• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cell. Author manuscript; available in PMC Dec 9, 2012.
Published in final edited form as:
PMCID: PMC3240866
NIHMSID: NIHMS338207

Polycomb Repressed Genes have Permissive Enhancers that Initiate Reprogramming

SUMMARY

Key regulatory genes, suppressed by Polycomb and H3K27me3, become active during normal differentiation and induced reprogramming. Using the well-characterized enhancer/promoter pair of MYOD1 as a model, we have identified a critical role for enhancers in reprogramming. We observed an unexpected nucleosome depleted region (NDR) at the H3K4me1-enriched enhancer at which transcriptional regulators initially bind, leading to subsequent changes in the chromatin at the cognate promoter. Exogenous Myod1 activates its own transcription by binding first at the enhancer leading to an NDR and transcription-permissive chromatin at the associated MYOD1 promoter. Exogenous OCT4 also binds first to the permissive MYOD1 enhancer, but has a different effect on the cognate promoter, where the monovalent H3K27me3-marks are converted to the bivalent state characteristic of stem cells. Genome-wide, a high percentage of Polycomb targets are associated with putative enhancers in permissive states, suggesting they may provide a widespread avenue for the initiation of cell-fate reprogramming.

INTRODUCTION

Epigenetic mechanisms regulate genomic output in normal tissue and are implicated in reprogramming (Maherali et al., 2007; Rideout et al., 2001). The roles of DNA methylation and histone modifications have been extensively studied in promoter regulation while the significance of nucleosome occupancy is increasingly being recognized (Hinshelwood et al., 2009; Kelly et al., 2010; Lin et al., 2007; Wolff et al., 2010; You et al., 2011). Distal regulatory regions such as enhancers also play important roles in regulating gene expression, though it has been difficult to identify enhancer/promoter pairs since they can be located at varied distances from transcriptional start sites (TSS) or act in trans (Atchison and Perry, 1988). Epigenome-wide studies have begun to establish chromatin signatures of active enhancers, which are DNase hypersensitive (Xi et al., 2007), have a moderate association with p300 (Heintzman et al., 2007; Visel et al., 2009; Wang et al., 2008), acetylation of Histone 3 Lysine 27 (H3K27Ac) (Creyghton et al., 2010; Rada-Iglesias et al., 2011) and a high correlation with Histone 3 Lysine 4 monomethylation (H3K4me1) (Heintzman et al., 2009; Heintzman et al., 2007; Koch et al., 2007). The presence of H3K4me1 and the absence of H3K27Ac characterises poised enhancers (Creyghton et al., 2010; Rada-Iglesias et al., 2011), which can be marked by H3K27me3 in embryonic stem cells (ES cells; Rada-Iglesias et al., 2011). Whether enhancers exist in a similar poised state when paired with promoters carrying only repressive marks (that is, H3K27me3 but not H3K4me3) has not been investigated. The relevance of enhancers paired with inactive genes and their effect on promoter epigenetic signatures is unclear.

In normal somatic cells, genes are typically expressed in a tissue-specific manner, or repressed by Polycomb repressive complex (PRC) and the associated H3K27me3 (Gal-Yam et al., 2008). Interestingly, PRC targets are usually repressed, yet poised for activation in ES cells, carrying the counteracting active (H3K4me3) and repressive (H3K27me3) modifications (Azuara et al., 2006; Bernstein et al., 2006). Therefore, repression of gene activity by PRC is reversible. Furthermore, various somatic cell types can be reprogrammed by over-expression of some key factors with essential roles in determining cellular identity (Boukamp et al., 1992; Hollenberg et al., 1993; Lassar et al., 1986; Weintraub et al., 1989). Reprogramming is an active area of interest (Daley et al., 2011) and it is not yet known how key transcriptional regulators initiate reprogramming, or if enhancers contribute to such events.

To investigate the role of enhancers in detail, we use the tissue-specific auto-regulatory MYOD1 gene as a model for understanding epigenetic interactions between enhancer/promoter pairs. MYOD1 has a well-characterized enhancer located ~20kB upstream of the TSS and contains a minimal core region of 258 base pairs (bp) that is necessary for promoter activity (Goldhamer et al., 1995). MYOD1 is expressed in myoblasts but repressed in normal non-muscle cells by PRC and H3K27me3 (Gal-Yam et al., 2008). The presence of a well-defined enhancer and a requirement for this transcription factor in muscle lineage determination make MYOD1 an optimal choice for investigation of chromatin structures of an enhancer/promoter pair in a variety of transcriptional contexts.

In this study, we used a high-resolution Nucleosome Occupancy and Methylome assay (NOMe-seq) to show that the MYOD1 minimal enhancer exhibits a striking nucleosome depleted region (NDR) that is bordered by H2A.Z containing nucleosomes marked with H3K4me1. This enhancer architecture can be associated with both active and repressed promoter states and is therefore more representative of a permissive state rather than just active enhancers. We found that the PRC occupied promoter exhibits a multivalent epigenotype in somatic cells and retains some regulatory flexibility, consistent with MYOD1 being in a transcriptionally competent state. In keeping with this observation, the forced expression of exogenous Myod1 results in binding first at the enhancer, followed by chromatin remodeling, causing the formation of a promoter NDR and endogenous expression. Importantly, we also show that reprogramming can be multi-directional. The binding of OCT4 to the MYOD1 enhancer also precedes occupancy at the promoter, which allows for establishment of a bivalent state that is characteristic of stem cells. Thus, in two distinct contexts, the permissive enhancer state orchestrates changes at the promoter through binding of master regulatory factors. We extended our analyses genome-wide and found that permissive enhancers marked by H3K4me1 not only regulate transcriptionally active promoters, but are also paired with PRC repressed promoters. We observed strikingly similar patterns in several somatic cell types, indicating that this is not a cell-type specific event. Our data suggest that the presence of counteracting epigenetic states at enhancer/promoter pairs ensures the correct tissue specific gene expression patterns of PRC targets, yet transcriptional flexibility is retained. These findings provide insight into the molecular events underlying reprogramming.

RESULTS

MYOD1, a model for studying epigenetic regulation of an enhancer/promoter pair

We performed detailed epigenetic analyses on the well-characterized regulatory regions of MYOD1 using two cell lines that contained this gene in distinct transcriptional contexts. Quantitative PCR confirmed that MYOD1 is expressed in a human rhabdomyosarcoma cell line (RD) but not normal human fibroblasts (LD419; Figure 1A). Bisulfite sequencing revealed that the MYOD1 enhancer and promoter were not permanently silenced by DNA methylation in fibroblasts (Figure 1A). ChIP assays confirmed enrichment of H3K4me3 (Figure 1B) and phosphorylated RNA polymerase II (Pol-IIP; Figure 1C) in RD cells, as well as Enhancer of Zeste 2 (EZH2; Figure 1D) and H3K27me3 (Figure 1E) in LD419 cells. Thus, the selected cell lines were suitable for studying the active (RD) and PRC repressed (LD419) contexts of MYOD1.

Figure 1
MYOD1 exhibits a multivalent epigenotype when repressed by PRC

The MYOD1 promoter exhibits a multivalent epigenotype when repressed by PRC in somatic cells

H2A.Z is associated with key regulatory regions of transcriptionally active and poised regulatory regions genome-wide (Barski et al., 2007; Creyghton et al., 2008). Consistent with this, H2A.Z was localized to the MYOD1 enhancer/promoter pair in expressing cells (RD; Figure 1F). We found that H2A.Z remained associated with the MYOD1 promoter even in the presence of EZH2 and H3K27me3 in fibroblasts (LD419; Figure 1F) and with the enhancer in these cells, though this distal region was devoid of H3K27me3 and EZH2 (Figure 1D, E). H2A.Z deposition at both proximal and distal gene regulatory regions may therefore serve as one mechanism by which PRC regulated genes remain permissive.

Previous studies have shown that H3K4me1 is localized to enhancers that act on transcriptionally active genes (Heintzman et al., 2007; Koch et al., 2007; Visel et al., 2009). We examined H3K4me1 profiles and as anticipated, detected this modification at the minimal enhancer in MYOD1 expressing cells (RD, Figure 1G). Unexpectedly, H3K4me1 was also detected at the enhancer in fibroblasts (LD419; Figure 1G). H3K4me1, but not the H3K4me3 mark characteristic of a bivalent promoter, extended into the TSS of MYOD1 in LD419 cells. Together, these data suggest that the repressed promoter exhibits a multivalent epigenetic signature marked by H2A.Z, H3K4me1 and H3K27me3 in somatic cells. The presence of counteracting histone modifications did not extend to the corresponding enhancer, which was only enriched for marks associated with transcriptional activity.

Nucleosome depleted regions characterize permissive enhancers

Nucleosomes are integral to epigenetic regulation and mapping their location is necessary to accurately view the epigenetic landscape. Taking advantage of the fact that mammalian cells are devoid of GpC methylation, we generate high-resolution nucleosome positioning maps by incubating intact nuclei with M.CviPI, a methyltransferase that recognizes GpC dinucleotides not associated with nucleosomes or tightly bound transcription factors (Kelly et al., 2010; Wolff et al., 2010; You et al., 2011). This technique provides a digital readout of nucleosome occupancy within individual DNA modules at CpG-rich and CpG-poor regions, while retaining information regarding the endogenous methylation states.

We examined nucleosome occupancy across the MYOD1 promoter and minimal enhancer in active and repressed states (Figure 2). A prominent NDR was detected in 80% of enhancer modules in cells expressing MYOD1 (RD; Figure 2A, left panel), overlapping exactly with the minimal enhancer and missing at least one nucleosome. Similar to the enhancer, we detected an NDR in approximately 70% of promoter modules in these cells (Figure 2A, right panel). Further, we found a highly positioned +1 nucleosome in 80% of promoter modules when an additional 75bp region downstream of the TSS was included in our analyses (data not shown). Together, these data show using a high-resolution, single molecule approach that enhancers have open configurations, similar to promoters of active genes.

Figure 2
Nucleosome depleted regions characterize permissive enhancers

To determine whether NDRs were characteristic of enhancers regulating PRC and H3K27me3 repressed promoters (Figure 1D, E), we examined nucleosome occupancy at the MYOD1 enhancer in fibroblasts (LD419 cells). We found a striking depletion of at least one nucleosome in approximately 40% of enhancer modules in LD419 cells, remarkably similar to actively expressing RD cells (compare Figure 2A and Figure 2B left panels). The presence of an NDR (Figure 2), and H2A.Z and H3K4me1 enrichment (Figure 1F, G), highlights the epigenetic similarities between the MYOD1 enhancer in two distinct transcriptional contexts. Unlike the enhancer, however, every promoter module was occupied by nucleosomes in LD419 cells (Figure 2B, right panel). Nucleosomes located immediately upstream of the TSS likely contribute to MYOD1 repression in somatic cells. These data show that although promoters repressed by PRC are occupied by nucleosomes, the corresponding enhancer maintains an unexpected open configuration similar to enhancers of transcriptionally active genes. Thus, specific histone modifications and transcription factors, at least that have been mapped to the enhancer in this study, do not predict the activity of the promoter, suggesting that the MYOD1 enhancer may exist in a permissive state in fibroblasts and retaining the potential for gene activation in response to the appropriate signals.

Transcriptional competence is determined by the epigenetic state of the enhancer

To determine whether the enhancer NDR underlies the permissive nature of PRC repressed genes and facilitates reprogramming of endogenous MYOD1, we transfected LD419 cells with a plasmid expressing TAP-tagged mouse Myod1 and examined the Myod1 binding kinetics using an anti-TAP antibody. At 6hr, Myod1 binding was not detected at either the MYOD1 enhancer or promoter (Figure 3A). Maximal association of Myod1 at the enhancer occurred by 24hr, at which time binding was not observed at the MYOD1 promoter (Figure 3B). These data are consistent with the presence of an NDR only at the enhancer (Figure 2), which would physically allow for Myod1 binding. By 48hr Myod1 not only associates with the enhancer but is also detected at the TSS of MYOD1 in fibroblasts (Figure 3C), suggesting that the promoter had been remodeled and an NDR generated, allowing Myod1 to bind.

Figure 3
The enhancer NDR underlies MYOD1 transcriptional competence

In support of this, we detected a newly formed NDR in ~34% of individual MYOD1 promoter modules in fibroblasts at 48hr post-transfection (LD419; Figure 3D, right panel). Interestingly, the proportion of enhancer modules exhibiting an NDR increased only marginally (<10%; Figure 3D, left panel) relative to untransfected cells (compare to Figure 2B, left panel). Despite Myod1 binding to its own regulatory regions (Figure 3B–C) and equal transfection (Figure S1A), endogenous MYOD1 mRNA expression was only detected when transfected cells were cultured in medium conditioned by RD cells (Figure S1B), suggesting a requirement for an additional factor that positively induces MYOD1 expression. Identifying specific factors is beyond the scope of this study, though possible candidates include thyroid hormones or insulin-like growth factors (Carnac et al., 1992; Pinset et al., 1988). Conditioned medium results in additional promoter chromatin remodeling in transfected cells, an increase in DNA modules with an NDR (60%; Figure S1C) and acquisition of active epigenetic marks (Figure S1D–F).

An enhancer NDR enables transcription factor binding and promoter chromatin remodeling

To investigate the requirement of the enhancer NDR in reprogramming events, we took advantage of the RKO colorectal cell line, which also does not express MYOD1. In contrast to fibroblasts, the enhancer and promoter of MYOD1 are occupied by nucleosomes in RKO cells (NOME-seq, data not shown), allowing us to determine the mechanistic significance of the permissive chromatin state. Thus, we transfected RKO cells with Myod1-TAP and compared Myod1 binding, as previously. We chose to examine Myod1 occupancy in these cells at 48hr post-transfection, which was at the time point when we saw maximal Myod1 binding at the MYOD1 enhancer/promoter pair in fibroblasts (Figure 3C). Similar to our kinetic study, we see Myod1 associating with the enhancer and promoter of MYOD1 in fibroblasts (Figure 3E, black bars). In striking contrast, the MYOD1 enhancer and promoter regions in RKO cells are devoid of Myod1 binding at 48hr post-transfection (Figure 3E, grey bars). Thus, Myod1 does not physically bind to the enhancer nor do we see subsequent binding or chromatin remodeling events at the MYOD1 promoter in these cells (Figure 3F).

MYOD1 gains a bivalent state after OCT4 initiates reprogramming through the enhancer

Induced pluripotent stem (iPS) cells can be generated from somatic cells by forced expression of key transcription factors, such as OCT4 (Maherali et al., 2007; Takahashi et al., 2007; Takahashi and Yamanaka, 2006). OCT4 association with PRC targets in ES cells correlates with gene inactivity and the presence of a bivalent epigenetic promoter signature (Bernstein et al., 2006). A bivalent signature (containing both H3K4me3 and H3K27me3) in ES cells, allows developmentally important genes to be poised for activation while remaining transcriptionally inactive. Interestingly, the MYOD1 enhancer contains an OCT4 consensus sequence. Critically, no such sequence exists at the promoter, providing a unique system to investigate changes in the enhancer/promoter epigenetic signature. By forcing expression of OCT4 in fibroblasts to initiate processes involved in iPS cell generation we sought to investigate the kinetic events associated with establishing a bivalent state. We transfected LD419 cells with a plasmid expressing OCT4 and examined the binding kinetics using an anti-OCT4 antibody. In mock transfected cells (0hr), only a minimal level of OCT4 binding was detected at both the MYOD1 enhancer and promoter (Figure 4A). OCT4 occupancy at the enhancer occurred by 24hr, at which time binding was not enriched at the MYOD1 promoter (Figure 4A). These data are highly similar to those observed for Myod1 (Figure 3B) and consistent with the presence of an NDR only at the enhancer (Figure 2). By 72hr we then detected OCT4 at the MYOD1 promoter (Figure 4A). Interestingly, 72hr after OCT4 over-expression, the H3K4me1 modification at the MYOD1 promoter was replaced by H3K4me3 while maintaining H3K27me3 (Figure 4B–D), indicating that the MYOD1 promoter reverted to a bivalent chromatin signature typical of developmentally important genes in iPS and ES cells. There is a lower level of H3K27me3 at both the promoter and enhancer in this state (Figure 4C, 72hr), consistent with previous reports that PRC activity is inhibited by the presence of H3K4me3 (Schmitges et al., 2011) and perhaps H3K4me1. We also analyzed the promoter and putative enhancer of GRP78, a constitutively expressed gene, to control for pulldown efficiency (Figure S2A). The enhancer was enriched for H3K4me1 in the basal state (0hr), and was not altered upon OCT4 overexpression (72hr). As expected, the promoter carried the H3K4me3 modification but was devoid of the repressive H3K27me3 mark even after OCT4 expression. Importantly, only a minimal level of OCT4 was present at the GRP78 enhancer and promoter after 72hr (Figure S2A), demonstrating that the presence of OCT4 is not generalized to all enhancers and promoters. Investigation of nucleosome occupancies showed that the bivalent state remains permissive in that the enhancer still exhibits an NDR, while the promoter is nucleosome occupied (Figure S2B). Together, these data offer insight into the molecular events that establish bivalent epigenetic signatures and provides additional evidence that reprogramming is likely initiated through nucleosome depleted enhancer regions.

Figure 4
MYOD1 gains a bivalent state after OCT4 initiates reprogramming through the enhancer

Enhancer/promoter crosstalk underlies reprogramming events

To understand the mechanisms underlying reprogramming of MYOD1, we next examined the involvement of protein complexes in facilitating enhancer/promoter interactions. To definitively demonstrate enhancer/promoter looping, a chromosome conformation capture assay (3C) can be performed. However, 3C can be unreliable when assessing genomic regions that are as close as 20kB apart (Dekker, 2006; Dekker et al., 2002; Miele and Dekker, 2009), as is the case for the MYOD1 enhancer and promoter. Nonetheless, the presence of the cohesin protein complex is consistent with DNA looping and enhancer/promoter interactions (Kagey et al., 2010). Thus, we measured the presence of the cohesin complex at the MYOD1 enhancer and promoter using an antibody raised against a highly conserved subunit of the complex, RAD21 (Schmidt et al., 2010). We observed RAD21 binding at the enhancer and promoter regions in expressing cells (RD, black bars; Figure S3A). An shRNA targeted knockdown of RAD21 caused a reduction in RAD21 protein (Figure S3B) and mRNA levels (Figure S3C, black bars). Importantly, this resulted in reduced MYOD1 expression in RD cells (Figure S3C, green bars), again consistent with the requirement of this complex for DNA looping. We next investigated the necessity for RAD21 in epigenetic reprogramming of MYOD1 in normal human fibroblasts, where MYOD1 is repressed. In LD419 cells a residual amount of RAD21 was bound to the MYOD1 promoter, while the enhancer and inter-regulatory regions were devoid of RAD21 (black bars; Figure 5). Following exogenous Myod1 expression, a small amount of RAD21 was detected at the enhancer at 24hr post-transfection, but promoter occupancy did not change (white bars, Figure 5), while at 48hr, RAD21 occupancy increases at both the enhancer and promoter (red bars, Figure 5). The presence of cohesin at both the enhancer and promoter is consistent with enhancer/promoter interaction (Kagey et al., 2010).

Figure 5
Enhancer/promoter interactions underlie reprogramming events

Taken together, our data demonstrate that transcription factor binding to the enhancer precedes binding to the promoter, providing evidence that the enhancer NDR underlies a permissive state and directs MYOD1 promoter chromatin remodeling in response to the appropriate signals (Figure 3D, S1B), and expression in the presence of the correct factors (Figure S1A). The ability of enhancers to drive changes at promoters provides a potential mechanism for reprogramming.

H3K4me1 marks enhancers of transcriptionally active, as well as PRC repressed genes

Epigenome-wide studies have identified correlations between enhancers marked by H3K4me1 and promoter transcriptional activity (Heintzman et al., 2007; Koch et al., 2007; Visel et al., 2009). However, these studies have not explicitly considered the epigenetic signatures of enhancers paired with inactive promoters. We sought to extend our investigation of MYOD1 and specifically asked whether enhancers marked by H3K4me1 could also be associated with PRC repressed promoters genome-wide, as indicated by the presence of the H3K27me3 mark.

Previous reports indicate that enhancers located within ~10kB of TSS are statistically linked to promoter activity and that those located upstream of the TSS are more likely to influence transcription (MacIsaac et al., 2010). While enhancers may be located several hundreds of kilobases away (Atchison and Perry, 1988), additional evidence suggests that enhancers are likely to act on the closest promoter in many situations (Xi et al., 2007). Therefore, we focused our analyses on those enhancers located in a 10kB window upstream of the TSS (from −2kB to −12kB) to encompass the same approximate region identified by MacIsaac et al., (2010), whilst eliminating the effect of epigenetic marks at adjacent promoters (<2kB). We probed publicly available ENCODE datasets using the epigenome-wide profiles of genomic regions containing either H3K4me1, H3K4me3 or H3K27me3 of the following human cell lines: ES cells (H1), B-lymphocytes (GM12878), human mammary epithelial cells (HMEC), normal human epidermal keratinocytes (NHEK) and normal human lung fibroblasts (NHLF) (Kellis et al., 2011).

We asked if TSS with promoters marked by either H3K4me3 or H3K27me3 had at least one putative enhancer marked by H3K4me1 within a 10kB window upstream of the TSS (−2kB to −12kB; Figure 6A, B). As expected, >70% of active promoters with H3K4me3 enriched promoters are paired with enhancers enriched for H3K4me1 across cell types, including ES cells (−2kB to −12kB; left y-axis, Figure 6A). The majority of these enhancers are DNase hypersensitive (i.e. likely depleted of nucleosomes) in normal somatic cells (56–64%; right y-axis, Figure 6A). We noted that the proportion of DNase hypersensitive enhancers, which are presumably nucleosome depleted, was lower in ES cells (49%) compared to somatic cells. H3K27me3 promoters could also have at least one putative enhancer enriched for H3K4me1 located in this 10kB window (Figure 6B). The percentage of H3K27me3 marked promoters with putative enhancers varied between somatic cell types (27 – 57%; Figure 6B, left y-axis) and was substantially higher in ES cells (87%; Figure 6B, left y-axis). Many enhancers paired with H3K27me3 marked promoters are sensitive to DNase (48–62%, Figure 6B, right y-axis), consistent with nucleosome depletion and as we found with MYOD1 in fibroblasts (Figure 2B). Together, these data suggest that H3K4me1 marked enhancers vary in their DNase hypersensitivity and are associated with both active and PRC repressed promoters.

Figure 6
H3K4me1 marks enhancers of transcriptionally active, as well as PRC repressed genes

To extend our findings beyond MYOD1 we examined two additional PRC target genes with validated enhancers PAX6 (Heintzman et al., 2009) and NODAL (Rada-Iglesias et al., 2011) within several ENCODE cell lines. The promoters of PAX6 and NODAL were enriched for H3K27me3 while their enhancers exhibited H3K4me1 in the cell types examined (Figure S4A, C). Examining the chromatin configurations of PAX6 and NODAL in more detail (Figure S4B, D), we showed that PAX6 and NODAL were devoid of H3K4me3 and Pol-IIP; yet enriched for EZH2, H3K27me3, H2A.Z and H3K4me1, consistent with expression (data not shown) in fibroblasts. These data confirm that the generality of our findings can be applied to additional genes and cell types.

Together, H3K4me1 is not a defining mark of “active” enhancers alone. Indeed, our data suggest that PRC repressed promoters can potentially be activated (Figures 3 and and4)4) and therefore, the enrichment of this histone modification more accurately represents “permissive” enhancers that co-exist with at least two distinct promoter contexts.

DISCUSSION

The ability of enhancers to regulate genes in cis and trans has made correlating enhancers with their promoters challenging. Our data emphasize the importance of understanding how enhancer/promoter pairs work together to obtain an accurate view of epigenetic gene regulation. Interestingly, we found that PRC repressed promoters have permissive enhancers in somatic cells. Conventionally, “active” enhancers (H3K4me1) are paired with transcriptionally active promoters (H3K4me3), yet we show that they can also potentially regulate multivalent promoters (H3K27me3/H2A.Z). Therefore, the term “active” to describe enhancers characterized by H2A.Z incorporation, H3K4me1 enrichment and nucleosome depletion may be misleading and thus, we propose the use of “permissive” to describe such enhancers. In this way, “permissive” describes the epigenetic signature, and also refers to the biological functions of these regulatory regions. While enhancers can regulate promoters at greater distances or on different chromosomes, Xi et al., (2007) showed that enhancers are likely to act on the closest promoter in many situations. We found that 27–57% of PRC repressed promoters have putative enhancers marked by H3K4me1 located −2 to −12kB upstream of their TSS and at least 48% of them are DNase hypersensitive. Therefore, we suggest that the majority of enhancers found to be non-functional in verification assays (~22%; Heintzman et al., 2009) might actually represent permissive enhancers that can potentially regulate a PRC occupied promoter.

It is formally possible that the permissive enhancers identified in our epigenome-wide analyses may be activating alternate gene promoters. Thus, we focused on human MYOD1, whose distal enhancer is extremely well characterized and to our knowledge has not been associated with any other gene. This enhancer is critical for MYOD1 activation during development and lineage commitment (Faerman et al., 1995; Goldhamer et al., 1995; Kablar et al., 1999). An additional enhancer located ~4.8kB upstream of the TSS also contributes to MYOD1 expression, yet this is restricted to a subset of cells that are already committed (Asakura et al., 1995; Kablar et al., 1997; Tapscott et al., 1992). The distal enhancer/promoter crosstalk we describe here points to a mechanism for this previous work. Using MYOD1 we show that an NDR at the minimal enhancer region allows reprogramming to be initiated, which occurs in response to signals such as the forced expression of Myod1 in fibroblasts. The nucleosome footprint at the minimal enhancer was only slightly altered in response to Myod1 expression, suggesting that additional chromatin remodeling of enhancer modules is not required for epigenetic crosstalk between the enhancer/promoter pair. In contrast, the endogenous MYOD1 promoter is dramatically remodeled and an NDR is formed in response to Myod1 expression. It is unclear how the enhancer NDR is established and maintained in the absence of an active promoter. Nucleosome assembly may be inhibited by the binding of an unidentified transcription factor(s) that establishes and maintains the NDR (eg. You et al., 2011). These events did not occur when a nucleosome occupies the MYOD1 enhancer and impedes transcription factor binding. Thus, it is possible that an unidentified factor associates with the MYOD1 enhancer in LD419 but not in RKO cells, explaining the difference in chromatin architecture and the inability of these cells to be reprogrammed. Nonetheless, our data highlight the importance of the underlying chromatin state in reprogramming events.

The MYOD1 enhancer also contains an OCT4 consensus sequence, which is not present at the promoter, providing an unequivocal system for investigating changes in the epigenetic signature. Thus, any epigenetic changes that are detected, and the appearance of OCT4 at the promoter (even though there is no binding site), is most likely explained by enhancer/promoter interactions. By forcing OCT4 expression in fibroblasts to initiate processes involved in iPS cell generation, we show the kinetic events associated with establishing a bivalent state. As a bivalent state is generated H3K27me3 is reduced at the enhancer. Whether H3K27me3 is replaced by H3K27Ac is yet to be determined. The replacement of H3K4me1 by H3K4me3 at the promoter, while maintaining H3K27me3, indicates that the MYOD1 promoter reverts to a bivalent chromatin signature in response to forced OCT4 expression. OCT4 may play a direct role in establishing bivalency at MYOD1, although it is also possible that OCT4 expression causes global changes that indirectly lead to this promoter state. A more detailed kinetic study might distinguish between these possibilities.

The OCT4 experiments, together with Myod1 show that reprogramming is multidirectional and is ultimately determined by the master transcription factor. Regardless, reprogramming is driven through the nucleosome depleted enhancer, representing a permissive state and supporting a model that transcription factor binding first to the enhancer is responsible for epigenetic changes at the promoter (Figure 7).

Figure 7
Polycomb repressed promoters maintain regulatory flexibility through epigenetic crosstalk with their permissive enhancers

Our study demonstrates that the MYOD1 promoter is multivalent, carrying epigenetic marks correlated with activation (H3K4me1 and H2A.Z) as well as repression (H3K27me3 and EZH2) in somatic cells. Similar multivalency has been described whereby H2A.Z localizes to H3K27me3 enriched genomic regions in ES cells (Creyghton et al., 2008). This is comparable to the bivalent signature described in ES cells (H3K4me3 and H3K27me3) (Azuara et al., 2006; Bernstein et al., 2006), which likely keeps genes poised for induction. The multivalent epigenotype may be generated from a bivalent state in ES cells, or generated during lineage commitment. The maintenance of multivalency and the ability to acquire these epigenetic modifications in the correct order may explain why some somatic cells are more suitable for generating iPS cells (Maherali et al., 2007; Takahashi et al., 2007; Takahashi and Yamanaka, 2006) and why some are resistant to reprogramming (Boukamp et al., 1992).

The specific roles of H2A.Z and H3K4me1 in maintaining a permissive environment are unclear. H2A.Z is localized to the MYOD1 enhancer and promoter in both RD cells and LD419 cells, consistent with its localization to active and poised regulatory regions. The incorporation of H2A.Z may lead to altered nucleosome stability in certain contexts (Jin et al., 2009), providing a reason for H2A.Z marked nucleosomes at the PRC occupied promoter. The role of H3K4me1 is less clear and the fact that it marks both the promoter and enhancer of PRC repressed genes is interesting. Moreover, H3K4me1 is resolved to H3K4me3 at the promoter in expressing cells. The resolution of H3K4me1 to a di- or tri-methyl group is not observed at the enhancer, irrespective of the transcriptional state, suggesting that it may have independent roles at distinct gene regulatory regions. It is possible that the H3K4me1 mark acts as a foundation for the addition of methyl groups, resulting in H3K4me2 and H3K4me3 and perhaps priming the promoter for activation. Such dynamic changes have been reported in studies of H3K4me kinetics in developing embryos (Lepikhov and Walter, 2004). Finally, the presence of H3K4me1 may simply prevent spurious epigenetic activity. It is feasible that enrichment of H3K4me1 interferes with epigenetic silencing mechanisms, much like H2A.Z is mutually exclusive with DNA methylation.

Particular emphasis was placed on nucleosome occupancy in our study, utilizing the high-resolution NOME-seq approach, which provides direct evidence that nucleosomes contribute to the repression of PRC target gene promoters in somatic cells, as well as the finding that an NDR is present at the enhancer in opposing transcriptional states. Notably, these data are presented in the context of an enhancer/promoter pair, which we show is necessary to draw accurate conclusions regarding the status of a gene. Our results also indicate that H3K4me1 marked enhancers contribute to the epigenetic regulation of PRC target genes across the genome. Biologically, the presence of an NDR might enable the enhancer to respond to activating signals, which triggers promoter chromatin remodeling and causes a promoter NDR to form. This may explain how transcription factors gain access to their targets and how repressive chromatin states impede reprogramming events. In fact, our data demonstrate that the enhancer of PRC target genes may allow these loci to remain permissive in somatic cells, retaining epigenetic plasticity and initiating reprogramming.

EXPERIMENTAL PROCEDURES

Cell Culture

The rhabdomyosarcoma (RD) and colorectal (RKO) cell lines were obtained from the American Type Culture Collection (ATCC) and cultured under recommended conditions. A normal human fibroblast cell line (LD419) was generated in our laboratory and cultured in McCoy's 5A medium supplemented with 20% FBS, 100U/mL penicillin and 100µg/mL streptomycin at 37°C and 5% CO2.

RNA Isolation and Quantitative PCR Analysis

RNA was isolated using Trizol reagent, digested with DNase I, and reverse transcribed (iScript, BioRad). cDNA was amplified using the CFX96 Real-time PCR detection system (BioRad) and SYBR Fast qPCR Mix (Kapa Biosystems) under the following conditions: 95°C for 3 mins, followed by 45 cycles of 95°C for 3 seconds then 60°C for 30 seconds. A melt curve analysis was performed (60–95°C, rising by 1°C every 5 seconds). Oligonucleotides are listed in Table S1. Analyses were conducted in parallel using human GAPDH for normalization. A standard curve was generated for each primer set to correlate threshold (Ct) values to copy number.

Nucleosome Occupancy and Methylome-Sequencing (NOME-seq)

Nuclei were extracted as described previously (Schreiber et al., 1989). Briefly, cells were trypsinized and centrifuged for 3 mins at 500g, then washed in ice-cold PBS and resuspended in 1mL ice-cold Nuclei Buffer (10mM Tris, pH 7.4, 10mM NaCl, 3mM MgCl2, 0.1mM EDTA and 0.5% NP-40, plus protease inhibitors) per 5×106 cells and incubated on ice for 5 min. Nuclei were recovered by centrifugation at 900g for 3 min and washed in Nuclei Wash Buffer (10mM Tris, pH 7.4, 10mM NaCl, 3mM MgCl2 and 0.1mM EDTA containing protease inhibitors). Freshly prepared nuclei (2×105 cells) were resuspended in 1X M.CviPI reaction buffer (NEB), then treated with 200U of M.CviPI (NEB) in 15µL 10× reaction buffer, 45µL 1M sucrose and 0.75µL SAM in a volume of 150uL. Reactions were quenched by the addition of an equal volume of Stop Solution (20nM Tris-HCl [pH 7.9], 600mM NaCl, 1% SDS, 10mM EDTA, 400µg/ml Proteinase K) and incubated at 55°C overnight. DNA was purified by phenol/chloroform extraction and ethanol precipitation. Bisulfite conversion was performed using the Epitect Bisulfite Kit (Qiagen). Molecules were cloned using the Topo TA Kit (Invitrogen), both according to the manufacturers’ instructions.

Transfections

The Myod1-TAP and OCT4 plasmids were obtained from Addgene. Transfections were performed using Lipofectamine LTX and Plus Reagent (Invitrogen) according to the manufacturers’ instructions. Mock transfections served as negative controls. Transfections were performed in 10cm tissue culture plates using 12.5µg of Myod1-TAP or OCT4 plasmid DNA. For conditioned medium experiments, RD cells (2 × 106) were cultured for 24hr in McCoy’s 5A supplemented with 20% FBS, without Penicillin/Streptomycin. Medium was sterile filtered to remove debris and detached cells. Medium was added to LD419 cells 6hr post-transfection and replaced at 24hr. Cells were harvested as required.

Chromatin Immunoprecipitation (ChIP)

ChIP assays were performed as previously (Kelly et al., 2010). For assays using Anti-H3K4me1, Anti-TAP and Anti-RAD21 antibodies, nuclei were purified after formaldehyde crosslinking, then resuspended in SDS Lysis buffer before sonication. Antibodies (10µg) used for ChIP experiments included Histone H3 (#ab1791, Abcam), H2A.Z (#ab4174, Abcam), H3K4me1 (#39298, Active Motif), H3K4me3 (#39160, Active Motif), H3K27me3 (#07-360, Millipore), EZH2 (#39639, Active Motif), RNA Polymerase II (phosphorylated) (#ab24758, Abcam), RNA Polymerase II (#ab817, Abcam), TAP (#CAB1001, Open Biosystems), RAD21 (#ab992, Abcam) and CD8 (#sc-32812, Santa Cruz). Quantitative PCR (qPCR) for ChIP was performed as described for mRNA analyses. Oligonucleotides are listed in Table S1. For each PCR, DNA standards were included for quantitation. Samples were also immunoprecipitated with a nonspecific antibody (CD8) to control for background enrichment. Immunoprecipitated DNA was calculated as a percentage of input DNA.

Epigenome-Wide Analyses of Enhancer/Promoter Pairs

From the UCSC Genome Browser (Raney et al., 2011) we downloaded the ENCODE epigenome-wide modification profiles of H4K3me1, H4K3me3 and H3K27me3 for five cell lines: ES cells (ES), B-lymphocytes (GM12878), human mammary epithelial cells (HMEC), normal human epidermal keratinocytes (NHEK) and normal human lung fibroblasts (NHLF) (Kellis et al., 2011). For the same cell lines, we also downloaded genome-scale mapping of DNase I sensitivity data (Stamatoyannopoulos et al., 2006). We used the UCSC KnownGene database to define the transcriptional start site (TSS). For 66803 KnownGenes, we defined putative promoter regions to be ±500 bases around the TSS, and putative enhancer regions to be −(X + 10kB) or −(X + 1kB) upstream of the TSS (where X = 2KB, 3KB, …, 10KB). For each cell line, we counted the number of KnownGenes with either H4K3me3 or H3K27me3 at the promoter, and the percentage of those with H4K3me1 at the enhancer, respectively. We also determined the proportion of H3K4me1 marked enhancers that exhibit DNase hypersensitivity.

Supplementary Material

01

02

ACKNOWLEDGEMENTS

We thank Dr. Adele Murrell for helpful advice and suggestions. We thank Dr. C. Andreu-Vieyra and Fides Lay for critical review of the manuscript. Funding for this work to P.A.J. was provided by NIH R37 CA-082422 and to X.J.Z by NSF Career award 0747475.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  • Asakura A, Lyons GE, Tapscott SJ. The regulation of MyoD gene expression: conserved elements mediate expression in embryonic axial muscle. Dev Biol. 1995;171:386–398. [PubMed]
  • Atchison ML, Perry RP. Complementation between two cell lines lacking kappa enhancer activity: implications for the developmental control of immunoglobulin transcription. EMBO J. 1988;7:4213–4220. [PMC free article] [PubMed]
  • Azuara V, Perry P, Sauer S, Spivakov M, Jorgensen HF, John RM, Gouti M, Casanova M, Warnes G, Merkenschlager M, et al. Chromatin signatures of pluripotent cell lines. Nat Cell Biol. 2006;8:532–538. [PubMed]
  • Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, Wei G, Chepelev I, Zhao K. High-resolution profiling of histone methylations in the human genome. Cell. 2007;129:823–837. [PubMed]
  • Bernstein BE, Mikkelsen TS, Xie X, Kamal M, Huebert DJ, Cuff J, Fry B, Meissner A, Wernig M, Plath K, et al. A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell. 2006;125:315–326. [PubMed]
  • Boukamp P, Chen J, Gonzales F, Jones PA, Fusenig NE. Progressive stages of "transdifferentiation" from epidermal to mesenchymal phenotype induced by MyoD1 transfection, 5-aza-2'-deoxycytidine treatment, and selection for reduced cell attachment in the human keratinocyte line HaCaT. J Cell Biol. 1992;116:1257–1271. [PMC free article] [PubMed]
  • Carnac G, Albagli-Curiel O, Vandromme M, Pinset C, Montarras D, Laudet V, Bonnieu A. 3,5,3'-Triiodothyronine positively regulates both MyoD1 gene transcription and terminal differentiation in C2 myoblasts. Mol Endocrinol. 1992;6:1185–1194. [PubMed]
  • Creyghton MP, Cheng AW, Welstead GG, Kooistra T, Carey BW, Steine EJ, Hanna J, Lodato MA, Frampton GM, Sharp PA, et al. Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proc Natl Acad Sci U S A. 2010;107:21931–21936. [PMC free article] [PubMed]
  • Creyghton MP, Markoulaki S, Levine SS, Hanna J, Lodato MA, Sha K, Young RA, Jaenisch R, Boyer LA. H2AZ is enriched at polycomb complex target genes in ES cells and is necessary for lineage commitment. Cell. 2008;135:649–661. [PMC free article] [PubMed]
  • Daley G, Surani A, Tanaka E, Plath K. Reprogramming: What's Unknown? Cell. 2011;145:811–812.
  • Dekker J. The three 'C' s of chromosome conformation capture: controls, controls, controls. Nat Methods. 2006;3:17–21. [PubMed]
  • Dekker J, Rippe K, Dekker M, Kleckner N. Capturing chromosome conformation. Science. 2002;295:1306–1311. [PubMed]
  • Faerman A, Goldhamer DJ, Puzis R, Emerson CP, Jr, Shani M. The distal human myoD enhancer sequences direct unique muscle-specific patterns of lacZ expression during mouse development. Dev Biol. 1995;171:27–38. [PubMed]
  • Gal-Yam EN, Egger G, Iniguez L, Holster H, Einarsson S, Zhang X, Lin JC, Liang G, Jones PA, Tanay A. Frequent switching of Polycomb repressive marks and DNA hypermethylation in the PC3 prostate cancer cell line. Proc Natl Acad Sci U S A. 2008;105:12979–12984. [PMC free article] [PubMed]
  • Goldhamer DJ, Brunk BP, Faerman A, King A, Shani M, Emerson CP., Jr Embryonic activation of the myoD gene is regulated by a highly conserved distal control element. Development. 1995;121:637–649. [PubMed]
  • Heintzman ND, Hon GC, Hawkins RD, Kheradpour P, Stark A, Harp LF, Ye Z, Lee LK, Stuart RK, Ching CW, et al. Histone modifications at human enhancers reflect global cell-type-specific gene expression. Nature. 2009;459:108–112. [PMC free article] [PubMed]
  • Heintzman ND, Stuart RK, Hon G, Fu Y, Ching CW, Hawkins RD, Barrera LO, Van Calcar S, Qu C, Ching KA, et al. Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat Genet. 2007;39:311–318. [PubMed]
  • Hinshelwood RA, Melki JR, Huschtscha LI, Paul C, Song JZ, Stirzaker C, Reddel RR, Clark SJ. Aberrant de novo methylation of the p16INK4A CpG island is initiated post gene silencing in association with chromatin remodelling and mimics nucleosome positioning. Hum Mol Genet. 2009;18:3098–3109. [PubMed]
  • Hollenberg SM, Cheng PF, Weintraub H. Use of a conditional MyoD transcription factor in studies of MyoD trans-activation and muscle determination. Proc Natl Acad Sci U S A. 1993;90:8028–8032. [PMC free article] [PubMed]
  • Jin C, Zang C, Wei G, Cui K, Peng W, Zhao K, Felsenfeld G. H3.3/H2A.Z double variant-containing nucleosomes mark 'nucleosome-free regions' of active promoters and other regulatory regions. Nat Genet. 2009;41:941–945. [PMC free article] [PubMed]
  • Kablar B, Krastel K, Ying C, Asakura A, Tapscott SJ, Rudnicki MA. MyoD and Myf-5 differentially regulate the development of limb versus trunk skeletal muscle. Development. 1997;124:4729–4738. [PubMed]
  • Kablar B, Krastel K, Ying C, Tapscott SJ, Goldhamer DJ, Rudnicki MA. Myogenic determination occurs independently in somites and limb buds. Dev Biol. 1999;206:219–231. [PubMed]
  • Kagey MH, Newman JJ, Bilodeau S, Zhan Y, Orlando DA, van Berkum NL, Ebmeier CC, Goossens J, Rahl PB, Levine SS, et al. Mediator and cohesin connect gene expression and chromatin architecture. Nature. 2010;467:430–435. [PMC free article] [PubMed]
  • Kellis M, Ernst J, Kheradpour P, Mikkelsen TS, Shoresh N, Ward LD, Epstein CB, Zhang XL, Wang L, Issner R, et al. Mapping and analysis of chromatin state dynamics in nine human cell types. Nature. 2011;473:43–U52. [PMC free article] [PubMed]
  • Kelly TK, Miranda TB, Liang G, Berman BP, Lin JC, Tanay A, Jones PA. H2A.Z Maintenance during Mitosis Reveals Nucleosome Shifting on Mitotically Silenced Genes. Mol Cell. 2010;39:901–911. [PMC free article] [PubMed]
  • Koch CM, Andrews RM, Flicek P, Dillon SC, Karaoz U, Clelland GK, Wilcox S, Beare DM, Fowler JC, Couttet P, et al. The landscape of histone modifications across 1% of the human genome in five human cell lines. Genome Res. 2007;17:691–707. [PMC free article] [PubMed]
  • Lassar AB, Paterson BM, Weintraub H. Transfection of a DNA locus that mediates the conversion of 10T1/2 fibroblasts to myoblasts. Cell. 1986;47:649–656. [PubMed]
  • Lepikhov K, Walter J. Differential dynamics of histone H3 methylation at positions K4 and K9 in the mouse zygote. BMC Dev Biol. 2004;4:12. [PMC free article] [PubMed]
  • Lin JC, Jeong S, Liang G, Takai D, Fatemi M, Tsai YC, Egger G, Gal-Yam EN, Jones PA. Role of nucleosomal occupancy in the epigenetic silencing of the MLH1 CpG island. Cancer Cell. 2007;12:432–444. [PubMed]
  • MacIsaac KD, Lo KA, Gordon W, Motola S, Mazor T, Fraenkel E. A quantitative model of transcriptional regulation reveals the influence of binding location on expression. PLoS Comput Biol. 2010;6:e1000773. [PMC free article] [PubMed]
  • Maherali N, Sridharan R, Xie W, Utikal J, Eminli S, Arnold K, Stadtfeld M, Yachechko R, Tchieu J, Jaenisch R, et al. Directly reprogrammed fibroblasts show global epigenetic remodeling and widespread tissue contribution. Cell Stem Cell. 2007;1:55–70. [PubMed]
  • Miele A, Dekker J. Mapping cis- and trans- chromatin interaction networks using chromosome conformation capture (3C) Methods Mol Biol. 2009;464:105–121. [PMC free article] [PubMed]
  • Pinset C, Montarras D, Chenevert J, Minty A, Barton P, Laurent C, Gros F. Control of myogenesis in the mouse myogenic C2 cell line by medium composition and by insulin: characterization of permissive and inducible C2 myoblasts. Differentiation. 1988;38:28–34. [PubMed]
  • Rada-Iglesias A, Bajpai R, Swigut T, Brugmann SA, Flynn RA, Wysocka J. A unique chromatin signature uncovers early developmental enhancers in humans. Nature. 2011;470:279–283. [PubMed]
  • Raney BJ, Cline MS, Rosenbloom KR, Dreszer TR, Learned K, Barber GP, Meyer LR, Sloan CA, Malladi VS, Roskin KM, et al. ENCODE whole-genome data in the UCSC genome browser (2011 update) Nucleic Acids Res. 2011;39:D871–D875. [PMC free article] [PubMed]
  • Rideout WM, Jr, Eggan K, Jaenisch R. Nuclear cloning and epigenetic reprogramming of the genome. Science. 2001;293:1093–1098. [PubMed]
  • Schmidt D, Schwalie PC, Ross-Innes CS, Hurtado A, Brown GD, Carroll JS, Flicek P, Odom DT. A CTCF-independent role for cohesin in tissue-specific transcription. Genome Res. 2010;20:578–588. [PMC free article] [PubMed]
  • Schmitges FW, Prusty AB, Faty M, Stutzer A, Lingaraju GM, Aiwazian J, Sack R, Hess D, Li L, Zhou S, et al. Histone methylation by PRC2 is inhibited by active chromatin marks. Molecular cell. 2011;42:330–341. [PubMed]
  • Schreiber E, Matthias P, Muller MM, Schaffner W. Rapid detection of octamer binding proteins with 'mini-extracts', prepared from a small number of cells. Nucleic Acids Res. 1989;17:6419. [PMC free article] [PubMed]
  • Stamatoyannopoulos JA, Sabo PJ, Kuehn MS, Thurman R, Johnson BE, Johnson EM, Hua C, Man Y, Rosenzweig E, Goldy J, et al. Genome-scale mapping of DNase I sensitivity in vivo using tiling DNA microarrays. Nature Methods. 2006;3:511–518. [PubMed]
  • Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, Yamanaka S. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131:861–872. [PubMed]
  • Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006;126:663–676. [PubMed]
  • Tapscott SJ, Lassar AB, Weintraub H. A novel myoblast enhancer element mediates MyoD transcription. Mol Cell Biol. 1992;12:4994–5003. [PMC free article] [PubMed]
  • Visel A, Rubin EM, Pennacchio LA. Genomic views of distant-acting enhancers. Nature. 2009;461:199–205. [PMC free article] [PubMed]
  • Wang Z, Zang C, Rosenfeld JA, Schones DE, Barski A, Cuddapah S, Cui K, Roh TY, Peng W, Zhang MQ, et al. Combinatorial patterns of histone acetylations and methylations in the human genome. Nat Genet. 2008;40:897–903. [PMC free article] [PubMed]
  • Weintraub H, Tapscott SJ, Davis RL, Thayer MJ, Adam MA, Lassar AB, Miller AD. Activation of muscle-specific genes in pigment, nerve, fat, liver, and fibroblast cell lines by forced expression of MyoD. Proc Natl Acad Sci U S A. 1989;86:5434–5438. [PMC free article] [PubMed]
  • Wolff EM, Byun HM, Han HF, Sharma S, Nichols PW, Siegmund KD, Yang AS, Jones PA, Liang G. Hypomethylation of a LINE-1 promoter activates an alternate transcript of the MET oncogene in bladders with cancer. PLoS Genet. 2010;6:e1000917. [PMC free article] [PubMed]
  • Xi H, Shulha HP, Lin JM, Vales TR, Fu Y, Bodine DM, McKay RD, Chenoweth JG, Tesar PJ, Furey TS, et al. Identification and characterization of cell type-specific and ubiquitous chromatin regulatory structures in the human genome. PLoS Genet. 2007;3:e136. [PMC free article] [PubMed]
  • You JS, Kelly TK, De Carvalho DD, Taberlay PC, Liang G, Jones PA. OCT4 establishes and maintains nucleosome-depleted regions that provide additional layers of epigenetic regulation of its target genes. Proc Natl Acad Sci U S A. 2011 [PMC free article] [PubMed]
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...