Cell Stem Cell
Multiple Epigenetic Modifiers Induce
Aggressive Viral Extinction
in Extraembryonic Endoderm Stem Cells
Michael C. Golding,1,2,3,4Liyue Zhang,3,5and Mellissa R.W. Mann1,2,3,5,*
1Department of Obstetrics & Gynecology
2Department of Biochemistry
University of Western Ontario, Schulich School of Medicine and Dentistry, London, Ontario N6A 4V2, Canada
3Children’s Health Research Institute, London, Ontario N6C 2V5, Canada
4Department of Veterinary Physiology, College of Veterinary Medicine, Texas A&M University, College Station, TX 77843, USA
5Lawson Health Research Institute, London, Ontario N6C 2V5, Canada
To prevent insertional mutagenesis arising from
retroviral reactivation, cells of embryonic origin
possess a unique capacity to silence retroviruses.
Given the distinct modes of X chromosome inactiva-
tion between embryonic and extraembryonic line-
ages, we investigated paradigms of viral extinction.
We show that trophectoderm stem cells do not
silence retroviral transcription, whereas extraembry-
onic endoderm stem cells aggressively extinguish
proviral transcription, even more rapidly than do
embryonic stem cells. By using a short hairpin RNA
extinction in extraembryonic endoderm stem cells.
sor complex proteins act to modulate integrated, as
well as endogenous, retroviral element silencing,
with a subset of factors displaying differential effects
between stem cell types. Furthermore, our data
suggest that small RNAs play a role in this process
through interactions with the Argonaute family. Our
results further the understanding of mechanisms
regulating retroviral transcription in different stem
To contend with the constant threat of retroviral infection,
lian genomes have evolved complex defense mechanisms to
impede the life cycle of invading parasitic nucleic acids. Histori-
cally, these defensive strategies have been separated into two
broad categories: viral restriction and viral extinction (Niwa
et al., 1983; Cherry et al., 2000).
Restriction is a process whereby protein factors encoded by
the host genome interact directly with viral elements to block
some aspect of the invading virus’s life strategy. Proteins such
as Friend virus susceptibility 1 (FV1), tripartite interaction motif
5a (TRIM5a), zinc finger antiviral protein (ZAP), TRIM19/PML
(promyelocytic leukemia), and the TRIM28-ZFP809 (zinc finger
antiviral protein 809) complex all restrict retroviral tropism via
direct biochemical interactions (Best et al., 1996; Kaiser et al.,
2007; Gao et al., 2002; Turelli et al., 2001; Wolf and Goff, 2009;
Rowe et al., 2010).
Less well understood is the process of retroviral extinction,
which is progressive silencing of proviral transcription that
occurs over long-term cellular growth or differentiation (Cherry
et al., 2000; Laker et al., 1998). This process is epigenetic in
nature and has been associated with acquisition of DNA methyl-
ation and other transcriptionally repressive chromatin modifica-
tions at the viral integration site (Harbers et al., 1981; Ja ¨hner
et al., 1982; Pannell et al., 2000; Poleshko et al., 2008; Matsui
et al., 2010).
Cells of embryonic origin are unique in their capacity to silence
retroviruses, which is understandable given the necessity of
preventing insertional mutagenesis that could arise from retro-
viral reactivation. g-retroviruses, like the mouse leukemia virus
(MLV), can integrate into embryonic carcinoma and embryonic
stem cells but are silenced by both restriction- and extinction-
based mechanisms (Loh et al., 1990; Niwa et al., 1983; Teich
et al., 1977). The early embryo possesses three stem cell line-
ages: embryonic stem (ES), trophectoderm stem (TS), and
extraembryonic endoderm (XEN) stem cells that give rise to the
embryo proper, placenta, and yolk sac, respectively. Although
viral extinction has been well documented in cells of embryonic
origin, the capacity of extraembryonic stem cells to epigeneti-
cally silence g-retroviruses has not been examined. Given the
distinct epigenetic mechanisms regulating X chromosome inac-
tivation in embryonic and extraembryonic cells (Kunath et al.,
2005; Takagi and Sasaki, 1975), we hypothesized that different
modes of viral extinction may also exist between separate
lineages of the preimplantation embryo. Specifically, we sought
to determine whether stem cells derived from trophoblast
and extraembryonic endoderm could epigenetically extinguish
a MLV variant modified to escape retroviral restriction but which
is susceptible to extinction (Cherry et al., 2000; Grez et al., 1990;
Laker et al., 1998).
To examine viral extinction in stem cells, we assayed tran-
scriptional activity of genetically marked mouse g-retroviruses
in primary ESCs, TSCs, and XEN stem cells. Unlike ESCs,
Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc. 457
TSCs did not silence proviral transcription but rather maintained
consistent and high expression levels over long-term culture.
Surprisingly, integration of g-retrovirus into XEN cells produced
rapid transcriptional silencing, by comparison to ESCs. We
present data that indicate XEN cell extinction is epigenetic in
nature and mediated by multiple chromatin remodeling and
polycomb repressor complexes. These complexes not only
act to modulate the silent state of integrated retroviruses but
also play significant roles in repressing endogenous retro-
element transcription. Furthermore, our results suggest that
these repressor complexes are recruited to sites of viral integra-
tion through interactions with the Argonaute family of proteins.
Distinct Patterns of Retroviral Extinction in Embryonic
and Extraembryonic Stem Cells
For decades, scientists have intensively studied mechanisms of
gene regulation in ESCs, including those involved in exogenous
and endogenous retroviral silencing. However, only a few epige-
netic factors employed in retroviral silencing have been identi-
fied. Furthermore, epigenetic mechanisms operating in cohabit-
ing extraembryonic stem cells have been virtually ignored, the
assumption being that all stem cells utilize similar mechanisms.
To determine whether stem cells derived from trophoblast and
extraembryonic endoderm possess similar or distinct capacities
to silence retroviruses, we chose to examine transcriptional
activity of aMLV variant thatisclosely related to theendogenous
type C retroviral element. To achieve this, we constructed a
series of genetically marked mouse embryonic stem cell viruses
(MSCV) (Figures 1A and 2A) and assayed their transcriptional
activity in infected primary ESCs, TSCs, and XEN cells.
Within 24 hr of infection, expression of virally delivered GFP
could be detected within each of the three stem cell lineages
(Figure 1B; Table S1 available online). To assay for viral extinc-
tion, cells were propagated for 6–15 weeks, with the percentage
of GFP-positive cells monitored via flow cytometry at each
passage (Figure 1C). Consistent with previous observations
(Cherry et al., 2000), ESCs exhibited a progressive decline in
the number of cells expressing GFP with near complete viral
extinction occurring after 15 weeks in culture (passage 40)
(Figure S1). Surprisingly, TSCs did not inactivate the g-retroviral
reporter but rather maintained a constant level of GFP-express-
ing cells. This trend was maintained through 12 weeks of obser-
vation (30 passages) and correlated with continued expression
of TSC-specific markers (data not shown). By comparison,
XEN cells exhibited aggressive silencing of viral transcription
such that within 3 weeks (6–7 passages), GFP expression was
Retroviral Silencing in XEN Cells Is Epigenetic in Origin
Given the surprising speed with which the g-retroviral reporter
was silenced in the XEN cell lineage, we examined viral extinc-
tion in this relatively uncharacterized cell type. To ensure that
the observed extinction was an epigenetic phenomenon, and
not an artifact of cellular differentiation or GFP toxicity, three
experiments were conducted. First, to determine whether loss
of viral expression was a result of cellular differentiation, RNA
was extracted from MSCV-GFP-infected XEN cells at passages
2, 7, and 10 and assayed for expression of XEN cell markers.
Abundant expression of transcripts encoding GATA binding
protein 4 (Gata4), Forkhead box protein A2 (Foxa2), Hepatocyte
Figure 1. Distinct Patterns of Retroviral Extinction between Embry-
onic and Extraembryonic Stem Cells that Is Epigenetic in Origin
(A) Schematic representation of g-retroviral reporters. MSCV GFP is a MSCV
variant containing GFP.
(B) Light micrograph and fluorescent images showing expression of g-retro-
viral delivered GFP in ESCs, TSCs, and XEN cells at passage three.
(C) Expression of viral-delivered GFP as measured by cytometry over a 6 week
period. ESCs exhibited a gradual decline in GFP-positive cells, whereas TSCs
displayed no measurable change in GFP-positive cell numbers. XEN cells
rapidly silenced the g-retroviral GFP reporter. No fluorescence was observed
in cells infected with an empty vector.
ysis of XEN cells containing the silenced MSCV GFP g-retroviral reporter
(passage 11) after treatment with vehicle, 5-Aza, and TSA aloneor in combina-
tion for 3 days or 1 week. Data represent percentage of cells expressing GFP.
Error bars, SEM.
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
458 Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc.
nuclear factor 4 (Hnf4), and Sex determining region Y-box 7
(Sox7) was observed via reverse transcription polymerase chain
reaction (RT-PCR) (Figure S1; Kunath et al., 2005). Maintenance
of XEN cell marker expression, together with consistent XEN cell
morphology throughout the culture period, indicated that
silencing of virally delivered GFP was not associated with loss
Second, to ensure that the observed extinction was not due to
loss of integrated virus, nor to GFP toxicity, infected XEN cell
DNA was examined for MSCV-GFP provirus at passage 2
through10 byquantitative PCR(qPCR). No difference in theratio
of GFP to Glyceraldehyde 3-phosphate dehydrogenase (Gapdh)
promoter amplification was detected throughout the experi-
ment, indicating that the number of cells containing the inte-
grated provirus was relatively unchanged (Figure S1).
Finally, to determine whether viral reporter silencing was
epigenetic in nature, pharmacological inhibition of DNA methyl-
ation and histone deacetylation was employed. To this end,
XEN cells were again infected with MSCV-GFP, cultured for 10
passages to allow g-retroviral reporter silencing, then treated
alone or in combination. Treatment with 5-Aza or TSA alone
induced a 5- to 15-fold increase in reactivation of proviral tran-
scription compared to vehicle alone. Treatment with both drugs
produced a synergistic response leading to transcriptional acti-
vation (Figure 1D). Thus, integration of g-retrovirus into the XEN
cell genome rapidly initiated an aggressive epigenetic-based
antiviral response, resulting in complete silencing of proviral
RNA Interference Screen for Epigenetic Modifiers
of Viral Extinction
To identify specific biochemical factors that mediate this
epigenetic response, we conducted a loss-of-function, RNA
interference (RNAi), positive selection screen. By sing a micro-
RNA-based short hairpin RNA library (shRNAmir) (Silva et al.,
2005) that was designed to target 250 known protein-coding
genes involved in epigenetic gene regulation (?3 shRNAmir/
gene; Table S2), we addressed the question of how g-retrovi-
ruses are silenced in mouse XEN cells. To conduct our screen,
we designed two separate g-retroviral reporters with distinct
fluorescent and drug-selectable markers (Figure 2A). These con-
structs were packaged into infectious retroviral particles and
withdrawn from drug selection, and passaged to allow for retro-
viral extinction. As previously observed, integrated virus was
rapidly silenced (Figure S2). Additionally, cultures were verified
for maintenance of stemness via XEN cell markers as described
above (Figure S1).
At passage 11, replication-deficient lentiviral particles (Fig-
ure 2B) were used to deliver the shRNAmirepigenetic library
into XEN cells containing a silenced g-retroviral reporter. These
constructs were specifically designed to facilitate delivery into
ESCs, TSCs, and XEN cells (unpublished data). shRNAmir
constructs were matched to contain distinct fluorescent and
As a control, a shRNAmirtargeting the firefly luciferase (Luc) gene
was stably transduced into XEN cells, and experiments were
conducted in parallel. Upon delivery of the shRNAmirlibrary, cells
were passaged for 1 week, then drug selected with either
neomycin or puromycin to enrich for XEN cells with reactivated
MSCV ChIN or MSCV PIG reporters, respectively. Surviving
colonies were picked and DNA isolated, and perspective
shRNAmirs conferring resistance to silencing were amplified
and sequenced (425 colonies in total). Results of the screen
based on a comprehensive scoring system (Figure S3) are
summarized in Table 1. Three of the top 25 identified factors
have known interactions with viral elements, which validates
our experimental design: SWI/SNF-related, matrix-associated,
actin-dependent regulator of chromatin, subfamily a, member
5 (SMARCA5); DNA methyltransferase 1 (DNMT1); and Helicase,
lymphoid-specific (HELLS) (Chong et al., 2007; De La Fuente
et al., 2006). In contrast, the majority of the candidates identified
in this screen are novel regulators of retroviral silencing.
Validation of mRNA Depletion of Individual
To validate the most promising candidates that emerged from
our screen, shRNAmirs targeting the top ten candidates based
on our scoring system were reintroduced into XEN cells. Target
mRNA depletion was verified with a combination of quantitative
RT-PCR (qRT-PCR) and western blot analysis (Figures S4 and
S5). As controls, candidate shRNAmir-targeted XEN cells were
compared to wild-type XEN cells, XEN cells infected with
MSCV ChIN or MSCV PIG, as well as Luc shRNAmir-targeted
XEN cells. For analysis, results were normalized to wild-type
expression levels. All tested shRNAmirs produced target mRNA
depletion. Our scoring system gives strong preference to genes
for which multiple independent shRNAs emerged. Although
arguably we may be bypassing strong candidates, it minimized
the potential for off-target shRNAs to be included. The observa-
screen then weaker ones validates the scoring system.
Differential Expression of Candidates between ESC,
TSC, and XEN Cell Lineages
To determine whether differences in proviral silencing between
ESCs, TSCs, and XEN cells could be explained by a lineage-
levels, relative transcript abundance was investigated via qRT-
PCR analysis. To ensure accurate quantitation of candidate
transcript levels between cell types, measurements were normal-
izedagainstthe geometric meanofb-actin, 7SK, and Hexokinase
transcript levels. Very little SWI/SNF-related, matrix-associated,
actin-dependent regulator of chromatin, subfamily a, member 1
(Smarca1) expression was detected in TSCs; both ESCs and
XEN cells possessed greater than 1000-fold higher transcript
abundance than did TSCs (Figure 3). The remaining factors
showed a 2- to 10-fold increase in transcript abundance in ESCs
compared to TSCs and XEN cells, including SWI/SNF-related,
matrix-associated, actin-dependent regulator of chromatin, sub-
family a, member 5 (Smarca5), Histone deactylases Hdac2,
Hdac7, and Hdac9, Polycomb group ring finger 6 (Pcgf6), DNA
methyltransferase 3a (Dnmt3a), Argonaute1 (Ago1), and Embry-
onic ectoderm development (Eed). The lone exception was SWI/
SNF-related, matrix-associated, actin-dependent regulator of
chromatin, subfamily c, member 2 (Smarcc2), which was 3 times
more abundantly expressed in XEN cells than in ESCs or TSCs.
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc. 459
Retroviral Extinction Correlates with a Switch from
Transcriptionally Permissive to Repressed Chromatin
Given that SMARCA5 was identified in a previous genetic screen
for modifiers of retroviral gene silencing (Chong et al., 2007), it is
not surprising that other gene family members (SMARCA1 and
SMARCC2) may be involved in proviral suppression. Chromatin
remodeling is associated with numerous heterogeneous pro-
tein complexes and diverse changes in chromatin structure,
histone acetylation changes associated with proviral silencing,
we conducted quantitative chromatin immunoprecipitation
(ChIP) analysis of the MSCV 50long terminal repeat (LTR) and
proximal packaging region (Psi) (Figure 4A) over the course of
XEN cell viral extinction. A 1000-fold reduction in signal between
passages 2 and 6 was observed when lysates were immunopre-
cipitated with antibodies recognizing histone 3 acetylation
(H3Ac) (Figure 4B). This dramatic drop in acetylated histones
strongly supports the involvement of histone deacetylases
(HDAC2, HDAC7, and/or HDAC9) in proviral extinction. In further
support of this data, Poleshko et al. (2008) and Keedy et al.
Figure 2. Viral Reporter Constructs and Screening Strategy
(A) Schematic representation of g-retroviral reporters used in the screen. MSCV ChIN encodes mCherry and neomycin. MSCV PIG contains GFP and puromycin.
(B) Schematic representation of lentiviral constructs used to deliver the shRNAmirlibrary. Two versions of the library were used to induce RNAi in XEN cells, one
containing puromycin and GFP (PEG) while the other contained neomycin and mCherry (NEC).
(C) Screening strategy for epigenetic modifiers of g-retroviral silencing. Left, infectious MSCV ChIN g-retroviral particles were delivered into XEN cells. Cells were
selected in G418 for 3 days to obtain pure populations, withdrawn from drug selection, then cultured for ten passages to induce viral reporter silencing. Once
silencing wasconfirmed,packaged, infectious lentiviral particles fromthePEG shRNAmirexpressioncassettewereusedtotransduce XENcells atlowmultiplicity
to achieve ?1 shRNAmirintegrant/cell. After 1 week in culture, cells were selected with G418 to enrich for MSCV ChIN activation. Right, a similar experiment was
conducted with the MSCV PIG g-retroviral reporter, NEC shRNAmirlentiviral construct, and puromycin drug selection. Surviving colonies from both experiments
were isolated after another week in culture, and shRNAs targeting candidate genes were identified via PCR amplification.
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
460 Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc.
(2009) identified a role for the class 1 histone deacetylases
HDAC1-3 in suppressing retroviral transcription. Here, we have
potentially expanded the family of HDACs involved in retroviral
extinction to include both class 1 (HDAC2) and class 2 (HDAC7
and HDAC9) histone deacetylases.
One interesting candidate to emerge from our screen is
ponent of the self-renewal process in ESCs (Hu et al., 2009), is
a putative subunit of the Polycomb Repressive Complex 1
(PRC1) (Lee et al., 2007). PRC1 is a multimeric protein complex
known to possess histone demethylating activity. Specifically,
PCGF6 directly associates with members of the Jumonji AT-rich
interactive domain (JARID) family of proteins, and thus the enzy-
matic removal of histone 3 lysine 4 trimethylation (H3K4me3)
(Lee et al., 2007). When lysates were precipitated with an anti-
body recognizing H3K4me3, a 60-fold reduction in signal was
observed between passages 2 and 6 (Figure 4B).
The third protein complex identified in our screen was PRC2,
a repressive complex composed minimally of EED, Enhancer
of zeste homolog 2 (EZH2), and Suppressor of zeste 12 homolog
(SUZ12) (Schwartz and Pirrotta, 2007; Sparmann and van Lohui-
zen, 2006). Two of the three core components (EED and EZH2)
were identified in our screen, strongly implicating this complex
in proviral silencing. Furthermore, PRC2 interacts with sup-
pressor of variegation 3-9 homolog 1 (SUV39H1), HDAC2, and
the DNMT family of proteins, all of which were identified in our
screen. Through these interactions, PRC2 mediates transcrip-
tional repression via histone 3 lysine 9 (H3K9) and lysine 27
(H3K27) methylation, as well as by directing DNA methylation
(Cao et al., 2002; Kuzmichev et al., 2002; Sewalt et al., 2002;
Vire ´ et al., 2006). To examine the role of PRC2 in viral extinction,
we examined changes in H3K9 trimethylation (H3K9me3) and
H3K27 trimethylation (H3K27me3) via ChIP. When lysates were
immunoprecipitated with antibodies recognizing H3K9me3,
a 300-fold increase in signal was detected between passages
2 and 6 (Figure 4B). By comparison, a slower, modest increase
in H3K27me3 signal was observed over the experimental time
PRC2 associates with all three members of the DNA methyl-
transferase family and directs DNA methylation of EZH2 target
promoters (Vire ´ et al., 2006). Given the strong association
between DNA methylation and retroviral transcriptional silencing
(Cherry et al., 2000; Harbers et al., 1981) and the identification of
both DNMT1 and DNMT3A in our screen, we examined the DNA
methylation status of the viral LTR in XEN cells undergoing viral
extinction by using the bisulphite mutagenesis and sequencing
analysis. As a control, NIH 3T3 cells were infected and time
points measured in parallel. De novo DNA methylation at the
MSCV 50LTR was progressively acquired over time, such that
by passage 7, DNA strands were hypermethylated (Figure 4C).
In contrast, MSCV LTR remained hypomethylated in NIH 3T3
cells over the period assayed. Collectively, these data indicate
that the significant change in local chromatin structure from
a transcriptionally permissive to silent state was probably due
to the identified repressive complexes.
Epigenetic Factors Localize to Retroviral Elements
in Silenced XEN cells
To further examine a potential role for the identified factors in
regulating g-retroviral extinction, it was necessary to establish
a link between the candidates and transcriptional regulation of
integrated proviral elements. To determine whether candidate
proteins were directly associated with 50MSCV viral LTR, ChIP
analysis was conducted with proteins for which ChIP-grade anti-
bodies were available: AGO1, HDAC2, HDAC7, SMARCA1, and
SMARCA5. All assayed proteins produced enrichment of the
by qPCR, with the highest enrichment for SMARCA1 and AGO1.
Candidate Epigenetic Factors Are Required
for Endogenous Retroviral Silencing
To further establish a link between candidate factors and tran-
to endogenous retroviral elements. Endogenous retroviruses
are remnants of ancient viral infections that persist within the
genome. Generally, endogenous retroviral elements are main-
tained in a silent state. To determine whether the identified
epigenetic factors were involved in retroviral transcriptional
repression, we conducted shRNA-mediated depletion of candi-
date factors in XEN cells and then assessed their effect on
endogenous retroviral transcription. To this end, transcript levels
Table 1. Top 25 Candidate Genes Mediating g-Retroviral
Silencing in XEN Cells
Gene NameOther Names Score Complex Association
Smarca1 Snf2L, Nurf140637NURF
Smarca5Snf2h, MommeD4349PRC1, SIN3a-HDAC,
Hdac2Rpd3, Yaf1, Yyibp 180PRC2, SIN3a, NcOR
Hdac9 Hdac7b, Hdrp, Mitr135?
Dnmt3a90 PRC2, PRC3
Eed 77 PRC2, PRC3, PRC4
Myst4 Morf, Moz2,
Smarcd1Baf60a 36 BRG, BAF
Myst2Hbo1, Hboa, Kat721?
18 DNMTs / HMTs
Ezh2Enx-1, Kmt614PRC2, PRC3, PRC4
Actl6bActl6, Baf53b, ArpNa8BAF
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc. 461
of the Line1, IAP, and virus-like 30S (VL30) retro-elements were
quantitated in cells containing shRNAmirs targeting the top ten
candidates, via qRT-PCR analysis. Comparisons were made to
controls, wild-type XEN cells, MSCV ChIN-infected XEN cells,
MCSV PIG-infected XEN cells, and Luc shRNAmir-targeted
XEN cells. Results were normalized to wild-type expression
levels. All tested shRNAmirs demonstrated transcriptional activa-
tion for atleast oneendogenous retroviral family (Figure 5B),with
preferential activation of specific retro-elements by some candi-
dates. For example, depletion of Pcgf6 mRNA produced the
strongest transcriptional activation of Line1 and VL30 elements,
whereas RNAi directedagainst Smarca5produced the strongest
activation of IAP elements (Figure 5B), consistent with the latter
as an agouti yellow gene modifier (Chong et al., 2007).
To determine whether candidates identified in XEN cells were
also involved in silencing retro-elements in the embryonic
lineage, RNAi depletion of each top ten candidate was con-
Figure 3. Comparison of Candidate Factor
Transcript Levels between ESCs, TSCs,
and XEN Cells
normalized to the geometric mean of transcripts
encoding housekeeping genes b-actin, 7SK RNA,
and Hexokinase. Error bars, SEM.
ducted in ESCs (Figure 5B). Similar to
XEN cells, depletion of Smarca5 induced
retro-elements. However, unlike XEN
cells, depletion of neither Pcgf6 nor
Smarca1 induced retro-element reactiva-
in ESCs showed the strongest activation
of retro-elements as opposed to the
more modest reactivation seen in XEN
cells. These data indicate that factors
identified in this screen not only interact
with and silence the MSCV g-retrovirus
but also play significant roles in repres-
sing transcriptional activity of endoge-
In this study, we investigated viral extinc-
tion in ESCs, TSCs, and XEN cells. Our
work shows that the three cell lineages
of the early mammalian embryo have
vastly different viral silencing strategies
retroviral activity. Cells derived from the
inner cell mass have long been shown
to repress retroviral activity (Harbers
et al., 1981; Ja ¨hner et al., 1982; Stewart
et al., 1982). This capacity was explained
retroviral activity, given that insertional
mutagenesis during early embryonic time points could poten-
tially create detrimental heritable mutations, reducing reproduc-
tive fitness (Loh et al., 1990; Niwa et al., 1983; Teich et al., 1977).
In comparison to embryonic stem cells, extraembryonic cells
possess vastly different responses to retroviral integration. We
observed that trophectoderm stem cells fail to silence retroviral
transcription. Consistent with this, early studies of placental
tissue via electron microscopy found an abundance of endoge-
nous type C retroviral particles in syncytial trophoblasts, sug-
gesting that robust retroviral activity occurs within this tissue
type (Gross et al., 1975; Kalter et al., 1973). Later, it was discov-
ered that long terminal repeats of active retroviral elements act
viral envelope genes confer fusogenic and immunosuppressive
functions (Cohen et al., 2009; Mi et al., 2000; Rawn and Cross,
2008). One explanation for exaptation of retroviral elements
is that relaxation of epigenetic control of specific retroviral
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
462 Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc.
elements may have coevolved with placental development,
enabling expression of these essential retroviral factors.
Prior to this study, one might have assumed that cells of
extraembryonic origin would have similar strategies for regu-
latingretroviralelements.Theunexpectedly aggressive retroviral
extinction observed in XEN cells indicates that the permissive
transcriptional environment seen in the trophectoderm lineage
is not a shared feature of extraembryonic stem cell types. Our
results indicate a potent antiviral response requiring the interac-
tion of at least three repressive complexes to initiate, establish,
and maintain retroviral silencing.
Despite the fact that viral extinction has been studied for
decades, the precise epigenetic factors employed in silencing
retroviruses had yet to be determined. To identify specific
biochemical factors that mediate this epigenetic response, we
utilized loss-of-function RNAi screens for epigenetic factors
that mediate silencing of integrated retroviruses. Interestingly,
18 of the top 25 candidates associate with 1 of 3 epigenetic
modifier complexes (Chromatin remodeling complex, Polycomb
repressive complex 1, and Polycomb repressive complex 2), all
of which have well-characterized roles in posttranslational modi-
fication of chromatin to a transcriptionally silent state (Roberts
and Orkin, 2004; Schwartz and Pirrotta, 2007; Sparmann and
van Lohuizen, 2006). Three of the identified factors have known
interactions with viral elements. SMARCA5 and DNMT1 were
identified as modifiers of intracisternal A particles within the
agouti variable yellow locus (Chong et al., 2007). Moreover,
HELLS was identified as a potent modulator of endogenous
retroviral methylation and expression (De La Fuente et al.,
2006). In contrast, the majority of candidates identified in this
screen are novel regulators of retroviral silencing. This work
represents the first step toward understanding how mammalian
genomes recognize integrated parasitic nucleic acids and the
biochemical responses initiated to silence these elements.
One striking observation from this study was that TSCs failed
to silence the retroviral reporter. Given that very little Smarca1
expression was detected in TSCs in comparison to ESCs and
XEN cells, it is tempting to speculate that SMARCA1 may play
Figure 4. MSCV Extinction Correlated with a Switch from Transcriptionally Permissive to Constrained Chromatin
(A) Schematic representation of 50MSCV viral long terminal repeat (LTR) and packaging signal (c, Psi) regions. Open circle, CpG dinucleotides. Arrow, transcrip-
tion start site.
(B) Loss of active and gain of repressive histone modification during retroviral silencing. ChIP analysis of integrated 50MSCV LTR and Psi regions. qPCR
measured enrichment after immunoprecipitation via antibodies to H3Ac, H3K4me3, H3K27me3, and H3K9me3 at passages 2, 6, and 10. Measurements
were normalized to the Gata4 promoter. Error bars, SEM.
(C) MSCV LTR became hypermethylated in XEN cells. Methylation status of LTR in individual DNA strands of cultured XEN cells (passage 1, 3, 5, 7, 10) as deter-
mined by bisulfite mutagenesis and sequencing analysis. Unmethylated CpGs are represented as empty circles while methylated CpGs are depicted as filled
circles. Each line denotes an individual DNA strand. Percent methylation was number of methylated CpGs over total number of CpGs. NIH 3T3 fibroblasts
were infected as a negative control. ND, not determined.
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc. 463
a major role in retroviral silencing. Experiments generating
ectopic Smarca1 expression in TSCs would determine whether
SMARCA1 confers the capacity to silence integrated virus. With
regard to the aggressive versus the slow viral extinction of XEN
cells and ESCs, respectively, the similar or greater expression
of the remaining candidate factors in ESCs indicates that they
are not likely to provide differential viral silencing capacities,
with the exception of SMARCC2. More abundant expression of
and greater silencing activity of endogenous retro-elements
by SMARCC2 in XEN cells is suggestive of a role in rapid viral
many complexes are common between the three stem cell line-
ages, significant differences also exist, potentially explaining
the varied capacities to extinguish proviral transcription. We
and others postulate that multimeric epigenetic complexes
have a core set of proteins that mediate their chromatin-modi-
fying activity while possessing a subset of ‘‘exchangeable’’
proteins that confer target specificity (Wolf and Goff, 2009). In
the case of XEN cells, SMARCC2 may provide specificity.
Another interesting finding to emerge from this study is that
multiple silencing pathways can act to silence proviral transcrip-
of endogenous retro-elements. Because different retro-element
promoter regions lack primary sequence homology, no single
sequence or binding protein could be utilized to direct broad
spectrum silencing (Mandal and Kazazian, 2008). This suggests
that multiple repressive complexes capable of targeting distinct
function for different retroviral elements. Furthermore, given the
importance of restraining endogenous retro-element transcrip-
tion to genome stability, it is likely that functional redundancy
has also evolved (Kazazian and Goodier, 2002). Support for
this hypothesis comes from the observation that depletion of
most candidate factors generated only a 2- to 12-fold reactiva-
tion of endogenous retro-elements. These results align with
those reported for Dnmt1?/?and Np95?/?as well as Suv39h1/
h2-and Dnmt3a/3b double null ESCs (Martens et al., 2005;
Dong et al., 2008; Sharif et al., 2007).
Although we and others have identified several candidate
proteins, which modify proviral chromatin (Chong et al., 2007;
De La Fuente et al., 2006), no mechanism has been identified
by which epigenetic modifiers are recruited to viral integration
sites. One candidate emerging from our screen, Eukaryotic
translation initiation factor 2C1 (EIF2C1), better known AGO1,
is suggestive of a potential recruitment mechanism. AGO1 is
a member of a clade of RNA-binding proteins that form the cata-
lytic core of the RNA-induced silencing complex (Carmell et al.,
2002). Thus, one potential mechanism of recruitment is through
an RNA interaction via homologous siRNAs processed from
virally derived, double-stranded RNA. Recognition of parasitic
nucleic acids through DNA:RNA or RNA:RNA pairing is a
potential mechanism for the genome to distinguish ‘‘self’’ from
parasitic ‘‘non-self’’ (Malone and Hannon, 2009; Tam et al.,
2008; Watanabe et al., 2006; Aravin et al., 2007).
The capacity of retroviral elements to modify expression of
protein coding genes and influence organism phenotype was
first described by Barbara McClintock (McClintock, 1956).
Clearly, eukaryotic cells encountered retroviral challenges very
early in evolution and evolved complex strategies for suppress-
ing these integrated elements. The subsequent molecular arms
race that ensued between the host and pathogen allowed
development of complex epigenetic mechanisms, which may
have been adapted to other areas of transcriptional regulation
between endogenous retroviral elements and the epigenome
will aid our understanding of basic biology of transcriptional
regulation and mammalian development.
Figure 5. Epigenetic Factors Localized to MSCV LTR in Silenced
XEN Cells and Were Required for Endogenous Retroviral Silencing
(A) Top candidate proteins localized to MSCV 50LTR and Psi elements. ChIP
analysis of integrated LTR was performed with antibodies specific for AGO1,
HDAC2, HDAC7, SMARCA1, and SMARCA5. All candidates demonstrated
strong enrichment of the provirus compared to input via qPCR.
(B) Distinct complexes repress different families of endogenous retroviral
elements. qRT-PCR analysis of Line1 (L1), IAP, and VL30 retro-element fami-
lies in controls and candidate-depleted ESCs and XEN cells. Transcriptional
activity of each retrovirus family via two independent primer sets was normal-
ized to Mrpl1. Error bars, SEM.
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
464 Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc.
Vector construction and oligonucleotide sequences can be found in Supple-
C57BL/6 female mice (B6) (Jackson Laboratory) were superovulated by intra-
peritoneal injection of 6.25 IU pregnant mare’s serum gonadotropin (Intervet
Canada) followed by 6.25 IU human chorionic gonadotropin (Intervet Canada)
44–48 hr later. Females were mated with Mus musculus castaneus (CAST)
males (Jackson Laboratory). Pregnancy was determined by a vaginal plug
the following morning (day 0.5). F1hybrid embryos were flushed from genital
tract of females at day 3.0 to recover morulae. Primary ESCs, TSCs, and
XEN cells were derived from B6XCAST F1embryos as previously described
(Kunath et al., 2005; Nagy et al., 1993; Tanaka et al., 1998). In brief,
ESC cultures were maintained in DMEM (Sigma D5671) supplemented with
50 mg/ml penicillin/streptomycin (Sigma), 100 mm b-mercaptoethanol, 13 LIF
(Sigma), 2 mM L-Glutamine (Sigma), 13 MEM nonessential amino acids
(Sigma), and 15% hyclone ESC grade fetal bovine serum (FBS). TSC and
XEN cell cultures were maintained as described (Kunath et al., 2005; Tanaka
et al., 1998) via RPMI (Sigma R0883) supplemented with 50 mg/ml penicillin/
streptomycin, 1 mM sodium pyruvate, 100 mm b-mercaptoethanol, 1 mg/ml
heparin (Sigma), 2 mM L-glutamine, 13 FGF basic, 13 FGF4 (R&D Systems),
and 20% hyclone ESC grade FBS. Cells were grown on a mitomycin C
(Sigma)-treated feeder fibroblast layer. For experiments involving DNA meth-
ylation and histone deacetylation inhibitor treatment, XEN cell were cultured
in media supplemented with 0.15 mm 5-Aza-20-deoxycytidine (Sigma) and/or
200 nM Trichostain A (Sigma). Experiments were performed in compliance
with the guidelines set by the Canadian Council for Animal Care, and the poli-
ciesand procedures approvedbytheUniversityof WesternOntario Council on
Production of Recombinant Virus
Recombinant gamma and lenti class retroviral particles were prepared as
described (Lois et al., 2002). In brief, viral vectors described above, along
with plasmids encoding a vesicular stomatitis virus glycoprotein pseudotype
(plasmid pMDG) plus either Psi (pPsi) or Delta (pCMV-Delta 8.9) packaging
elements (for gamma or lenti viruses, respectively), were transfected into
6-well dishes containing NIH HEK293 cells (ATCC) at 70% confluency, via
calcium phosphate precipitation. Ratio of plasmids was 2.3 mg viral transfer
vector, 1.0 mg pMDG, and 1.7 mg pPsi or pCMV-DeltaR 8.91. Media was
replaced 24 hr posttransfection, and cells were cultured for an additional
48–72 hr. After which, virus-containing media was collected, syringe filtered
through a 0.45 mm filter, and aliquoted into 2 ml snap cap tubes.
Somatic and Stem Cell Infection
NIH 3T3 (ATCC) cells, ESCs, TSCs, and XEN cells were seeded into 24-well
plates to give a density of 50% confluence after 12–18 hr of growth. The
next day, cells were infected by delivery of filtered virus directly into the culture
media along with 13 polybrene solution (final concentration). Cells were spun
at 1000 3 g for 1 hr and cultured in viral media overnight. The following day,
media was changed, and cells were incubated for 24–48 hr before assessing
fluorescence transgene expression. Once GFP/mCherry expression was
established, cells were subpassed via the standard protocol for each cell
type, and selection with the appropriate antibiotic was initiated. Cells were
infected with a dilution series of viral particles to ensure that less then 20%
of cells became infected, as measured by GFP or mCherry expression. Low
levels of infection were crucial to ensuring approximately a single viral integra-
tion event per cell. NIH 3T3 cells were selected with 2 mg/ml puromycin or
400 mg/ml neomycin, while XEN cells were selected with 1 mg/ml or 200 mg/ml
puromycin and neomycin, respectively (Sigma Aldrich). Studies examining
candidate suppression in ESCs utilized siRNAs (QIAGEN) (Table S2) trans-
ected into cells via Lipofectamine 2000 (Invitrogen) according to the manufac-
For cytometry analysis, cells were washed twice with warm PBS and trypsi-
nized with 13 trypsin (Sigma) into a single-cell suspension. Cells were spun
and resuspended in 10% FBS PBS solution. Cells were then analyzed on
a Beckman Coulter Epics XL-MCL Flow Cytometer by normalizing readings
first to nontransgenic wild-type cells, and then measuring fluorescence of 3
independent groups of 50,000 cells for each sample. Results from three
independent measurements were averaged and standard error of the mean
RNA Isolation and Reverse Transcription
dissociated with 13 trypsin (Sigma). Cells were spun down and washed once
in cold PBS, then RNA was isolated with Trizol (Invitrogen) according to the
manufacturer’s protocol. 1 mg of purified total RNA was treated with amplifica-
tion-grade DNaseI (Invitrogen) according to the manufacturer’s protocol, and
then reverse transcribed with the SuperScriptII system (Invitrogen) by
combining 1 ml random hexamer oligonucleotides (Invitrogen), 1 ml 10 mM
dNTP (Invitrogen), and 11 ml RNA plus water. This mixture was brought to
70?C for 5 min and then cooled to room temperature. SuperScriptII reaction
buffer, DTT (Invitrogen), and SuperScriptII were then added according to
manufacturer’s protocol and the mixture was brought to 25?C for 5 min,
42?C for 50 min, 45?C for 50 min, 50?C for 30 min, then 70?C for 5 min.
Real-Time PCR Ampification
Real-time PCR analysis of mRNA levels was carried out with the iQ SYBR
Green Supermix (BioRad) according to the manufacturer’s instructions. Reac-
tions were performed on a MJ Thermocycler Chromo4 Real Time PCR system
(BioRad). Samples were normalized to the mouse ribosomal binding protein
(Mrpl1) gene or the geometric mean of transcripts encoding b-actin, 7SK,
and Hexokinase (Mamo et al., 2007).
PCR Analysis of Drug-Resistant Colonies
Cells undergoing selection during the screening process were grown in 20 cm
dishes, emerging drug-resistant colonies were picked, and then subpassaged
into 24-well plates under continued selection to maintain clonal populations.
Upon confluence, colonies were harvested and DNA isolated via the QIAGEN
DNeasy Blood & Tissue Kit, according to manufacturer’s protocol. To identify
shRNAmirs present in surviving colonies, 100 ng genomic DNA was seeded
into a PCR reaction (PCR hot start Ready-to-go beads, GE) containing a 50
GFP or mCherry primer and a 30woodchuck response element (WRE) primer
located within the lentiviral backbone. Amplicons were sequenced at the
London Regional Genomic Centre with GFP or mCherry primers.
Chromatin Immunoprecipitation Analysis
Cultured cells were grown to 80% confluence, washed twice in warm PBS,
trypsinized, and then resuspended in warm growth media containing
0.1 volume of crosslinking solution (Kondo et al., 2004). Subsequently, ChIP
reactions were performed as described (Martens et al., 2005), which was fol-
lowed by DNA purification with a QIAGEN PCR Cleanup kit. Antibodies used in
the immunoprecipitation of modified histones and candidate proteins were
anti-Rabbit IGG (Santa Cruz SC-2027), anti-Acetylated-Histone H3 (Millipore
06-599), anti-Trimethyl Histone H3 Lysine 4 (Millipore 04-745), anti-Trimethyl
Histone H3 Lysine 9 (Abcam Ab8898), anti-Trimethyl Histone H3 Lysine 27
(Millipore 17-622), anti-AGO1, monocolonal (Millipore 04-083), anti-AGO1
polyclonal (Millipore 07-599) (Janowski et al., 2006; Kim et al., 2006), anti-
HDAC2 (Abcam Ab7029), anti-HDAC7 (Abcam Ab50212), anti-SMARCA1
(Abcam Ab37003), and anti-SMARCA5 (SNF2H Abcam Ab3749). Antibodies
for modified histones were used at 1 mg/ChIP reaction while antibodies to
candidate proteins were used at 5 mg/ChIP reaction. The concentration of
IGG was adjusted from 1 mg to 5 mg as appropriate. For quantitative analysis,
real-time PCR was carried out with the iQ SYBR Green Supermix according to
the recommended protocol. Reactions were performed on aMJ Thermocycler
Chromo4 Real Time PCR system. Samples were normalized to measurements
taken for the Gata4 promoter, and data were analyzed with formula previously
described (Mukhopadhyay et al., 2008).
Sodium Bisulfite Mutagenesis and Sequencing Assay
Bisulfite mutagenesis and sequencing with agarose embedding was per-
formed (Market-Velker et al., 2010) with modification. Lysed cells (10 mL)
were embedded in 20 ml 2% low melting point agarose (Sigma). After bisulfite
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc. 465
mutagenesis, 22 ml diluted agarose was added to Ready-to-go PCR Beads
tions were halved, allowing for two independent PCR reactions. First round
product (5 mL) was seeded into each second round PCR reaction with
MSCV BIS F2 and R primers. Samples were sent to the Nanuq Sequencing
Facility or BioBasic Inc. for sequencing. To obtain representative number of
DNA strands, 15–20 clones were sequenced. Sequences with less than 95%
conversion rates were not included. Identical clones (identical location and
number of methylated CpGs and unconverted non-CpG cytosines) were
Supplemental Information includes Supplemental Experimental Procedures,
six figures, and three tables and can be found with this article online at
The authors thank Joe Mymryk and Peter Pelka for critical review of the manu-
script and Morgan McWilliam, Michelle Gabriel, and Fatima Bab’bad for tech-
nical assistance. Authors would also like to thank G.J. Hannon for reagents.
This work was supported by Research Grant MOP-81167 from the Canadian
Instituteof Health Research and grants from Lawson Health Research Institute
and Department of Obstetrics and Gynecology, University of Western Ontario.
M.R.W.M. was supported by the Ontario Women’s Health Council/CIHR
InstituteofGender and HealthNew Investigator Award. M.C.G.wassupported
by the Ontario Women’s Health Council/CIHR Institute of Gender and
Health Fellowship Award and the Dr. David Whaley Postdoctoral Fellowship
in Maternal/Fetal and Neonatal Research.
Received: August 7, 2009
Revised: January 30, 2010
Accepted: March 5, 2010
Published: May 6, 2010
Aravin, A.A., Hannon, G.J., and Brennecke,J.(2007). The Piwi-piRNA pathway
provides an adaptive defense in the transposon arms race. Science 318,
Best, S., Le Tissier, P.,Towers,G., and Stoye, J.P. (1996). Positional cloning of
the mouse retrovirus restriction gene Fv1. Nature 382, 826–829.
Cao, R., Wang, L., Wang, H., Xia, L., Erdjument-Bromage, H., Tempst, P.,
Jones, R.S., and Zhang, Y. (2002). Role of histone H3 lysine 27 methylation
in Polycomb-group silencing. Science 298, 1039–1043.
Carmell, M.A., Xuan, Z., Zhang, M.Q., and Hannon, G.J. (2002). The Argonaute
family: tentacles that reach into RNAi, developmental control, stem cell main-
tenance, and tumorigenesis. Genes Dev. 16, 2733–2742.
Cherry, S.R., Biniszkiewicz, D., van Parijs, L., Baltimore, D., and Jaenisch, R.
(2000). Retroviral expression in embryonic stem cells and hematopoietic
stem cells. Mol. Cell. Biol. 20, 7419–7426.
Chong, S., Vickaryous, N., Ashe, A., Zamudio, N., Youngson, N., Hemley, S.,
Stopka, T., Skoultchi, A., Matthews, J., Scott, H.S., et al. (2007). Modifiers of
epigenetic reprogramming show paternal effects in the mouse. Nat. Genet.
Cohen, C.J., Lock, W.M., and Mager, D.L. (2009). Endogenous retroviral
LTRs as promoters for human genes: A critical assessment. Gene 448,
De La Fuente, R., Baumann, C., Fan, T., Schmidtmann, A., Dobrinski, I., and
Muegge, K. (2006). Lsh is required for meiotic chromosome synapsis and
retrotransposon silencing in female germ cells. Nat. Cell Biol. 8, 1448–1454.
Dong, K.B., Maksakova, I.A., Mohn, F., Leung, D., Appanah, R., Lee, S., Yang,
H.W., Lam, L.L., Mager, D.L., Schu ¨beler, D., et al. (2008). DNA methylation in
ES cells requires the lysine methyltransferase G9a but not its catalytic activity.
EMBO J. 27, 2691–2701.
Gao, G., Guo, X., and Goff, S.P. (2002). Inhibition of retroviral RNA production
by ZAP, a CCCH-type zinc finger protein. Science 297, 1703–1706.
Grez, M., Akgu ¨n, E., Hilberg, F., and Ostertag, W. (1990). Embryonic stem cell
virus, a recombinant murine retrovirus with expression in embryonic stem
cells. Proc. Natl. Acad. Sci. USA 87, 9202–9206.
Gross, L., Schidlovsky, G., Feldman, D., Dreyfuss, Y., and Moore, L.A. (1975).
C-type virus particles in placenta of normal healthy Sprague-Dawley rats.
Proc. Natl. Acad. Sci. USA 72, 3240–3244.
Harbers, K., Schnieke, A., Stuhlmann, H., Ja ¨hner, D., and Jaenisch, R. (1981).
DNA methylation and gene expression: endogenous retroviral genome
becomes infectious after molecular cloning. Proc. Natl. Acad. Sci. USA 78,
Hu, G., Kim, J., Xu, Q., Leng, Y., Orkin, S.H., and Elledge, S.J. (2009).
A genome-wide RNAi screen identifies a new transcriptional module required
for self-renewal. Genes Dev. 23, 837–848.
Ja ¨hner, D., Stuhlmann, H., Stewart, C.L., Harbers, K., Lo ¨hler, J., Simon, I., and
Jaenisch, R.(1982).Denovomethylationand expression ofretroviralgenomes
during mouse embryogenesis. Nature 298, 623–628.
Janowski, B.A., Huffman, K.E., Schwartz, J.C., Ram, R., Nordsell, R., Shames,
D.S., Minna, J.D., and Corey, D.R. (2006). Involvement of AGO1 and AGO2 in
mammalian transcriptional silencing. Nat. Struct. Mol. Biol. 13, 787–792.
Kaiser, S.M., Malik, H.S., and Emerman, M. (2007). Restriction of an
extinct retrovirus by the human TRIM5alpha antiviral protein. Science 316,
Kalter, S.S., Helmke, R.J., Panigel, M., Heberling, R.L., Felsburg, P.J., and
Axelrod, L.R. (1973). Observations of apparent C-type particles in baboon
(Papio cynocephalus) placentas. Science 179, 1332–1333.
Kazazian, H.H., Jr. (2004). Mobile elements: Drivers of genome evolution.
Science 303, 1626–1632.
genome instability. Cell 110, 277–280.
Keedy, K.S., Archin, N.M., Gates, A.T., Espeseth, A., Hazuda, D.J., and
Margolis, D.M. (2009). A limited group of class I histone deacetylases acts
to repress human immunodeficiency virus type 1 expression. J. Virol. 83,
Kim, D.H., Villeneuve, L.M., Morris, K.V., and Rossi, J.J. (2006). Argonaute-1
directs siRNA-mediated transcriptional gene silencing in human cells. Nat.
Struct. Mol. Biol. 13, 793–797.
Kondo, Y., Shen, L., Yan, P.S., Huang, T.H., and Issa, J.P. (2004). Chromatin
immunoprecipitation microarrays for identification of genes silenced by
histone H3 lysine 9 methylation. Proc. Natl. Acad. Sci. USA 101, 7398–7403.
Kunath, T., Arnaud, D., Uy, G.D., Okamoto, I., Chureau, C., Yamanaka, Y.,
Heard, E., Gardner, R.L., Avner, P., and Rossant, J. (2005). Imprinted X-inac-
tivation in extra-embryonic endoderm cell lines from mouse blastocysts.
Development 132, 1649–1661.
Kuzmichev, A., Nishioka, K., Erdjument-Bromage, H., Tempst, P., and
Reinberg, D. (2002). Histone methyltransferase activity associated with
a human multiprotein complex containing the Enhancer of Zeste protein.
Genes Dev. 16, 2893–2905.
Laker, C., Meyer, J., Schopen, A., Friel, J., Heberlein, C., Ostertag, W., and
Stocking, C. (1998). Host cis-mediated extinction of a retrovirus permissive
for expression in embryonal stem cells during differentiation. J. Virol. 72,
Lee, M.G., Norman, J., Shilatifard, A., and Shiekhattar, R. (2007). Physical and
functional association of a trimethyl H3K4 demethylase and Ring6a/MBLR,
a polycomb-like protein. Cell 128, 877–887.
Loh, T.P., Sievert, L.L., and Scott, R.W. (1990). Evidence for a stem cell-
specific repressor of Moloney murine leukemia virus expression in embryonal
carcinoma cells. Mol. Cell. Biol. 10, 4045–4057.
Lois, C., Hong, E.J.,Pease, S., Brown, E.J.,and Baltimore, D. (2002). Germline
transmissionand tissue-specific expression of transgenes delivered by lentivi-
ral vectors. Science 295, 868–872.
Malone, C.D., and Hannon, G.J. (2009). Small RNAs as guardians of the
genome. Cell 136, 656–668.
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
466 Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc.
Mamo, S., Gal, A.B., Bodo, S., and Dinnyes, A. (2007). Quantitative evaluation
and selection of reference genes in mouse oocytes and embryos cultured
in vivo and in vitro. BMC Dev. Biol. 7, 14.
Mandal, P.K., and Kazazian, H.H., Jr. (2008). SnapShot: Vertebrate transpo-
sons. Cell 135, 192–192, e1.
Market-Velker, B.A., Zhang, L., Magri, L.S., Bonvissuto, A.C., and Mann, M.R.
(2010). Dual effects of superovulation: Lossof maternal and paternal imprinted
methylation in a dose-dependent manner. Hum. Mol. Genet. 19, 36–51.
Martens, J.H., O’Sullivan, R.J., Braunschweig, U., Opravil, S., Radolf, M.,
Steinlein, P., and Jenuwein, T. (2005). The profile of repeat-associated histone
lysine methylation states in the mouse epigenome. EMBO J. 24, 800–812.
Matsui, T., Leung, D., Miyashita, H., Maksakova, I.A., Miyachi, H., Kimura, H.,
Tachibana, M., Lorincz, M.C., and Shinkai, Y. (2010). Proviral silencing in
embryonic stem cells requires the histone methyltransferase ESET. Nature,
in press. Published online February 17, 2010. 10.1038/nature08858.
McClintock, B. (1956). Controlling elements and the gene. Cold Spring Harb.
Symp. Quant. Biol. 21, 197–216.
Mi, S., Lee, X., Li, X., Veldman, G.M., Finnerty, H., Racie, L., LaVallie, E., Tang,
X.Y., Edouard, P., Howes, S., et al. (2000). Syncytin is a captive retroviral
envelope protein involved in human placental morphogenesis. Nature 403,
Mukhopadhyay, A., Deplancke, B., Walhout, A.J., and Tissenbaum, H.A.
(2008). Chromatin immunoprecipitation (ChIP) coupled to detection by quanti-
tative real-time PCR to study transcription factor binding to DNA in Caeno-
rhabditis elegans. Nat. Protoc. 3, 698–709.
Nagy, A., Rossant, J., Nagy, R., Abramow-Newerly, W., and Roder, J.C.
(1993). Derivation of completely cell culture-derived mice from early-passage
embryonic stem cells. Proc. Natl. Acad. Sci. USA 90, 8424–8428.
Niwa, O., Yokota, Y., Ishida, H., and Sugahara, T. (1983). Independent mech-
anisms involved in suppression of the Moloney leukemia virus genome during
differentiation of murine teratocarcinoma cells. Cell 32, 1105–1113.
Pannell, D., Osborne, C.S., Yao, S., Sukonnik, T., Pasceri, P., Karaiskakis, A.,
Okano, M., Li, E., Lipshitz, H.D., and Ellis, J. (2000). Retrovirus vector silencing
is de novo methylase independent and marked by a repressive histone code.
EMBO J. 19, 5884–5894.
Poleshko, A., Palagin, I., Zhang, R., Boimel, P., Castagna, C., Adams, P.D.,
tain retroviral epigenetic silencing: evidence for an antiviral response. J. Virol.
Rawn, S.M., and Cross, J.C. (2008). The evolution, regulation, and function of
placenta-specific genes. Annu. Rev. Cell Dev. Biol. 24, 159–181.
cancer. Nat. Rev. Cancer 4, 133–142.
Rowe, H.M., Jakobsson, J., Mesnard, D., Rougemont, J., Reynard, S., Aktas,
T., Maillard, P.V., Layard-Liesching, H., Verp, S., Marquis, J., et al. (2010).
KAP1 controls endogenous retroviruses in embryonic stem cells. Nature
Schwartz, Y.B., and Pirrotta, V. (2007). Polycomb silencing mechanisms and
the management of genomic programmes. Nat. Rev. Genet. 8, 9–22.
Sewalt, R.G., Lachner, M., Vargas, M., Hamer, K.M., den Blaauwen, J.L.,
Hendrix, T., Melcher, M., Schweizer, D., Jenuwein, T., and Otte, A.P. (2002).
Selective interactions between vertebrate polycomb homologs and the
SUV39H1 histone lysine methyltransferase suggest that histone H3-K9 meth-
ylation contributes to chromosomal targeting of Polycomb group proteins.
Mol. Cell. Biol. 22, 5539–5553.
Sharif, J., Muto, M., Takebayashi, S., Suetake, I., Iwamatsu, A., Endo, T.A.,
Shinga, J., Mizutani-Koseki, Y., Toyoda, T., Okamura, K., et al. (2007). The
SRA protein Np95 mediates epigenetic inheritance by recruiting Dnmt1 to
methylated DNA. Nature 450, 908–912.
Silva, J.M., Li, M.Z., Chang, K., Ge, W., Golding, M.C., Rickles, R.J., Siolas, D.,
Hu, G., Paddison, P.J., Schlabach, M.R., et al. (2005). Second-generation
shRNA libraries covering the mouse and human genomes. Nat. Genet. 37,
Sparmann, A., and van Lohuizen, M. (2006). Polycomb silencers control cell
fate, development and cancer. Nat. Rev. Cancer 6, 846–856.
Stewart, C.L., Stuhlmann, H., Ja ¨hner, D., and Jaenisch, R. (1982). De novo
methylation, expression, and infectivity of retroviral genomes introduced into
embryonal carcinoma cells. Proc. Natl. Acad. Sci. USA 79, 4098–4102.
Takagi, N., and Sasaki, M. (1975). Preferential inactivation of the paternally
derived X chromosome in the extraembryonic membranes of the mouse.
Nature 256, 640–642.
Tam, O.H., Aravin, A.A., Stein, P., Girard, A., Murchison, E.P., Cheloufi, S.,
Hodges, E., Anger, M., Sachidanandam, R., Schultz, R.M., and Hannon, G.J.
(2008). Pseudogene-derived small interfering RNAs regulate gene expression
in mouse oocytes. Nature 453, 534–538.
Tanaka, S., Kunath, T., Hadjantonakis, A.K., Nagy, A., and Rossant, J. (1998).
Promotion of trophoblast stem cell proliferation by FGF4. Science 282,
Teich, N.M.,Weiss,R.A., Martin,G.R.,and Lowy, D.R.(1977).Virus infection of
murine teratocarcinoma stem cell lines. Cell 12, 973–982.
Turelli, P., Doucas, V., Craig, E., Mangeat, B., Klages, N., Evans, R., Kalpana,
G., and Trono, D. (2001). Cytoplasmic recruitment of INI1 and PML on
incoming HIV preintegration complexes: Interference with early steps of viral
replication. Mol. Cell 7, 1245–1254.
Vire ´,E.,Brenner,C., Deplus, R.,Blanchon,L.,Fraga,M.,Didelot,C.,Morey,L.,
Van Eynde, A., Bernard, D., Vanderwinden, J.M., et al. (2006). The Polycomb
group protein EZH2 directly controls DNA methylation. Nature 439, 871–874.
Watanabe, T., Takeda, A., Tsukiyama, T., Mise, K., Okuno, T., Sasaki, H.,
Minami, N.,and Imai,H.(2006).Identification and characterizationoftwonovel
classes of small RNAs in the mouse germline: Retrotransposon-derived
siRNAs in oocytes and germline small RNAs in testes. Genes Dev. 20,
Wolf, D., and Goff, S.P. (2009). Embryonic stem cells use ZFP809 to silence
retroviral DNAs. Nature 458, 1201–1204.
Xue, Y., Wong, J., Moreno, G.T., Young, M.K., Co ˆte ´, J., and Wang, W. (1998).
NURD, a novel complex with both ATP-dependent chromatin-remodeling and
histone deacetylase activities. Mol. Cell 2, 851–861.
Cell Stem Cell
Epigenetic Modifiers of Viral Extinction
Cell Stem Cell 6, 457–467, May 7, 2010 ª2010 Elsevier Inc. 467