Anatomic Pathology / HYPERMETHYLATION OF ZHX2 IN HCC
Promoter Hypermethylation of a Novel Gene, ZHX2,
in Hepatocellular Carcinoma
Zili Lv, PhD,1,2 Meng Zhang, MMed,1 Jiong Bi, PhD,1 Fangping Xu, PhD,1 Shaowei Hu, PhD,1
and Jianming Wen, MD, PhD1
Key Words: Liver neoplasm; Promoter methylation; Methylation-sensitive restriction fingerprinting; Zinc fingers and homeobox protein 2;
α-Fetoprotein
DOI: 10.1309/09B452V7R76K7D6K
Abstract
We used methylation-sensitive restriction
fingerprinting (MSRF) to identify novel CpG (5'-CG-3'
palindrome; p, phosphate group)–rich sequences that
are methylated differentially between the hepatocellular
carcinoma (HCC) genomes and adjacent nontumorous
liver tissues. A 199-base-pair sequence methylated in
HCC tumor tissue was isolated and showed high
homology to the 5'-CpG island of the zinc fingers and
homeoboxes protein 2 (ZHX2) gene. By using bisulfite
sequencing, we confirmed that hypermethylation of the
5'-CpG island of ZHX2 occurred in some HCC and
HepG2 cell lines but not in 6 normal liver tissue
samples. By using methylation-specific polymerase
chain reaction, we detected methylation of the 5'-CpG
island of ZHX2 in 46.9% of the HCCs. Reverse
transcription–polymerase chain reaction demonstrated
that ZHX2 messenger RNA (mRNA) was expressed in
all 6 normal liver tissue samples but in only 13.3% of
the methylated HCCs. Treatment of HepG2 with 5-azadeoxycytidine could demethylate the promoter and
increase ZHX2 mRNA expression. These results suggest
that hypermethylation-mediated silencing of ZHX2 is
an epigenetic event involved in HCC.
740
740
Am J Clin Pathol 2006;125:740-746
DOI: 10.1309/09B452V7R76K7D6K
Hepatocellular carcinoma (HCC) is an aggressive malignancy with a poor prognosis. It is the fifth most common cancer in the world and the third most common cause of cancerrelated death.1 The development of HCC is a chronic process
that involves multiple genetic alternations. One unique feature
of the cancer is its close association with a high level of α-fetoprotein (AFP) in the serum and with hepatitis B virus infection.2 Recent advances in molecular genetics indicate that activation of oncogenes or inactivation of tumor suppressor genes
might have an important role in HCC tumorigenesis.3,4
Many studies demonstrated that abnormal changes in DNA
methylation (including methylation content, level, and pattern)
lead to the inactivation of some tumor suppressor genes, such as
p16, p15, p14ARF, and RASSF,5-8 which are involved in HCC
carcinogenesis. These changes usually occur in 5'-CpG (5'-CG3' palindrome; p, phosphate group) dinucleotides that are
clustered frequently in regions approximately 1 to 2 kb long,
called 5'-CpG islands, in or near the promoter and the first
exon region of genes.9 Aberrant hypermethylation may contribute to tumorigenesis through C-to-T mutation at the methylated CpG dinucleotides. This epigenetic alteration frequently
is associated with transcriptional silencing of their cognate
gene.10 Therefore, identification of methylated CpG
sequences in HCC genomes could lead to the discovery of a
cancer-related gene.
In this study, we used a technique known as methylationsensitive restriction fingerprinting (MSRF)11 to screen for specific methylation sequences in HCC tissue samples. Initially,
genomic DNA is digested with a 4-base restriction endonuclease that cuts bulk DNA into small fragments but that rarely
cuts into the CG-rich regions. The digested DNA is treated
with a second restriction endonuclease, which discriminates
© American Society for Clinical Pathology
Anatomic Pathology / ORIGINAL ARTICLE
between methylated and unmethylated CpG sites, and then is
amplified by polymerase chain reaction (PCR) with short arbitrary primers (10 mers) at low stringency. The aberrant methylated sequences were detected in the amplified HCC tumor
DNA relative to the amplified adjacent nontumorous liver tissue DNA of the same patient by high-resolution gel electrophoresis. One aberrant methylated sequence was identified
and showed high homology to the promoter of the zinc fingers
and homeoboxes protein 2 (ZHX2) gene, which acts as a novel
transcriptional repressor.12,13 By using bisulfite sequencing
and methylation-specific PCR (MSP), we confirmed the
methylation of the promoter ZHX2 gene. We also studied the
relationship between the expression of messenger RNA
(mRNA) and the methylation of the ZHX2 promoter. These
results indicate that hypermethylation-mediated silencing of
ZHX2 may take part in the progression of HCC.
Materials and Methods
Tissue Samples
HCC and corresponding adjacent nontumorous liver tissue samples were obtained from 32 patients (28 men and 4
women) who underwent hepatectomy at the First Affiliated
Hospital, Sun Yat-sen University, Guangzhou, China. Clinical
data were obtained from medical records. For 26 patients, the
AFP serum level was 25 ng/mL (25 µg/L) or more. The mean
± SD age was 46.03 ± 12.33 years. Normal liver tissue samples were from 6 patients (5 men and 1 woman) with a cavernous hemangioma of the liver or cholangitis. None of them
had an AFP serum level of 25 ng/mL (25 µg/L) or more. All
of the tissue samples obtained were immediately frozen at
–80°C. All patients had given informed consent for their participation, and the ethics committee approved the study. All
tumor tissues were classified by histologic examination and
graded according to Edmondson-Steiner.14 The 32 tumor tissue samples were of 19 well-differentiated (grades I and II)
and 13 poorly differentiated (grades III and IV) tumors. The
tumor cell content in the tissue samples was estimated by histologic examination to be approximately 70% to 80%.
Cell Culture and 5-Aza-deoxycytidine Treatment
Human hepatocellular carcinoma cell line HepG2 cells
were incubated in fresh culture medium, RPMI 1640 (Gibco
BRL, Grand Island, NY), supplemented with 10% heatedinactivated fetal bovine serum (Hyclone, South Logan, UT), 2
mmol/L L-glutamine, 1 × 105 U/L of penicillin, and 1 × 105
µg/L of streptomycin (Promega, Madison, WI) at 37°C in a
humidified atmosphere containing 5% carbon dioxide. For
demethylation experiments, cells were plated at a density of
2.0 × 105 cells per 100-mm dish and cultured for 24 hours.
Cells then were treated with 0.5, 1, or 5 µmol/L of 5-azadeoxycytidine (5-Aza-CdR, Sigma, St Louis, MO) for 9 days
and harvested for DNA and RNA extraction.
DNA Extraction and MSRF
High-molecular-weight DNA was extracted using the
DNeasy tissue kit (Omega, Atlanta, GA) and quantified by spectrophotometer. To identify novel CpG-rich DNA sequences that
are methylated differentially between HCC and adjacent nontumorous liver tissue samples, we performed MSRF using
genomic DNA isolated from HCC and adjacent nontumorous
liver tissue samples from the same patient. For additional control
samples, we included genomic DNA from 6 normal liver tissue
samples for comparison. Genomic DNA isolated from different
tissue samples was subjected to MseI restriction digestion.
Because the restriction site of MseI is TTAA, which is rare in CGrich regions, genomic DNA was digested into small fragments,
but the integrity of the CpG islands was preserved. The MseIdigested DNA then was subjected to digestion with BstUI, a
methylation-sensitive endonuclease. This endonuclease was chosen because its recognition sequence (CGCG) occurs frequently
within CpG islands but is rare in bulk DNA. The methylated CpG
islands protected from the BstUI digestion then were PCR-amplified with short arbitrary primers, whereas sequences with unmethylated BstUI sites were digested and yielded no PCR products.
Genomic DNA was digested with MseI alone or with MseI
and BstUI at 10 U/µg of DNA according to the conditions recommended by the supplier (New England Biolabs, Beverly,
MA). The digested DNA (100-300 ng) was added to the PCR
reaction reagent in a total volume of 25 µL containing 0.4
µmol/L of arbitrary primers (Bs1, 5'-AGCGGCCGCG-3'; Bs18,
5'-ACCCCACCCG-3', Sangon, Shanghai, China) and 0.75 U of
Taq DNA polymerase (Promega). Initial denaturation was at
94°C for 5 minutes, and the DNA then was subjected to 35
cycles of amplification consisting of 2 minutes of denaturation at
94°C, 1 minute of annealing at 40°C, and 2 minutes of extension
at 72°C and a final extension step at 72°C for 10 minutes. The
products of each amplification reaction were size-fractionated on
8% nondenaturing polyacrylamide gel (Sangon) electrophoresis
for 3 hours at 10 V/cm. The bands were visualized by silver staining (Sangon). The aberrant methylated-DNA was eluted from the
gel using an elution kit (Poly-Gel DNA Extraction Kit, Omega),
ligated directly into the pGEM-T Easy vector (Promega), and
transformed into DH5α. Six to eight white colonies were selected and sequenced (Bioasia, Shanghai, China). The resulting
nucleotide sequence was compared with GenBank using the
BLAST program (http://www.ncbi.nlm.nih.gov/BLAST).
Bisulfite Sequencing
Genomic DNA was obtained from fresh HCC, adjacent
noncancerous tissue samples, normal liver tissue samples, and
HepG2 cells with or without treatment with 5-Aza-CdR. All
Am J Clin Pathol 2006;125:740-746
© American Society for Clinical Pathology
741
DOI: 10.1309/09B452V7R76K7D6K
741
741
Lv et al / HYPERMETHYLATION OF ZHX2 IN HCC
the genomic DNA was modified with sodium bisulfite according to Herman et al.15 DNA (2 µg in a volume of 50 µL) was
denatured by sodium hydroxide (final concentration, 0.2
mol/L) for 10 minutes at 37°C. Then 30 µL of a 10-mmol/L
concentration of hydroquinone (Sigma) and 520 µL of a 3mol/L concentration of sodium bisulfite (Sigma), both freshly
prepared (pH 5.0), was added and mixed.
The mixtures were incubated under mineral oil at 50°C for
16 hours. Modified DNA was purified using the Wizard DNA
purification resin (Promega) according to the manufacturer’s
instructions and then eluted into 50 µL of Milli-Q water
(Millipore, Billerica, MA). Modification was completed by
treatment with sodium hydroxide (final concentration, 0.3
mol/L) for 5 minutes at room temperature, followed by ethanol
precipitation. DNA was resuspended in 20 µL of Milli Q water
and used immediately or stored at –20°C. A 226-base-pair (bp)
sequence including 17 CpG islands of the ZHX2 gene promoter was amplified by the forward primer, 5'-TTATTATTTTTTTGAGTTTGAAGT- 3' (–342 nucleotides), and the
reverse primer 5'-ATTTTCTTTATAAAAAATTCAAAATATCAC-3' (–567 nucleotides) under the following conditions: after
an initial 10-minute preincubation step at 95°C, 35 amplification cycles were run, each consisting of 95°C for 30 seconds,
52°C for 45 seconds, and 72°C for 45 seconds. The PCR product was extracted from the polyacrylamide gel with the appropriate kit (Poly-Gel DNA Extraction Kit) and subcloned into the
pGEM-T Easy vector. To determine the methylation status of
the CpG islands in the ZHX2 gene, 8 to 10 clones were picked
for sequencing (Bioasia).
Methylation-Specific PCR
The methylation status of the ZHX2 promoter was detected in 32 HCCs, adjacent nontumorous liver tissue samples, and
HepG2 cells with or without treatment with 5-Aza-CdR by
MSP according to Herman et al.15 A 115-bp containing 12 CpG
islands was amplified with a pair of methylation-specific or
nonmethylation-specific primer sets. The methylation-specific
forward primer, 5'-CGTATAGTTTTACGTAAAGGGTTTCTGT-3' (–414 nucleotides); the reverse primer, 5'-AACAAACTATTAATCTTAACCACGTA-3' (–528 nucleotides); the unmethylation-specific forward primer, 5'-TTTATGTAAAGGGTTTTGG3' (–413 nucleotides), and the reverse primer, 5'-AACAAACTATTAATCTTAACCACATA-3' (–527 nucleotides), were amplified under the following conditions: 95°C for 30 seconds, 58°C for
30 seconds, and 72°C for 30 seconds.
The PCR reaction reagent contained 1× PCR buffer (16.6
mmol/L of ammonium sulfate; 67 mmol/L of tris(hydroxymethyl)aminomethane, pH 8.8; 6.7 mmol/L of magnesium
chloride; and 10 mmol/L of 2-mercaptoethanol), deoxynucleoside triphosphates (1.25 mmol/L), primers (300 ng), and modified
DNA, 100 ng, in a final volume of 50 µL. Reactions were hot-started at 95°C for 5 minutes before adding 1.25 U of Taq polymerase
742
742
Am J Clin Pathol 2006;125:740-746
DOI: 10.1309/09B452V7R76K7D6K
(Promega). Control experiments without DNA were performed
for each set of PCR. Placental DNA, treated with SssI methyltransferase and subsequently treated with sodium bisulfite, was
used as a positive control for methylated alleles, and DNA from
normal lymphocytes treated with sodium bisulfite was used as a
positive control for unmethylated alleles. The PCR product was
loaded directly onto nondenaturing 8% polyacrylamide gels.
After electrophoresis the gel was stained with silver nitrate.
Reverse Transcription–Polymerase Chain Reaction
Total RNA was isolated from HCC, adjacent noncancerous liver, and normal liver tissue samples and HepG2 cells with
Trizol (Invitrogen, Carlsbad, CA) according to the protocol
provided by the manufacturer. After washing twice with 70%
ethanol, the RNA was dissolved in diethylpyrocarbonate-treated water. The quantity and quality of the RNA samples were
measured carefully by spectrophotometry and electrophoresis.
Four micrograms of total RNA was reverse transcribed
with MuLV reverse transcriptase (Fermentas, Hanover, MD) at
42°C for 1 hour, and the reaction was terminated by incubation
at 75°C for 10 minutes. Of the 20 µL of resulting complementary DNA, 1 µL was used in each PCR. Intron-spanning primers
were designed with the Primer3 output program
(http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi;
Whitehead Institute, Cambridge, MA) and identified by the
BLAST program. The primers used to amplify ZHX2
(NM_014943) mRNA from complementary DNA were forward, 5'-GGTAGCGACGAGAACGAG-3', and reverse, 5'AGGACTTTGGCACTATGAAC- 3'. PCR was performed in a
25-µL reaction volume for 35 cycles under the following conditions: initial denaturation at 95°C for 5 minutes; then 35 amplification cycles were run, consisting of 95°C for 30 seconds,
58°C for 45 seconds; and 72°C for 45 seconds; and a final
extension at 72°C for 7 minutes. The product was 389 bp.
Glyceraldehyde-3-phosphate dehydrogenase was used as an
internal control; the forward primer was 5'-TGAGTACGTCGTGGAGTCCA-3' and the reverse primer, 5'CAAAGTTGTCATGGATGACC-3'. The product was 230 bp.
Statistical Analysis
Results were analyzed by using the χ2 test to compare the
differences between the groups using SPSS 10.0 software
(SPSS, Chicago, IL). A P value of less than .05 was considered statistically significant.
Results
MSRF Screening
From the MSRF screening, we identified 3 abnormally
methylated sequences in 32 cases of HCC by using a pair of
primers. Only 1 of the abnormally methylated sequences
belonged to the promoter region of a novel gene. (The other 2
© American Society for Clinical Pathology
Anatomic Pathology / ORIGINAL ARTICLE
sequences are not described herein.) This gene was detected positively in 3 of 32 HCC tissue samples, accounting for 9.4% of the
cases. The aberrantly methylated DNA fragment (199 bp) is shown
in ❚Image 1❚ and subsequently was cloned and sequenced. By the
BLAST program, this isolated fragment showed high homologies
to from –601 nucleotides to –404 nucleotides in the 5' upstream
region of the ZHX2 gene (gene ID, 22882). Computational sequence
analyses showed that this sequence has a GC content of 64%.
Methylation of the ZHX2 Gene
To study the methylation status of the 5'-CpG islands of the
ZHX2 gene in HCC, we performed bisulfite sequencing on 6
normal liver tissue samples, the HepG2 cell line, and the 3 HCC
tissue samples in which aberrant methylation of the ZHX2 gene
was positively detected by MSRF screening. Multiple methylation of CpG sites was shown in the sequences from all 3 HCC
M
+
–
+
–
+
–
+
+
+
+
Mse I
Bst U I
+
+
200 bp
100 bp
A
B
M
U
20T
M
M
U
U
M
27T
M
C
2C
1C
U
❚Image 1❚ Methylation-sensitive restriction fingerprinting
(MSRF) screening. MSRF analysis of genomic DNA isolated
from hepatocellular (HCC) tumor tissue (T), nontumorous liver
tissue (NT), and normal liver tissue (control [C]). MseI, DNA
was digested with MseI. BstUI, DNA was digested with
BstUI. A DNA fragment (199-base-pair [bp]) was amplified by
using an arbitrary primer in (MseI + BstUI)-digested HCCs but
not in double-digested nontumorous liver tissue or normal
liver tissue, indicating that the fragment is methylated at a
BstUI site. M, marker, 100-bp ladder.
7T
3T
U
tissue samples and the HepG2 cell line. CG is protected from
bisulfite modification because of methylation in the CpG
islands, but in 6 normal liver tissue samples, CG was changed
to TG by bisulfite modification because of unmethylation in
the CpG islands.
Of the 32 HCCs, 15 (46.9%) were positively amplified with
methylation-specific primers in MSP ❚Image 2A❚. By
histopathologic grade, 9 (47.4%) of 19 grade I or II and 6
(46.2%) of 13 grade III and IV tumors were positive for amplification (P = .9) ❚Table 1❚. Only 5 adjacent nontumorous liver
tissue samples (15.6%) showed partial methylation in the
sequence (P = .007; Table 1). No methylation was detected in 6
normal liver tissue samples ❚Image 2B❚. HepG2 cells also
showed partial hypermethylation in the ZHX2 promoter, but after
treatment with 5-Aza-CdR, the methylation could not be detected ❚Image 2C❚.
U
M
Untreated
Treated
U
M
U
M
❚Image 2❚ Methylation-specific polymerase chain reaction of the ZHX2 promoter. A, Hepatocellular carcinoma (HCC) tumor
tissue DNA from cases 3, 7, 20, and 27 (3T, 7T, 20T, and 27T). The HCC DNA can be amplified with methylation-specific and
unmethylation-specific primers. B, Normal liver tissue from cases 1 and 2 (1C and 2C). The normal liver DNA can be amplified
only with unmethylation-specific primers. C, HepG2 cells treated and untreated with 5-aza-deoxycytidine (5-Aza-CdR). M,
reaction specific for methylated DNA; U, reaction specific for unmethylated DNA.
Am J Clin Pathol 2006;125:740-746
© American Society for Clinical Pathology
743
DOI: 10.1309/09B452V7R76K7D6K
743
743
Lv et al / HYPERMETHYLATION OF ZHX2 IN HCC
Expression of ZHX2 mRNA
All 6 normal liver tissue samples expressed ZHX2
mRNA ❚Image 3A❚. The expression of ZHX2 mRNA was
higher in adjacent nontumorous liver tissue samples
(62.5%) than in HCC tissue samples (34.4%; P = .02;
Table 1) ❚Image 3B❚. The 11 HCCs with positive ZHX2
mRNA included 7 grade I or II and 4 grade III or IV
tumors (P = .7; Table 1). Furthermore, the expression of
ZHX2 mRNA was higher in HCCs associated with a concentration of AFP in the serum of less than 25 ng/mL (<25
µg/L; 83.3%) than in HCCs with an AFP of 25 ng/mL or
more (≥25 µg/L; 23.1%; P = .005; Table 1). Of the 15
HCCs with methylation in the promoter of the ZHX2 gene,
2 (13.3%) showed positive transcripts for the ZHX2 gene
at a significantly lower level than in HCCs without methylation (P = .047) ❚Table 2❚.
Demethylation by 5-Aza-CdR
The HepG2 cell line showed only trace amounts of
ZHX2 transcript expression. After treatment with 0.5, 1, or 5
µmol/L of 5-Aza-CdR for 9 days, the expression of ZHX2
mRNA increased ❚Image 4❚.
Discussion
DNA methylation in promoter areas is estimated to have
a crucial role in the control of gene expression and chromosome structure in mammalian cells.16 The methylation of
CpG dinucleotide sequences is a common occurrence, and
roughly 70% of CpG dinucleotides in mammalian genomes
seem to be methylated. It is reported that 40% to 60% of
human genes have CpG-rich sequences in the 5' regulatory
regions, called 5'-CpG islands, which normally remain
unmethylated. The methylation of CpG islands is deeply
associated with inactivation of the X-chromosome in females
and genomic imprinting.17,18 Moreover, hypermethylation of
5'-CpG islands in HCC is associated with transcriptional
silencing of many genes, such as E-cadherin, collagen type I
alpha 2 (COLIA2), insulin-like growth factor binding protein
2 (IGFBP2), connective tissue growth factor (CTGF), and
fibronectin 1,19 hsMAD2,20 HDPR1,21 NADPH,22 RASSF1A,23
RUNX3,24 and GSTP1.25 These genes regulate liver cell
growth, apoptosis, migration, and cell motility and are involved
in HCC carcinogenesis.
MSRF was first described by Huang et al.11 This technique has been used to screen aberrant methylation in
❚Table 1❚
Methylation of ZHX2 Promoter and RT-PCR of ZHX2 mRNA in HCC and Nontumorous Liver Tissue*
Methylation+
Tissue type
Nontumorous (n = 32)
HCC (n = 32)
AFP in serum, ng/mL (µg/L)
≥25 (≥25) (n = 26)
<25 (<25) (n = 6)
Differentiation
Well-differentiated (n = 19)
Poorly differentiated (n = 13)
χ2
7.27
P
χ2
mRNA+
.007
5 (15.6)
15 (46.9)
P
5.07
.02
7.84
.005
0.13
.7
20 (62.5)
11 (34.4)
4.40
.01
15 (57.7)
0 (0.0)
6 (23.1)
5 (83.3)
0.01
.9
9 (47.4)
6 (46.2)
7 (36.8)
4 (30.8)
AFP, α-fetoprotein; HCC, hepatocellular carcinoma; mRNA, messenger RNA; RT-PCR, reverse transcription–polymerase chain reaction.
* Data are given as number (percentage).
A
B
3
M
M
1C
2C
3C
4C
5C
NT
7
T
NT
20
T
NT
27
T
NT
T
6C
ZHX2 (389 bp)
ZHX2 (389 bp)
GAPDH (230 bp)
❚Image 3❚ Reverse transcription–polymerase chain reaction of ZHX2. A, Normal liver tissue (C) can express ZHX2 messenger
RNA (mRNA; 389 base pairs [bp]). B, ZHX2 mRNA is detected in adjacent nontumorous liver tissue (NT) but not in methylated
hepatocellular carcinoma tumor tissue (T). M, 100-base-pair DNA ladder.
744
744
Am J Clin Pathol 2006;125:740-746
DOI: 10.1309/09B452V7R76K7D6K
© American Society for Clinical Pathology
Anatomic Pathology / ORIGINAL ARTICLE
nasopharyngeal carcinoma and prostate carcinoma26,27 and is
considered a rapid and efficient method to screen changes of
methylation in genomic DNA. To identify CpG-rich methylated DNA sequences that are expressed differentially between
HCC and adjacent noncancerous liver tissues, we carried out
comparative MSRF using genomic DNA from HCCs, corresponding normal tissue samples, and additional normal tissue
samples from patients with hemangioma or cholangitis.
The majority of amplified products showed no methylation or normal methylation. Only 3 DNA fragments were
detected in the MseI/BstUI-digested HCC DNA, but not in the
corresponding adjacent noncancerous tissue DNA or in normal liver tissue DNA. By using the BLAST program, only 1
DNA fragment was confirmed to be part of the promoter
region of the novel gene ZHX2, suggesting that the promoter
of the ZHX2 gene might be hypermethylated in HCC.
By studying the methylation status of the promoter of the
ZHX2 gene by bisulfite sequencing, we found that the 3 HCCs
that exhibited hypermethylation in the ZHX2 gene promoter
by MSRF contained methylated CpG islands in the promoter
of the ZHX2 gene. By using MSP, we observed a high frequency (46.9%) of promoter methylation of the ZHX2 gene in
HCC tissue samples, which was higher than in adjacent nontumorous liver tissue samples, but there was no methylation in
normal liver tissue samples. Furthermore, the HCC cell line,
HepG2, also showed methylation in the promoter of the ZHX2
gene, but the promoter could be demethylated by 5-Aza-CdR,
suggesting that promoter methylation is related to HCC. It is
interesting that ZHX2 promoter methylation already was present in low-grade HCCs and some adjacent nontumorous liver
tissue samples, suggesting that this molecular abnormality
might be a relatively early step in the development of HCC.
The expression of ZHX2 mRNA in HCC was decreased in
HCC compared with adjacent noncancerous tissue samples, but
there was no significant difference in the expression between
low- and high-grade HCCs, indicating that the negative expression of ZHX2 mRNA is related to the carcinogenesis of HCC
and might have no relationship to progression. In addition, only
13.3% of the transcripts of ZHX2 could be detected in HCCs
with methylated promoters of the ZHX2 gene. Therefore, one of
the mechanisms of ZHX2 gene silencing in HCC should be promoter hypermethylation. Hypermethylation of a promoter
region might interfere with transcription of the gene and even
hold back transcription. For further studies on methylation and
demethylation in HCC, we treated HepG2 cells with the
demethylation agent, 5-Aza-CdR. The expression of ZHX2
transcription was increased after treatment, indicating that 5Aza-CdR could remove ZHX2 gene hypermethylation in
HepG2 cells and, thereby, activate the gene. The role of ZHX2
in HCC carcinogenesis needs to be studied.
ZHX2, located in chromosome 8q24.13, is one of the members of the zinc fingers and homeobox family. It is a ubiquitous
❚Table 2❚
Relationship Between Methylation of ZHX2 Promoter and
Expression of mRNA*
Methylation
mRNA+
mRNA–
Positive
Negative
2
9
13
8
mRNA, messenger RNA.
* χ2 = 3.93; P = .047; r = –0.42.
0.5
µmol/L
0
µmol/L
1
µmol/L
5
µmol/L Marker
ZHX2 (389 bp)
300 bp
200 bp
GAPDH (230 bp)
❚Image 4❚ Reactivation of ZHX2 expression in HepG2 cells
after treatment with 5-aza-deoxycytidine (5-Aza-CdR). Cells
were treated with 0.5, 1, or 5 µmol/L of 5-Aza-CdR for 9 days.
Untreated and treated cells were used to extract total RNA
for reverse transcription–polymerase chain reaction analysis.
Glyceraldehyde-3-phosphate dehydrogenase was used as an
internal control.
transcriptional repressor. Recently, in vitro and in vivo proteinprotein interaction assays revealed that the ZHX family of proteins is involved in the expression of a number of nuclear factor-Y (NF-Y)–regulated genes via an organized transcription
network.12,13 In this study, we found that HCC tissue samples,
especially methylated HCCs, have low expression of ZHX2
mRNA. This implies that methylated HCCs lost the transcription repressor activity of ZHX2. Perhaps this results in the
uncontrolled proliferation of liver cells by the loss of regulation
of NF-Y–mediated genes, such as DEK,28 p53 and its homologues,29 and cyclin-dependent kinases.30,31 Thus, the low
expression of ZHX2 may be involved in HCC carcinogenesis
through the loss of regulation of these genes. Perhaps the ZHX2
gene may be regarded as a tumor suppressor.
Perincheri and colleagues32 recently confirmed that the
ZHX2 gene was responsible for the hereditary persistence of
AFP and H19. Although AFP is silenced in the adult liver, it
can be reactivated in HCC, explaining why a high serum AFP
level is an important clinical diagnostic indicator of HCC. In
the present study, we found that ZHX2 mRNA expression was
decreased in HCCs with a high serum AFP level, implying
that the silenced ZHX2 mRNA can reactivate the expression of
AFP in HCC. Whether ZHX2 can inhibit the expression of
AFP by interacting with NF-Y and p53 demands further study.
We found that the ZHX2 promoter is hypermethylated in
HCCs. ZHX2 mRNA expression was decreased significantly in
methylated HCCs, especially in those with high serum AFP
Am J Clin Pathol 2006;125:740-746
© American Society for Clinical Pathology
745
DOI: 10.1309/09B452V7R76K7D6K
745
745
Lv et al / HYPERMETHYLATION OF ZHX2 IN HCC
levels. The expression of ZHX2 mRNA in HepG2 cells could be
increased by demethylation with 5-Aza-CdR. These findings
indicate that promoter methylation is one of the causes of ZHX2
mRNA silencing in HCC and in HepG2 cells and is associated
with AFP expression. The negative expression of ZHX2 mRNA
might be related to early steps in HCC development.
From the Departments of Pathology, 1First Affiliated Hospital, Sun
Yat-sen University, Guangzhou; and 2First Affiliated Hospital,
Guangxi Medical University, Nanning, People’s Republic of China.
Address reprint requests to Dr Wen: Dept of Pathology, the
First Affiliated Hospital, Sun Yat-sen University, Guangzhou
510080, P.R. China.
Acknowledgment: We thank Lei Yi-Xiong, PhD, Institute of
Chemical Carcinogenesis, Guangzhou Medical College, for
practical and productive technical advice regarding the MSRF.
References
1. Llovet JM, Bruix J. Systematic review of randomized trials for
unresectable hepatocellular carcinoma: chemoembolization
improves survival. Hepatology. 2003;37:429-442.
2. Garcia VA, Castillo R. Asymptomatic advanced
hepatocellular carcinoma presenting with spinal cord
compression. Dig Dis Sci. 2005;50:308-311.
3. Fujii K, Kondo T, Yokoo H, et al. Proteomic study of human
hepatocellular carcinoma using two-dimensional difference gel
electrophoresis with saturation cysteine dye. Proteomics.
2005;5:1411-1422.
4. Chan KL, Guan XY, Ng IO. High-throughput tissue microarray
analysis of c-myc activation in chronic liver disease and
hepatocellular carcinoma. Hum Pathol. 2004;35:1324-1331.
5. Lee S, Lee HJ, Kim JH, et al. Aberrant CpG island
hypermethylation along multistep hepatocarcinogenesis. Am J
Pathol. 2003;163:1371-1378.
6. Liu L, Vo A, Mckeehan WL. Specificity of the methylationsuppressed A isoform of candidate tumor suppressor RASSF1
for microtubule hyperstabilization is determined by cell death
inducer C19ORF5. Cancer Res. 2005;65:1830-1838.
7. Anzola M, Cuevas N, Lopez-Martinez M, et al. P14ARF gene
alterations in human hepatocellular carcinoma. Eur J
Gastroenterol Hepatol. 2004;16:19-26.
8. Li X, Hui AM, Sun L, et al. p16INK4A hypermethylation is
associated with hepatitis virus infection, age, and gender in
hepatocellular carcinoma. Clin Cancer Res. 2004;10:7484-7489.
9. Heisler LE, Torti D, Boutros PC, et al. CpG island microarray
probe sequences derived from a physical library are
representative of CpG islands annotated on the human
genome. Nucleic Acids Res. 2005;33:2952-2961.
10. Jones PA, Baylin SB. The fundamental role of epigenetic
events in cancer. Nat Rev Genet. 2002;3:415-428.
11. Huang TH, Laux DE, Hamlin BC, et al. Identification of
DNA methylation markers for human breast carcinomas using
the methylation-sensitive restriction fingerprinting technique.
Cancer Res. 1997;57:1030-1034.
12. Kawata H, Yamada K, Shou Z, et al. The mouse zinc-fingers
and homeoboxes (ZHX) family; ZHX2 forms a heterodimer
with ZHX3. Gene. 2003;323:133-140.
13. Kawata H, Yamada K, Shou Z, et al. Zinc-fingers and homeoboxes
(ZHX) 2, a novel member of the ZHX family, function as a
transcriptional repressor. Biochem J. 2003;373:747-757.
746
746
Am J Clin Pathol 2006;125:740-746
DOI: 10.1309/09B452V7R76K7D6K
14. Edmondson HA. Tumor of the Liver and Intrahepatic Bile Ducts.
Washington, DC: Armed Forces of Institute of Pathology;
1958. Koss LG, ed. Atlas of Tumor Pathology. First series,
fascicle 25.
15. Herman JG, Graff JR, Myohanen S, et al. Methylation-specific
PCR: a novel PCR assay for methylation status of CpG
islands. Proc Natl Acad Sci U S A. 1996;93:9821-9826.
16. Jones PA, Takai D. The role of DNA methylation in
mammalian epigenetics. Science. 2001;293:1068-1070.
17. Riggs AD, Pfeifer GP. X-chromosome inactivation and cell
memory. Trends Genet. 1992;8:169-174.
18. Razin A, Cedar H. DNA methylation and genomic
imprinting. Cell. 1994;77:473-476.
19. Chiba T, Yokosuka O, Fukai K, et al. Identification and
investigation of methylated genes in hepatoma. Eur J Cancer.
2005;41:1185-1194.
20. Jeong SJ, Shin HJ, Kim SJ, et al. Transcriptional abnormality
of the hsMAD2 mitotic checkpoint gene is a potential link to
hepatocellular carcinogenesis. Cancer Res. 2004;64:8666-8673.
21. Yau TO, Chan CY, Chan KL, et al. HDPR1, a novel inhibitor
of the WNT/beta-catenin signaling, is frequently
downregulated in hepatocellular carcinoma: involvement of
methylation-mediated gene silencing. Oncogene.
2005;24:1607-1614.
22. Tada M, Yokosuka O, Fukai K, et al. Hypermethylation of
NAD(P)H: quinone oxidoreductase 1 (NQO1) gene in
human hepatocellular carcinoma. J Hepatol. 2005;42:511-519.
23. Yeo W, Wong N, Wong WL, et al. High frequency of promoter
hypermethylation of RASSF1A in tumor and plasma of
patients with hepatocellular carcinoma. Liver Int.
2005;25:266-272.
24. Mori T, Nomoto S, Koshikawa K, et al. Decreased expression
and frequent allelic inactivation of the RUNX3 gene at 1p36 in
human hepatocellular carcinoma. Liver Int. 2005;25:380-388.
25. Zhang YJ, Chen Y, Ahsan H, et al. Silencing of glutathione Stransferase P1 by promoter hypermethylation and its
relationship to environmental chemical carcinogens in
hepatocellular carcinoma. Cancer Lett. 2005;221:135-143.
26. Lo KW, Tsang YS, Kwong J, et al. Promoter hypermethylation
of the EDENRB gene in nasopharyngeal carcinoma. Int J
Cancer. 2002;98:651-655.
27. Wu M, Ho SM. PMP24, a gene identified by MSRF, undergoes
DNA hypermethylation–associated gene silencing during
cancer progression in an LNCaP model. Oncogene.
2004;23:250-259.
28. Sitwala KV, Adams K, Markovitz DM. YY1 and NF-Y binding
sites regulate the transcriptional activity of the dek and dekcan promoter. Oncogene. 2002;21:8862-8870.
29. Innocente SA, Lee JM. p53 is a NF-Y– and p21-independent,
Sp1-dependent repressor of cyclin B1 transcription. FEBS Lett.
2005;579:1001-1007.
30. Chae HD, Yun J, Bang YJ, et al. Cdk2-dependent
phosphorylation of the NF-Y transcription factor is essential
for the expression of the cell cycle–regulatory genes and cell
cycle G1/S. Oncogene. 2004;23:4084-4088.
31. Manni I, Mazzaro G, Gurtner A, et al. NF-Y mediates the
transcriptional inhibition of the cyclin B1, cyclin B2, and
cdc25C promoters upon induced G2 arrest. J Biol Chem.
2001;276:5570-5576.
32. Perincheri S, Dingle RW, Peterson ML, et al. Hereditary
persistence of alpha-fetoprotein and H19 expression in liver of
BALB/cJ mice is due to a retrovirus insertion in the ZHX2
gene. Proc Natl Acad Sci U S A. 2005;102:396-401.
© American Society for Clinical Pathology