Download Epigenetic mechanisms regulate placental c-myc

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Extrachromosomal DNA wikipedia , lookup

Non-coding DNA wikipedia , lookup

Gene expression programming wikipedia , lookup

Nucleosome wikipedia , lookup

Designer baby wikipedia , lookup

X-inactivation wikipedia , lookup

No-SCAR (Scarless Cas9 Assisted Recombineering) Genome Editing wikipedia , lookup

DNA vaccination wikipedia , lookup

Gene therapy of the human retina wikipedia , lookup

History of genetic engineering wikipedia , lookup

Gene expression profiling wikipedia , lookup

Long non-coding RNA wikipedia , lookup

Epitranscriptome wikipedia , lookup

Site-specific recombinase technology wikipedia , lookup

Vectors in gene therapy wikipedia , lookup

DNA methylation wikipedia , lookup

Fetal origins hypothesis wikipedia , lookup

Artificial gene synthesis wikipedia , lookup

RNA-Seq wikipedia , lookup

Genomic imprinting wikipedia , lookup

Primary transcript wikipedia , lookup

Telomerase wikipedia , lookup

Epigenetics of human development wikipedia , lookup

Oncogenomics wikipedia , lookup

Therapeutic gene modulation wikipedia , lookup

NEDD9 wikipedia , lookup

Epigenetics of depression wikipedia , lookup

Bisulfite sequencing wikipedia , lookup

Transgenerational epigenetic inheritance wikipedia , lookup

Epigenetic clock wikipedia , lookup

Epigenetics wikipedia , lookup

Epigenetics of neurodegenerative diseases wikipedia , lookup

Epigenetics in learning and memory wikipedia , lookup

Behavioral epigenetics wikipedia , lookup

Epigenomics wikipedia , lookup

Cancer epigenetics wikipedia , lookup

Epigenetics of diabetes Type 2 wikipedia , lookup

Cell-free fetal DNA wikipedia , lookup

Epigenetics in stem-cell differentiation wikipedia , lookup

Mir-92 microRNA precursor family wikipedia , lookup

Polycomb Group Proteins and Cancer wikipedia , lookup

Nutriepigenomics wikipedia , lookup

Transcript
Molecular Human Reproduction, Vol.20, No.10 pp. 1026–1040, 2014
Advanced Access publication on July 14, 2014 doi:10.1093/molehr/gau053
ORIGINAL RESEARCH
Epigenetic mechanisms regulate
placental c-myc and hTERT in normal
and pathological pregnancies; c-myc as
a novel fetal DNA epigenetic marker
for pre-eclampsia
Beenish Rahat1, Abid Hamid 2, Rauf Ahmad Najar 2, Rashmi Bagga 3,
and Jyotdeep Kaur 1,*
1
Department of Biochemistry, Postgraduate Institute of Medical Education and Research, Chandigarh 160012, India 2Cancer Pharmacology
Division, Indian Institute of Integrative Medicine, Jammu 180001, India 3Department of Obstetrics and Gynaecology, Postgraduate Institute
of Medical Education and Research, Chandigarh 160012, India
*Correspondence address. Tel: +91-172-2747585-5181; E-mail: [email protected]
Submitted on April 8, 2014; resubmitted on May 23, 2014; accepted on July 1, 2014
abstract: Placental development is known for its resemblance with tumor development, such as in the expression of oncogenes (c-myc) and
telomerase (hTERT). The expression of c-myc and hTERT is up-regulated during early pregnancy and gestational trophoblastic diseases (GTDs). To
determine the role of DNA methylation [via methylation-sensitive high resolution melting (MS-HRM)] and histone modifications [via chromatin
immunoprecipitation (ChIP assay)] in regulating the differential expression of c-myc and hTERT during normal gestation and their dysregulation
during placental disorders, we obtained placental samples from 135 pregnant women, in five groups: normal first, second and third trimester
(n ¼ 30 each), pre-eclamptic pregnancy (n ¼ 30) and molar pregnancy (n ¼ 15). Two placental cell lines (JEG-3 and HTR-8/SVneo) and isolated
first-trimester cytotrophoblasts were also studied. Quantitative RT –PCR revealed decreased mRNA expression levels of c-myc and hTERT, which
were associated with a higher level of H3K9me3 (1.5-fold, P , 0.05) and H3K27me3 (1.9-fold, P , 0.05), respectively, in third-trimester placental villi versus first-trimester villi. A significantly lower level of H3K27me3 in molar placenta was associated with a higher mRNA expression of c-myc
and hTERT. The development of pre-eclampsia (PE) was associated with increased methylation (P , 0.001) and H3K27me3 (P , 0.01) at the
c-myc promoter and reduced H3K9me3 (P , 0.01) and H3K27me3 (P , 0.05) at the hTERT promoter. Further, mRNA expression of c-myc
and hTERT was strongly correlated in molar villi (r ¼ 0.88, P , 0.01) and JEG-3 cells (r ¼ 0.99, P , 0.02). Moreover, on the basis of methylation
data, we demonstrate the potential of c-myc as a fetal DNA epigenetic marker for pre-eclamptic pregnancies. Thus we suggest a role for epigenetic
mechanisms in regulating differential expression of c-myc and hTERT during placental development and use of the c-myc promoter region as a
potential fetal DNA marker in the case of PE.
Key words: placenta / DNA methylation / histone modifications / gestational trophoblastic diseases / pre-eclampsia
Introduction
The functioning of the placenta as a fetomaternal organ is of paramount
importance throughout pregnancy in terms of intrauterine development
of the fetus. Placental development involves a series of complex structural and physiological changes, to physically connect the mammalian
embryo to its mother (Cross et al., 1994) and functionally synchronize
fetal development. Such events involve invasion of the constituent cells
of the placenta (the trophoblasts) into the uterine wall (Pijnenborg
et al., 1981), high cell proliferation, migration and many malignancylike phenotypes that have led to the definition of the placenta as a
‘pseudo-malignant’ type of tissue (Strickland and Richards, 1992;
Redman, 1997; Even-Ram et al., 1998; Ferretti et al., 2007). Placental trophoblasts share several features with malignant cells in terms of their biological behavior, such as rapid proliferation and invasiveness (Bischof
et al., 2001) and the gene expression profile, especially the expression of
certain proto-oncogenes (Ohlsson, 1989; Janneau et al., 2002; Chiu
et al., 2007), activation of telomerase and the acquisition of a rich blood
supply (Soundararajan and Rao, 2004). Several oncogenes are similarly
expressed by both normal trophoblasts and cancer cells, especially the
oncogenes encoding transcription factors such as c-myc and human
telomerase reverse transcriptase (hTERT). However, the specialized
& The Author 2014. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: [email protected]
1027
Epigenetic regulation of c-myc and hTERT in human placenta
phenotypic behavior of placental trophoblasts is highly regulated, unlike
most tumor tissues, involving a highly complex differential global gene expression profile varying with gestational age. The importance of this regulatory mechanism in placental development becomes clearer in the case of
aberrant trophoblastic proliferation and invasion, resulting in various
pregnancy-related disorders, such as gestational trophoblastic diseases
(GTDs: hydatidiform mole and choriocarcinoma), which result from uncontrolled trophoblast proliferation, while inadequate trophoblast invasion
has been implicated in the pathophysiology of pre-eclampsia (PE).
C-myc encodes a transcription factor containing the bHLH/LZ (basic
Helix-Loop-Helix Leucine Zipper) domain and can bind to Enhancer Box
sequences (E-boxes) via its bHLH domain recruiting histone acetyltransferases and further, dimerizing with its partner Max, with its leucine
zipper domain, thus regulating the expression of a large number of downstream target genes (Gearhart et al., 2007) involved in cell growth and
proliferation. Overexpression of c-myc resulting in the up-regulation of
many other genes is known to be involved in the development of
cancer. The catalytic subunit of hTERT is one such c-myc downstream
target gene (Horikawa et al., 1999). C-myc overexpression thus has a
role in telomerase activation, which in turn would allow permanent proliferation. hTERT is involved in the maintenance of the telomeric length
(Gomez et al., 2012) and in the cellular replicative life span (Nakamura
and Cech, 1998) and is generally repressed in most adult human
somatic cells, limiting their proliferative potential due to induced
chromosomal instability, leading to cellular senescence after a few cell
divisions. However, strong telomerase activity has been detected in
human cancer cells, with unlimited replication potential (Kim et al.,
1994). hTERT also has oncogenic properties, whereby its expression
transforms some cell lines by promoting tumorogenesis, inhibiting apoptosis and stimulating mitogen-independent growth (Morales et al., 1999).
Like most progressive cancers, and in contrast to the majority of human
somatic tissues, the human placenta expresses telomerase activity.
Studies have shown a high level of telomerase expression in early placental villi with a drastic decrease with advancing gestation (Chen et al., 2002;
Nishi et al., 2004). Further, in vitro studies using models for trophoblast
differentiation have demonstrated a negative regulation of telomerase
activity during placental differentiation (Rama et al., 2001). These
studies highlight the high proliferative potential of trophoblastic cells
during early pregnancy (Soundararajan and Rao, 2004).
Increased telomerase activity and c-myc expression have already been
reported in the early weeks of pregnancy and in cases of GTDs, such as
hydatidiform mole and choriocarcinoma (Pfeifer-Ohlsson et al., 1984;
Chen et al., 2002), suggesting that these genes may be important in the
pathogenesis of a hydatidiform mole and choriocarcinoma (Fulop
et al., 1998). Further, c-myc expression has been found both in
villous and extravillous cytotrophoblasts (Pfeifer-Ohlsson et al., 1984;
Goustin et al., 1985; Roncalli et al., 1994).
Although the expression of these genes varies with advancing gestation and plays a role in placental disorders, the molecular mechanisms
regulating their differential expression have not been identified. Recently,
studies have shown a role for DNA methylation (Rahnama et al., 2006)
and histone modifications (Dokras et al., 2006) in the regulation of the
differential expression of E-cadherin and maspin in placenta, respectively.
Therefore, we hypothesize that epigenetic regulation of c-myc and hTERT
could be the possible regulatory mechanism for their differential expression. Further, the expression of these genes has not been studied in
reference to the known invasive potential of placental derived cell
lines, such as HTR-8/SVneo (a highly invasive cell line derived from
transformed extravillous trophoblasts) and JEG-3 cells (a comparatively
less invasive cell line derived from placental choriocarcinoma) (Suman
and Gupta, 2012). Moreover, the presence of circulating fetal DNA
in maternal blood has led the way to the development of fetal DNA
epigenetic markers, for example RASSF1A and maspin (Chim et al.,
2005; Chan et al., 2006), that can be used for physiological pregnancies
but there is no known epigenetic marker specific for pathological
pregnancies.
Based on the previous literature, in this study we analyzed the mRNA
expression of c-myc and hTERT in reference to the epigenetic mechanisms of DNA methylation and histone modification, in order to find
the regulatory mechanisms behind their differential expression and
further investigate the correlation between c-myc and hTERT mRNA expression in normal and pathological pregnancies. mRNA expression of
these genes was further assessed in placental-derived cell lines with different invasive potentials. Finally, based on our DNA methylation data,
we aimed to search for a fetal DNA epigenetic marker that is specific
for pathological pregnancies.
Materials and Methods
Ethical approval
This study was carried out in the department of Biochemistry in collaboration
with the department of Obstetrics and Gynecology at the Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh, India. The
protocol of the study was approved by the Institute Ethics Committee
(IEC) and a written informed consent was obtained from all participants.
Subjects and sample collection
A total of 135 pregnant women attending the Department of Obstetrics
and Gynaecology with the clinical diagnosis of normal pregnancy or
pregnancy-related disorders (PE, hydatidiform mole) were included in this
study. These pregnant women with a singleton pregnancy were classified
into three normal groups and two disorder groups based on the gestational
age and clinical conditions (all data here are mean + SD): the normal groups
were first-trimester (6– 11 weeks, who desired to undergo medical termination of pregnancy (MTP) by suction evacuation, n ¼ 30, age 27.90 + 3.44
years and gestation 8.22 + 1.44 weeks), second-trimester (16– 20 weeks,
who desired to undergo MTP, n ¼ 30, age 27.65 + 3.91 years and gestation
18 + 1.62 weeks) and third-trimester (37– 40 weeks, n ¼ 30, age 28 + 3.5
years and gestation 38.06 + 0.877 weeks), while the disorder groups were
PE (clinical symptoms of systolic pressure 140 mmHg and diastolic pressure
90 mmHg, proteinuria . 300 mg in 24 h, n ¼ 30, age 26 + 3.38 years
and gestation 35.27 + 2.53 weeks) and hydatidiform mole (in first or
early second trimester, diagnosed by ultrasonography and confirmed by
histopathology, n ¼ 15, age 25.09 + 2.63 years and gestation 13.68 + 2.9
weeks).
Placental villi samples were collected after elective pregnancy terminations
from pregnant women with normal first, second trimester and hydatidiform
mole pregnancy, or after cesarean section or vaginal deliveries in case of
normal third-trimester and pre-eclamptic pregnancies. A maternal peripheral blood sample (10 ml) was collected in each case just before these obstetric procedures; however, in case of a normal third-trimester pregnancy 10 ml
of maternal peripheral blood was also collected 24 h after delivery and was
processed similarly. Maternal blood samples were processed to separate leukocytes and plasma.
1028
Rahat et al.
Sample processing
Quantitative Real-time PCR
Maternal blood samples were centrifuged at 1600 g for 10 min (min) at 48C
and supernatant was collected and recentrifuged at 16 000 g for 10 min at
48C for harvesting cell-free plasma samples (Chiu et al., 2001), which were
stored at 2808C for isolation of cell-free circulating DNA, while the remaining peripheral blood cell portion was processed for red blood cell (RBC) lysis
and isolation of maternal leukocytes using RBC lysis buffer (0.82% NH4Cl)
and phosphate-buffered saline (PBS). Placental tissue obtained from each
subject was processed in chilled PBS for the separation of clear placental
villi free from fibroid tissue, blood clots, amnion and basement membrane.
Total RNA was isolated from placental villi, maternal blood cells, and the
JEG-3 and HTR-8/SVneo cell lines using TRIzol (Ambion, Life Technologies
Corporation, CA, USA), and 1 mg was reverse-transcribed using RevertAidTM M-MuLV-RT kit (MBI Fermentas, Life Sciences, USA). The resulting
cDNA was quantified using Applied Biosystems StepOnePlusTM Real-Time
PCR System under the following conditions: 1 cycle of 958C for 10 min,
40 cycles of 958C for 15 s, primer-specific annealing temperature for 1 min
and followed by a melt curve of 958C for 10 s, 608C for 1 min and 958C
for 15 s. The PCR mixture contained 10 ml of SYBR Green master mix
(Applied Biosystems, Inc., Life Technologies Corporation, CA, USA),
500 nM of each primer and 1.5 mM of MgCl2 and 1 ml (almost 60 ng) of
cDNA template. Primers were designed using the online Primer-BLAST tool
(http://www.ncbi.nlm.nih.gov/tools/primer-blast/) and cross-checked for
primer self-dimerization and potential hairpin formation using the online
software Oligonucleotide Properties Calculator (http://www.basic.
northwestern.edu/ BIOTOOLS/OLIGOCALC.HTML) (Table I). The
mRNA expression levels of two housekeeping genes glyceraldehyde-3phosphate dehydrogenase (GAPDH) and 18S ribosomal RNA (18S rRNA)
were analyzed initially in different groups for their stability and, as both
showed no statistically significant variation between groups (data not
shown), the expression of GAPDH was used as the endogenous control.
The comparative Ct method (DDCT) (Livak and Schmittgen, 2001) was
used for quantification of the transcripts. Measurement of DCt was performed in triplicate.
Isolation of cytotrophoblasts from
first-trimester villi
First-trimester cytotrophoblasts were isolated from placentas (6–11 weeks)
by modifying the method of Fisher et al. (1989). Briefly, a sample of 2 g of firsttrimester villous tissue was dissected from chorionic membranes and washed
in sterile ice-cold tissue washing buffer (PBS 1×, antibiotic– antimycotic
Solution 1×, Gentamycin 50 mg/ml, D-Glucose 0.1% w/v), then was subjected to the first enzymatic digestion (PBS 1×, antibiotic–antimycotic Solution 1×, Gentamycin 50 mg/ml, HEPES 15 mM, MgSO4 5 mM, Trypsin
0.25%, DNaseI 0.02% w/v) for 10 min at 378C without shaking and then
passed through a pre-equilibrated 60 mm mesh to remove the supernatant
containing syncytiotrophoblasts. The rest of the tissue was collected from
the mesh and subjected to further rounds of enzymatic digestion at 378C initially for 30 min and then for another two sequential rounds of 10 min each.
After each step the cell suspension was passed through pre-equilibrated
60 mm mesh to remove cellular debris and collect the filtrate containing isolated cytotrophoblasts. Fetal bovine serum (FBS) was used to inactivate
trypsin. The filtrates were then centrifuged, the cell pellets were resuspended
in sterile PBS and subjected to enrichment on a preformed 10–70% Percoll
gradient (all reagents were purchased from Sigma Chemical Co., St. Louis,
MO, USA) at 800 g for 20 min at 208C. The white buffy layer (the mononuclear cells containing cytotrophoblasts) below the top layer was collected
carefully by micropipette and was immunopurified to a pure population of
cytotrophoblasts by the negative selection method using EasySepTM Human
CD45 Depletion Kit and EasySepTM magnet (STEMCELL Technologies, Inc.,
Vancouver, BC, Canada) based on immunomagnetic depletion of CD45+
cells. CD45+ labeled cells were removed, leaving behind the pure population
of cytotrophoblasts, which was then confirmed by flow cytometry using the
trophoblast marker cytokeratin-7 [CK-7, rabbit monoclonal (Epitomics, Inc.,
Burlingame, California, cat. no. 2303-1)], showing .90% cells as CK-7 positive.
Cell culture
Purified cytotrophoblasts were resuspended in cyto-culture media (1× Dulbecco’s Modified Eagle’s Medium-(DMEM)-F12, 20% FBS, 1× Antibiotic –
Antimycotic Solution, 14.28 mM NaHCO3, 15 mM HEPES) and then
seeded at the density of 2 × 105 cells/3 ml/well in a 6-well plate precoated
with type 4 collagen (Becton Dickinson, Franklin Lakes, NJ, USA) for 24 h and
incubated in a humidified atmosphere of 5% CO2 at 378C. In addition, two
adherent cell lines, JEG-3- a human placental choriocarcinomic cell line,
and HTR-8/SVneo- a transformed extravillous trophoblast cell line
[Source: American Type Culture Collection (ATCC), kindly provided by
Dr. S.K. Gupta, National Institute of Immunology, Delhi, India] were used
in this study. JEG-3 and HTR-8/SVneo cells were cultured in DMEMHG,
with 4500 mg/l glucose and RPMI-1640 medium, respectively, supplemented with L-glutamine and 25 mM HEPES, 10% FBS, sodium bicarbonate
(3.7 g/l and 2 g/l, respectively), penicillin (100 U/ml), streptomycin
(100 mg/ml) and amphotericin B (0.25 mg/ml) under standard conditions
(378C, 5% CO2, humidified atmosphere). Cell culture media and other
reagents were purchased from Sigma Chemical Co.
Methylation-sensitive high resolution melting
DNA was extracted for methylation analysis from the placental villi, the
corresponding maternal blood leukocytes, cultured cytotrophoblasts, and
cultured JEG-3 and HTR-8/SVneo cell lines by using the genomic DNA isolation kit (Real Genomics, Real Biotech Corporation, Taipei, Taiwan), while
circulating DNA was isolated from maternal plasma using the Miniprep DNA
isolation kit (Bioserve, MD, USA) according to the manufacturer’s instructions. Promoter region CpG methylation analysis of the cmyc and hTERT
was carried out by methylation-sensitive high resolution melting (MSHRM), which is a real-time PCR-based technique that uses new generation
saturating HRM dye that intercalates specifically to dsDNA. MS-HRM can differentiate sequences on the basis of the GC content, which determines the
temperature at which the dsDNA sequence denatured (Wojdacz and
Dobrovic, 2007). A standard curve comprising 0, 0.5, 5, 10, 20, 40, 60, 80
and 100% methylation standards was generated by mixing 0 and 100%
methylated standards in different proportions, for the analysis of the percentage of methylation of unknown samples. For the 0% methylated standard, a
commercial unmethylated DNA (Qiagen – EpiTectw Control DNA) was
used, while the 100% methylated standard was generated by treating
genomic DNA with M.SssI enzyme (CpG Methyltransferase, New England
Biolabs, Beverly, MA, USA), according to the manufacturer’s instructions.
An aliquot of 1 mg of DNA from each sample and methylation standards
was treated with sodium bisulfite using the EZ DNA Methylation-Gold Kit
(Zymo Research, USA), according to the manufacturer’s instructions.
Sodium bisulfite treatment converts all unmethylated cytosines into uracil,
leaving methylated cytosines unchanged and thus creating single nucleotide
base differences in the amplicons. Primers were designed for the promoter
region of c-myc and hTERT on the basis of parameters set earlier (Wojdacz
et al., 2008) to amplify an 80 bp sequence containing 9 CpG for c-myc, and
139 bp sequence containing 19 CpG for hTERT (Table I). MS-HRM analysis
was performed on the Applied Biosystemsw StepOnePlusTM Real-Time
PCR System using the following conditions: 1 cycle of 958C for 10 min,
40 cycles of 958C for 15 s, primer-specific annealing temperature for 1 min
and followed by an HRM step of 958C for 10 s, 608C for 1 min and continuous
acquisition to 958C for 15 s with a ramp rate of 0.3%. PCR was performed in a
1029
Epigenetic regulation of c-myc and hTERT in human placenta
Table I Primer sequences and annealing temperature used in this study.
Gene/
assay
Forward primer 5′ -3′
Reverse primer 5′ -3′
Amplicon size
bp/annealing
temperature (88 C)
Genbank
accession
number
.............................................................................................................................................................................................
c-myc
qRT– PCR
TGAGGAGACACCGCCCAC
CAACATCGATTTCTTCCTCATCTTC
71/65
NM_002467.4
hTERT
qRT– PCR
CGTGGTTTCTGTGTGGTGTC
CCTTGTCGCCTGAGGAGTAG
214/65
NM_001193376.1
GAPDH
qRT– PCR
CGACCACTTTGTCAAGCTCA
AGGGGTCTACATGGCAACTG
228/60-65
NM_001256799.2
c-myc
MS-HRM
TGAGGATTTTCGAGTTGTGTTGT
CTTCTCGAAACAAAAAAAAAACCAAAA
80/58
NG_007161.1
hTERT
MS-HRM
GGGAAGTTTTGGTTTCGGTTATTTT
CGCCAACCCTAAAACCCCAAA
139/60
AF097365.1
c-myc
qChIP-PCR
ACGCGCGCCCATTAATAC
GCCGCGCTTTGATCAAGA
63/62
NG_007161.1
hTERT
qChIP-PCR
AGCCCCTCCCCTTCCTTT
CAGCGCTGCCTGAAACTC
64/62
AF097365.1
qRT –PCR, quantitative reverse transcriptase polymerase chain reaction; hTERT, human telomerase reverse transcriptase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase;
MS-HRM, methylation-sensitive high resolution melting; qChIP, quantitative chromatin immunoprecipitation assay.
final volume of 20 ml, containing 10 ml of master mix (MeltDoctorTM
MS-HRM Master Mix, Applied Biosystems, Inc., Life Technologies Corporation), 250 nM of each primer, 1.5 mM MgCl2 and 1 ml (almost 30 ng) of
bisulfite modified DNA template. The data obtained via the Stepone plus
software were then analyzed with the Applied Biosystemsw MS-HRM
software (v.3.0.1.). Standard curves were included in each assay and to
deduce the methylation percentage values for each sample according to
the method of Migheli et al. (2013) by imputing the average aligned fluorescence value of each sample in the Polyfit interpolation curve (MatLab
program-The MathWorks, Inc., USA) obtained for methylation standards
ranging from 0 to 100% used in the particular MS-HRM assay. Each aligned
fluorescence percentage value for a single methylation standard was calculated as an average of fluorescence units of the temperatures relative to
the melting status in a specific temperature range (from 72.1 to 78.98C for
c-myc and from 75.9 to 82.98C for hTERT).
Chromatin immunoprecipitation assay
The chromatin immunoprecipitation assay (ChIP) assay was used to evaluate
the presence of a particular histone modification within the promoter region
of c-myc and hTERT. Histones were cross-linked to DNA by incubating placental villi samples (25 mg) from normal gestational groups, PE and hydatidiform mole (4 samples in each group) in 1.5% formaldehyde on a rocker for
10 min, and cross-linking was stopped by adding glycine to a final concentration of 125 mM. Ice-cold PBS was used to wash placental villi followed by disaggregation using the Dounce homogenizer in 1 ml lysis buffer (NaCl
150 mM, Tris HCl (pH 7.5), 25 mM, EDTA 5 mM, Trition X 100 1%, SDS
0.1%, Sodium deoxycholate 0.5%, PMSF 1 mM, Sodium butyrate 10 mM,
Protease inhibitor cocktail [PIC] 1X) to obtain a single-cell suspension. The
suspension was centrifuged and the cell pellet was washed sequentially in
wash buffer 1 (Triton X 2100 0.25%, EDTA 10 mM, EGTA 0.5 mM,
HEPES (pH 7.5) 10 mM, PMSF 1 mM, Sodium butyrate 10 mM, PIC 1×)
and wash buffer 2 (NaCl 0.2M, EDTA 1M, EGTA 0.5 mM, HEPES
(pH 7.5), 10 mM, PMSF 1 mM, Sodium butyrate 10 mM, PIC 1×) followed
by resuspension in lysis buffer for sonication using Bioruptor XL (Diagenode,
Belgium) at 20 pulses of 10 s each. The sonicated sample was centrifuged
at 13 000 g for 10 min at 48C and then incubated with protein
A/G-polyacrylamide beads alone (UltraLinkw Resin, Thermo Scientific,
Rockford, IL, USA) for 4 h at 48C on a rocker to preclear the lysate. Supernatant was then divided into three aliquots: the first aliquot was used for input
DNA (10% of total supernatant); the other two aliquots were used for immunoprecipitation with anti-trimethyl H3-K9 (rabbit polyclonal) and H3-K27
(mouse monoclonal) antibody (45% of total supernatant, from Abcam,
Cambridge, England, UK, cat. no. ab8898 and ab6002, respectively) and
normal rabbit immunoglobulin G (IgG) (45% of total supernatant, from
Santa Cruz Biotechnology, Santa Cruz, CA, USA, cat. no. sc-2027) for
isotype control, at 48C overnight. To collect immunoprecipitated complexes, the suspension was incubated with protein A/G-polyacrylamide
beads for 4 h at 48C. Beads were harvested and washed with RIPA and TE
buffers (10 mM Tris 1 mM EDTA, pH 8.0). The cross-links were reversed
by heating the samples with 4 M NaCl at 658C for 4 h, followed by treatment
with RNase (100 mg/ml final concentration) for 30 min at 378C and then
proteinase K (100 mg/ml final concentration) overnight (all ChIP reagents
were purchased from Sigma Chemical Co.). DNA was extracted by the
phenol – chloroform method, ethanol precipitated and resuspended in deionized water. Quantitative real-time PCR was carried out to estimate the
amount of immunoprecipitated DNA using the primers designed by
Applied Biosystems Primer Express Software v3.0.1 for the promoter
region of c-myc and hTERT (Table I). The PCR program was set as: 1 cycle
of 958C for 10 min, 40 cycles of 958C for 15 s, primer-specific annealing temperature for 1 min and followed by a melt curve of 958C for 10 s, 608C for
1 min and 958C for 15 s. PCR was performed in a final volume of 20 ml,
containing 10 ml of SYBR Green master mix (Applied Biosystems, Inc., Life
Technologies Corporation), 500/150 nM primers for c-myc and hTERT,
respectively, 1.5 mM MgCl2 and 1 ml of immunoprecipitated DNA. The
assays were performed in triplicate and the levels of histone tail modifications
were calculated by the fold enrichment method relative to the non-specific
antibody and normalized to the input DNA.
Statistical analysis
All statistical analyses were performed using the SPSS software for Windows
version 16.0 and GraphPad Prism software version 5.00.288. Between-group
comparisons were made using one-way ANOVA; if found to be significant,
1030
the Fisher post hoc test was applied. Student’s t-test and the Mann– Whitney
U-test were used to analyze the data between two variables. Pearson’s correlation analysis was used to estimate the correlations between different
parameters. The effect of different epigenetic mechanisms on mRNA expression level was assessed by multiple regression analysis. All statistical tests
were two sided, and differences were considered statistically significant at
P , 0.05. Unless otherwise stated, all data are expressed as a mean (SEM).
Results
Quantification of mRNA expression
of c-myc and hTERT
Quantification of c-myc mRNA expression revealed a decrease in mRNA
levels with advancing gestation, a significant decrease being observed in
the normal third-trimester versus the first (3.5-fold) and second
(2.5-fold) trimester (P , 0.001) groups. In both disorder groups, c-myc
expression was dysregulated, with a 4.1-fold (P , 0.001) decrease in
PE versus normal third-trimester placenta and 1.4 (P , 0.05) and
2-fold (P , 0.001) higher in molar placenta when compared with firstand second-trimester normal placentas, respectively. Progression to
choriocarcinoma was found to be associated with a further rise in
c-myc mRNA levels as depicted by 1.9 (P , 0.05) and 2.8 (P , 0.001)
fold increased c-myc expression in choriocarcinomic cell line JEG-3,
when compared with molar placenta and normal first-trimester placenta,
respectively (Fig. 1a).
The c-myc mRNA expression level in maternal leukocytes (Fig. 1b)
increased by 3.1 (P , 0.01) and 6.2 (P , 0.001) fold in the secondand third-trimester, respectively, when compared with first-trimester
maternal leukocytes. The c-myc mRNA levels were also raised in molar
maternal leukocytes by 2.7-fold (P , 0.01) versus the normal firsttrimester group. A comparison of c-myc mRNA levels between placental
villi and their corresponding maternal leukocytes (Fig. 1c) showed higher
relative expression in the first-trimester placenta (12.8-fold, P , 0.001),
second-trimester placenta (2.9-fold, P , 0.05) and molar placenta
(6.8-fold, P , 0.001). In normal third-trimester placenta and PE there
was a non-significant difference in c-myc expression between placental
villi and maternal leukocytes. Analysis of the placental-derived cell lines
showed a significantly raised level (P , 0.001) of c-myc mRNA in
HTR-8/SVneo cells when compared with JEG-3 cells (4.4-fold) and
normal first-trimester placenta (12.7-fold) (Fig. 1d).
mRNA expression level of hTERT in placental villi was 5.4-fold (P ,
0.001) higher in pre-eclamptic villi compared with its respective
control. Decreasing levels of hTERT mRNA were observed with increasing gestation age, as shown by the 4.9-fold (P , 0.001) decrease in thirdtrimester placenta when compared with first-trimester. Progression to
choriocarcinoma may be associated with raised hTERT mRNA levels,
as shown by 6.5- to 8.9-fold higher hTERT transcript levels in JEG-3
cells versus normal first- and second-trimester and molar placentas
(Fig. 2a). The overall expression of hTERT is very low in all normal gestational villi groups in reference to HTR-8/SVneo cells, with a difference in
DCt values ranging from 8 to 10 between first-trimester and thirdtrimester villi versus HTR/8SVneo cells, respectively.
In maternal leukocytes advanced gestation age was associated with
decreased hTERT mRNA levels, which decreased by around 2-fold in
the second and third trimester groups (P , 0.001). Further, the molar
group showed a significant increase by 1.98-fold (P , 0.01) in hTERT
Rahat et al.
mRNA levels compared with second-trimester maternal leukocytes
(Fig. 2b). hTERT mRNA expression was significantly lower (P , 0.001)
in all maternal leukocyte groups compared with their respective placental
groups (Fig. 2c). This difference was least in the third-trimester group
among the normal groups. However, pre-eclamptic and molar placental
villi had a 17-fold and 6.2-fold higher level of hTERT mRNA, respectively,
than their respective maternal leukocytes. HTR-8/SVneo cells showed a
higher expression of hTERT mRNA (33-fold) relative to JEG-3 cells
and 217-fold higher than that of the normal first-trimester placenta
(P , 0.001) (Fig. 2d).
Methylation status of CpG islands in the
promoter region of c-myc and hTERT
On the basis of the methylation percentage as revealed by MS-HRM, we
defined the placental villi groups as hypomethylated and hypermethylated. The promoter region methylation pattern for c-myc is shown in
Fig. 3a–d. With the exception of pre-eclamptic villi, all groups showed
hypomethylation at the c-myc promoter (Fig. 4a). In pre-eclamptic villi
there was 71.5% (P , 0.001) increase in c-myc promoter methylation
versus normal third-trimester placenta. The Pearson correlation analysis
of the correlation between the methylation percentage for the c-myc promoter in pre-eclamptic villi and its corresponding mRNA expression
showed a high negative correlation coefficient value of 20.9, P , 0.01.
Although first-trimester placental villi showed statistically significant
methylation for c-myc promoter compared with other placental villi
groups (except pre-eclamptic villi), methylation of the c-myc promoter
was not observed to correlate with c-myc mRNA expression in these
placental villi groups.
The c-myc promoter region within maternal leukocyte groups was also
hypomethylated, with a similar comparatively higher methylation in the
first-trimester compared with the third-trimester (P , 0.01) and
molar group (P , 0.05); however, no other significant differences
were observed (Fig. 4b). Comparing the percentage of methylation
between placental villi samples and their corresponding maternal leukocytes revealed a highly significant difference only in the preeclamtic
group, with 76.6% higher methylation in pre-eclamptic villi (P , 0.001)
versus its maternal group (Fig. 4c). In placental cell lines and isolated cytotrophoblasts (Fig. 4d), the c-myc promoter was significantly less methylated (by 19.5%) in JEG-3 cells (P , 0.05), 24% in HTR-8/SVneo cells
(P , 0.01) and 16.2% (P , 0.05) in cytotrophoblasts when compared
with the normal first-trimester placenta. The promoter region methylation analysis of hTERT revealed a completely unmethylated promoter in
all categories (Fig. 5a–c).
Histone modifications within the promoter
regions of c-myc and hTERT
Histone modification was quantified as the relative amount of each
histone modification in different placental villi groups above the background. We targeted two main silencing modifications, i.e. trimethylation of H3K9 and H3K27. In case of c-myc, H3K9me3 was significantly
raised 6.7-fold (P , 0.05) in the third-trimester placenta when compared with the 4.4-fold increase above the background in the firsttrimester placenta, while molar placenta showed a lower level of
H3K9me3 when compared with normal first- and second-trimester
placenta (Fig. 6a). The H3K27me3 pattern in c-myc was raised in PE
2.4-fold (P , 0.01) relative to normal third-trimester placenta, while
Epigenetic regulation of c-myc and hTERT in human placenta
1031
Figure 1 Quantitative reverse transcriptase PCR (qRT– PCR) analysis of c-myc normalized with glyceraldehyde-3-phosphate dehydrogenase (GAPDH).
(a – d) Relative fold change in mRNA levels among different (a) villi samples and JEG-3 cells [*P , 0.05, ***P , 0.01 with respect to 1st Trim, ###P , 0.001
with respect to 2nd Trim, E E EP , 0.001 and ¥P , 0.05 with respect to molar], (b) maternal blood leukocytes [**P , 0.01, ***P , 0.001 with respect to
1st Trim and #P , 0.05 with respect to 2nd Trim], (c) villi samples with respect to that in their corresponding maternal blood leukocytes [*P , 0.05,
***P , 0.001 with respect to corresponding maternal leukocytes] and (d) different placental cells namely JEG-3, HTR-8/Svneo and 1st Trim. villi
[‡‡P , 0.01, ‡‡‡P , 0.001 with respect to JEG-3 and ^^^P , 0.001 with respect to HTR-8/SVneo]. The data were compared using one-way analysis
of variance (ANOVA), followed by Fisher’s post hoc test between ≥3 variables, while Student’s t-test was used for 2 variables and are expressed as the
mean fold change + SEM, for 30 patients in each group. Trim ¼ trimester.
being lower by 2.6-fold (P , 0.05) in molar placenta versus normal
second-trimester placenta (Fig. 6b); however, it was almost equal in all
normal gestational groups at around 3.8-fold. Pearson’s correlation
analysis (used to study the independent relation of each epigenetic modification and mRNA expression of c-myc) revealed a significantly high
correlation between different epigenetic modifications and mRNA
expression of c-myc. The Pearson’s correlation coefficient value
between its different parameters, being 20.99, P , 0.001 and 20.82,
P , 0.05 for mRNA expression and H3K9me3 and H3K27me3 levels,
respectively, thus suggesting a strong regulation of c-myc mRNA expression by epigenetic modifications. Further, c-myc promoter region DNA
methylation was also found to be strongly correlated with H3K27me3
levels (0.89, P , 0.05). This was further validated by a significantly
higher regression coefficient (r 2 ¼ 1, P , 0.01), showing the combined
effect of various epigenetic modifications on the mRNA expression of
c-myc (Table II).
As revealed by ChIP analysis, the hTERT promoter region showed significant difference in the level of H3K9me3 only in PE placenta, being
2.2-fold lower (P , 0.01) than in the normal third-trimester placenta,
while no significant difference was observed in the other categories
(Fig. 7a). Levels of H3K27me3 in the hTERT promoter decreased significantly by 1.8-fold (P , 0.05) in PE placenta when compared with normal
third-trimester placenta. The third-trimester placenta exhibited a
1.9-fold (P , 0.05) higher level of H3K27me3 when compared with
the first-trimester placenta. However, H3K27me3 was significantly
lower by 2.9-fold (P , 0.001) in molar placenta when compared with
the second-trimester placenta (Fig. 7b). Correlation analysis between
its mRNA expression and various histone modifications revealed a
strong inverse correlation with H3K9me3 (r ¼ 20.77) and
H3K27me3 (r ¼ 20.78). On multiple regression analysis, a stronger
impact was observed on mRNA expression of hTERT, as shown by significantly higher r 2 values of 0.99, P , 0.05 (Table II).
1032
Rahat et al.
Figure 2 qRT – PCR analysis of hTERT normalized with GAPDH. (a – d) the relative fold change in mRNA levels among different (a) villi samples and JEG-3
cells [***P , 0.001 with respect to 1st Trim, ###P , 0.001 with respect to 2nd Trim, EEEP , 0.001 with respect to 3rd Trim and ¥¥¥P , 0.001 with
respect to molar], (b) maternal blood leukocytes [**P , 0.01 with respect to 1st Trim and ##P , 0.01 with respect to 2nd Trim], (c) villi samples with
respect to that in their corresponding maternal blood leukocytes [**P , 0.01, ***P , 0.001 with respect to corresponding maternal leukocytes] and
(d) the different placental cells JEG-3, HTR-8/Svneo and 1st Trim. villi [‡‡‡P , 0.001 with respect to JEG-3 and ^^^P , 0.001 with respect to HTR-8/
SVneo]. The data were compared using one-way ANOVA, followed by Fisher’s post hoc test between ≥3 variables, while Student’s t-test was used for
2 variables and are expressed as the mean fold change + SEM, for 30 patients in each group.
Correlation between mRNA expression
of c-myc and hTERT
Pearson’s correlation analysis revealed a strong positive correlation
between the expression profiles of c-myc and hTERT (r ¼ 0.94,
P , 0.02) between normal placental villi groups and the choriocarcinoma cell line JEG-3 (with the exception of pre-eclamptic villi). This was
further validated by analyzing the correlation of c-myc and hTERT
mRNA expression within molar placental villi samples and JEG-3 cells,
showing a strong positive Pearson correlation coefficient of 0.88,
P , 0.01 and 0.99, P , 0.02, respectively. These data strongly suggest
the coordination of c-myc and hTERT mRNA expression in the development of molar pregnancy and further its progression to choriocarcinoma.
Fetal epigenetic markers in maternal plasma
To validate the methylated promoter region of a particular gene as a potential fetal epigenetic marker, the placental epigenetic pattern of that
gene should be found in the maternal plasma. We studied the promoter
region of c-myc in maternal plasma in the pre-eclamptic group based on
the observation of a significant difference of 76.6% in promoter region
methylation between pre-eclamptic placental villi and the corresponding
maternal blood leukocytes. Our analysis (Fig. 8a and b) revealed an
average of 53 + 16% methylation for the c-myc promoter in maternal
plasma samples, which was significantly higher (P , 0.001) than that in
maternal leukocytes. This hypermethylated pattern of the c-myc promoter in maternal plasma mimics the methylation pattern of c-myc in preeclamptic placenta, thus indicating the presence of hypermethylated fetal
c-myc in maternal plasma and supporting the use of c-myc as an epigenetic
marker in PE.
Discussion
Development of the placenta is a complex molecular event that involves
proliferation, migration and invasion of placental trophoblasts into the
myometrium of the uterus and its vasculature in order to nourish the
developing fetus, in a way that is similar to that in most malignant
tumors (Soundararajan and Rao, 2004; Ferretti et al., 2007). Several
Epigenetic regulation of c-myc and hTERT in human placenta
1033
Figure 3 Methylation specific-high resolution melting (MS-HRM) analysis for the c-myc promoter region. (a) Aligned melt curve and (b) Difference
plots normalized to the 0%-methylated standard DNA, for methylated standard curves of M100 to 0% (i.e. DNA standards with 100 to 0% methylation)
represented in different colors, (c and d) Difference plots with the methylated standard curve of M100 to 0% shown as black curves and differential
fluorescence among few representative samples of different, (c) placental villi groups and placental-derived cell lines, (d) maternal leukocyte groups.
oncogenes, such as c-fos, c-jun and c-myc, are similarly expressed by both
normal trophoblasts and cancer cells (Bamberger et al., 2004) and are
known to play an important role in the proliferative and invasive nature
of the placenta. Human placenta also shows high telomerase activity,
as in most progressive cancers (Soundararajan and Rao, 2004). In this
study we analyzed the role of epigenetic mechanisms in the regulation
of the differential expression of c-myc and hTERT in normal villous
tissue from each trimester and in molar and pre-eclamptic villi, based
on recent studies highlighting the possibility of common epigenetic modifications in trophoblasts and cancer cells which facilitate their proliferative, migratory and invasive properties (Ferretti et al., 2007; Serman et al.,
2007).
The expression data for c-myc and hTERT in our study, suggesting the
highest expression in the first trimester relative to other normal gestational groups and being abnormally high in a molar pregnancy and choriocarcinoma cell line, are consistent with the previously reported studies
(Pfeifer-Ohlsson et al., 1984; Fulop et al., 1998; Chen et al., 2002).
These observations suggest a role for c-myc and hTERT in the proliferation
and invasive potential of placental trophoblasts in these cases, supported
by the fact that gametes, stem and tumor cells show unlimited replicative
potential due to activation of telomerase that compensates for telomere
shortening (Kim et al., 1994). The observed significantly lower level of
hTERT expression in normal early trimester placenta with limited proliferative and invasive potential relative to choriocarcinoma (showing uncontrolled invasive and proliferative potential) also supports the role of
higher hTERT expression, and thus maintenance of telomeric length,
for indefinite proliferation and invasion. The role of c-myc and hTERT in
promoting trophoblast invasiveness and proliferation is further supported by the decreasing expression of these genes in the order of decreasing invasive and proliferative potential: in HTR-8/SVneo cells
followed by JEG-3 cells and the least invasive first-trimester villi
(Suman and Gupta, 2012).
The observed high positive Pearson’s correlation value between c-myc
and hTERT expression in normal placental villi groups and GTD groups
(molar villi and JEG-3 cells) in this study strongly emphasizes the role of
c-myc in enhancing the expression of hTERT, which is one of its downstream target genes (Horikawa et al., 1999), and further supports the observation that abnormally higher c-myc expression might therefore be a
1034
Rahat et al.
Figure 4 MS-HRM analysis for the c-myc promoter region. (a – d) Graphical representation of 5′ CpG Methylation (%), among different (a) villi samples
and JEG-3 cells [**P , 0.01, ***P , 0.001 with respect to 1st Trim and EEEP , 0.001 with respect to 3rd Trim], (b) maternal blood leukocytes [*P , 0.05,
**P , 0.01 with respect to 1st Trim] (a and b) the values are summarized using a box-and-whisker plot. (c) Villi samples with respect to that in their corresponding maternal blood leukocytes [*P , 0.05, ***P , 0.001 with respect to corresponding maternal leukocytes] and (d) placental cells JEG-3, HTR-8/
Svneo, cytotrophoblasts and 1st Trim. villi [‡P, 0.05 with respect to JEG-3, ^^P , 0.01 with respect to HTR-8/SVneo and £P , 0.05 with respect to
cytotrophoblasts], (c and d) the data are expressed as the mean percentage of methylation + SEM. The data were compared using one-way ANOVA,
followed by Fisher’s post hoc test between ≥3 variables, while Student’s t-test was used for 2 variables, for 30 patients in each group.
factor responsible for uncontrolled up-regulated expression of hTERT in
GTD. Further, hydatidiform moles are genetically abnormal (complete
hydatidiform moles being diploid and androgenic in origin (Hoffner and
Surti, 2012)), while partial hydatidiform moles are triploid (Lawler
et al., 1979). Thus, based on our observations and genetic abnormalities
linked to the hydatidiform mole, we can suggest that higher c-myc expression inducing abnormal expression of hTERT might be a contributing
factor for the uncontrolled proliferation and invasion observed in the
hydatidiform mole.
To delineate the epigenetic mechanisms regulating the expression of
c-myc and hTERT under these conditions, we studied the role of DNA
methylation and post-translational histone modifications, which are the
two main aspects of epigenetic regulation working in concert for regulating gene expression (D’Alessio and Szyf, 2006; Klose and Bird, 2006).
Maintenance of these epigenetic marks is necessary for the normal
physiological behavior of the cells, while in case of many tumors there
is an aberrant change in these epigenetic marks, resulting in an abnormal
transcriptional profile (Costello et al., 2000; Fraga et al., 2005; Song et al.,
2005; Bernstein et al., 2007; Esteller, 2007). In view of the close resemblance between placental development and the process of tumorigenesis, some tumor-associated methylation and histone modifications
have been observed in the placenta (Dokras et al., 2006; Novakovic
et al., 2008; Wong et al., 2008). Our analysis of the role of DNA methylation in regulation of c-myc and hTERT differential expression in placenta
villi samples and their corresponding maternal blood samples revealed
either unmethylated or hypomethylated promoter regions, thus ruling
out DNA methylation of these genes as a regulatory mechanism for
their differential expression. However, in case of pre-eclamptic villi,
hypermethylation of the c-myc promoter region was associated with
repressed c-myc expression. Such abnormal DNA methylation regulating
the enzymes modulating vitamin D homeostasis at the fetomaternal
interface has also been reported for Vitamin D deficiency associated
Epigenetic regulation of c-myc and hTERT in human placenta
1035
Figure 5 MS-HRM analysis for the human Telomerase reverse transcriptase (hTERT) promoter region. (a) Aligned melt curve and (b) Difference plots
normalized to the 0%-methylated standard DNA, for methylated standard curves of M100 to 0% (i.e. DNA standards with 100 to 0% methylation) represented in different colors, (c) Difference plots with methylated standard curve of M100 to 0% shown as black curves and differential fluorescence among few
representative samples of different placental villi groups, placental-derived cell lines and maternal leukocyte groups.
Figure 6 Chromatin immunoprecipitation (ChIP) for the c-myc promoter region in different placental villi groups. Relative fold enrichment in (a)
H3K9me3 and (b) H3K27me3, relative to negative control (normal rabbit IgG) and normalized with input DNA. The values are expressed as the mean
fold change + SEM; n ¼ 4 per group. *P , 0.05 with respect to 1st Trim, #P , 0.05 with respect to 2nd Trim and EEP , 0.01 with respect to 3rd
Trim. The data were compared using the Student’s t-test.
1036
Rahat et al.
Table II Correlation and regression analysis of DNA methylation, histone modifications and mRNA expression levels of
c-myc and hTERT in five placental villi groups.
Gene
Pearson’s correlation analysis: r
...........................................................................
H3K9me3
H3k27me3
mRNA versus DNA
methylation
Regression analysis with mRNA
expression as dependent
variable: r 2 (SEM)
.............................................................................................................................................................................................
C-MYC
mRNA
DNA
Methylation
hTERT
20.99***
0.46
20.82*
20.52
1** (0.000084)
UD
0.99* (0.004)
0.89*
mRNA
20.77
20.78
DNA
Methylation
UD
UD
The five placental villi groups included were normal first, second and third trimester, pre-eclampsia and molar villi. r ¼ Pearson’s coefficient, r 2 ¼ regression determination coefficient,
UD ¼ undetermined due to the unmethylated gene promoter. In case of regression analysis, the equation for prediction for each gene was ‘mRNA expression ¼ b1H3K9me3 +
b2H3K27me3 + b3DNA methylation + a’, where b1, b2, b3 are regression coefficients and a is the intercept of the test.
*P , 0.05, **P , 0.01 and ***P , 0.001.
Figure 7 ChIP for hTERT promoter region in different placental villi groups. Relative fold enrichment in (a) H3K9me3 and (b) H3K27me3, relative to
negative control (normal rabbit IgG) and normalized with input DNA. The values are expressed as the mean fold change + SEM; n ¼ 4 per group.
*P , 0.05 with respect to 1st Trim, ###P , 0.001 with respect to 2nd Trim and EP , 0.05, E EP , 0.01 with respect to 3rd Trim. The data were compared
using the Student’s t-test.
with PE (Bodnar et al., 2007; Novakovic et al., 2009). Lower expression of
c-myc in PE further supports the abnormal cytotrophoblast differentiation and invasion observed in PE (Brosens et al., 1972; Uzan et al.,
2011). Although methylation was observed in the c-myc promoter
region in first-trimester villi, it could not be correlated with the mRNA
expression of first-trimester villi, therefore emphasizing the role of
other epigenetic mechanisms in the regulation of differential expression.
Like DNA methylation, the level of histone modification is known to
be lower in the trophoblast lineage compared with the embryonic
lineage (Torres-Padilla et al., 2007) but the histone modifications in
normal extraembryonic development is indispensable, supported by
the failure of invasive trophoblast giant cell differentiation (Wang et al.,
2002) and amnion and chorion formation (Pasini et al., 2004) due to
mutations in PcG repressive complex (PRC) 2 mediating H3K27 methylation (Hemberger, 2007). Based on the importance of histone modifications in placental development, we analyzed H3K9me3 and H3K27me3
in the transcriptional regulation of c-myc and hTERT. These modifications
are known to play an important role in transcriptional silencing, being the
hallmarks of inactive DNA (Lachner and Jenuwein, 2002; Kondo et al.,
2008).
Histone modifications have been reported to regulate the placentalspecific expression of five genes of the human growth hormone cluster
with different roles for HAT and HMT co-activator complexes (Alsat
et al., 1998; Kimura et al., 2004), thus highlighting the special importance
of histone modifications in placental development. In support of the
role of histone modifications in placental gene regulation, we were able
to identify increased levels of H3K9me3 and H3K27me3 in normal placental villi at the c-myc and hTERT promoter, respectively, with advancing
gestation resulting in a decrease in the expression of these genes in
human placenta. Methylation of H3K9 has also been reported to be associated with a lack of hTERT expression in telomerase-negative cell lines
(Atkinson et al., 2005).
Epigenetic regulation of c-myc and hTERT in human placenta
1037
Figure 8 MS-HRM based evaluation of c-myc as an epigenetic marker. (a) Difference plots normalized to the 0%-methylated standard DNA, with methylated standard curves of M100 to 0% (i.e. DNA standards with 100 to 0% methylation), showing the differential fluorescence among few representative
samples of pre-eclampsia (PE) group. (b) Graphical representation of 5′ CpG Methylation among different pre-eclamptic placental villi samples (PE villi),
their corresponding maternal blood leukocytes (PE blood) and plasma samples collected previous to the obstetric procedure from pre-eclamptic
women (PE plasma) and women with normal pregnancy in the 3rd trimester (3rd Trim plasma). The data were compared using the Student’s t-test and
are expressed as the mean percentage of methylation + SEM. ***P , 0.001 with respect to the respective maternal blood leukocytes, for 30 patients
in each group.
Further, our study highlighted the role of trimethylation of H3 lysine residues in the development of PE and molar pregnancy. A lower level of
H3K27me3 at the promoter region of c-myc as well as hTERT was associated with molar pregnancy. Lower H3K9 methylation at the c-myc and
hTERT gene promoter is reported to be involved in their activation in leukemia and hepatocellular carcinoma, respectively (Foltankova et al., 2012;
Zhang et al., 2014). Lower methylation of H3K9 or H3K27 at the hTERT
promoter is also supported by the reported low level of repressive
histone marks and higher level of activating acetylation histone marks on
c-myc-target genes (Guccione et al., 2006; Martinato et al., 2008).
The contrasting mRNA expression levels of c-myc and hTERT in preeclamptic villi highlight some possible factor that induces opposite
effects on c-myc and hTERT expression in these cases. In our study
we have observed contrasting histone modifications as one such possible factor, with higher H3K27me3 levels at the c-myc promoter and
lower trimethylated H3 lysine residues at the hTERT promoter. Thus,
in case of PE, both promoter region DNA methylation and higher
H3K27me3 levels at the c-myc promoter might be the combined
epigenetic factors regulating the lower expression of c-myc in preeclamptic villous tissue. These results thus emphasize the possible
role of epigenetic dysregulation at the c-myc promoter in the development of PE, but further studies are required to confirm these findings.
Furthermore, PE (Caniggia et al., 1999) is reported to be associated
with hypoxia-like conditions, resulting in aberrant gene expression.
Hypoxia is associated with the expression of hypoxia-inducible factor
1a (HIF-1a) (Caniggia et al., 1999), which is involved in enhancing
hTERT expression (Zhang et al., 2003) and inducing the expression of
hTERT in a choriocarcinoma cell line (Nishi et al., 2004). Moreover,
hypoxia and HIF have also been reported to decrease c-myc expression
in human lung carcinoma and in renal cell carcinoma (Zhang et al.,
2007). Furthermore, in lung carcinoma hypoxia associated with
lower global acetylated histone4 was mediated by lower c-myc expression (Li and Costa, 2009). These studies thus support the higher
expression of hTERT reported previously (Geifman-Holtzman et al.,
2010) and lower expression of c-myc in PE as observed in the present
study. Although the inverse relation between expression of c-myc and
hTERT in pre-eclamptic placental villi seems to be associated with
hypoxia as well as the opposite epigenetic modifications for these
genes in this particular group, in case of early trimester villi where a
physiologically low oxygen environment prevails, the higher expression
of c-myc, as observed in this study and as reported previously (PfeiferOhlsson et al., 1984; Fulop et al., 1998), seems to be regulated by
histone modifications. Moreover, there is co-localization and higher
expression of the platelet-derived growth factor early in pregnancy,
which is known to stimulate the expression of c-myc (Goustin et al.,
1985; Osterlund et al., 1996). C-myc is part of the postreceptor intracellular signaling pathway for the stimulation of cell proliferation by
growth factors. Furthermore, the higher expression of these genes in
early trimester villi is also consistent with the higher invasive potential
of first-trimester trophoblasts (Goustin et al., 1985).
Analysis of hTERT and c-myc expression in placental tissue in reference
to maternal blood revealed significantly higher expression in placental
villi, which supports the known higher proliferative and malignant phenotype of placental villi. This observation is substantiated by a previous
study where longer telomeres are observed in placenta tissue than in
cord blood from the same donor (Allsopp et al., 2007).
1038
Based on these findings, we suggest a role for epigenetic mechanisms
in the regulation of c-myc and hTERT differential expression, with higher
expression in the first trimester, promoting its proliferative capacity and
thus its higher cancer-like phenotype. We also suggest that further dysregulation of this epigenetic modification is a potential factor in the development of PE and molar pregnancy, as supported by the high regression
coefficient observed in our study. Future studies establishing the role of
histone modifications in placental villi and maternal blood may further
elucidate the role of epigenetic mechanisms in placental development.
The presence of cell-free fetal DNA in maternal plasma holds promise
for the development of non-invasive prenatal genetic diagnostics since it
is a better source of fetal DNA than fetal cells in maternal plasma (which
are low in number) and is also unique to the current pregnancy, as it is
cleared from maternal plasma within 24 h after delivery. Cell-free fetal
DNA in maternal plasma has recently been analyzed to search for fetal
DNA epigenetic markers, such as RASSF1A and maspin (Chim et al.,
2005; Chan et al., 2006), with the potential of being used in physiological
pregnancies as well as pathological pregnancies; however, no specific epigenetic marker for pathological pregnancies has been identified to date.
In this regard, we studied the promoter region of c-myc in pre-eclamptic
maternal plasma because of its considerably higher level of methylation in
pre-eclamptic villi than in maternal DNA. Placenta is the most likely organ
for releasing fetal nucleic acids into maternal plasma (Alberry et al.,
2007), while maternal blood cells are the main source of background maternal DNA (Lui et al., 2002). The methylation pattern observed in circulating fetal DNA should be the same as that of the placental villi. Based on
our finding of a significantly higher methylation level of c-myc promoter in
maternal plasma, mimicking the methylation status of pre-eclamptic villi,
we suggest c-myc as a possible specific fetal epigenetic marker for preeclamptic pregnancy. However, in order to be of clinical value, our findings need to be validated and also repeated in early pregnancy. An earlier
study used the difference in median concentration of hypermethylated
RASSF1A to differentiate pre-eclamptic pregnancies from normal pregnancies (Zhao et al., 2010). This approach makes use of RASSF1A as an
epigenetic marker of PE and is dependent on the concentration of fetal
DNA in maternal plasma, which shows high between-subject variation,
whereas in our study we used the exact percentage of methylation of
c-myc in maternal plasma to differentiate pre-eclamptic pregnancies
from normal pregnancies. Furthermore, c-myc was observed to be
hypermethylated only in PE, making it a specific marker for PE.
In conclusion, this is a preliminary study reporting, for the first time, a
hypermethylated c-myc promoter in maternal plasma and requires
further validation in a larger population and in early pregnancy in order
to establish the use of c-myc as an early diagnostic marker for preeclamptic pregnancies in the future.
Acknowledgements
The authors acknowledge Prof. S.K. Gupta (Deputy Director, National
Institute of Immunology, Delhi, India) for providing the JEG-3 and
HTR-8/SVneo cell line. The authors also thank the pregnant women
included in this study for their cooperation.
Authors’ roles
The authors were responsible for the following aspects of the study. B.R.
contributed to the concept and study design, and participated in the
Rahat et al.
acquisition, analysis and interpretation of data from MS-HRM, quantitative real-time PCR, ChIP assay and cell culturing. A.H. contributed to the
study design and participated in the analysis of methylation data. R.A.N.
contributed to the acquisition of data from MS-HRM. R.B. contributed to
the study design and patient recruitment, and participated in the clinical
data collection. J.K. contributed to the concept, study design and interpretation of all data. The manuscript was prepared by B.R. and J.K. and
further critically revised and approved by all authors.
Funding
This work was supported by the Indian Council of Medical Education and
Research (ICMR, Project reference number: 5/10/FR/3/2010-RHN).
B.R. was a recipient of Senior Research Fellowship (reference number:
45/2/2010-BMS/CMB) from the Division of Basic Medical Science,
ICMR.
Conflict of interest
None declared.
References
Alberry M, Maddocks D, Jones M, Abdel Hadi M, Abdel-Fattah S, Avent N,
Soothill PW. Free fetal DNA in maternal plasma in anembryonic
pregnancies: confirmation that the origin is the trophoblast. Prenat Diagn
2007;27:415 – 418.
Allsopp R, Shimoda J, Easa D, Ward K. Long telomeres in the mature human
placenta. Placenta 2007;28:324– 327.
Alsat E, Guibourdenche J, Couturier A, Evain-Brion D. Physiological role of
human placental growth hormone. Mol Cell Endocrinol 1998;
140:121– 127.
Atkinson SP, Hoare SF, Glasspool RM, Keith WN. Lack of telomerase gene
expression in alternative lengthening of telomere cells is associated with
chromatin remodeling of the hTR and hTERT gene promoters. Cancer
Res 2005;65:7585 – 7590.
Bamberger AM, Bamberger CM, Aupers S, Milde-Langosch K, Loning T,
Makrigiannakis A. Expression pattern of the activating protein-1 family of
transcription factors in the human placenta. Mol Hum Reprod 2004;
10:223 – 228.
Bernstein BE, Meissner A, Lander ES. The mammalian epigenome. Cell 2007;
128:669– 681.
Bischof P, Meisser A, Campana A. Biochemistry and molecular biology of
trophoblast invasion. Ann N Y Acad Sci 2001;943:157– 162.
Bodnar LM, Catov JM, Simhan HN, Holick MF, Powers RW, Roberts JM.
Maternal vitamin D deficiency increases the risk of PE. J Clin Endocrinol
Metab 2007;92:3517– 3522.
Brosens IA, Robertson WB, Dixon HG. The role of the spiral arteries in the
pathogenesis of PE. Obstet Gynecol Annu 1972;1:177– 191.
Caniggia I, Grisaru-Gravnosky S, Kuliszewsky M, Post M, Lye SJ. Inhibition of
TGF-beta 3 restores the invasive capability of extravillous trophoblasts in
preeclamptic pregnancies. J Clin Invest 1999;103:1641 – 1650.
Chan KC, Ding C, Gerovassili A, Yeung SW, Chiu RW, Leung TN, Lau TK,
Chim SS, Chung GT, Nicolaides KH et al. Hypermethylated RASSF1A in
maternal plasma: a universal fetal DNA marker that improves the reliability
of noninvasive prenatal diagnosis. Clin Chem 2006;52:2211–2218.
Chen RJ, Chu CT, Huang SC, Chow SN, Hsieh CY. Telomerase activity in
gestational trophoblastic disease and placental tissue from early and late
human pregnancies. Hum Reprod 2002;17:463– 468.
Epigenetic regulation of c-myc and hTERT in human placenta
Chim SS, Tong YK, Chiu RW, Lau TK, Leung TN, Chan LY, Oudejans CB,
Ding C, Lo YM. Detection of the placental epigenetic signature of the
maspin gene in maternal plasma. Proc Natl Acad Sci USA 2005;102:
14753–14758.
Chiu RW, Poon LL, Lau TK, Leung TN, Wong EM, Lo YM. Effects of
blood-processing protocols on fetal and total DNA quantification in
maternal plasma. Clin Chem 2001;47:1607– 1613.
Chiu RW, Chim SS, Wong IH, Wong CS, Lee WS, To KF, Tong JH, Yuen RK,
Shum AS, Chan JK et al. Hypermethylation of RASSF1A in human and
rhesus placentas. Am J Pathol 2007;170:941 – 950.
Costello JF, Fruhwald MC, Smiraglia DJ, Rush LJ, Robertson GP, Gao X,
Wright FA, Feramisco JD, Peltomaki P, Lang JC et al. Aberrant
CpG-island methylation has non-random and tumour-type-specific
patterns. Nat Genet 2000;24:132– 138.
Cross JC, Werb Z, Fisher SJ. Implantation and the placenta: key pieces of the
development puzzle. Science 1994;266:1508 –1518.
D’Alessio AC, Szyf M. Epigenetic tete-a-tete: the bilateral relationship
between chromatin modifications and DNA methylation. Biochem Cell
Biol 2006;84:463 – 476.
Dokras A, Coffin J, Field L, Frakes A, Lee H, Madan A, Nelson T, Ryu GY,
Yoon JG. Epigenetic regulation of maspin expression in the human
placenta. Mol Hum Reprod 2006;12:611 – 617.
Esteller M. Cancer epigenomics: DNA methylomes and histone-modification
maps. Nat Rev Genet 2007;8:286 – 298.
Even-Ram S, Uziely B, Cohen P, Grisaru-Granovsky S, Maoz M, Ginzburg Y,
Reich R, Vlodavsky I, Bar-Shavit R. Thrombin receptor overexpression in
malignant and physiological invasion processes. Nat Med 1998;
4:909 – 914.
Ferretti C, Bruni L, Dangles-Marie V, Pecking AP, Bellet D. Molecular circuits
shared by placental and cancer cells, and their implications in the
proliferative, invasive and migratory capacities of trophoblasts. Hum
Reprod Update 2007;13:121 – 141.
Fisher SJ, Cui TY, Zhang L, Hartman L, Grahl K, Zhang GY, Tarpey J,
Damsky CH. Adhesive and degradative properties of human placental
cytotrophoblast cells in vitro. J Cell Biol 1989;109:891 – 902.
Foltankova V, Legartova S, Kozubek S, Bartova E. Tumor-specific histone
signature and DNA methylation in multiple myeloma and leukemia cells.
Neoplasma 2012;59:450 – 462.
Fraga MF, Ballestar E, Villar-Garea A, Boix-Chornet M, Espada J, Schotta G,
Bonaldi T, Haydon C, Ropero S, Petrie K et al. Loss of acetylation at
Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark
of human cancer. Nat Genet 2005;37:391 – 400.
Fulop V, Mok SC, Genest DR, Szigetvari I, Cseh I, Berkowitz RS. c-myc,
c-erbB-2, c-fms and bcl-2 oncoproteins. Expression in normal placenta,
partial and complete mole, and choriocarcinoma. J Reprod Med 1998;
43:101– 110.
Gearhart J, Pashos EE, Prasad MK. Pluripotency redux—advances in
stem-cell research. N Engl J Med 2007;357:1469– 1472.
Geifman-Holtzman O, Xiong Y, Holtzman EJ, Hoffman B, Gaughan J,
Liebermann DA. Increased placental telomerase mRNA in hypertensive
disorders of pregnancy. Hypertens Pregnancy 2010;29:434 – 445.
Gomez DE, Armando RG, Farina HG, Menna PL, Cerrudo CS,
Ghiringhelli PD, Alonso DF. Telomere structure and telomerase in
health and disease (review). Int J Oncol 2012;41:1561 – 1569.
Goustin AS, Betsholtz C, Pfeifer-Ohlsson S, Persson H, Rydnert J, Bywater M,
Holmgren G, Heldin CH, Westermark B, Ohlsson R. Coexpression of the
sis and myc proto-oncogenes in developing human placenta suggests
autocrine control of trophoblast growth. Cell 1985;41:301– 312.
Guccione E, Martinato F, Finocchiaro G, Luzi L, Tizzoni L, Dall’ Olio V,
Zardo G, Nervi C, Bernard L, Amati B. Myc-binding-site recognition in
the human genome is determined by chromatin context. Nat Cell Biol
2006;8:764 – 770.
1039
Hemberger M. Epigenetic landscape required for placental development. Cell
Mol Life Sci 2007;64:2422 – 2436.
Hoffner L, Surti U. The genetics of gestational trophoblastic disease: a rare
complication of pregnancy. Cancer Genet 2012;205:63 – 77.
Horikawa I, Cable PL, Afshari C, Barrett JC. Cloning and characterization of
the promoter region of human telomerase reverse transcriptase gene.
Cancer Res 1999;59:826 – 830.
Janneau JL, Maldonado-Estrada J, Tachdjian G, Miran I, Motte N, Saulnier P,
Sabourin JC, Cote JF, Simon B, Frydman R et al. Transcriptional expression
of genes involved in cell invasion and migration by normal and tumoral
trophoblast cells. J Clin Endocrinol Metab 2002;87:5336 – 5339.
Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PL,
Coviello GM, Wright WE, Weinrich SL, Shay JW. Specific association of
human telomerase activity with immortal cells and cancer. Science 1994;
266:2011– 2015.
Kimura AP, Liebhaber SA, Cooke NE. Epigenetic modifications at the human
growth hormone locus predict distinct roles for histone acetylation and
methylation in placental gene activation. Mol Endocrinol 2004;18:1018–1032.
Klose RJ, Bird AP. Genomic DNA methylation: the mark and its mediators.
Trends Biochem Sci 2006;31:89 –97.
Kondo Y, Shen L, Cheng AS, Ahmed S, Boumber Y, Charo C, Yamochi T,
Urano T, Furukawa K, Kwabi-Addo B et al. Gene silencing in cancer by
histone H3 lysine 27 trimethylation independent of promoter DNA
methylation. Nat Genet 2008;40:741 – 750.
Lachner M, Jenuwein T. The many faces of histone lysine methylation. Curr
Opin Cell Biol 2002;14:286 – 298.
Lawler SD, Pickthall VJ, Fisher RA, Povey S, Evans MW, Szulman AE. Genetic
studies of complete and partial hydatidiform moles. Lancet 1979;2:580.
Li Q, Costa M. c-Myc mediates a hypoxia-induced decrease in acetylated
histone H4. Biochimie 2009;91:1307 – 1310.
Livak KJ, Schmittgen TD. Analysis of relative gene expression data using
real-time quantitative PCR and the 2(-Delta Delta C(T)) Method.
Methods 2001;25:402– 408.
Lui YY, Chik KW, Chiu RW, Ho CY, Lam CW, Lo YM. Predominant
hematopoietic origin of cell-free DNA in plasma and serum after sexmismatched bone marrow transplantation. Clin Chem 2002;48:421–427.
Martinato F, Cesaroni M, Amati B, Guccione E. Analysis of Myc-induced
histone modifications on target chromatin. PLoS One 2008;3:e3650.
Migheli F, Stoccoro A, Coppede F, Wan Omar WA, Failli A, Consolini R,
Seccia M, Spisni R, Miccoli P, Mathers JC et al. Comparison study of
MS-HRM and pyrosequencing techniques for quantification of APC and
CDKN2A gene methylation. PLoS One 2013;8:e52501.
Morales CP, Holt SE, Ouellette M, Kaur KJ, Yan Y, Wilson KS, White MA,
Wright WE, Shay JW. Absence of cancer-associated changes in
human fibroblasts immortalized with telomerase. Nat Genet 1999;21:
115 – 118.
Nakamura TM, Cech TR. Reversing time: origin of telomerase. Cell 1998;
92:587– 590.
Nishi H, Nakada T, Kyo S, Inoue M, Shay JW, Isaka K. Hypoxia-inducible
factor 1 mediates upregulation of telomerase (hTERT). Mol Cell Biol
2004;24:6076 – 6083.
Novakovic B, Rakyan V, Ng HK, Manuelpillai U, Dewi C, Wong NC,
Morley R, Down T, Beck S, Craig JM et al. Specific tumour-associated
methylation in normal human term placenta and first-trimester
cytotrophoblasts. Mol Hum Reprod 2008;14:547 – 554.
Novakovic B, Sibson M, Ng HK, Manuelpillai U, Rakyan V, Down T, Beck S,
Fournier T, Evain-Brion D, Dimitriadis E et al. Placenta-specific
methylation of the vitamin D 24-hydroxylase gene: implications for
feedback autoregulation of active vitamin D levels at the fetomaternal
interface. J Biol Chem 2009;284:14838– 14848.
Ohlsson R. Growth factors, protooncogenes and human placental
development. Cell Differ Dev 1989;28:1– 15.
1040
Osterlund C, Wramsby H, Poussete A. Temporal expression of PDGF-A and
its receptor in human pre-implantation embryos. Mol Hum Reprod 1996;
2:507 – 512.
Pasini D, Bracken AP, Jensen MR, Lazzerini Denchi E, Helin K. Suz12 is
essential for mouse development and for EZH2 histone methyltransferase
activity. EMBO J 2004;23:4061–4071.
Pfeifer-Ohlsson S, Goustin AS, Rydnert J, Wahlstrom T, Bjersing L,
Stehelin D, Ohlsson R. Spatial and temporal pattern of cellular myc
oncogene expression in developing human placenta: implications for
embryonic cell proliferation. Cell 1984;38:585– 596.
Pijnenborg R, Bland JM, Robertson WB, Dixon G, Brosens I. The pattern of
interstitial trophoblastic invasion of the myometrium in early human
pregnancy. Placenta 1981;2:303 –316.
Rahnama F, Shafiei F, Gluckman PD, Mitchell MD, Lobie PE. Epigenetic
regulation of human trophoblastic cell migration and invasion.
Endocrinology 2006;147:5275 – 5283.
Rama S, Suresh Y, Rao AJ. Regulation of telomerase during human
trophoblast differentiation: a role for TGF b1. Mol Cell Endocrinol 2001;
182:233– 248.
Redman CW. Cytotrophoblasts: masters of disguise. Nat Med 1997;
3:610 – 611.
Roncalli M, Bulfamante G, Viale G, Springall DR, Alfano R, Comi A,
Maggioni M, Polak JM, Coggi G. C-myc and tumour suppressor gene
product expression in developing and term human trophoblast. Placenta
1994;15:399 – 409.
Serman L, Vlahovic M, Sijan M, Bulic-Jakus F, Serman A, Sincic N, Matijevic R,
Juric-Lekic G, Katusic A. The impact of 5-azacytidine on placental weight,
glycoprotein pattern and proliferating cell nuclear antigen expression in rat
placenta. Placenta 2007;28:803– 811.
Song J, Noh JH, Lee JH, Eun JW, Ahn YM, Kim SY, Lee SH, Park WS, Yoo NJ,
Lee JY et al. Increased expression of histone deacetylase 2 is found in
human gastric cancer. APMIS 2005;113:264 – 268.
Soundararajan R, Rao AJ. Trophoblast ‘pseudo-tumorigenesis’: significance
and contributory factors. Reprod Biol Endocrinol 2004;2:15.
Strickland S, Richards WG. Invasion of the trophoblasts. Cell 1992;
71:355 – 357.
Rahat et al.
Suman P, Gupta SK. Comparative analysis of the invasion-associated genes
expression pattern in first trimester trophoblastic (HTR-8/SVneo) and
JEG-3 choriocarcinoma cells. Placenta 2012;33:874 – 877.
Torres-Padilla ME, Parfitt DE, Kouzarides T, Zernicka-Goetz M. Histone
arginine methylation regulates pluripotency in the early mouse embryo.
Nature 2007;445:214 – 218.
Uzan J, Carbonnel M, Piconne O, Asmar R, Ayoubi JM. Pre-eclampsia:
pathophysiology, diagnosis, and management. Vasc Health Risk Manag
2011;7:467 – 474.
Wang J, Mager J, Schnedier E, Magnuson T. The mouse PcG gene eed is
required for Hox gene repression and extraembryonic development.
Mamm Genome 2002;13:493 – 503.
Wojdacz TK, Dobrovic A. Methylation-sensitive high resolution melting
(MS-HRM): a new approach for sensitive and high-throughput
assessment of methylation. Nucleic Acids Res 2007;35:e41.
Wojdacz TK, Dobrovic A, Hansen LL. Methylation-sensitive high-resolution
melting. Nat Protoc 2008;3:1903 – 1908.
Wong NC, Novakovic B, Weinrich B, Dewi C, Andronikos R, Sibson M,
Macrae F, Morley R, Pertile MD, Craig JM et al. Methylation of
the adenomatous polyposis coli (APC) gene in human placenta
and hypermethylation in choriocarcinoma cells. Cancer Lett 2008;
268:56– 62.
Zhang P, Chan SL, Fu W, Mendoza M, Mattson MP. TERT suppresses
apoptotis at a premitochondrial step by a mechanism requiring reverse
transcriptase activity and 14 – 3 – 3 protein-binding ability. FASEB J 2003;
17:767 – 769.
Zhang H, Gao P, Fukuda R, Kumar G, Krishnamachary B, Zeller KI, Dang CV,
Semenza GL. HIF-1 inhibits mitochondrial biogenesis and cellular
respiration in VHL-deficient renal cell carcinoma by repression of
C-MYC activity. Cancer Cell 2007;11:407 – 420.
Zhang B, Chen J, Cheng AS, Ko BC. Depletion of sirtuin 1 (SIRT1) leads to
epigenetic modifications of telomerase (TERT) gene in hepatocellular
carcinoma cells. PLoS One 2014;9:e84931.
Zhao F, Wang J, Liu R, Yang J, Cui K, Wu Y, Guo J, Mu Y, Wang X.
Quantification and application of the placental epigenetic signature
of the RASSF1A gene in maternal plasma. Prenat Diagn 2010;30:778– 782.