to obtain evidence that SOX4 is
overexpressed significantly in human HCC compared with the corre-
sponding non-tumorous tissue (in 63.8 and 70% of HCC cases by
immunohistochemistry and western blot, respectively). We also evalu-
ated the correlation between the SOX4 expression pattern (cytoplasm
and nuclear) and clinicopathological features in HCC after surgical
resection. Interestingly, the nuclear expression of SOX4 in HCC was
associated significantly with better prognosis in both disease-free sur-
vival and overall survival rates (P 5 0.014 and P 5 0.045, respec-
tively), indicating that patients with SOX4-related HCC have a better
Fig. 4. SOX4 modulates p53-mediated transcription activity. (A) A dose–response analysis of SOX4 on the Bax promoter containing p53-responsive elements.
The indicated promoter-luciferase plasmid and Renilla luciferase plasmid (pRL-TK) were transiently transfected into HepG2 cells expressing WT p53. (B) The
p53-deficient Hep3B cells were transiently co-transfected with the Bax-luciferase plasmid, pRL-TK reporter plasmid and 25 ng of pDS332-p53 in combination
with increasing amounts of pFLAG-SOX4 (WT) plasmid. At 24 h after transfection, the cells were lysed, and luciferase activity was analyzed. The firefly
luminescence signal was normalized to the Renilla luminescence signal. All transfections were performed in triplicate; the data are presented as means ± SDs.
?P , 0.05 versus groups expressing SOX4.??P , 0.01 versus groups expressing SOX4. (C) Chromatin immunoprecipitation with anti-FLAG antibody beads or
anti-p53 antibody or an immunoprecipitation control IgG in HepG2 cells or anti-p53 antibody or an immunoprecipitation control IgG in HepG2 cells transfected
with mock vector or SOX4 (WT) vector. The DNAs were then amplified by polymerase chain reaction using primers that cover a region of human Bax promoter
(-250/-530), which contains the p53-binding sites. The panel shows polymerase chain reaction amplification of the immunoprecipitated DNA using primers
for the glyceraldehyde 3-phosphate dehydrogenase gene promoter.
W.Hur et al.
by guest on November 5, 2015
prognosis than those with SOX4-unrelated HCC. Thus, SOX4 protein
expression could be a potentially useful prognostic indicator of HCC
patient outcome. To the best of our knowledge, this is the first study to
report such a relationship between SOX4 expression and survival rates
inhumanHCC.Recently, Liao et al. (25)have showed thatintrahepatic
metastasis was significantly associated with SOX4 expression in mes-
senger RNA level. However, the expression level of SOX4 was not
verified by immunoblot or immunohistochemistry. This contradictory
finding to our result may be explained by different methodology for
confirmation of SOX4 expression because protein expression may be
modified by posttranscriptional modification. In addition, it is known
that SOX4 exerts its effect on transcription via cooperative binding to
DNA with transcription factor partner (8). Therefore, the overexpres-
sion of SOX4 on tumor initiation and metastatic progression may be
involved in regulating tumor microenvironment.
Previous microarray analyses have revealed SOX4 up-regulation
in pineoblastoma (36), medulloblastoma (20), lung (37), bladder can-
cers (17), prostate cancer (38) and colorectal cancer (39). In bladder
cancer patients, the overexpression of SOX4 protein in the nucleus
and cytoplasm is significantly correlated with longer overall survival
rates (17). In colorectal cancer, patients with high SOX4 transcript
levels have a higher recurrence rate (39); however, they do not dem-
onstrate the association of SOX4 protein level and recurrence in co-
lorectal cancer. As mentioned previously, our results indicate that low
nuclear SOX4 expression might reflect the malignant potential of
HCC. These clinicopathological findings prompted us to investigate
the role of SOX4 in human hepatocarcinogenesis. To clarify the un-
derlying mechanism that connects SOX4 expression to HCC devel-
opment, we performed an in vitro functional characterization of
SOX4 in HCC cells. The mutation and inactivation of p53 have been
reported for .50% of human cancers. In particular, it has been re-
ported that the most common mutation associated with HCC occurs at
codon 249 of p53, which is causally related to high AFB1(40). The
inactivation of p53 owing to point mutation or allelic deletion is
a crucial step during carcinogenesis and a critical event during all
stages of HCC development (41,42). In many cases, these genetic
alterations have contributed to the progression but not the initiation
of HCC (43). Although HCC has been associated with the somatic
mutation and inactivation of the p53 gene, no p53 mutation has been
identified in the majority of HCC cases (44,45), suggesting that an-
other mechanism may be involved in the development or progression
of HCC. The functions of p53 are regulated by protein stabilization,
posttranscriptional modifications and protein subcellular localization
through interactions with numerous proteins (46). Pan et al. (30) were
the first to report that SOX4 is a novel mediator for p53 activation in
response to DNA damage, and it interacts with and stabilizes p53
protein by blocking Mdm2-mediated p53 ubiquitination and degra-
dation. Their findings, together with our results demonstrating the
overexpression of SOX4 in human HCC, led us to examine the ability
of SOX4 to regulate p53 transcriptional activity in HCC cells.
Furthermore, in order to determine whether SOX4 nuclear overex-
pression is correlated with the inactivation of p53 gene function
through p53 somatic mutations, we performed p53 gene sequencing
in the HCC collection. As shown in Table S1 (Supplementary data are
available at Carcinogenesis Online), 5 of our 12 samples showed WT
p53 and the other (7 out of 12) showed Arg72Pro polymorphisms
without mutation, which together shows functionally competent
p53 in our samples. These results support the notion that SOX4
nuclear overexpression inhibits WT p53 transcriptional activity
without p53 somatic mutations.
Fig. 5. SOX4 inhibits the pro-apoptotic activity of p53. (A) Flow cytometric analysis of annexin V–fluorescein isothiocyanate/propidium iodide double-stained
cells. HepG2 cells transfected with mock vector or SOX4 (WT) vector were treated with 7.5 Gy c-irradiation to activate endogenous p53. Apoptosis was assessed
24 h after c-irradiation. The percentage of apoptotic cells was expressed as the percentage of annexin V-positive cells divided by total cells in the gate.
(B) Quantitative analysis of the percentage of apoptotic cells in HepG2 and Hep3B cells transfected with mock vector or SOX4 (WT) vector. The results are
presented as percentage of apoptotic cells relative to mock-transfected cells. Data are representative of three independent experiments. (C) SOX4 regulates Bax
protein expression.Total homogenates ofHepG2 and Hep3Bcells transfected with mockvector or SOX4 (WT)vector, which had been treated with c-irradiation at
doses of7.5 Gy, were analyzed by immunoblottingfor the presenceof SOX4,p53,Bax and greenfluorescence protein(GFP).The expressionlevelwasnormalized
to that of GFP. Values are reported as means ± SDs from three separate experiments.
Role of SOX4 overexpression in HCC
by guest on November 5, 2015
In this study, we identified p53 as a SOX4-interacting protein and
demonstrated that the SOX4–p53 interaction binds onto the Bax pro-
moter and inhibits p53 transcriptional activity on the Bax promoter.
The transcription factor SOX4 regulatesSOX4-mediated transcription
activity through multiple protein interactions (22,47–49). Usually,
these interactions are mediated by the HMG domain of SOX4; this
domain is highly conserved among all SOX proteins (10), binds DNA
in a site-specific manner and facilitates protein dimerization and other
protein–protein interactions (50). Here, we demonstrated that the
HMG domain of SOX4 is required for the interaction between
SOX4 and p53.
Although SOX4 has been well characterized in many different de-
velopmental processes, the proteins that interact with SOX4 and their
target gene promoters are not well defined. Among the known
SOX4-interacting proteins, syntenin is associated with interleukin
5-mediated activation via its interaction with SOX4 in B cell devel-
opment (47). In addition, Pan et al. (48) demonstrated that human
ubiquitin-conjugating enzyme 9, which regulates Bcl-2 expression,
interacts with the HMG domain of SOX4 to repress SOX4 transcrip-
tional activity. Moreover, SOX4 forms complexes with the predicted
binding motifs of at least 31 unidentified target genes (25), and among
the identified SOX4 target genes, semaphorin (SEMA3C) and neuro-
pilin-1 (NRP1) play important roles in tumorigenesis or tumor
progression (51,52). Various SOX4 target genes may to be associated
with tumorigenesis or tumor progression in other cancer types; how-
ever, only their role in HCC progression is known (25). This identi-
fication of SOX4-interacting proteins and SOX4 target genes in
HCC may provide clues to the identities and roles of SOX4-binding
partners in other cancer types.
The depletion of SOX4 messenger RNAvia small interfering RNA
can induce apoptosis in cancer cells and can regulate p53 stability
(21,30), suggesting that SOX4 may regulate p53-mediated apoptosis,
as indicated by the interaction between SOX4 and p53 proteins dem-
onstrated in the present study.
In luciferase reporter assays using the p53-responsive promoters
of Bax, Mdm2 and p21waf1, we demonstrated that overexpressed
SOX4 protein regulated p53-mediated transcription from the Bax
promoter. The heterogeneity of p53-mediated transcription activity
on p53-responsive genes (53) indicates that individual pathways are
transactivated via different p53 target promoters at different
sequence-specific DNA-binding motifs to accomplish the various
biological functions of p53. Thus, our results suggest that p53-
mediated transcriptional regulation of Bax and apoptosis could
be further modulated by SOX4–p53 interaction.
In the present study, we found a significant reduction in ionizing
radiation-induced apoptosis in the cells that had been transfected with
was accompanied by a reduction of Bax expression in the c-irradiated
SOX4-expressingcells(Figure 5C). These results suggest that the mod-
ulation of p53 transcriptional activity by SOX4 may be responsible for
addition, decreased apoptosis in the SOX4-expressing cells following
irradiation indicates that SOX4 expression in any cancer cells may be
related to radioresistance in a clinical setting. Although little is known
about the biological functions of SOX4 in apoptosis or tumorigenesis,
on the protein that dimerizes with SOX4 and the target gene promoter
that is affected. Additional studies are required to elucidate the molec-
ular mechanisms underlying the many possible functions of SOX4.
In conclusion, our data indicate that the overexpression of SOX4
protein is closely associated with hepatocarcinogenesis in human
HCC. Thus, the positivity or negativity in SOX4 expression in surgi-
cally excised HCC tissues might help to predict prognosis such as
disease-free survival or overall survival. Notably, our findings suggest
that SOX4 interacts with p53 and that this association in turn modu-
lates p53-mediated transcription at the Bax promoter, leading to the
inhibition of apoptosis via the suppression of Bax gene expression.
These results offer a novel mechanism of SOX4 in the p53 signaling
pathway that can be explored to identify new target drugs for HCC.
21C Frontier Functional Human Genome Project from the Ministry of
Science and Technology, Korea (FG-08-12-05).
Supplementary material can be found at http://carcin.oxfordjournals
Conflict of Interest Statement: None declared.
1.Lodato,F. et al. (2006) Hepatocellular carcinoma prevention: a worldwide
emergence between the opulence of developed countries and the economic
constraints of developing nations. World J. Gastroenterol., 12, 7239–7249.
2.Aravalli,R.N. et al. (2008) Molecular mechanisms of hepatocellular carci-
noma. Hepatology, 48, 2047–2063.
3.Wegner,M. (1999) From head to toes: the multiple facets of Sox proteins.
Nucleic Acids Res., 27, 1409–1420.
4.Smith,J.M. et al. (2004) The ins and outs of transcriptional control: nucle-
ocytoplasmic shuttling in development and disease. Trends Genet., 20, 4–8.
5.Gubbay,J. et al. (1990) A gene mapping to the sex-determining region of
the mouse Y chromosome is a member of a novel family of embryonically
expressed genes. Nature, 346, 245–250.
6.van de Wetering,M. et al. (1992) Sequence-specific interaction of the HMG
box proteins TCF-1 and SRYoccurs within the minor groove of a Watson-
Crick double helix. EMBO J., 11, 3039–3044.
7.Schepers,G.E. et al. (2002) Twenty pairs of sox: extent, homology, and
nomenclature of the mouse and human sox transcription factor gene fam-
ilies. Dev. Cell, 3, 167–170.
8.Wilson,M. et al. (2002) Matching SOX: partner proteins and co-factors of
the SOX family of transcriptional regulators. Curr. Opin. Genet. Dev., 12,
9.Kamachi,Y. et al. (2000) Pairing SOX off: with partners in the regulation of
embryonic development. Trends Genet., 16, 182–187.
10.Farr,C.J. et al. (1993) Characterization and mapping of the human SOX4
gene. Mamm. Genome, 4, 577–584.
11.van de Wetering,M. et al. (1993) Sox-4, an Sry-like HMG box protein, is
a transcriptional activator in lymphocytes. EMBO J., 12, 3847–3854.
12.Hur,E.H.et al. (2004)Functionalidentificationofthe pro-apoptotic effector
domain in human Sox4. Biochem. Biophys. Res. Commun., 325, 59–67.
13.van Houte,L.P. et al. (1995) Solution structure of the sequence-specific
HMGbox ofthe lymphocyte transcriptional activatorSox-4.J. Biol. Chem.,
14.Schilham,M.W. et al. (1997) Sox-4 facilitates thymocyte differentiation.
Eur. J. Immunol., 27, 1292–1295.
15.Schilham,M.W. et al. (1996) Defects in cardiac outflow tract formation and
pro-B-lymphocyte expansion in mice lacking Sox-4. Nature, 380, 711–714.
16.Busslinger,M. (2004) Transcriptional control of early B cell development.
Annu. Rev. Immunol., 22, 55–79.
17.Aaboe,M. et al. (2006) SOX4 expression in bladder carcinoma: clinical
aspects and in vitro functional characterization. Cancer Res., 66, 3434–
18.Ahn,S.G. et al. (1999) Identification of cDNAs for Sox-4, an HMG-Box
protein, and a novel human homolog of yeast splicing factor SSF-1 differ-
entially regulated during apoptosis induced by prostaglandin A2/delta12-
PGJ2 in Hep3B cells. Biochem. Biophys. Res. Commun., 260, 216–221.
19.Ahn,S.G. et al. (2002) Sox-4 is a positive regulator of Hep3B and HepG2
cells’ apoptosis induced by prostaglandin (PG)A(2) and delta(12)-PGJ(2).
Exp. Mol. Med., 34, 243–249.
20.Lee,C.J. et al. (2002) Differential expression of SOX4 and SOX11 in
medulloblastoma. J. Neurooncol., 57, 201–214.
21.Liu,P. et al. (2006) Sex-determining region Y box 4 is a transforming
oncogene in human prostate cancer cells. Cancer Res., 66, 4011–4019.
22.McCracken,S.et al. (1997)Analternativepathwayforexpressionofp56lck
from type I promoter transcripts in colon carcinoma. Oncogene, 15,
23.McGowan,E.M. et al. (1999) Effect of overexpression of progesterone
receptor A on endogenous progestin-sensitive endpoints in breast cancer
cells. Mol. Endocrinol., 13, 1657–1671.
W.Hur et al.
by guest on November 5, 2015
24.Hunt,S.M. et al. (1999) Expression and hormonal regulation of the Download full-text
Sox4 gene in mouse female reproductive tissues. Biol. Reprod., 61,
25.Liao,Y.L. et al. (2008) Identification of SOX4 target genes using phyloge-
netic footprinting-based prediction from expression microarrays suggests
that overexpression of SOX4 potentiates metastasis in hepatocellular
carcinoma. Oncogene., 27, 5578–5589.
26.Haupt,Y. et al. (1995) Induction of apoptosis in HeLa cells by trans-
activation-deficient p53. Genes Dev., 9, 2170–2183.
27.Fridman,J.S. et al. (2003) Control of apoptosis by p53. Oncogene, 22,
28.Vogelstein,B. et al. (2004) Cancer genes and the pathways they control.
Nat. Med., 10, 789–799.
29.Greenblatt,M.S. et al. (1994) Mutations in the p53 tumor suppressor gene:
clues to cancer etiology and molecular pathogenesis. Cancer Res., 54,
30.Pan,X. et al. (2009) Induction of SOX4 by DNA damage is critical for p53
stabilization and function. Proc. Natl Acad. Sci. USA, 106, 3788–3793.
31.Liver Cancer Study Group of Japan. (1992) The General Rules for the
Clinical and Pathological Study of Primary Liver Cancer. Kanehara Press,
32.Poeta,M.L. et al. (2007) TP53 mutations and survival in squamous-cell
carcinoma of the head and neck. N. Engl. J. Med., 357, 2552–2561.
33.Killian,A. et al. (2007) A simple method for the routine detection of
somatic quantitative genetic alterations in colorectal cancer. Gastroenter-
ology, 132, 645–653.
34.Ryu,M.H. et al. (2007) Prognostic significance of p53 gene mutations and
protein overexpression in localized gastrointestinal stromal tumours.
Histopathology, 51, 379–389.
35.Roos,W.P. et al. (2006) DNA damage-induced cell death by apoptosis.
Trends Mol. Med., 12, 440–450.
36.Fevre-Montange,M. et al. (2006) Microarray analysis reveals differential
gene expression patterns in tumors of the pineal region. J. Neuropathol.
Exp. Neurol., 65, 675–684.
37.Bangur,C.S. et al. (2002) Identification of genes over-expressed in
small cell lung carcinoma using suppression subtractive hybridi-
zation and cDNA microarray expression analysis. Oncogene, 21,
38.Haram,K.M. et al. (2008) Geneexpression profile of mouse prostate tumors
reveals dysregulations in major biological processes and identifies potential
murine targets for preclinical development of human prostate cancer
therapy. Prostate, 68, 1517–1530.
39.Andersen,C.L. et al. (2009) Dysregulation of the transcription factors
SOX4, CBFB and SMARCC1correlates with outcome of colorectalcancer.
Br. J. Cancer, 100, 511–523.
40.Katiyar,S. et al. (2000) P53 tumor suppressor gene mutations in hepatocel-
lular carcinoma patients in India. Cancer, 88, 1565–1573.
41.Bressac,B. et al. (1991) Selective G to T mutations of p53 gene in hepato-
cellular carcinoma from southern Africa. Nature, 350, 429–431.
42.Ghebranious,N. et al. (1998) Hepatitis B injury, male gender, aflatoxin, and
p53 expression each contribute to hepatocarcinogenesis in transgenic mice.
Hepatology, 27, 383–391.
43.Teramoto,T. et al. (1994) P53 gene abnormalities are closely related to
hepatoviral infections and occur at a late stage of hepatocarcinogenesis.
Cancer Res., 54, 231–235.
44.Hsu,I.C. et al. (1991) Mutational hotspot in the p53 gene in human
hepatocellular carcinomas. Nature, 350, 427–428.
45.Hosono,S. et al. (1993) Infrequent mutation of p53 gene in hepatitis B virus
positive primary hepatocellular carcinomas. Oncogene, 8, 491–496.
46.Horn,H.F. et al. (2007) Coping with stress: multiple ways to activate p53.
Oncogene, 26, 1306–1316.
47.Geijsen,N. et al. (2001) Cytokine-specific transcriptional regulation
through an IL-5Ralpha interacting protein. Science, 293, 1136–1138.
48.Pan,X. et al. (2006) Ubc9 interacts with SOX4 and represses its transcrip-
tional activity. Biochem. Biophys. Res. Commun., 344, 727–734.
49.Wotton,D. et al. (1995) The high mobility group transcription factor,
SOX4, transactivates the human CD2 enhancer. J. Biol. Chem., 270,
50.Wissmuller,S. et al. (2006) The high-mobility-group domain of Sox pro-
teins interacts with DNA-binding domains of many transcription factors.
Nucleic Acids Res., 34, 1735–1744.
51.Bielenberg,D.R. et al. (2006) Neuropilins in neoplasms: expression, regu-
lation, and function. Exp. Cell Res., 312, 584–593.
52.Herman,J.G. et al. (2007) Increased class 3 semaphorin expression modu-
lates the invasive and adhesive properties of prostate cancer cells. Int.
J. Oncol., 30, 1231–1238.
53.Forrester,K. et al. (1995) Effects of p53 mutantson wild-typep53-mediated
transactivation are cell type dependent. Oncogene, 10, 2103–2111.
Received November 11, 2009; revised March 10, 2010; accepted April 2, 2010
Role of SOX4 overexpression in HCC
by guest on November 5, 2015