Cell Stem Cell

Perspective

Epigenetic Mechanisms that Regulate Cell Identity

Marı́a J. Barrero,1 Stephanie Boué,1 and Juan Carlos Izpisúa Belmonte1,2,*
1Center of Regenerative Medicine in Barcelona, Dr. Aiguader, 88, 08003 Barcelona, Spain
2Salk Institute for Biological Studies, 10010 North Torrey Pines Road, La Jolla, CA 92037, USA
*Correspondence: belmonte@salk.edu or izpisua@cmrb.eu (J.C.I.B.)
DOI 10.1016/j.stem.2010.10.009

Individual cell fate decisions can vary according to changes in gene expression in response to environmental,
developmental, or metabolic cues. This plasticity is tightly regulated during embryonic development and
mediated by the exquisitely coordinated activation and repression of groups of genes. Genes that become
repressed are immersed in a condensed chromatin environment that renders them refractory to stimulation.
This mechanism is responsible for both the loss of cell plasticity during differentiation and the preservation of
cell identity. Understanding the molecular events involved in the establishment and maintenance of these
restrictive domains will benefit the design of strategies for cellular reprogramming, differentiation, and cancer
treatment.

Despite sharing the same genome, different cell types from
a given organism respond differently to environmental, develop-
mental, or metabolic cues. This variable property, or plasticity of
responsiveness, is a defining aspect of a cell’s identity and is
mainly interpreted at the level of gene expression. Stem and
progenitor cells are a special case, given that their potential for
plasticity is, by definition, more pronounced than for other
tissue-specific cells. During differentiation, stem cells lose their
plasticity and narrow down their identity into a particular differen-
tiated cell type. For many years this differentiated state was
thought to be irreversible but recent findings indicate that it is
possible to manipulate cells in order to change cell identity or
to regain plasticity. Understanding how fate decisions are regu-
lated and maintained by different cell types, both during devel-
opment and under pathological or experimental conditions,
represents an important step toward the successful manipula-
tion of cell plasticity in clinical settings.
We have known for a long time that genetic material associ-
ates with structural proteins to constitute chromatin, which plays
important roles structurally and functionally. Hence, genes
immersed in highly packed areas of chromatin remain largely
inaccessible, fail to recruit the RNA polymerase to their pro-
moters, and thus remain silent and refractory to stimulation.
However, genes having a more open and accessible chromatin
structure are more likely to recruit and engage the RNA poly-
merase into productive rounds of transcription and their expres-
sion is more prone to be modulated by signaling events. Impor-
tantly, the chromatin environment in which a particular gene is
embedded can differ from one cell type to another, contributing
to cell diversity and identity.
The fundamental unit of chromatin is the nucleosome, which is
composed of two copies each of four core histones, H2A, H2B,
H3, and H4, and wrapped by 146 bp of DNA. The N-terminal tails
of histones are relatively accessible to enzymatic modifications
such as acetylation, methylation, phosphorylation, ubiquitina-
tion, and sumoylation. Of all the known histone modifications,
acetylation is the only modification that directly causes a struc-
tural relaxation of chromatin by introducing a negative charge.
Other modifications seem to act as docking sites that promote
the recruitment and stabilization of effector protein complexes
rather than altering the chromatin structure per se. In a similar
fashion, DNA methylation, which takes place at cytosine resi-
dues within CpG dinuclotides at gene promoters, is usually
correlated with transcriptional repression and mediates its
effect by blocking the binding of transcription factors and/or
facilitating the assembly of repressor complexes at the methyl-
ated regions. In general, the presence of histone modifications
involved in gene activation, such as trimethylation of histone
H3 at lysine 4 (H3K4me3), promotes the recruitment and
stabilization of effector complexes with histone acetyltransfer-
ase (HAT) and ATP-dependent remodeling activities. These
complexes mediate the acetylation of histones and stimulate
the mobility of the nucleosomes, or even their eviction, to facili-
tate the relaxation of chromatin. Ultimately, this change renders
DNA more accessible and facilitates the recruitment of DNA-
binding transcription factors to target genes, which in turn
contribute to the recruitment of RNA polymerase and the stabi-
lization of the preinitiation complex (PIC). Modifications involved
in gene repression, such as DNA methylation and trimethylation
of histone H3 at lysine 9 (H3K9me3) and at lysine 27 (H3K27me3),
serve as docking sites for repressor complexes that contain
histone deacetylase (HDAC) and ATP-dependent remodeling
activities. The action of these complexes, together with the
recruitment of structural nonhistone proteins, leads to chromatin
compaction and reduced DNA accessibility.
Molecular Basis of Pluripotency in Embryonic
Stem Cells
The chromatin in pluripotent embryonic stem cells (ESCs) dis-
plays distinctive features when compared to differentiated cells.
In ESCs the chromatin structural proteins display hyperdynamic
interactions with chromatin (Meshorer et al., 2006) and the over-
all transcriptional activity is hyperactive compared to differenti-
ated cells (Efroni et al., 2008). ESCs also express high levels of
chromatin-remodeling factors involved in maintaining the open
chromatin status. Moreover, variations in the expression levels
of specific subunits of remodeling complexes, such as the BAF
complex, in ESCs relative to differentiated cells leads to the
formation of unique remodeling complexes that differ in subunit
composition and potentially in function (Ho et al., 2009). These

Cell Stem Cell 7, November 5, 2010 ª2010 Elsevier Inc. 565


specialized complexes could contribute to sustain high levels of
transcriptional activity in pluripotent cells. This idea is further
supported by the reported loss of self-renewal caused by the
depletion of the remodeling factors Chd1 or Brg1 in ESCs (Efroni
et al., 2008; Gaspar-Maia et al., 2009).
ESCs feature a unique plasticity that allows them to differen-
tiate into virtually any cell type of the adult organism. This prop-
erty, called pluripotency, relies on the fact that critical genes
involved in differentiation, despite remaining silent, have a
permissive chromatin structure that makes them sensitive to
differentiation-inducing signals. This permissive chromatin envi-
ronment is characterized by the presence of large H3K27me3
domains harboring smaller regions of H3K4me3 around the tran-
scriptional start site of critical developmental genes. The coexis-
tence of these two antagonistic marks is the defining feature of
so-called bivalent genes and has been suggested to play
a role in silencing differentiation genes in ESCs while keeping
them poised for activation upon initiation of specific develop-
mental pathways (Azuara et al., 2006; Bernstein et al., 2006).
Moreover, the poised nature of these bivalent domains is further
reinforced by the absence of DNA methylation, despite the pres-
ence of numerous CpG islands (Fouse et al., 2008; Meissner
et al., 2008) and the presence of poised RNA polymerase II at
the transcription initiation site of the marked genes (Guenther
et al., 2007). Ultimately, the physiological function of bivalent
domains might be to maintain important regulatory sequences
accessible and responsive at very early stages of differentiation
(Gargiulo et al., 2009). Importantly, genes marked with bivalent
domains often encode master transcription factors that are
induced very early during differentiation, usually in a lineage-
restricted manner, and that are able to orchestrate whole
programs of gene expression in differentiated cells.
Although the bivalent marks are faithfully transmitted through
cell division in self-renewing cells, they should not be regarded
as static but as the result of a highly dynamic equilibrium that
is controlled by the balance of histone-modifying enzymes
recruited to these areas. The polycomb group of proteins (PcGs)
has been described to play an essential role in maintaining the
pluripotent state of ESCs by mediating H3K27 methylation at
the bivalent domains (Surface et al., 2010). PcG proteins localize
at genes encoding developmental regulators and correlate with
the presence of H3K27 trimethylation both in mouse and human
ESCs. Moreover, mouse ESCs null for specific PcG proteins
exhibit decreased H3K27 methylation and aberrant expression
of key developmental genes (Boyer et al., 2006; Lee et al.,
2006). On the other hand, the H3K4me3 marks present at the
bivalent domains of silent developmental genes could be cata-
lyzed by trithorax (trxG) homologs that belong to the MLL family,
although their putative contribution to the establishment of these
domains remains unknown. As part of a highly dynamic and
responsive equilibrium, it is likely that histone lysine demethy-
lases also contribute to keep the appropriate balance of marks
at these domains. Confirming this idea, is the observation that
the PcGs recruit the H3K4 demethylase RBP2 to its target genes
to maintain the proper balance of H3K4 and H3K27 methylation
at the bivalent domains in mouse ESCs (Pasini et al., 2008).
To fully understand the role of bivalent domains in ESCs, it
is important to know how these domains are first established.
Bivalent marks have been reported to exist in pluripotent cells
566 Cell Stem Cell 7, November 5, 2010 ª2010 Elsevier Inc.
Cell Stem Cell
Perspective
in the embryo and to appear after fertilization during genomic
activation at the maternal-zygotic transition (Vastenhouw et al.,
2010). In addition, recent reports describe the presence of
microregions of H3K27me3 at the transcriptional start sites of
developmental regulators in sperm. The persistence of these
discrete regions after fertilization and the presence of CpG
islands at the promoters of the marked developmental genes
might serve as primers for the establishment of broad
H3K27me3 domains during genome activation in embryos
(Brykczynska et al., 2010). However, it is likely that specific tran-
scription factors also contribute to establish and/or maintain
these domains. Indeed, genome-wide correlation studies have
revealed that the self-renewal transcription factors Oct4, Sox2,
and Nanog occupy not only promoters of genes involved in
self-renewal but also of a high number of genes that contain
bivalent domains (Boyer et al., 2005), suggesting that they play
a role in keeping this genes silenced. In agreement with this
model, it has been reported that Nanog and Oct4 interact with
complexes involved in transcriptional repression, including
PcG subunits (Pardo et al., 2010; Wang et al., 2006). Another
potential transcription factor that could be involved in the molec-
ular arrangement of these domains is JARID2, which has been
recently suggested to participate in the recruitment of PcG to
the regulatory regions of target genes in mouse ESCs (Landeira
et al., 2010; Li et al., 2010; Pasini et al., 2010; Peng et al., 2009).
Chromatin Dynamics during Differentiation
The permissive chromatin structure present in multipotent cells
is progressively and selectively closed during differentiation.
Accessible regulatory areas, such as bivalent domains, close
up and are no longer accessible to transcription factors, leading
to a loss of regulatory potential (Gargiulo et al., 2009).
The analysis of the genome-wide distribution of histone and
DNA methylation marks and their correlation with gene expres-
sion in ESCs, in vitro differentiated ESCs, and adult cells has
provided a better understanding of gene plasticity in pluripotent
and differentiated tissues (Fouse et al., 2008; Hawkins et al.,
2010; Meissner et al., 2008; Mikkelsen et al., 2007). These
studies have led researchers to focus on two types of genes:
ones with promoters enriched in CpG islands (HCP) and those
showing poor representation of CpGs (LCP) (Figure 1). HCP
genes are usually devoid of DNA methylation in ESCs and
include self-renewal and housekeeping genes usually marked
with H3K4me3 and developmental regulators containing bivalent
domains. This pattern suggests that the targets of trxG are pro-
tected against DNA methylation and are transcriptionally active
unless actively repressed by PcG proteins (Mikkelsen et al.,
2007). LCP genes are not expressed, show DNA methylation,
and are mostly devoid of H3K4 or H3K27 methylation in ESCs
(Meissner et al., 2008). These genes are likely to be targets of
early master transcription factors and thus they become induced
late during differentiation in a tissue-specific manner to carry out
specialized functions in particular tissues.
Differentiation to one particular lineage implies the permanent
and irreversible silencing of genes involved in alternative line-
ages as well as those necessary for pluripotent cells. For exam-
ple, pluripotency genes lose H3K4me3 and gain distinct combi-
nations of repressive marks (H3K27me3, H3K9me3, and DNA
methylation). Bivalent genes that become irreversibly silenced
Cell Stem Cell

Perspective

Figure 1. Epigenetic Map of Fate

Genome-wide and correlation studies (Boyer et al., 2005; Doi et al., 2009; Fouse et al., 2008; Hawkins et al., 2010; Meissner et al., 2008; Mikkelsen et al., 2007)
have allowed the classification of genes according to their genetic, epigenetic, and transcriptional status in different cell types and physiological situations. This
classification scheme derives from statistical correlations of broad data sets, and thus not all genes will follow the flowchart precisely. Blue arrows denote tran-
sitions from self-renewal to differentiation, red arrows denote transitions to cancer, and gray arrows indicate critical changes to regain pluripotency during
reprogramming. ON and OFF refer to transcriptional status; DNAme refers to DNA methylation; OSN targets refers to genes occupied by Oct4, Sox2, and Nanog;
and trimethylation at lysine 4, 9, or 27 of histone H3 is referred to as H3K4me3, H3K27me3, and H3K9me3, respectively. Marks separated by a slash indicate
co-occurrence of marks.

lose H3K4me3 and preserve H3K27me3 (Cui et al., 2009; Mikkel-
sen et al., 2007). Usually, repression is further reinforced by the
expansion of the H3K27 methylated areas or the accumulation
of a second repressive mark, such as H3K9me3 or DNA methyl-
ation (Hawkins et al., 2010). This ‘‘double lock’’ mechanism
ensures permanent repression of developmental and pluripo-
tency genes and thus preserves cell identity.
How chromatin modifying and remodeling activities partici-
pate in the process of differentiation is not yet fully understood.
The switch of signaling events that characterizes the transition
from self-renewal to differentiation induces dramatic changes
in gene expression, including the downregulation of genes
involved in self-renewal and changes in the expression of critical
chromatin regulators (Pardo et al., 2010; Wang et al., 2006). This
last event can lead to variations in the subunit composition of
effector complexes, allowing them to carry out specialized func-
tions during differentiation. Contrary to previous perceptions, in
which chromatin marks could be passively diluted, several
reports point out the importance of active and targeted removal
of marks. Such is the role of the H3K27 demethylases UTX and
Jmjd3 in the activation of Hox genes during development (Agger
et al., 2007) and in neuronal commitment (Burgold et al., 2008).
Both demethylases associate with MLL complexes, suggesting
that removal of the H3K27me3 mark and maintenance of the
H3K4me3 at bivalent genes that become activated during differ-
entiation are coordinated events. Less is known about the
molecular regulation and physiological relevance of a small
percentage of bivalent domains that remain unresolved upon
differentiation and that are thus bivalent in adult cells (Figure 1).
Perhaps more intriguing is the fact that a few new bivalent
regions are established during differentiation (Mikkelsen et al.,
2007). Regarding the activities that participate in repression
during differentiation, the methyltransferase G9a is needed for
the proper silencing of genes, among which are self-renewal
factors such as Oct4 (Feldman et al., 2006). G9a catalyzes the
methylation of H3K9 at the regulatory regions of these genes
during differentiation, which in turn facilitates the binding of
heterochromatin protein 1 (HP1) and de novo DNA methylation.
Reprogramming and Cell Plasticity
For many years the differentiated state was thought to be irre-
versible. However, several experimental strategies have been
developed to induce the reactivation of the embryonic program
in differentiated cells, a process known as nuclear reprogram-
ming. Historically, nuclear transfer experiments showed that
the genetic content of adult and embryonic cells is equivalent

Cell Stem Cell 7, November 5, 2010 ª2010 Elsevier Inc. 567


and that the differences lie in gene expression (Wilmut et al.,
1997). Later, cell fusion experiments demonstrated that the
epigenetic activities present in the nucleus of ESCs are able to
induce the reactivation of the embryonic program in differenti-
ated cells (Do and Scholer, 2004). More recently, it has been
described that it is possible to generate induced pluripotent
stem cells (iPSCs) from somatic cells by overexpressing specific
transcription factors, most commonly Oct4, Sox2, Klf4, and
c-Myc (Takahashi and Yamanaka, 2006). This methodology
enables the generation of patient-specific pluripotent cells that
could overcome the risk of rejection in future cell-based thera-
pies and provide a valuable model for mechanistic studies. In a
similar approach, transdifferentiation of one adult cell type to
another can also be achieved by overexpression of appropriate
combinations of transcription factors (Davis et al., 1987).
Both nuclear transfer and transcription factor-induced reprog-
ramming are very inefficient processes, and the few animals that
are successfully generated by nuclear transfer often suffer from
numerous abnormalities. Importantly, the efficiency of reprog-
ramming differs depending on the somatic cell type and seems
to be inversely correlated with its degree of differentiation
(Maherali and Hochedlinger, 2008). Overall, these findings sug-
gest the existence of important epigenetic barriers that are
imposed during differentiation, probably to preserve cell identity,
and that need to be overcome to successfully reprogram a cell
back to the pluripotent state. Certainly, the process of reprog-
ramming entails dramatic changes in gene expression including
the transcriptional activation of endogenous self-renewal factors
and the silencing of differentiation genes. Both events involve
profound changes in the chromatin environment at the regula-
tory regions of these genes. Critical differentiation genes must
reacquire their poised status, as suggested by the reestablish-
ment of bivalent domains at their proximal promoters (Hawkins
et al., 2010), and self-renewal genes must lose the repressive
chromatin marks that kept them silent in somatic cells (Takaha-
shi and Yamanaka, 2006).
Factor-induced reprogramming of fibroblasts is a gradual
process that includes several critical stages (Figure 2). An early
stage is characterized by the appearance of alkaline phospha-
tase activity and the stem cell marker SSEA-1. At this stage cells
568 Cell Stem Cell 7, November 5, 2010 ª2010 Elsevier Inc.
Cell Stem Cell
Perspective
Figure 2. Critical Events during
Reprogramming and Differentiation
Transduction of somatic cells with the four factors
causes stress, apoptosis, and replicative senes-
cence. Cells able to overcome this initial barrier
are able to self-renew but still depend on the
expression of the ectopic factors. A critical second
step consists of overcoming the epigenetic barrier
that prevents the reactivation of the endogenous
pluripotency genes. However, after reactivation
of the pluripotency network, several cell divisions
might be needed to erase the adult cellular
memory, a process that is characterized by the re-
gaining of H3K4 and H3K27 trimethylation and
erasing of DNA methylation at bivalent domains.
Proper reestablishment of these domains ensures
the adequate induction of developmental genes
during differentiation and thus is a critical determi-
nant of the quality of the reprogrammed cells. Blue
and orange backgrounds indicate stages depen-
dent on transgene or endogenous pluripotency
factor expression, respectively.
can be trapped in a partially reprogrammed state in which they
self-renew but are dependent on the expression of the trans-
genes. A second stage of reprogramming is characterized by
the activation of the endogenous pluripotency factors Oct4,
Sox2, and Nanog. This event allows the maintenance of pluripo-
tency in an autonomous way, independent of the expressed
transgenes. Recent findings suggest that the rare occurrence
of late events during reprogramming is due to the inability of
the transfactors to bind and activate the regulatory regions of
the endogenous pluripotency genes, resulting from the presence
of an inaccessible repressive chromatin environment (Sridharan
et al., 2009). Several lines of evidence further support this idea.
First, it has been recently demonstrated that the overexpression
of members of the chromatin remodeling complex BAF enhance
reprogramming by facilitating the binding of Oct4 to target pro-
moters (Singhal et al., 2010). Second, treatment with DNA
methylation inhibitors, which render chromatin more accessible,
facilitates the completion of reprogramming in partially reprog-
rammed cells (Mikkelsen et al., 2008). Increased reprogramming
efficiency has also been reported in the presence of HDACs,
DNA methyltransferases, and G9a inhibitors (Huangfu et al.,
2008; Mikkelsen et al., 2008). Finally, it has been suggested
that the increased efficiency of reprogramming observed in
cord blood cells compared to adult fibroblasts may be due to
the presence of a more permissive chromatin organization at
the promoters of self-renewal genes in this population (Giorgetti
et al., 2009). Thus, a more complete understanding of the activ-
ities and mechanisms that keep pluripotency genes silenced in
differentiated cells appears critical to develop strategies to
improve the efficiency of reprogramming.
Analysis of genome-wide data shows that important changes
at the level of DNA methylation occur during reprogramming in
bivalent developmental regulators and more generally in genes
occupied by Oct4, Sox2, and Nanog in ESCs. These genes
become hypomethylated in iPSCs compared to the fibroblast
starting population. Indeed, comparison of the methylomes of
ESCs and iPSCs revealed that they are very similar, although
some differentially methylated regions can be found, mainly in
genes involved in developmental processes. In these differen-
tially methylated regions, the methylation of iPSCs was found
Cell Stem Cell
Perspective
to be intermediate between that of ESCs and fibroblasts, sug-
gesting incomplete reprogramming, but also different from
both ESCs and fibroblasts, which might imply aberrant estab-
lishment of marks during reprogramming (Doi et al., 2009).
Regarding histone modifications, it appears that H3K9me3 con-
tributes more than H3K27me3 to the differences between iPSCs
and ESCs. Both failure to erase K9me3 methylation marks at
particular regions and the establishment of aberrant H3K9
methylated areas during reprogramming have been reported
(Hawkins et al., 2010). The impact of aberrant epigenetic reprog-
ramming is clearly illustrated by a recent report describing that
the persistent aberrant silencing of a particular imprinted region
after reprogramming renders mouse iPSCs developmentally
challenged (Stadtfeld et al., 2010). Consistently, aberrant
imprinted loci have been also reported in cloned embryos (Liu
et al., 2008).
In essence, the compact chromatin structure present at self-
renewal genes in differentiated cells seems at least partly
responsible for the low efficiency of the reprogramming process.
However, a larger negative impact may stem from the potential
establishment of aberrant chromatin marks at developmental
genes during reprogramming, because these mistakes might
affect the quality of the reprogrammed cells. Importantly, a
recent study suggests that the aberrant expression of bivalent
genes in iPSCs can be inversely correlated with the degree of
pluripotency that these cells display (Boue et al., 2010). Thus,
improper establishment of chromatin structure at these develop-
mental genes might lead to defective differentiation, including
the inability to generate a particular differentiated cell type or,
more importantly, the production of immature cell types in which
cell identity is not fully established and perhaps prone to rever-
sion. In line with this hypothesis, a recent study found that iPSCs
derived from nonhematopoietic cells display reduced blood-
forming potential resulting from residual DNA methylation at
loci required for hematopoietic lineage specification (Kim et al.,
2010). Interestingly, partial reprogramming of bivalent domains
could bestow iPSCs with an epigenetic memory, which might
make them more prone to differentiate back into cells of their
tissue of origin. Indeed, early passage iPSCs have been proven
to retain degree of expression of cell-of-origin-specific genes,
which facilitates their differentiation into their original lineage
(Polo et al., 2010). However, cell-of-origin differences tend to
disappear after successive passages in culture, suggesting
that complete epigenetic reprogramming to pluripotency is a
slow and gradual process. Thus, after the reactivation of the
endogenous pluripotency genes, a third critical step, the full
reestablishment of bivalent domains, involving both histone
marks and DNA demethylation, is essential to achieve pluripo-
tent status (Figure 2). Although the efficiency of reprogramming
is often reported based on the detection of alkaline phosphatase
activity or the reactivation of pluripotency genes, quality will defi-
nitely need to be judged based on the complete epigenetic
reprogramming of critical differentiation genes (Maherali and
Hochedlinger, 2008).
Differentiation, Dedifferentiation, and Cancer: A Stem
Cell Link to Cancer?
Although malignant transformation is usually associated with
mutations and translocations, aberrant DNA methylation is also
a hallmark of cancer. De novo DNA methylation of CpG islands
in tumor cells is associated with the transcriptional silencing of
many cancer-related genes, among them tumor suppressors.
It is not clear how this aberrant DNA methylation takes place at
the molecular level, but the fact that some cancers present
abnormal expression of chromatin modifiers and that DNA meth-
yltransferase and histone deacetylase (HDAC) inhibitors have
been successfully used to treat cancer brings to light the rele-
vance of the epigenetic machinery in this process (Esteller,
2008).
Promoters marked with bivalent domains in ESCs frequently
undergo DNA methylation in a tissue-specific manner during
differentiation (Meissner et al., 2008; Mohn et al., 2008), become
hypomethylated during reprogramming, and have a tendency to
become hypermethylated in cancer (Doi et al., 2009; Widsch-
wendter et al., 2007). More striking is the finding that these
hypermethylated domains display bivalent marks in certain
cancers (Rodriguez et al., 2008). Cancer-dependent DNA hyper-
methylation of these regions can be reverted during reprogram-
ming, highlighting the importance of the epigenetic context in the
control of these domains (Li et al., 2003; Ron-Bigger et al., 2010).
Despite the heterogeneity that cancer cells display, overall these
data suggest that bivalent regions are critical epigenetic targets
for defining cell fate and that deregulation of their epigenetic
state can lead to important disorders such as cancer.
Although cancer cells might resemble ESCs regarding the
silencing of tissue-specific genes and the subsequent adoption
of a more undifferentiated phenotype, evidence suggests that
these parallel outcomes are mediated by distinct mechanisms.
Although in ESCs developmental genes are repressed through
bivalent domains and probably the action of Oct4, Sox2,
and Nanog, in cancer cells these genes become permanently
silenced by DNA methylation. However, the bivalent domain
and cancer DNA hypermethylation connection makes it tempting
to speculate about the potential stem cell origin of cancer. Thus,
in adult stem cells, bivalent domain-containing genes could be
prone to acquire DNA methylation and as a result, generate
cancer-like cells able to self-renew, but with impaired differenti-
ation potential (Widschwendter et al., 2007).
In summary, gene silencing plays critical roles during develop-
ment, differentiation, reprogramming, and cancer. Thus, a better
understanding of the activities, mechanisms, and signals that
control the silencing of critical genes will help to design strate-
gies to improve the efficiency of reprogramming and to treat
cancer.
ACKNOWLEDGMENTS
We apologize for those authors whose work could not be cited because of
space limitation. We would like to thank all the members of the J.C.I.B. lab
at The Salk Institute for Biological Studies and the CMRB for their insightful
comments and input. This work was partially supported by grants RYC-
2007-01510 and SAF2009-08588 from the Ministerio de Ciencia e Innovación
of Spain to M.J.B. and grants from the G. Harold and Leila Y. Mathers Chari-
table and Cellex Foundations, Sanofi-Aventis, and MICINN to J.C.I.B.
REFERENCES
Agger, K., Cloos, P.A., Christensen, J., Pasini, D., Rose, S., Rappsilber, J.,
Issaeva, I., Canaani, E., Salcini, A.E., and Helin, K. (2007). Nature 449,
731–734.
Cell Stem Cell 7, November 5, 2010 ª2010 Elsevier Inc. 569

Azuara, V., Perry, P., Sauer, S., Spivakov, M., Jorgensen, H.F., John, R.M.,
Gouti, M., Casanova, M., Warnes, G., Merkenschlager, M., et al. (2006). Nat.
Cell Biol. 8, 532–538.
Bernstein, B.E., Mikkelsen, T.S., Xie, X., Kamal, M., Huebert, D.J., Cuff, J., Fry,
B., Meissner, A., Wernig, M., Plath, K., et al. (2006). Cell 125, 315–326.
Boue, S., Paramonov, I., Barrero, M.J., and Izpisua Belmonte, J.C. (2010).
PLoS ONE 5, e12664.
Boyer, L.A., Lee, T.I., Cole, M.F., Johnstone, S.E., Levine, S.S., Zucker, J.P.,
Guenther, M.G., Kumar, R.M., Murray, H.L., Jenner, R.G., et al. (2005). Cell
122, 947–956.
Boyer, L.A., Plath, K., Zeitlinger, J., Brambrink, T., Medeiros, L.A., Lee, T.I.,
Levine, S.S., Wernig, M., Tajonar, A., Ray, M.K., et al. (2006). Nature 441,
349–353.
Brykczynska, U., Hisano, M., Erkek, S., Ramos, L., Oakeley, E.J., Roloff, T.C.,
Beisel, C., Schubeler, D., Stadler, M.B., and Peters, A.H. (2010). Nat. Struct.
Mol. Biol. 17, 679–687.
Burgold, T., Spreafico, F., De Santa, F., Totaro, M.G., Prosperini, E., Natoli, G.,
and Testa, G. (2008). PLoS ONE 3, e3034.
Cui, K., Zang, C., Roh, T.Y., Schones, D.E., Childs, R.W., Peng, W., and Zhao,
K. (2009). Cell Stem Cell 4, 80–93.
Davis, R.L., Weintraub, H., and Lassar, A.B. (1987). Cell 51, 987–1000.
Do, J.T., and Scholer, H.R. (2004). Stem Cells 22, 941–949.
Doi, A., Park, I.H., Wen, B., Murakami, P., Aryee, M.J., Irizarry, R., Herb, B.,
Ladd-Acosta, C., Rho, J., Loewer, S., et al. (2009). Nat. Genet. 41, 1350–1353.
Efroni, S., Duttagupta, R., Cheng, J., Dehghani, H., Hoeppner, D.J., Dash, C.,
Bazett-Jones, D.P., Le Grice, S., McKay, R.D., Buetow, K.H., et al. (2008). Cell
Stem Cell 2, 437–447.
Esteller, M. (2008). N. Engl. J. Med. 358, 1148–1159.
Feldman, N., Gerson, A., Fang, J., Li, E., Zhang, Y., Shinkai, Y., Cedar, H., and
Bergman, Y. (2006). Nat. Cell Biol. 8, 188–194.
Fouse, S.D., Shen, Y., Pellegrini, M., Cole, S., Meissner, A., Van Neste, L., Jae-
nisch, R., and Fan, G. (2008). Cell Stem Cell 2, 160–169.
Gargiulo, G., Levy, S., Bucci, G., Romanenghi, M., Fornasari, L., Beeson, K.Y.,
Goldberg, S.M., Cesaroni, M., Ballarini, M., Santoro, F., et al. (2009). Dev. Cell
16, 466–481.
Gaspar-Maia, A., Alajem, A., Polesso, F., Sridharan, R., Mason, M.J., Heiders-
bach, A., Ramalho-Santos, J., McManus, M.T., Plath, K., Meshorer, E., et al.
(2009). Nature 460, 863–868.
Giorgetti, A., Montserrat, N., Aasen, T., Gonzalez, F., Rodriguez-Piza, I., Vas-
sena, R., Raya, A., Boue, S., Barrero, M.J., Corbella, B.A., et al. (2009). Cell
Stem Cell 5, 353–357.
Guenther, M.G., Levine, S.S., Boyer, L.A., Jaenisch, R., and Young, R.A.
(2007). Cell 130, 77–88.
Hawkins, R.D., Hon, G.C., Lee, L.K., Ngo, Q., Lister, R., Pelizzola, M., Edsall,
L.E., Kuan, S., Luu, Y., Klugman, S., et al. (2010). Cell Stem Cell 6, 479–491.
Ho, L., Ronan, J.L., Wu, J., Staahl, B.T., Chen, L., Kuo, A., Lessard, J., Nesvizh-
skii, A.I., Ranish, J., and Crabtree, G.R. (2009). Proc. Natl. Acad. Sci. USA 106,
5181–5186.
Huangfu, D., Maehr, R., Guo, W., Eijkelenboom, A., Snitow, M., Chen, A.E.,
and Melton, D.A. (2008). Nat. Biotechnol. 26, 795–797.
Kim, K., Doi, A., Wen, B., Ng, K., Zhao, R., Cahan, P., Kim, J., Aryee, M.J., Ji,
H., Ehrlich, L.I., et al. (2010). Nature 467, 285–290.
Landeira, D., Sauer, S., Poot, R., Dvorkina, M., Mazzarella, L., Jorgensen, H.F.,
Pereira, C.F., Leleu, M., Piccolo, F.M., Spivakov, M., et al. (2010). Nat. Cell Biol.
12, 618–624.
Lee, T.I., Jenner, R.G., Boyer, L.A., Guenther, M.G., Levine, S.S., Kumar, R.M.,
Chevalier, B., Johnstone, S.E., Cole, M.F., Isono, K., et al. (2006). Cell 125,
301–313.
570 Cell Stem Cell 7, November 5, 2010 ª2010 Elsevier Inc.
Cell Stem Cell
Perspective
Li, L., Connelly, M.C., Wetmore, C., Curran, T., and Morgan, J.I. (2003). Cancer
Res. 63, 2733–2736.
Li, G., Margueron, R., Ku, M., Chambon, P., Bernstein, B.E., and Reinberg, D.
(2010). Genes Dev. 24, 368–380.
Liu, J.H., Yin, S., Xiong, B., Hou, Y., Chen, D.Y., and Sun, Q.Y. (2008). Mol.
Reprod. Dev. 75, 598–607.
Maherali, N., and Hochedlinger, K. (2008). Cell Stem Cell 3, 595–605.
Meissner, A., Mikkelsen, T.S., Gu, H., Wernig, M., Hanna, J., Sivachenko, A.,
Zhang, X., Bernstein, B.E., Nusbaum, C., Jaffe, D.B., et al. (2008). Nature
454, 766–770.
Meshorer, E., Yellajoshula, D., George, E., Scambler, P.J., Brown, D.T., and
Misteli, T. (2006). Dev. Cell 10, 105–116.
Mikkelsen, T.S., Ku, M., Jaffe, D.B., Issac, B., Lieberman, E., Giannoukos, G.,
Alvarez, P., Brockman, W., Kim, T.K., Koche, R.P., et al. (2007). Nature 448,
553–560.
Mikkelsen, T.S., Hanna, J., Zhang, X., Ku, M., Wernig, M., Schorderet, P.,
Bernstein, B.E., Jaenisch, R., Lander, E.S., and Meissner, A. (2008). Nature
454, 49–55.
Mohn, F., Weber, M., Rebhan, M., Roloff, T.C., Richter, J., Stadler, M.B., Bibel,
M., and Schubeler, D. (2008). Mol. Cell 30, 755–766.
Pardo, M., Lang, B., Yu, L., Prosser, H., Bradley, A., Babu, M.M., and Choudh-
ary, J. (2010). Cell Stem Cell 6, 382–395.
Pasini, D., Hansen, K.H., Christensen, J., Agger, K., Cloos, P.A., and Helin, K.
(2008). Genes Dev. 22, 1345–1355.
Pasini, D., Cloos, P.A., Walfridsson, J., Olsson, L., Bukowski, J.P., Johansen,
J.V., Bak, M., Tommerup, N., Rappsilber, J., and Helin, K. (2010). Nature 464,
306–310.
Peng, J.C., Valouev, A., Swigut, T., Zhang, J., Zhao, Y., Sidow, A., and
Wysocka, J. (2009). Cell 139, 1290–1302.
Polo, J.M., Liu, S., Figueroa, M.E., Kulalert, W., Eminli, S., Tan, K.Y., Aposto-
lou, E., Stadtfeld, M., Li, Y., Shioda, T., et al. (2010). Nat. Biotechnol. 28,
848–855.
Rodriguez, J., Munoz, M., Vives, L., Frangou, C.G., Groudine, M., and Peinado,
M.A. (2008). Proc. Natl. Acad. Sci. USA 105, 19809–19814.
Ron-Bigger, S., Bar-Nur, O., Isaac, S., Bocker, M., Lyko, F., and Eden, A.
(2010). Stem Cells 28, 1349–1354.
Singhal, N., Graumann, J., Wu, G., Arauzo-Bravo, M.J., Han, D.W., Greber, B.,
Gentile, L., Mann, M., and Scholer, H.R. (2010). Cell 141, 943–955.
Sridharan, R., Tchieu, J., Mason, M.J., Yachechko, R., Kuoy, E., Horvath, S.,
Zhou, Q., and Plath, K. (2009). Cell 136, 364–377.
Stadtfeld, M., Apostolou, E., Akutsu, H., Fukuda, A., Follett, P., Natesan, S.,
Kono, T., Shioda, T., and Hochedlinger, K. (2010). Nature 465, 175–181.
Surface, L.E., Thornton, S.R., and Boyer, L.A. (2010). Cell Stem Cell 7,
288–298.
Takahashi, K., and Yamanaka, S. (2006). Cell 126, 663–676.
Vastenhouw, N.L., Zhang, Y., Woods, I.G., Imam, F., Regev, A., Liu, X.S., Rinn,
J., and Schier, A.F. (2010). Nature 464, 922–926.
Wang, J., Rao, S., Chu, J., Shen, X., Levasseur, D.N., Theunissen, T.W., and
Orkin, S.H. (2006). Nature 444, 364–368.
Widschwendter, M., Fiegl, H., Egle, D., Mueller-Holzner, E., Spizzo, G., Marth,
C., Weisenberger, D.J., Campan, M., Young, J., Jacobs, I., et al. (2007). Nat.
Genet. 39, 157–158.
Wilmut, I., Schnieke, A.E., McWhir, J., Kind, A.J., and Campbell, K.H. (1997).
Nature 385, 810–813.