Content uploaded by Marcos J Araúzo-Bravo
Author content
All content in this area was uploaded by Marcos J Araúzo-Bravo
Content may be subject to copyright.
Author contributions: D.W.H.: Conception and design, data collection and/or assembly, data analysis and interpretation, manuscript writing; J.T.D.:
Data analysis and interpretation, data collection and/or assembly; M.J.A.-B.: Data analysis and interpretation; S.H.L.: Data analysis and
interpretation; A.M.: Data analysis and interpretation, provision of study materials; H.T.L.: Data analysis and interpretation; R.J.: Data analysis and
interpretation, provision of study materials; H.R.S.: Conception and design, data analysis and interpretation, financial support, manuscript writing.
4Corresponding author: Hans R. Schöler, Ph. D., Email: h.schoeler@mpi-muenster.mpg.de, Tel: 49-251-70365-300, Fax: 49-251-70365-399.
Received August 14, 2008; accepted for publication February 10, 2009; first published online in Stem Cells Express February 19, 2009. ©AlphaMed
Press 1066-5099/2009/$30.00/0 doi: 10.1002/stem.43
STEM CELLS®
E
MBRYONIC
S
TEM
C
ELLS
/I
NDUCED
P
LURIPOTENT
S
TEM
C
ELLS
Epigenetic Hierarchy Governing Nestin Expression
Dong Wook Han1,2, Jeong Tae Do1, Marcos J. Araúzo-Bravo1, Sung Ho Lee2, Alexander
Meissner3, Hoon Taek Lee2, Rudolf Jaenisch3, and Hans R. Schöler1,4
1Department of Cell and Developmental Biology, Max Planck Institute for Molecular Biomedicine, Röntgenstrasse 20,
48149 Münster, Germany; 2Department of Bioscience and Biotechnology, Bio-Organ Research Center, Konkuk
University, 1 Hwayang-dong, Gwangjin-Gu, Seoul 143-701, South Korea; 3Whitehead Institute and Department of
Biology, Massachusetts Institute of Technology, 9 Cambridge Center, Cambridge, MA 02142, USA
Key words. DNA methylation/epigenetic regulation/histone acetylation/Nestin
ABSTRACT
Nestin is an intermediate filament protein expressed
specifically in neural stem cells and progenitor cells
of the central nervous system. DNA demethylation
and histone modifications are two types of epigenetic
modifications working in a coordinate or synergistic
manner to regulate the expression of various genes.
This study investigates and elucidates the epigenetic
regulation of Nestin gene expression during
embryonic differentiation along the neural cell
lineage. Nestin exhibits differential DNA methylation
and histone acetylation patterns in Nestin-expressing
and non-expressing cells. In P19 embryonic
carcinoma (EC) cells, activation of Nestin expression
is mediated by both trichostatin A (TSA) and 5-aza-
2’-deoxycytidine (5-aza-dC) treatment, concomitant
with histone acetylation, but not with DNA
demethylation. Nestin transcription is also mediated
by treatment with retinoic acid (RA), again in the
absence of DNA demethylation. Thus, histone
acetylation is sufficient to mediate activation of
Nestin transcription. This study proposes that
regulation of Nestin gene expression can be used as a
model to study the epigenetic regulation of gene
expression mediated by histone acetylation, but not
by DNA demethylation.
INTRODUCTION
Nestin is an intermediate filament protein
expressed specifically in neural stem cells
(NSCs) and neural progenitor cells of the
central nervous system (CNS) ranging from
embryonic to adult stages [1] and thus has been
widely used as a marker gene to identify and
isolate NSCs [2-4]. As NSCs differentiate into
neuronal and glial cells during CNS
development, Nestin expression diminishes and
the protein is gradually replaced by
neurofilaments in neurons and glial fibrillary
acidic proteins (GFAP) in glial cells [5].
However, Nestin is also expressed in myogenic
precursor cells [6] and in developing heart [7],
differentiating testis [8], and developing tooth
bud [9] tissues during embryogenesis.
The Nestin gene is well-conserved among
species containing four exons and three introns
[2, 4, 10]. Two different tissue-specific
enhancers located in the first and second
introns of the Nestin gene drive Nestin
expression in somatic muscle progenitor cells
and NSCs, respectively. The CNS-specific
second intron enhancer is considered to be an
important element in the regulation of Nestin
expression [11]. In addition, an enhancer
element located in the 3’ half of the second
intron enhancer called the 3’enhancer is also
known to be involved in Nestin regulation in
the CNS [12-14]. A recent study has shown
that activation of Nestin expression is
dependent upon the binding of transcription
factors, such as SOX and POU factors, to the
Nestin 3’ enhancer [15]. However, although
factors involved in the transcriptional
regulation of Nestin expression have been
identified, evidence for their participation in an
epigenetic mechanism has remained elusive.
Epigenetic modifications such as DNA
methylation and chromatin remodeling are
important components in the transcriptional
regulation of many genes. DNA demethylation
and histone modifications have been shown to
play a synergic role in the activation of gene
transcription [16-19]. Despite the intense
characterization and the widespread use of
Nestin as a marker for NSCs, relatively little is
known about the epigenetic mechanism
governing Nestin transcription. The current
study aims to determine whether DNA
methylation and histone modifications are
involved in the regulation of Nestin regulation.
To this end, the study investigated 1) the effect
of inhibiting DNA methylation and histone
deacetylation (i.e. of using histone deacetylase
[HDAC]) on Nestin transcription; 2) the
methylation status of the Nestin regulatory
elements in both Nestin-expressing and non-
expressing cells; 3) the effect of inhibiting
DNA methylation and histone deacetylation on
DNA methylation patterns of Nestin regulatory
elements; 4) the transcriptional status of Nestin
in DNA methyltransferases knock-out (Dnmts
KO) embryonic stem (ES) cells; 5) the histone
acetylation status of Nestin regulatory elements
in both Nestin-expressing and non-expressing
cells; and 6) the epigenetic changes in Nestin
mediated by inhibition of DNA methylation
and histone deacetylation and by
differentiation induced by retinoic acid (RA).
MATERIALS AND METHODS
Mice
All mouse strains were either bred and housed
at the mouse facility of the Max Planck
Institute (MPI) or were bought from Halan or
Jackson laboratories. Animal handling was in
accordance with the MPI animal protection
guidelines and the German animal protection
laws.
Cell culture and treatment with reagents
P19 EC cells were grown on gelatin-coated
(0.1% in PBS) dishes in standard EC cell
media: high-glucose DMEM (Gibco BRL)
containing 15% fetal calf serum (FCS; Gibco
BRL), 1X penicillin/streptomycin/glutamine,
and 1X nonessential amino acids (Gibco BRL).
For the derivation of NSCs, brain tissue was
collected from 12.5- to 16.5-dpc
OG2/ROSA26 heterozygous female mice
according to a previous protocol [31]. Isolated
NSCs were grown in NS-A medium
(Euroclone) supplemented with N2
supplement, 10 ng/ml EGF, 10 ng/ml bFGF
(both from Invitrogen), 50 µg/ml BSA
(Fraction V; Gibco BRL), and 1X
penicillin/streptomycin/glutamine and 1X
nonessential amino acids (Gibco BRL).
P19 EC cells were first pre-cultured for 48 hrs
in standard EC cell media and then cultured
with either 5-aza-dC (5 ȝM and 10 ȝM) for 8,
16, 24, 48, and 72 hrs or with TSA (200 nM)
for 4, 6, 8, and 16 hrs. Pre-cultured P19 EC
cells were also treated with retinoic acid (RA;
5ȝM) for 12, 24, and 48 hrs.
RNA extraction, cDNA synthesis, and RT-
PCR
Total RNA was isolated from each sample
using the ToTALLY RNA kit (Ambion).
Reverse transcription was performed with 1 ȝg
of DNase-treated total RNA using the
RETROscript kit (Ambion) following the
manufacturer’s protocol. cDNA was amplified
by PCR (35 cycles) using SuperTaq
polymerase (Ambion) with the primers
described in a previous publication by this
laboratory [32].
Bisulfite sequencing analysis
To assess the DNA methylation status of
Nestin regulatory elements, genomic DNA was
treated with sodium bisulfite to convert all
unmethylated cytosine residues into uracil ones
using the One Day MSP kit (In2Gen, Korea)
according to a previous protocol [33]. In brief,
purified genomic DNA (0.5-1 ȝg) was
2
Epigenetic hierarchy governing
Nestin
expression
denatured with 3N NaOH (Sigma-Aldrich,
USA) at 37ºC for 10 min. Sodium bisulfite (5
M) was then added and the mixture was
incubated at 50ºC for 16 hrs in the dark.
Following desulfonation, neutralization, and
desalting, the modified DNA was dissolved in
20 ȝl of distilled water. Bisulfite PCR
amplification was then carried out using
aliquots of 1 ȝl of modified DNA per PCR
reaction.
PCR amplification and sequencing of the
bisulfite-treated genomic DNA
In this study, the methylation status of the
Nestin promoter and the 5’ and 3’enhancers
was compared in Nestin-expressing and non-
expressing cells. Following bisulfite treatment,
all genomic regions selected were amplified
using a nested primer approach. PCR
amplifications were performed using SuperTaq
polymerase (Ambion) in a total volume of 25
ȝl. All PCR amplifications consisted of a total
of 50 cycles of denaturation at 94ºC for 30s,
annealing at the appropriate temperature for
each target region for 30s, extension at 72ºC
for 30s with a 1st denaturation at 94ºC for 5
min, and a final extension at 72ºC for 10 min.
The primer sequences and annealing
temperatures used were as follows: Nestin
promoter 1st sense 5’-
GTTTTGGGTTGTTTGGTTGTATT-3’,
Nestin promoter 1st antisense 5’-
ATAACCCTTAAAACTTTTAAAAAAAA-3’
(298 bp, 45ºC), Nestin promoter 2nd sense 5’-
TTTGTTGAGTTGGGATGATGTAG-3’,
Nestin promoter 2nd antisense 5’-
TAACCCTTAAAACTTTTAAAAAAAA-3’
(202 bp, 50ºC), 5’enhancer 1st sense 5’-
TAAAGAGGTTGTTTGGTTTGGTAGT-3’,
5’enhancer 1st antisense 5’-
CTATTCCACTCAACCTTCCTAAAAC-3’
(394 bp, 45ºC), 5’enhancer 2nd sense 5’-
TAGTTTTTAGGGAGGAGATTAGAGG-3’,
5’enhancer 2nd antisense 5’-
CTCTTACCCCAAACACAACTAAAAC-3’
(188 bp, 58ºC), 3’enhancer 1st sense 5’-
TTTGTTATTAGTTTTGGGGGTTTAG-3’,
3’enhancer 1st antisense 5’-
CATATCCTACCACTACAAAATCACTCTT
-3’ (387 bp, 45ºC), 3’enhancer 2nd sense 5’-
TTTTTAGATGGTAGTGTGGATAAAAG-
3’, and 3’enhancer 2nd antisense 5’-
CCCATATTTAAAACTCAAAAAAACC-3’
(203 bp, 58ºC). For each primer set, 3 ȝl of
product from the first round of PCR was used
in the second round of PCR. The amplified
products were verified by electrophoresis on
1% agarose gel. The PCR products were
subcloned using the PCR 2.1-TOPO vector
(Invitrogen) according to the manufacturer’s
protocol. Reconstructed plasmids were purified
using the QIAprep Spin Miniprep kit (Qiagen)
and individual clones were sequenced (GATC-
biotech, Germany). Clones were only accepted
if there was at least 90% cytosine conversion
and all possible clonalities were excluded
based on the criteria from BiQ Analyzer
software (Max Planck Society, Germany). For
each fusion hybrid, results were confirmed by
performing at least 10 replicates per selected
genomic region and at least three separate
bisulfite treatments.
ChIP analysis
A chromatin immunoprecipitation assay was
performed using anti-acetylated histone H3
(06-599) and H4 (06-598) antibodies (Upstate
Biotechnology). In brief, cells were rinsed with
phosphate-buffered saline (PBS) and treated
with 1% formaldehyde at 37ºC for 10 min to
cross-link histones and proteins. Each sample
was first sonicated to obtain DNA fragments
that were 200- to 1,000-bp long and then
incubated with or without the antibodies
overnight at 4ºC to detect specific or
nonspecific binding, respectively. Heating at
65ºC for 4 hrs reversed covalent histone-DNA
bonds. DNA was recovered by
phenol/chloroform extraction and amplified by
PCR (35 cycles) with the following primer
sets: Hprt sense 5’-
TCAGGCCCACCTAGTCAGAT-3’, Hprt
antisense 5’-
CGGAAAGCAGTGAGGTAAGC-3’ (198 bp,
62ºC), Nestin promoter sense 5’-
TCCCTTTCTCCCCCTTAAAA-3’, Nestin
promoter antisense 5’-
TGAGCTCCCACATCTGAAAA-3’ (133 bp,
52ºC), 5’enhancer-1 sense 5’-
AGGGAGGAAACCAGAGGGTA-3’,
5’enhancer-1 antisense 5’-
AAGGAAGGCAGAAGGCTAGG-3’ (207 bp,
62ºC), 5’enhancer-2 sense 5’-
AACCACTGAGCCATCTGTCC-3’,
3
5’enhancer-2 antisense 5’-
TGCCTGCCTTTCAAAGAACT-3’ (270 bp,
62ºC), 3’enhancer-1 sense 5’-
GCCCCAGTCAGTCTTCTGAG-3’,
3’enhancer-1 antisense 5’-
TGGGAATTCTCAGGCTGTTC-3’ (181 bp,
62ºC), 3’enhancer-2 sense 5’-
CATGGGAACAGCCTGAGAAT-3’,
3’enhancer-2 antisense 5’-
AGGGCCAAGTGAATTGCTAA-3’ (205 bp,
58ºC), Oct4 sense 5’-
CCTCCGTCTGGAAGACACAGGCAGATA
GCG-3’, and Oct4 antisense 5’-
CGAAGTCTGAAGCAGGTGTCCAGCCAT
GG-3’ (221 bp, 55ºC). All PCR reactions were
performed in triplicate with at least two
independently prepared samples. Each PCR
product was quantitated using an ethidium
bromide–stained gel image and fold
enrichment was calculated using the
Alphaimager HP software (Hewlett Packard).
RESULTS
Activation of Nestin expression mediated by
epigenetic modifications
Nestin is expressed by cells of neural origin—
such as neural stem cells and neural progenitor
cells—but not by non-neural embryonic
carcinoma (EC) cells as determined by RT-
PCR (Fig. 1), consistent with previous results
[20]. Since P19 EC cells are frequently used as
an in-vitro model system to study
developmental milestone events, such as the
differentiation of neurons [21], P19 EC cells
were treated with reagents that induce
epigenetic modifications to determine whether
Nestin undergoes epigenetic regulation. Cells
were treated with either 5-aza-2’-deoxycytidine
(5-aza-dC), an inhibitor of DNA methylation,
for 8, 16, 24, 48, and 72 hrs or trichostatin A
(TSA), an inhibitor of histone deacetylation,
for 4, 6, 8, and 16 hrs. Nestin mRNA was first
detected after 4 hrs of TSA treatment in Nestin
non-expressing P19 EC cells; i.e. time-
dependent expression (Fig. 1A, Fig. S2A). F9
EC cells, another type of EC cell, also
exhibited the same expression pattern (Fig.
1A). Activation of Nestin expression was also
observed in P19 EC cells treated with different
concentrations of 5-aza-dC for 4, 6, 8, 16, 24,
48, and 72 hrs. Nestin mRNA was first
detected after 24 hrs of 5-aza-dC treatment, in
a time- and dose-dependent fashion (Fig. 1B,
Fig. S2B). These data were also confirmed
with F9 EC cells (Fig. 1B). The time-
dependent expression of Nestin mRNA
following inhibition of DNA methylation (Fig.
1B) and histone deacetylation (Fig. 1A) attests
to the multitude of regulatory mechanisms
governing Nestin expression. However, since
both TSA and 5-aza-dC could mediate
activation of Nestin expression in Nestin non-
expressing P19 EC cells, it was postulated that
Nestin expression may be epigenetically
regulated at both the DNA and histone levels.
DNA methylation status of promoter and
second intron enhancer regions of Nestin
DNA methylation is thought to play an
important regulatory role in Nestin expression,
suggesting that Nestin regulatory elements are
differentially methylated in Nestin-expressing
and non-expressing cells. To confirm this
hypothesis, the DNA methylation status of
Nestin-expressing NSCs was compared with
that of non-expressing P19 EC cells by
bisulfite sequencing analysis. The results
showed that the Nestin promoter region (-161
to + 183 bp), which exhibits the highest
activity among promoter regions [22], was
completely unmethylated in both NSCs and
P19 EC cells (Fig. 2A). Previous studies had
determined the second intron enhancer (i.e.
CNS enhancer) located in the second intron of
Nestin to be a critical regulatory element of
Nestin expression, as the 5’ and 3’enhancer
elements of the second intron enhancer play
different roles in the regulation of Nestin
transcription [11]. In particular, the 3’enhancer
element, which harbors SOX and POU binding
sites, has been shown to be essential in Nestin
regulation [15]. Thus, the DNA methylation
status of both 5’ and 3’enhancers in Nestin-
expressing NSCs was compared with that of
non-expressing P19 EC cells. The 3’enhancer,
along with the promoter region, was found to
be completely unmethylated in both NSCs and
P19 EC cells, while the 5’enhancer was
differentially methylated in NSCs and P19 EC
cells (Fig. 2A). Nestin-expressing NSCs were
found to be nearly completely unmethylated
(1.5% methylation), but P19 EC cells, which
do not express Nestin, were considerably
4
Epigenetic hierarchy governing
Nestin
expression
methylated (68.6% methylation). This result
was also confirmed with other embryonic
cells—F9 EC cells (methylation rate: 58.6%)
and ES cells (methylation rate: 73.3%). Taken
together, these data suggest that DNA
methylationídependent gene silencing
regulates Nestin expression.
DNA demethylation is not sufficient to
mediate activation of Nestin transcription in
EC cells
To delve deeper into the series of events
involved in Nestin regulation, first the DNA
methylation status of Nestin regulatory
elements was assessed in 5-aza-dCítreated P19
EC cells. Onset of demethylation of the Nestin
5’enhancer element in P19 EC cells was
observed only after 16 hrs of 5-aza-dC
treatment, suggesting that 5-aza-dC treatment
caused cell cycle–dependent passive
demethylation of the Nestin 5’enhancer (Fig.
2B). This demethylation pattern was similarly
observed for the Nestin 5’enhancer in 5-aza-
dCítreated F9 EC cells (Fig. 2B). Next, the
DNA methylation status of the Nestin
5’enhancer was assessed in TSA-treated P19
EC cells. TSA can induce not only gene-
specific, but also global DNA demethylation
[23, 24]; TSA-induced demethylation has been
attributed to Dnmt1 and Dnmt3b down-
regulation [25, 26]. Onset of Nestin 5’enhancer
demethylation was observed in P19 EC cells
after 16 hrs of TSA treatment, with a
methylation rate of only 30%, which is about
2-fold lower than that in untreated P19 EC
cells and even lower than that in P19 EC cells
after 24 hrs of 5-aza-dC treatment (Fig. 2B, C).
The same pattern was observed in F9 EC cells
(Fig. 2B, C). But since the onset of Nestin
expression in P19 EC cells occurs after 4 hrs of
TSA treatment, demethylation of Nestin
regulatory element may not be essential for
activation of Nestin transcription in P19 EC
cells. Moreover, since Nestin expression was
first detected after 24 hrs of 5-aza-dC treatment
and 8 hours after the onset of 5’enhancer
demethylation, DNA demethylation does not
appear to be sufficient to mediate activation of
Nestin transcription. However, we cannot
exclude that at a later stage activator
complexes interact with the demethylated
enhancer and result in delayed activation of
Nestin expression in 5-aza-dC treated cells. In
such a case the recruitment of an activator
complex would take several hours, and the
onset of Nestin expression would occur at
about the same time in TSA-treated cells as in
5-aza-dC–treated cells. However the onset of
Nestin expression was observed after the cells
had been treated with TSA for only 4 hours.
Therefore, DNA demethylation is not sufficient
to cause Nestin expression.
Transcription status of Nestin in Dnmts-
deficient ES cells
To further discern any potential involvement of
DNA methylation in the regulation of Nestin
transcription—as suggested by the differential
methylation pattern of the 5’enhancer in
Nestin-expressing and non-expressing cells—
the DNA methylation pattern was assessed in
DNA methyltransferase knock-out ES cell
lines (Dnmt KO ES cells). Genome-wide
demethylation has been observed in the Dnmts
KO ES cells [27]. If DNA methylation were to
play a role in Nestin expression, the Nestin
gene would be expected to be transcriptionally
active in Dnmts KO ES cells, which exhibit
Nestin enhancer demethylation. The Nestin
5’enhancer was assessed in wild-type ES cells
and in three different Dnmt KO ES cells lines:
Dnmt1 KO ES (Dnmt1-deficient), Dnmt 3a, 3b
KO ES (Dnmt3a- and 3b-deficient), and Dnmt
total knock-out (TKO) ES cells [28].
Hypermethylation was found in the wild-type
ES cells and hypomethylation in all Dnmt KO
ES cells—3.1% methylation rate in Dnmt1 KO
ES, 0.0% in Dnmt3 KO ES, and 0.0% in Dnmt
TKO ES cells (Fig. 3A). Although the degree
of Nestin enhancer hypomethylation in all
Dnmts KO ES cells was similar to that of
Nestin-expressing NSCs, Nestin expression
could not be detected in untreated Dnmts KO
ES cells, but only in TSA-treated cells (Fig.
3B). This demonstrates that DNA
demethylation of the Nestin enhancer element
is not sufficient to mediate activation of Nestin
transcription. It is possible that differentiation
of Dnmt KO ES lines into NSCs is
accompanied by enhanced Nestin expression
due to their unmethylated promoter. To test
this we have attempted to derive NSCs from
each Dnmt KO ES cell line based on the
published protocol [31]. However, although
5
differentiation into early precursors of NSCs
was possible, we were unable to generate NSC
lines from any of the Dnmt KO ES cell lines.
These early precursors of NSCs all stained
positive for Nestin and Sox2 as determined by
immunocytochemistry, similar to derivatives of
control ES cells (data not shown). However,
we were unable to obtain evidence for the
earlier activation of Nestin. Thus, enhanced
induction of Nestin activation could not be
observed in differentiated Dnmt KO ES cells,
indicating that DNA demethylation alone does
not support the activation of Nestin expression.
Chromatin structure of Nestin promoter
and enhancer elements
Since both TSA and 5-aza-dC treatment could
mediate the activation of Nestin transcription in
EC cells, the chromatin status of Nestin was
assessed in NSCs and P19 EC cells. Histone
acetylation of the Nestin regulatory elements in
P19 EC cells and NSCs was examined by
chromatin immunoprecipitation (ChIP) assay
using antibodies that recognize acetylated
versions of histones H3 and H4. The
acetylation level of histone H3 on the Nestin
promoter element was not found to differ
between NSCs and P19 EC cells. The same
region was completely unmethylated in both
NSCs and P19 EC cells (Fig. 2A), while
histone H4 on the promoter element was
slightly more acetylated in NSCs compared
with P19 EC cells (Fig. 4, Fig. S4B). However,
the acetylation level of histones H3 and H4 on
both 5’ and 3’enhancers was significantly
higher in NSCs compared with P19 EC cells
(Fig. 4, Fig. S4A, B) and F9 EC cells (Fig. S3).
Moreover, there was a nearly 3-fold difference
in histone H4 acetylation on the Nestin second
intron enhancer elements between NSCs and
P19 EC cells. On the other hand, there was a
2.4- to 3.3-fold higher acetylation of Oct4, a
marker for pluripotent cells, in P19 EC cells
compared with NSCs (Fig. 4, Fig. S4A, B).
Taken together, these data demonstrate the
epigenetic regulation of histones on the Nestin
promoter and enhancer elements.
Histone acetylation is sufficient to mediate
activation of Nestin transcription
Since activation of Nestin transcription can
occur without DNA demethylation after 4 hrs
of TSA treatment (Fig. 1A, Fig. 2C), histone
acetylation—which can be induced by TSA—
may be sufficient to mediate activation of
Nestin transcription. To test this possibility, the
acetylation of histones H3 and H4 on the
Nestin enhancer after TSA treatment was
assessed in a time-dependent fashion (4, 6, 8,
and 16 hrs). Strikingly, a dramatic increase in
the acetylation levels of both histones H3 and
H4 was found after 4 hrs of TSA treatment,
comparable with levels in NSCs, thereby
suggesting that TSA treatment induces rapid
histone acetylation on the Nestin enhancer,
followed by activation of Nestin transcription
(Fig. 5A, Fig. S4C). In a previous study,
histone hyperacetylation was observed after 5-
aza-dC treatment [29]. Thus, histone
acetylation on the Nestin enhancer after 5-aza-
dC treatment was assessed to discern any
association between histone acetylation on the
Nestin enhancer and Nestin transcription.
Although DNA demethylation of the Nestin
enhancer was observed after 16 hrs of 5-aza-
dC treatment (Fig. 2B), acetylation of histone
H3 on the 5’enhancer was seen to increase
after 24 hrs (Fig. 5B, Fig. S4D), consistent
with the onset of Nestin transcription, thus
suggesting that histone modification is a
prerequisite for Nestin activation. A significant
increase in the acetylation of histone H4—but
not histone H3—on the 3’enhancer was found.
While an increase in histone H4 acetylation on
one of the 3’enhancer elements was not
observed, there was a dramatic increase in
histone H4 acetylation on the other enhancer
elements, including 5’ and 3’enhancers, after
24 hrs of 5-aza-dC treatment (Fig. 5B, Fig.
S4D). Taken together, these data suggest that
histone acetylation is sufficient to mediate
activation of Nestin transcription. Moreover,
histone acetylation on the Nestin gene in P19
EC cells after 8 to 16 hrs of TSA treatment was
comparable to that in NSCs, but the histone
acetylation level in 5-aza-dCítreated P19 EC
cells was still lower than that in NSCs even
after 48 hrs of 5-aza-dC treatment (Fig. 5, Fig.
S4C, D). Similarly, since there was lower
Nestin expression after 48 hrs of 5-aza-dC
treatment than after 4 hrs of TSA treatment
(Fig. S2), DNA demethylation is likely to
occur subsequent to Nestin transcription.
6
Epigenetic hierarchy governing
Nestin
expression
Epigenetic hierarchy in the regulation of
Nestin expression
The data presented herein show that Nestin
transcription could be activated following
acetylation of histones in the absence of DNA
demethylation; however, DNA demethylation
after 5-aza-dC treatment is not sufficient to
mediate Nestin activation. Moreover, TSA and
5-aza-dC influence the epigenetic status of the
Nestin gene by completely different
mechanisms—TSA inhibits HDAC and 5-aza-
dC inhibits DNA methylation. Therefore, to
exclude the possibility of any other epigenetic
modifications induced by TSA or 5-aza-dC
treatment, the role of DNA methylation
following random activation of Nestin
transcription without specific inhibition of
either histone deacetylation or DNA
methylation was next investigated. First, Nestin
non-expressing P19 EC cells were treated with
retinoic acid (RA) for 12, 24, and 48 hrs to
induce cellular differentiation, and the
expression pattern of Nestin was then
examined by RT-PCR. Onset of Nestin
expression in P19 EC cells was observed after
24 hrs of RA treatment, with expression
increasing gradually over time (Fig. 6A).
Down-regulation of Oct4 expression was
detected after RA treatment, which is
consistent with RA-induced differentiation of
pluripotent cells (Fig. S1B). Nestin
transcription was also examined by real-time
RT-PCR (Fig. S1A). Since activation of Nestin
expression was observed after 24 hrs of RA
treatment, but not after 12 hrs, it was
postulated that passive demethylation might
occur after 24 hrs of RA treatment, accounting
for Nestin activation. However, DNA
demethylation of the Nestin enhancer could not
be observed after up to 48 hrs of RA treatment,
suggesting that Nestin activation was mediated
by RA in the absence of DNA demethylation
of the Nestin enhancer elements (Fig. 6B).
Again, the acetylation status of histones H3
and H4 was assessed following RA treatment.
All selected regions exhibited a gradual
increase in H3 acetylation and a more dramatic
increase (1.4-6.8-fold) in H4 acetylation after
24 hrs of RA treatment (Fig. 6C, Fig. S4E).
Taken together, these findings suggest that
DNA demethylation is not sufficient to mediate
Nestin activation, as Nestin expression could
not be detected in P19 EC cells after 5-aza-dC
treatment, i.e. in cells with unmethylated
enhancer elements. However, histone
modification (acetylation) is sufficient to
mediate Nestin activation, as Nestin expression
could be observed after TSA and RA
treatment, i.e. in cells without DNA
demethylation of the Nestin enhancer elements
(Fig. 7C). Therefore, histone modification (i.e.
acetylation) is postulated to be at the top of the
epigenetic regulatory hierarchy governing
Nestin expression, while DNA demethylation
appears to be a dispensable event.
DISCUSSION
Although Nestin has been widely used as a
marker gene to identify and isolate NSCs [2-4],
its regulation remains relatively poorly
understood. The current study was undertaken
to determine the impact of the key epigenetic
features histone deacetylation and DNA
methylation on Nestin gene transcription. An
earlier study examining the regulation of
Nestin expression in transgenic mice [11]
showed that the Nestin upstream promoter
element does not harbor any regulatory
elements specific to NSCs, but it identified two
cell type-specific enhancers in the intron
regions of Nestin: a myogenic precursor
cellspecific enhancer within the first intron
and a CNS-specific enhancer within the second
intron of the rat Nestin gene. The CNS-specific
second intron enhancer is a crucial element in
the regulation of Nestin expression in NSCs
[13]. Moreover, the 3’ part of the second intron
enhancer (3’enhancer) harbors SOX and POU
binding sites, which when mutated lead to
silencing of Nestin expression, indicating that
the Nestin 3’enhancer is the main regulatory
element involved in the positive regulation of
Nestin transcription. The current study showed
that the Nestin gene is differentially methylated
in Nestin-expressing and non-expressing cells
and, moreover, that only the 5’ part of second
intron enhancer (5’enhancer) is differentially
methylated, indicating that 5’enhancer may
also be a crucial element in Nestin
transcription. Nestin transcription was
therefore assessed in Nestin non-expressing EC
cells that had been treated with TSA and 5-aza-
dC—TSA inhibits histone deacetylation and 5-
7
aza-dC inhibits DNA methylation—to
determine the role of DNA methylation in the
regulation of Nestin expression. DNA
demethylation and histone modifications
typically act in a synergistic manner to regulate
the transcription of many genes [16-19].
However, it was found that activation of Nestin
expression could be mediated by the inhibition
of histone acetylation (i.e. of histone
deacetylase [HDAC]) through TSA treatment
in the absence of DNA demethylation on the
5’enhancer. Other neural marker genes were
also found to display similar transcriptional
activation patterns (Fig. S5), indicating that not
all genes are regulated by either equal or
synergic contributions of DNA methylation
and histone modifications. Moreover, onset of
Nestin expression was observed only after 24
hours of 5-aza-dC treatment, even though the
5’enhancer had already been demethylated.
These data suggest two possible roles of DNA
methylation in the mechanisms underlying
Nestin transcription. First, DNA methylation
simply reflects the transcriptional status of
Nestin. Second, and more feasibly, DNA
methylation may be a process secondary to the
activation of Nestin transcription. To verify
this hypothesis, an attempt was made to discern
the presence of an association between Nestin
transcription and DNA methylation of Nestin
in Dnmts KO ES cells, wherein genome-wide
DNA demethylation has normally been
observed. Although undermethylation of the
Nestin 5’enhancer (methylation rate: 0.0-3.1%)
was observed in all Dnmt KO ES cell lines,
there was no activation of Nestin expression in
the absence of histone deacetylation inhibition
induced by TSA treatment (Fig. 3). Therefore,
DNA methylation is not sufficient to mediate
Nestin transcription; it appears to be
dispensable for Nestin activation. Activation of
Nestin transcription was observed
concomitantly with the absence of DNA
demethylation in P19 EC cells that had been
treated with TSA for 4 hrs. However, DNA
demethylation of the 5’enhancer element—
observed after 16 hrs of TSA treatment—was
found to lead to enhanced Nestin transcription
(Fig. S2A), indicating that although DNA
methylation is dispensable for Nestin
activation, it may still represent an important
process secondary to Nestin regulation.
Furthermore, DNA demethylation of the Nestin
5’enhancer was indirectly mediated by TSA
treatment, suggesting that DNA methylation
may play a role in ensuring maximal Nestin
expression. Moreover, upon induction of
further differentiation of Olig2-GFP–positive
NSCs, which had been generated from Olig2-
GFP ES cells that had differentiated into
Olig2-GFP–negative cells, including neurons,
astrocytes, and oligodendrocyte (Fig. 7A),
hypermethylation of the Nestin 5’enhancer
element was observed. This result suggests that
although DNA methylation is not important for
Nestin activation, it may be a secondary
mechanism for suppressing Nestin expression
(Fig. 7B). The histone acetylation status of the
Nestin regulatory elements was then assessed
in both Nestin-expressing NSCs and non-
expressing EC cells. As expected, P19 EC cells
and NSCs were found to display completely
different acetylation patterns of both histones
H3 and H4 on the second intron enhancer,
including the 5’ and 3’enhancers (Fig. 4, Fig.
S4A, B). Although the Nestin upstream
promoter element does not harbor any specific
regulatory machinery in NSCs, significantly
more acetylation (1.8-fold) was observed on
histone H4 on the Nestin promoter in Nestin-
expressing NSCs, indicating that the Nestin
promoter may also be involved in the
epigenetic regulation of Nestin transcription.
Interestingly, histone acetylation of Nestin was
always observed concomitantly with activation
of Nestin transcription in both TSA- and 5-aza-
dC–treated cells. Moreover, Nestin silencing
could be disturbed by RA treatment in the
absence of DNA demethylation, and activation
of Nestin transcription was also observed
concomitantly with histone acetylation of the
Nestin second intron enhancer (Fig. 6, Fig.
7C). A recent study showed that histone
acetylation precedes DNA demethylation, with
a time lag between histone acetylation and
DNA demethylation in the transcription of an
in-vitro methylated reporter plasmid, while the
DNA demethylation of a reporter plasmid
always occurred in a replication-dependent
manner [30]. However, since histone
acetylation appears to be sufficient for
activation of Nestin transcription in the absence
of DNA demethylation, investigations into the
regulation of Nestin expression provide a
8
Epigenetic hierarchy governing
Nestin
expression
model to study the role of the epigenetic
modification histone acetylation in the absence
of DNA demethylation. To elucidate the roles
played by HDACs and histone
acetyltransferase (HATs) in Nestin regulation,
the levels of all HDACs or HATs were
determined in NSCs, ES cells, and MEF cells.
The expression patterns of all HDACs and
HATs were unexpectedly found to be very
similar between cell lines, suggesting that
Nestin activation may be mediated by the
combined effects of many HDACs and HATs
(Fig. S6). The results reviewed in the current
study suggest that histone modifications are
sufficient to mediate activation of Nestin
transcription and thus histone modification is a
higher-order epigenetic feature necessary for
Nestin transcription. Although we were unable
to show that Nestin activation is merely
dependent on the acetylation induced by TSA,
the finding that Nestin activation is always
accompanied by acetylation of the Nestin
enhancer indicates that this acetylation is
indeed both critical and a prerequisite for
Nestin transcription.
ACKNOWLEDGMENTS
We are indebted to all members of the Schöler
laboratory for helpful discussions of the
results. We are especially grateful to Dr.
Natalia Tapia for valuable comments on the
manuscript. This work was supported by the
Federal Ministry of Education and Research
(BMBF) initiative “Cell-Based Regenerative
Medicine” (Grant 01GN0539).
REFERENCES
1. Temple S. The development of neural stem cells.
Nature 2001;414:112-117.
2. Dahlstrand J, Zimmerman LB, McKay RD et al.
Characterization of the human nestin gene reveals a
close evolutionary relationship to neurofilaments. J
Cell Sci 1992;103 (Pt 2):589-597.
3. Kawaguchi A, Miyata T, Sawamoto K et al. Nestin-
EGFP transgenic mice: visualization of the self-
renewal and multipotency of CNS stem cells. Mol
Cell Neurosci 2001;17:259-273.
4. Lendahl U, Zimmerman LB, McKay RD. CNS stem
cells express a new class of intermediate filament
protein. Cell 1990;60:585-595.
5. Frederiksen K, McKay RD. Proliferation and
differentiation of rat neuroepithelial precursor cells
in vivo. J Neurosci 1988;8:1144-1151.
6. Kachinsky AM, Dominov JA, Miller JB. Myogenesis
and the intermediate filament protein, nestin. Dev
Biol 1994;165:216-228.
7. Kachinsky AM, Dominov JA, Miller JB. Intermediate
filaments in cardiac myogenesis: nestin in the
developing mouse heart. J Histochem Cytochem
1995;43:843-847.
8. Frojdman K, Pelliniemi LJ, Lendahl U et al. The
intermediate filament protein nestin occurs
transiently in differentiating testis of rat and mouse.
Differentiation 1997;61:243-249.
9. Terling C, Rass A, Mitsiadis TA et al. Expression of
the intermediate filament nestin during rodent tooth
development. Int J Dev Biol 1995;39:947-956.
10. Jin ZG, Liu L, Zhong H et al. Second intron of
mouse nestin gene directs its expression in
pluripotent embryonic carcinoma cells through POU
factor binding site. Acta Biochim Biophys Sin
(Shanghai) 2006;38:207-212.
11. Zimmerman L, Parr B, Lendahl U et al. Independent
regulatory elements in the nestin gene direct
transgene expression to neural stem cells or muscle
precursors. Neuron 1994;12:11-24.
12. Josephson R, Muller T, Pickel J et al. POU
transcription factors control expression of CNS stem
cell-specific genes. Development 1998;125:3087-
3100.
13. Lothian C, Lendahl U. An evolutionarily conserved
region in the second intron of the human nestin gene
directs gene expression to CNS progenitor cells and
to early neural crest cells. Eur J Neurosci
1997;9:452-462.
14. Yaworsky PJ, Kappen C. Heterogeneity of neural
progenitor cells revealed by enhancers in the nestin
gene. Dev Biol 1999;205:309-321.
15. Tanaka S, Kamachi Y, Tanouchi A et al. Interplay of
SOX and POU factors in regulation of the Nestin
gene in neural primordial cells. Mol Cell Biol
2004;24:8834-8846.
16. Ghoshal K, Datta J, Majumder S et al. Inhibitors of
histone deacetylase and DNA methyltransferase
synergistically activate the methylated
metallothionein I promoter by activating the
transcription factor MTF-1 and forming an open
chromatin structure. Mol Cell Biol 2002;22:8302-
8319.
17. Hwang CK, Song KY, Kim CS et al. Evidence of
endogenous mu opioid receptor regulation by
epigenetic control of the promoters. Mol Cell Biol
2007;27:4720-4736.
18. Yamada N, Hamada T, Goto M et al. MUC2
expression is regulated by histone H3 modification
and DNA methylation in pancreatic cancer. Int J
Cancer 2006;119:1850-1857.
19. Zhang Y, Fatima N, Dufau ML. Coordinated
changes in DNA methylation and histone
9
modifications regulate silencing/derepression of
luteinizing hormone receptor gene transcription. Mol
Cell Biol 2005;25:7929-7939.
20. Lin P, Kusano K, Zhang Q et al. GABAA receptors
modulate early spontaneous excitatory activity in
differentiating P19 neurons. J Neurochem
1996;66:233-242.
21. McBurney MW, Reuhl KR, Ally AI et al.
Differentiation and maturation of embryonal
carcinoma-derived neurons in cell culture. J Neurosci
1988;8:1063-1073.
22. Cheng L, Jin Z, Liu L et al. Characterization and
promoter analysis of the mouse nestin gene. FEBS
Lett 2004;565:195-202.
23. Ou JN, Torrisani J, Unterberger A et al. Histone
deacetylase inhibitor Trichostatin A induces global
and gene-specific DNA demethylation in human
cancer cell lines. Biochem Pharmacol 2007;73:1297-
1307.
24. Selker EU. Trichostatin A causes selective loss of
DNA methylation in Neurospora. Proc Natl Acad Sci
U S A 1998;95:9430-9435.
25. Januchowski R, Dabrowski M, Ofori H et al.
Trichostatin A down-regulate DNA
methyltransferase 1 in Jurkat T cells. Cancer Lett
2007;246:313-317.
26. Xiong Y, Dowdy SC, Podratz KC et al. Histone
deacetylase inhibitors decrease DNA
methyltransferase-3B messenger RNA stability and
down-regulate de novo DNA methyltransferase
activity in human endometrial cells. Cancer Res
2005;65:2684-2689.
27. Li E, Bestor TH, Jaenisch R. Targeted mutation of
the DNA methyltransferase gene results in
embryonic lethality. Cell 1992;69:915-926.
28. Meissner A, Gnirke A, Bell GW et al. Reduced
representation bisulfite sequencing for comparative
high-resolution DNA methylation analysis. Nucleic
Acids Res 2005;33:5868-5877.
29. Takebayashi S, Nakao M, Fujita N et al. 5-Aza-2'-
deoxycytidine induces histone hyperacetylation of
mouse centromeric heterochromatin by a mechanism
independent of DNA demethylation. Biochem
Biophys Res Commun 2001;288:921-926.
30. D'Alessio AC, Weaver IC, Szyf M. Acetylation-
induced transcription is required for active DNA
demethylation in methylation-silenced genes. Mol
Cell Biol 2007;27:7462-7474.
31. Conti L, Pollard SM, Gorba T et al. Niche-
independent symmetrical self-renewal of a
mammalian tissue stem cell. PLoS Biol 2005;3:e283.
32. Do JT, Scholer HR. Comparison of neurosphere cells
with cumulus cells after fusion with embryonic stem
cells: reprogramming potential. Reprod Fertil Dev
2005;17:143-149.
33. Han DW, Do JT, Gentile L et al. Pluripotential
reprogramming of the somatic genome in hybrid
cells occurs with the first cell cycle. Stem Cells
2008;26:445-454.
See www.StemCells.com for supporting information available
online.
10
Epigenetic hierarchy governing
Nestin
expression
Figure 1. Activation of Nestin in P19 and F9 EC cells after TSA and 5-aza-dC treatments. A,B.
The expression profiles of Nestin in TSA- (A) and 5-aza-dC(B) treated P19 and F9 EC cells were
determined by conventional RT-PCR. NSCs were used as a positive control for Nestin expression.
ȕ-actin was used as an RT control.
11
Figure 2. DNA methylation status of the Nestin regulatory elements in P19 and F9 EC cells. A.
DNA methylation status of Nestin regulatory elements, including the promoter and the 5’ and
3’enhancers, was assessed using bisulfite sequencing, with selected regions indicated with gray
boxes. B,C. Demethylation patterns of 5’enhancer after 5-aza-dC (B) and TSA (C) treatments.
Open and filled circles indicate unmethylated and methylated CpGs, respectively.
12
Epigenetic hierarchy governing
Nestin
expression
Figure 3. Transcription and DNA methylation status of Nestin in Dnmts KO ES cell lines. A. DNA
methylation status of the 5’enhancer was assessed using bisulfite sequencing in Dnmts KO ES cell
lines, with the selected region examined indicated with a gray box. Open and filled circles indicate
unmethylated and methylated CpGs, respectively. B. Expression profiles of Nestin in Dnmts KO
ES cell lines were determined by conventional RT-PCR. Wild-type ES cells were used as a
negative control for Nestin expression and ȕ-actin was used as an RT control. To activate Nestin
expression in Dnmts KO ES cell lines, each cell line was treated with TSA for 16 hrs.
13
Figure 4. ChIP analysis of the Nestin regulatory elements in NSCs and P19 EC cells. The upper
panel is a schematic diagram of the Nestin regulatory elements. The gray boxes indicate the primer
positions used for PCR of the ChIP assay. NSCs and P19 EC cells were subjected to ChIP analysis
using antibodies of acetylated versions of histone H3 (AcH3) and H4 (AcH4). Immunoprecipitation
in the absence of antibodies served as negative control. Aliquots of the sonicated cell supernatant
were also analyzed as an input control. PCR of Hprt served as positive control for the antibodies,
and the starting materials (templates) from each sample were empirically determined by measuring
the intensities of Hprt bands. The fold-enrichment values are shown below the respective lanes. In
addition, the acetylation status of Oct4 gene in P19 EC cells was examined, and was consistent the
ChIP analysis.
14
Epigenetic hierarchy governing
Nestin
expression
Figure 5. ChIP analysis of the Nestin enhancer after TSA and 5-aza-dC treatments. A. Histone
acetylation status of P19 EC cells was examined after TSA treatment. Pre-cultured P19 EC cells
were treated with TSA for 4, 6, 8, and 16 hrs. B. Histone acetylation status of P19 EC cells was
examined after 5-aza-dC treatment for 8, 16, 24, 48, and 72 hrs (10 ȝM). The fold-enrichment
values are shown below the respective lanes.
15
Figure 6. Activation of Nestin expression after RA treatment and sequential epigenetic
modifications in P19 EC cells. A. The effect of RA on Nestin expression was determined by
conventional RT-PCR. B. Demethylation pattern of the 5’enhancer after RA treatment as
evidenced by bisulfite sequencing analysis. Open and filled circles indicate unmethylated and
methylated CpGs, respectively. C. Histone acetylation status of Nestin in P19 EC cells was
examined after RA treatment for 12, 24, and 48 hrs (5 ȝM). The fold-enrichment values are shown
below the respective lanes.
16
Epigenetic hierarchy governing
Nestin
expression
Figure 7. Schematic representation of the regulatory mechanism governing Nestin expression. A.
Olig2-GFP-positive NSCs were derived from Olig2-GFP-negative ES cells and further
differentiated into Olig2-GFP-negative cells including neurons (Tuj1), astrocytes (GFAP), and
oligodendrocytes (O4). B. DNA methylation status of 5’enhancer was assessed in Olig2-GFP ES
cells, Olig2-GFP NSCs and differentiated cells. C. The relationship between sequential epigenetic
modifications, such as gene expression, DNA demethylation, and histone acetylation, and treatment
with TSA, 5-aza-dC, and RA is depicted.
17
18
