Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Aug 8.
Published in final edited form as: Cell Rep. 2017 Aug 8;20(6):1307–1318. doi: 10.1016/j.celrep.2017.07.046

Dynamics of RNA polymerase II pausing and bivalent histone H3 methylation during neuronal differentiation in brain development

Jiancheng Liu 1,5, Xiwei Wu 2,5, Heying Zhang 1, Gerd P Pfeifer 3,4, Qiang Lu 1,6,*
PMCID: PMC5564459  NIHMSID: NIHMS895588  PMID: 28793256

Summary

During cellular differentiation, genes important for differentiation are expected to be silent in stem/progenitor cells yet can be readily activated. RNA polymerase II (Pol II) pausing and bivalent chromatin marks are two paradigms suited for establishing such a poised state of gene expression, however, their specific contributions in development are not well understood. Here, we characterized Pol II pausing and H3K4me3/H3K27me3 marks in neural progenitor cells (NPCs) and their daughter neurons purified from the developing mouse cortex. We show that genes paused in NPCs or neurons are characteristic of respective cellular functions important for each cell type, although pausing and pause release was not correlated with gene activation. Bivalent chromatin marks poised the marked genes in NPCs for activation in neurons. Interestingly, we observed a positive correlation between H3K27me3 and paused Pol II. This study thus reveals cell-type specific Pol II pausing and gene activation-associated bivalency during mammalian neuronal differentiation.

Keywords: Pol II pausing, pause release, H3K4me3 and H3K27me3, bivalency, neuronal differentiation, cerebral cortical development

Graphical abstract

graphic file with name nihms895588u1.jpg

Introduction

Cellular differentiation is accompanied by global changes in gene expression from patterns of maintaining stem/progenitor cells to patterns for supporting differentiation, controlled by key networks of transcription regulators and epigenetic mechanisms (Lagha et al., 2012; Lomvardas and Maniatis, 2016; Martynoga et al., 2012; Reik, 2007; Suzuki and Bird, 2008; Tee and Reinberg, 2014; Zhou et al., 2011). How such global changes of gene expression are orchestrated in development, however, is not well understood. The core enzyme in transcription, RNA polymerase II (Pol II) is at the center of multiple regulations. Recruitment of Pol II and formation of the pre-initiation complex involving gene-specific transcription factors is known to be a critical step for initiating successful transcription. In addition to this key regulation, it was recently recognized that post-recruitment regulation has a crucial impact on whether or not Pol II can achieve a productive elongation of transcription (Adelman and Lis, 2012; Gaertner and Zeitlinger, 2014; Levine, 2011; Scheidegger and Nechaev, 2016). In this mechanism, Pol II could initiate transcription of a short product but then pause at about 30–50 nucleotides downstream of the transcription start site (TSS). The balance between negative and positive regulatory factors determines whether Pol II is paused or released from pause into elongation (Zhou et al., 2012). Pausing of Pol II was originally discovered in the study of transcriptional regulation of heat shock genes in Drosophila (Gilmour and Lis, 1986; Rougvie and Lis, 1988), and was later observed in a large number of genes (Core et al., 2008; Guenther et al., 2007; Muse et al., 2007; Nechaev et al., 2010; Rahl et al., 2010; Zeitlinger et al., 2007) in many different species, suggesting potential functions of pausing in developmental programs. Available data have documented involvement of pausing of Pol II in facilitating response to changes in environmental conditions (such as the heat shock response), to coordinate synchronous expression of genes across cells of a tissue during development, and/or to maintain a basal level of gene expression in the cell (Adelman and Lis, 2012; Gaertner and Zeitlinger, 2014; Levine, 2011; Scheidegger and Nechaev, 2016). However, it remains to be further investigated to what extend pausing and pause release is involved in gene expression dynamics or contributes to specific developmental programs during cellular specifications.

Bivalent histone H3 modification is another key feature implicated in regulation of gene expression (Bernstein et al., 2006; Zhou et al., 2011). The histone H3 lysine 4 trimethylation (H3K4me3) is in general linked to genes with active expression, whereas the histone H3 lysine 27 trimethylation (H3K27me3) is associated with genes in a repressed state. Interestingly, genes containing overlapping H3K4me3 and H3K27me3 marks only express at a low level, which leads to the thought that such bivalent genes may be poised for activation or repression (or maintained in the bivalent silent state) during the progression of embryogenesis. In support of this idea, the bivalent H3K4me3 and H3K27me3 chromatin domains are found to be preferentially enriched at developmental regulatory genes in the mouse ESCs (Bernstein et al., 2006), germline cells (Sachs et al., 2013), pre-implantation embryos (Liu et al., 2016b) and several adult tissues (Cui et al., 2009; Cui et al., 2012; Hammoud et al., 2009; Kinkley et al., 2016; Mikkelsen et al., 2007). Therefore, bivalent histone H3 modification also presents a suitable regulatory mechanism for modulating gene expression levels during development. However, available data documenting the in vivo status of H3K4me3 and H3K27me3 bivalent chromatin during development is limited due to the technical difficulties associated with scarce amounts of embryonic tissues. It remains to be further examined whether the histone bivalency and a change of bivalency state is linked to major changes of gene expression during developmental progressions such as during cell fate specifications.

In the developing mouse cerebral cortex, neural progenitor cells (NPCs) first expand the progenitor cell pool by consecutive rounds of proliferation. Then NPCs will initiate a transition from proliferation to differentiation and produce descendant neurons that populate different layers of a functional cortex (Gotz and Huttner, 2005; Greig et al., 2013; Kriegstein and Alvarez-Buylla, 2009; Lodato and Arlotta, 2015; Rakic, 2009; Taverna et al., 2014). During neurogenesis, NPCs divide asymmetrically to generate daughter neurons directly or indirectly. At the same time they undergo self-renewal to keep the neural progenitor pool. Young daughter neurons migrate outwards to the superficial part of the cortex, while NPCs keep their location in the inner part of the cortex. This pattern of progression during neuronal differentiation provides an excellent system for characterizing regulation of gene expression and epigenetic modifications during developmental transitions. We previously designed a dual reporter strategy by labeling NPCs with EGFP and differentiated neurons with mRFP in the same transgenic animals to help alleviate the problem of “carryover” of reporter (EGFP in this case) from NPCs to progeny neurons, thus allowing effective simultaneous purification of NPCs and daughter neurons from the embryonic cortices (Wang et al., 2011). Using this system, we have performed comparative studies of transcriptomes (Liu et al., 2016a; Wang et al., 2011) and several key epigenetic marks (Hahn et al., 2013) between the purified NPCs and neurons. In this study, we sought to characterize global patterns of gene expression regulation during neuronal differentiation by studying genomic DNA binding sites of Pol II as well as the status of bivalent chromatin domains using chromatin immunoprecipitation and sequencing (ChIP-seq). Our data uncovered distinct functions of Pol II promoter-proximal pausing and bivalent promoters during neuronal differentiation. While Pol II pausing in NPCs and neurons displayed features associated with cell type-specific functions, the change of bivalent promoter status between NPCs and neurons correlated with neuronal gene activation.

Results

Profiles of Pol II chromatin-binding in NPCs and neurons

We accumulated purified NPCs and their daughter neurons from the embryonic day 15.5 (E15.5) mouse cortices using a previously established dual transgenic reporter strategy (Wang et al., 2011). Because the amount of purified NPCs that could be obtained from the embryonic mouse brains was limited, we adopted a method of amplification for ChIP-seq using purified cells for each experiment (see Materials and Methods). To characterize the genomic binding sites of Pol II, we used antibodies that recognize Pol II in different states: antibodies that recognize (1) the N terminus of the largest subunit of Pol II (N-20); (2) the hypophosphorylated state of the Pol II; (3) the phosphorylated C-terminal domain of Pol II at Ser5; and (4) the phosphorylated C-terminal domain of Pol II at Ser2. The first antibody could detect all forms of Pol II (Pol II-total) regardless of the phosphorylation status, whereas the second antibody could specifically bind to the hypophosphorylated form of Pol II (Pol II-nonP). The latter two antibodies could specifically recognize an early elongation form of Pol II (Pol II-Ser5P) that is often seen bound at a promoter-proximal regions (Rahl et al., 2010) and an elongating Pol II (Pol II-Ser2P), respectively. Chromatin immunoprecipitation by each of the above mentioned antibodies was done using 2.5 × 105 sorted cells (NPCs or neurons) per antibody and amplified libraries were sequenced by next generation sequencing (Supplemental Table S1). Figure 1 and Supplemental Figure S1 summarizes the heat maps and composite metagene analyses of ChIP-seq data obtained from these antibodies comparing purified NPCs and neurons. In both NPCs and neurons, Pol II-total predominantly occupies promoter regions and peaked around the transcription start site (TSS) (Figure 1A and 1B). The binding signals extend from the TSS to the transcription end site (TES), reflecting Pol II in active elongation mode (Figure 1B). Genomic binding by Pol II-Ser5P (Figure 1A and 1B), or Pol II-Ser2P and Pol II-nonP (Supplemental Figure S1) similarly showed higher signal intensity around the TSS and extended intensity along the gene body.

Figure 1. ChIP-seq of Pol II in NPCs and neurons.

Figure 1

Open in a new tab

A. Heat maps of ChIP-seq data of two different forms of Pol II in NPCs and neurons. Rows were sorted by decreasing Pol II-total occupancy in the promoter regions (−2kb to +2kb of transcription start site) in NPCs. Scaled intensities are in units of log2 fold change between antibody enriched sample and IgG control.

B. Metagene analyses of Pol II-total and Pol II-Ser5P on gene body region (upper panels) and on promoter region (lower panels) in NPCs (green lines) and neurons (red lines). TSS: transcription start site; TES: transcription end site. The y-axis represents the log2 fold change between antibody enriched sample and IgG control.

C. Metagene profiles of Pol II-total (yellow line) and Pol II-Ser5P (blue line) in NPCs. Arrows showed peak positions of profiles, indicating Pol II-Ser5P signals shifted a little more downstream of TSS comparing with Pol II-total in NPCs.

In contrast to Pol II-total, the two antibodies to phosphorylated forms of Pol II appeared to show noticeable difference in composite metagene profiles between NPCs and neurons (Figure 1 and Supplemental Figure S1). Particularly, binding by Pol II-Ser5P displayed significant difference in signal intensity between NPCs and neurons around the TSS site, suggesting more genes in NPCs having the early elongation form of Pol II accumulated to a promoter-proximal site than those in neurons. An independent ChIP-seq of purified NPCs and neurons with the Pol II-total and Pol II-Ser5P antibodies obtained similar results (Supplemental Figure S2). In addition, there was an apparent shift between the Pol II binding sites detected by the two antibodies (Figure 1C), where Pol II-Ser5P appeared to be located further downstream of Pol II-total from the TSS site. Pol II-total signal peaked around nucleotide 30–40 downstream of the TSS site, while Pol II-Ser5P signals centered around nucleotides 60–70. The separation of Pol II-total and Pol II-Ser5P binding peaks at the promoters suggested that the majority of Pol II engaged with promoters in NPCs and neurons might reflect a state of docking as also observed in other systems (Maxwell et al., 2014). The binding of Pol II-Ser5P, on the other hand, might indicate a paused Pol II after initiation of transcription as suggested by previous studies (Nechaev et al., 2010; Rahl et al., 2010).

Profiles of H3K4me3 and H3K27me3 marks in NPCs and neurons

We next analyzed the H3K4me3 and H3K27me3 marks in purified NPCs and neurons using 2.5 × 105 sorted cells per antibody. Duplicate ChIP samples yielded consistent sequencing data (Supplemental Figure S3A). The heat maps (Figure 2A) and metagene analyses (Figure 2B) showed that similar to Pol II, both chromatin marks also primarily peaked just downstream of the TSS region in NPCs and neurons. The promoter signal intensities of the two marks were markedly different between NPCs and neurons (Figure 2B), suggesting developmental changes of these marks associated with the neuronal differentiation process.

Figure 2. ChIP-seq of H3K4me3 and H3K27me3 in NPCs and neurons.

Figure 2

Open in a new tab

A. Heat maps of ChIP-seq data of H3K4me3 and H3K27me3 in NPCs and neurons. Rows were sorted independently by H3K4me3 or H3K27me3 occupancy in the promoter regions (−2kb to +2kb of TSS) in each cell type. Scaled intensities are in units of log2 fold change between antibody enriched sample and IgG control.

B. Metagene analyses of H3K4me3 and H3K27me3 on gene body region (upper panels) and on promoter region (lower panels) in NPCs (green lines) and neurons (red lines). TSS: transcription start site; TES: transcription end site. The y-axis represents the log2 fold change between antibody enriched samples and IgG control.

Pol II occupancy and chromatin marks in relation to gene expression level in NPCs and neurons

In both NPCs and neurons, distribution of genes as a function of expression levels revealed a similar pattern (Figure 3A), which consisted of a bell-shaped curve of higher expressers (active genes) and a shoulder curve of lower expressers (inactive genes). We asked whether and how gene expression levels in NPCs and neurons were influenced by promoter Pol II-binding or by the H3K4me3 and H3K27me3 marks. With respect to Pol II, our data showed that the promoter Pol II-total signal was positively correlated with gene expression levels in both NPCs and neurons (Figure 3B), whereas Pol II-Ser5P signal showed an apparent biphasic curve of correlation (Figure 3B), suggesting that promoter Pol II-total occupancy is more closely associated with the overall strength of expression of the bound genes. With respect to the two chromatin marks, the promoter H3K4me3 signal was positively associated with gene expression levels in both NPCs and neurons (Figure 3C), consistent with previous studies of ESCs and other cell types. Interestingly, the promoter H3K27me3 signal displayed a biphasic relationship with gene expression level (Figure 3D), with lower H3K27me3 levels positively correlated with gene expression but higher H3K27me3 levels inversely correlated with gene expression. Further examination of the status of H3K4me3 in genes marked by H3K27me3 revealed that the biphasic curve of H3K27me3 was mainly attributed to the presence and contribution of H3K4me3 in these genes (Supplemental Figure S3B).

Figure 3. Promoter Pol II occupancy and H3K4me3 and H3K27me3 marks in relation to gene expression level.

Figure 3

Open in a new tab

A. Global gene expression distribution patterns in NPCs (green) and neurons (red).

B. Promoter Pol II occupancy with reference to gene expression. Pol II occupancy was divided into 6 equal sized groups from low signal to high signal. The normalized gene expression data were plotted for each promoter signal group.

C. Promoter H3K4me3 and H3K27me3 with reference to gene expression. H3K4me3 or H3K27me3 ChIP-seq signals were divided into 6 equal sized groups from low signal to high signal. The normalized gene expression data were plotted for each promoter signal group.

Pol II pausing in NPCs and neurons

We next sought to characterize the group of paused genes in NPCs and neurons. For this purpose, we applied promoter occupancy by Pol II-Ser5P as an indicator of polymerase pausing and pause release (Nechaev et al., 2010; Rahl et al., 2010), using an operational definition in which a paused gene was defined by the following three features: (1) significant promoter Pol II-Ser5P signal (intensity >= 4); (2) pausing index (PI) or traveling ratio (TR) calculated by relative ratio of density of Pol II-Ser5P in the promoter-proximal region (−30 bp to +300 bp) and the gene body (defined as from +300 to the TES) >=1; and (3) transcript expression level defined by RNA-seq >=2. Using this operational definition, we identified 7401 genes in NPCs and 2212 genes in neurons showing these features of pausing. Within the groups of genes paused in NPCs and neurons, genes appeared to be expressed at a wide range of levels (Supplemental Table S2–S4), indicating that pausing did not predict a suppressed state (low expression) of gene expression. This was in agreement with previous observations that many paused genes were actively expressed in resting cells (Gilchrist et al., 2012). As 1712 paused genes were shared by both NPCs and neurons, we further divided the paused genes into the following three groups (Figure 4A): genes specifically paused in NPCs (5689 genes; Supplemental Table S2), genes specifically paused in neurons (500 genes; Supplemental Table S3), and genes paused in both NPCs and neurons (1712 genes; Supplemental Table S4). DAVID bioinformatics analyses (DAVID 6.7) showed that the top biological processes in NPC-specific paused genes highlighted different aspects of catabolic processes and cell cycle (Figure 4B and 4E), functions that appear to be required for supporting the proliferative state of NPCs. Those of neuron-specific paused genes revealed enrichment of activities involving DNA damage response and repair and cellular stress responses (Figure 4C and 4F), functions important for maintaining the integrity of post-mitotic neuronal cells (Narciso et al., 2016; Pan et al., 2014). Interestingly, with respect to DNA damage response and repair, different groups of paused genes were identified between NPCs and neurons (Supplemental Table S5). The NPC-specific paused group, but not the neuron-specific group, was highly enriched with homologous recombination genes which are particularly important for the S/G2 phases of the cell cycle. This suggested that paused genes were closely associated with cell type-specific functions. The top biological processes in paused genes shared by both NPCs and neurons included common basic cellular functions such as molecule transport, protein translation and RNA processing (Figure 4D and 4G).

Figure 4. Cell type-specific Pol II pausing in neuronal differentiation.

Figure 4

Open in a new tab

A. Paused genes identified in NPCs and neurons. Pausing index was calculated by the average promoter Pol II-Ser5P signal (read counts in −30bp to +300bp of TSS) and average gene body Pol II-Ser5P signal (read counts in +300 of TSS to TES). 5689 genes were paused in NPCs but not in neurons, 500 genes were paused in neurons but not in NPCs, while 1712 genes were paused in both NPCs and neurons.

B. Top biological processes of NPC-specific paused genes displayed features of proliferating cells, including different aspects of catabolic processes and cell cycle. The top 3000 genes ranked by pausing index were analyzed by DAVID.

C. Top biological processes of neuron-specific paused genes showed enrichment of activities important for maintaining the integrity of post-mitotic neuronal cells, involving DNA damage response and repair and cellular stress responses.

D. Top biological processes of paused genes shared by NPCs and neurons included common basic cellular functions such as molecule transport, protein translation and RNA processing.

E. Usp38 and Ppp1cb represent the group of NPC-specific paused genes. The numbers in square brackets represent track heights. Scale bars represent 5 kb for both genes.

F. Gtf2h5 and Esco2 represent the group of neuron-specific paused genes. The numbers in square brackets represent track heights. Scale bars represent 1 kb for Gtf2h5 and 2 kb for Esco2.

G. Hnrnpa2b1 and Ddx5 represent the group of paused genes shared by NPC and neuron. The numbers in square brackets represent track heights. Scale bars represent 1 kb for both genes.

H3K4me3 and H3K27me3 marks in NPCs and neurons

We next characterized the genes marked by the H3K4me3 and H3K27me3 modifications in NPCs and neurons, using the promoter signals of the two chromatin modifications for selection of marked genes. Figure 5A summarized the dynamic changes of the chromatin modification state of four subgroups of promoters from NPCs to neurons, including promoters containing bivalent marks, prominent H3K4me3 alone, prominent H3K27me3 alone or neither of the two marks. To assess the potential involvement of the H3K4me3 and H3K27me3 marks in neuronal differentiation, we looked more closely at changes of the H3K4me3 alone and bivalent groups of promoters (Figure 5B). The top biological processes of NPC-specific H3K4me3 marked genes (Figure 5C; Supplemental Table S6) highlighted DNA metabolic process and cell cycle, reflecting prominent features of proliferating cells. This group of genes contained most NPCspecific genes whose protein products are known to be involved in supporting the progenitor cell state of NPCs, for instances, the Notch family (Notch 1, 2, 3, and Hes1, 5, 6), Sonic Hedgehog family (Gli2, 3), Wnt family (Wnt7a and Tcf3, 4), Hippo pathway genes (Tead1, 2), TAM receptor tyrosine kinases (Axl and Tyro-3), Brca1, Ephrin-B1, Pax6 and Sox2 (Figure 5D; Supplemental Table S6 and Supplemental Figure S4A). On the other hand, the top biological processes of NPC-specific bivalent genes (Figure 5C; Supplemental Table S7) displayed embryonic organ development, embryonic morphogenesis and neuron differentiation, reflecting functions related to differentiation and tissue patterning. These NPC-specific bivalent genes included many genes known to be critical for proper differentiation and lineage specification, migration and/or function of cortical neurons, such as Tbr1, Fezf2, SatB2, NeuroD2, Trnp1, Crmp1, Nav2, Prdm8, Reln and Rtn1 (Figure 5E; Supplemental Table S7 and Supplemental Figure S4B). Most of these neuronal specification and function-related bivalent genes showed a transition from a state of bivalency in NPCs to a state of more prominent H3K4me3 promoter abundance in neurons (Figure 5E and Supplemental Figure S4B), which also well correlated with the up-regulation of their gene expression levels (Supplemental Table S7). Interestingly, multiple oligodendrocyte progenitor cell-specific genes, including Olig1, Olig2 and Pdgfra, contained bivalent H3K4me3 and H3K27me3 marks in NPCs (Figure 5F and Supplemental Figure S4C). These genes were switched from bivalent to more abundant H3K27me3 mark in neurons, correlating with the down-regulation of their expression.

Figure 5. Dynamics of H3K4me3 and H3K27me3 in neuronal differentiation.

Figure 5

Open in a new tab

A. Differentially marked genes in NPCs and neurons. Four different groups of promoters were categorized: Bivalent: promoters with both H3K4me3 and H3K27me3 peaks; K4me3 only: promoters with H3K4me3 but no H3K27me3 peaks; K27me3 only: promoters with H3K27me3 but no H3K4me3 peaks; None: promoters without H3K4me3 and H3K27me3 peaks.

B. H3K4me3 marked genes in NPCs and neurons (upper panel) and bivalent genes in NPCs and neurons (lower panel).

C. Top biological processes of NPC-specific H3K4me3-marked genes (left panel) and bivalent genes (right panel). The numbers in the brackets represent gene number in each process.

D. Notch2 and Sox2 represent the group of H3K4me3 marked genes in NPCs, which contains many players crucial for maintaining the progenitor cell state and are down-regulated in neurons. The numbers in square brackets represent track heights. Scale bar: 1 kb.

E. Tbr1 and Neurod2 represent the group of bivalent genes in NPCs, which includes many genes important for neuronal differentiation and lineage specification and are up-regulated in neurons. The numbers in square brackets represent track heights. Scale bar: 1 kb.

F. Several oligodendrocyte progenitor cell-specific genes display bivalent marks in NPCs but show H3K27me3 mark only in neurons. These genes are down-regulated in neurons. The numbers in square brackets represent track heights. Scale bar: 1 kb.

Dynamic changes of pausing status or chromatin marks in relation to gene activation during neuronal differentiation

The original discovery of polymerase pausing in Drosophila revealed that paused genes were likely highly inducible genes such as heat shock genes, suggesting that pausing might prepare genes for quick induction in response to environmental or developmental stimuli, such as during cell fate specification. Similarly, bivalency of chromatin marks was also thought to reflect a poised state of genes for inducible expression during developmental progressions. We therefore asked whether polymerase pausing and pause release or changes of bivalency state was linked to gene activation during differentiation of NPCs into neurons during brain development.

To assess whether pausing and pause release from NPCs to neurons might set the genes for activation during differentiation, we first looked at pause release in relation to gene expression change. Pause Release Ratio (PRR) (Chen et al., 2015), an inverse of PI or TR, was calculated by relative ratio of density of Pol II-Ser5P in the gene body and the promoter-proximal region. As shown in Figure 6A, with the increase of the PRR_Neuron/NPC ratio, log2FC became more dynamic, but did not change significantly from NPCs to neurons, suggesting that pause release is not indicative of gene activation. To further examine this, we tested if activated genes during neuronal differentiation were characterized by increased pause release ratio. We defined the group of activated genes (gene expression in NPCs or neurons >=2 and Log2FC−Neuron/NPC>=0.8) and the group of non-activated genes (gene expression in NPCs or neurons >=2 and |Log2FC−Neuron/NPC|<=0.1) (Figure 6B). Between these two gene groups, the ratio of PRR (neurons/NPCs) was lower in the activated genes than in the non-activated genes (Figure 6C), further suggesting that pause release is not inherently linked to gene activation.

Figure 6. Pausing and pause release or dynamics of bivalent marks in relation to gene activation from NPCs to neurons.

Figure 6

Open in a new tab

A. The changes of Pausing Release Ratio (PRR) of Pol II-Ser5P between NPCs and neurons (PRR Ratio_Neuron/NPC) were divided into 6 groups from low to high (x-axis). Gene expression fold changes between NPCs and neurons were calculated and compared across groups.

B. Two defined gene groups: Activated (555 genes) and Non-Activated (2845 genes).

C. PRR ratio (Neuron/NPC) changes between activated and non-activated gene groups. Pvalue: 3.1×10−5 (Welch Two Sample t-test).

D. Bivalent to H3K4me3 switch from NPCs to neurons accompanied with gene activation. P-value: 0.004 (Welch Two Sample t-test).

E. Activated gene groups enriched with bivalent to H3K4me3 switch genes during neuronal differentiation. P-value: P<2.2e-16 (Fisher’s exact test).

To assess whether changes of bivalency from NPCs to neurons might be associated with gene activation during differentiation, we analyzed the group of NPC-specific bivalent genes, particularly the subgroup of genes displaying a change from bivalency in NPCs to prominent H3K4me3 alone in neurons. This subgroup included many genes critical for proper neuronal differentiation and lineage specification (Figure 5E). Our data found that these genes showed significantly higher expression levels in neurons than in NPCs (Figure 6D), suggesting that a switch from the state of bivalency in NPCs to H3K4me3 in neurons was well correlated with gene activation during neuronal differentiation. Furthermore, we found that within the group of activated genes (Figure 6B), there were a significantly higher ratio of genes (138/555) that showed a change from bivalency in NPCs to H3K4me3 in neurons, comparing to a much lower ratio of such genes (345/10872) within the group of active genes in NPCs or neurons (gene expression level >=2 in NPCs or neurons). Together, these data indicated that the bivalency to H3K4me3 switch is positively associated with gene induction during neuronal differentiation.

Potential interplay between H3K27me3 and Pol II pausing

The predominant domains of H3K4me3 and H3K27me3 around the TSS site in both NPCs and neurons (Figure 2) raised an interesting question as to whether these histone modifications may function in connection with Pol II, which occupies a neighboring or somewhat overlapping domain (Figure 1). To address this issue, we compared promoter signals between Pol II and the chromatin marks in genes specifically enriched with H3K4me3, H3K27me3 or both marks (bivalent). Our data showed that Pol II-total is more abundant in H3K4me3 promoters in both NPCs and neurons, whereas promoter signal of H3K27me3 displayed a significantly higher correlation with that of Pol II-Ser5P (Figure 7A), suggesting that H3K27me3 was positively correlated with paused Pol II. Furthermore, when sorted based on promoter Pol II-Ser5P signal, higher levels of H3K27me3 signal were associated with higher levels of Pol II-Ser5P around the TSS region, whereas H3K4me3 did not appear to show such a correlation (Figure 7B). To address whether the observed correlation represented co-presence of H3K27me3 and Pol II-Ser5P in single cells rather than due to cellular heterogeneity, we further performed sequential ChIP of the two followed by qPCR analysis of several target genes of high correlation. Our results revealed that reciprocal sequential ChIP could detect co-enrichment of H3K27me3 and Pol II-Ser5P on these target genes (Figure 7C and Supplemental Table S8), consistent with the idea that H3K27me3 and Pol II-Ser5P co-existing in the same cells. Together, these analyses showed that paused Pol II and H3K27me3 often co-exist on the promoter proximal regions.

Figure 7. Potential interplay between H3K27me3 and Pol II pausing.

Figure 7

Open in a new tab

A. Pol II-total and Pol II-Ser5P promoter occupancy in different histone mark groups.

B. Association of H3K27me3, but not H3K4me3, with Pol II-Ser5P. Pol II-Ser5P signal at promoter region was divided into 5 groups from high to low both in NPCs and neurons (upper panels), H3K4me3 or H3K27me3 marked promoters were divided accordingly (middle and lower panels).

C. Reciprocal sequential ChIP analysis (Pol II-Ser5P -> K27me3 indicated first IP with anti-Pol II-Ser5P followed by second IP with anti-H3K27me3) showed co-presence of Pol II-Ser5P and H3K27me3 on indicated promoters, which were among the top 20% genes showing strong positive correlation between H3K27me3 and Pol II-Ser5P in purified neurons (panel B). Assays were performed on FACS purified E15.5 neurons. The sequential ChIP data is given as fold over IgG control in the second IP. Error bars represent standard deviations.

Discussion

Developmental progression of stem/progenitor cell differentiation is accompanied by global changes of gene expression patterns in a precise temporal and/or spatial order, in which genes required for maintaining stem/progenitor cell state are tuned down while genes crucial for supporting the specification and function of differentiated cells are tuned up. Such an exquisite transition of global gene expression patterns requires a concerted action of various regulatory mechanisms working at different levels including DNA methylation, chromatin modification, specific DNA-binding factors, etc. To enable this transition to proceed smoothly, it is conceivable that the cell type-specific genes might be somehow placed in a standby state in stem/progenitor cells and can readily change transcriptional output upon being instructed to differentiate. Pol II polymerase pausing and bivalent chromatin modification are two mechanisms that are particularly suitable for such a poised state for gene induction, however, whether and how they are involved in controlling gene expression dynamics during cell fate specification are not clear. Combining purified cortical neural progenitor cells and their daughter neurons with ChIP-seq for limited number of cells, we determined the global profile of genomic binding by Pol II and the H3K4me3 and H3K27me3 marks. Our data revealed some unique features of Pol II pausing and bivalent chromatin marks in the regulation of neural progenitor cell differentiation during brain development.

Paused genes have been characterized by both the global run-on sequencing (GRO-seq) method, which detects the functional Pol II engaged in RNA synthesis and the ChIP-seq method, which detects promoter occupancy by Pol II. While GRO-seq would more accurately reveal the pausing status of Pol II, previous studies also indicated that the magnitude of GRO-seq signal correlated well with the density of Pol II at the promoter proximal region based on ChIP-seq data (Core et al., 2012), suggesting that the abundance of Pol II, particularly the form of Pol II-Ser5P (Rahl et al., 2010), at the promoters could be one index to reflect pausing. Due to the limited number of cortical NPCs derived from embryonic mouse brains (Wang et al., 2011), we chose to use ChIP-seq of Pol II to examine paused gene in this study, as GRO-seq would need to use significantly more (>40 fold more) purified cortical NPCs. Our ChIP-seq data revealed two interesting features: First, a global change of paused genes accompanied neuronal differentiation. Second, the change was highly correlated with cell type-specific functional requirements for NPCs or neurons. Thus pausing appeared to be a mechanism to globally orchestrate expression of thousands of genes in order for the cell to function properly for its cell state. The paused genes showed a wide range of expression levels. This perhaps reflected distinct requirement of respective functions of individual genes at the given moment of the cell state. For instance, we would envision that NPCs in a faster or slower growing state may impose higher or lower expression levels of genes of nucleotide metabolic pathways, respectively. In this context, although pausing is not required for rapid gene induction as originally thought, it seems to be a crucial mechanism for maintaining or sustaining concerted expression of genes essential for the characteristics of a cell type and allowing fine tuning of gene levels to suit the need of the state of the cell.

Our ChIP-seq data on H3K4me3 and H3K27me3 revealed that dynamic changes of these two chromatin marks correlated well with cell fate determination during development. In particular, the H3K4me3 mark alone and the bivalent H3K4me3/H3K27me3 marks showed close association with distinct molecular programs of cell fate specification in NPCs. Most genes known to be crucial for the maintenance of NPCs displayed abundant H3K4me3 promoter signals with little H3K27me3, suggesting a crucial function of H3K4me3 in promoting active expression of these genes in NPCs. After neuronal differentiation, these genes mostly lost the promoter H3K4me3 mark, which correlated with their silencing in neurons. In contrast, many neuronal lineage specification and neuron function-related genes were characterized by co-existence of the promoter H3K4me3 and H3K27me3 signals in NPCs. Most of these genes showed a change of chromatin marks from the bivalency in NPCs to a more prominent H3K4me3 mark status in neurons during neurogenesis. This also correlated with a lower expression level of many of these genes in NPCs but an up-regulated level in neurons. Thus, the H3K4me3 and H3K27me3 bivalent marks appeared to function to prepare the marked genes in NPCs to be induced for an active expression in neurons. Upon differentiation, the switch of bivalency to predominant H3K4me3 mark would maintain these genes in an active state to support the progression of neuronal differentiation.

Whether Pol II pausing and the bivalent chromatin marks work independently or they may coordinate with each other in transcriptional regulation is an intriguing question, giving the observations that the paused Pol II and chromatin marks occupy neighboring or overlapping domains in the promoter proximal region. Interestingly, our data revealed that H3K27me3 was positively correlated with Pol II-Ser5P in both NPCs and neurons. In this regard, it is intriguing to note that Drosophila PRC (Polycomb-repressive complex) could physically interact with paused Pol II (Tie et al., 2016) and was preferentially localized to paused promoters (Enderle et al., 2011). In addition, between the two classes of bivalent domains in embryonic stem cells (Ku et al., 2008), it was found that promoters bound by PRC2 alone could allow Pol II pausing (Min et al., 2011), although PRC1 and PRC2 co-occupied promoters were devoid of paused Pol II. Together, these observations suggest that H3K27me3 may not simply be a permissive mark but likely to either play an active role in Pol II pausing or coordinate with paused Pol II in controlling the transcriptional output during development. Future studies are required to further investigate this potential interplay between paused Pol II and the H3K27me3 mark.

In summary, our data indicate that Pol II promoter pausing and the bivalent H3K4me3 and H3K27me3 marks are two crucial but distinct regulatory mechanisms of transcriptional control for proper progression of neuronal differentiation during brain development. The bivalent marks are important for cell fate specification transitioning from NPC to neuron, while Pol II pausing is important for supporting the established state of the two cell types.

Experimental Procedures

Animals

Nestin-EGFP and Dcx-mRFP reporter mice were previously described (Hahn et al., 2013; Wang et al., 2011). Animal procedures were approved by the Institutional Animal Care and Use Committee (IACUC) and were conducted in accordance with NIH guideline and the Guide for the Care and Use of Laboratory Animals. All experiments were conducted with stage matched embryos (E15.5), which contained both males and females.

Purification of NPCs and neurons

Purification of E15.5 cortical cells using a double reporter strategy was done as previously described (Hahn et al., 2013; Wang et al., 2011).

RNA-seq

Purified E15.5 NPCs and neurons were accumulated to 0.5 – 1 million, and total RNAs were isolated using Trizol reagent (Invitrogen). Single End libraries were prepared, size selected, gel purified and sequenced using Illumina HiSeq2000 system.

ChIP-seq

ChIP experiments were modified from previous protocols (Hahn et al., 2013) using antibodies: Pol II total (sc-899, Santa Cruz); Pol II nonP (ab817, Abcam); Pol II Ser2P (04-1571, EMD Millipore); Pol II Ser5P (ab5131, Abcam); H3K4me3 (39159, Active Motif); H3K27me3 (07-449, EMD Millipore). Specifically, cells collected from FACS were cross-linked by adding 37% formaldehyde to a final concentration of 1%. Cells were incubated on a rotator for 10 minutes at room temperature (RT), and formaldehyde was quenched by adding 1.25 M glycine to a final concentration of 125 mM. Cells were rocked for 5 minutes at RT, washed with cold PBS and flash-frozen in liquid nitrogen. Cells were stored at −80 °C and accumulated to a desired number. Cells were lysed in buffer containing 1% SDS, 5 mM EDTA and 50 mM Tris-HCl (pH8.0) with freshly added protease inhibitor (04693116001, Roche). The chromatin was fragmented to 200–500 bp with a Misonix S3000 Sonicator at 4 °C. Afte r centrifugation, around 100 µl supernatant from 2.5 × 105 cells lysate was diluted to 1 ml with a dilution buffer containing 1% Triton X-100, 2mM EDTA, 150mM NaCl, and 20mM Tris-HCl (pH8.0) with freshly added protease inhibitor. Antibody (1 µg) was added to chromatin and the mixture was incubated at 4 °C overnight. 10 µl Dynabeads Protein A (10001D, Invitrogen) was added to the chromatin/antibody mixture and incubated for additional 2 hours at 4 °C. Beads wer e washed sequentially with 700 µl each of washing buffer I (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 150 mM NaCl, 20 mM Tris-HCl, pH8.0), washing buffer II (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 500 mM NaCl, 20 mM Tris-HCl, pH8.0), washing buffer III (0.25 M LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl, pH8.0) and TE buffer (10mM Tris-HCl, pH8.0, 1 mM EDTA). The sample was eluted with 250 µl elution buffer containing 0.5% SDS, 25 mM Tris-HCl, pH8.0 and 10 mM EDTA. The eluted sample was treated with 0.2 M NaCl and 1 mg/ml Protease K (P-2308, Sigma) at 65 °C overnight. DNA was purified with phenol/chloroform and was precipitated with cold ethanol and glycogen. The dissolved DNA samples after ChIP were subjected to amplification with MicroPlex Library Preparation kit (AB-004-0012, Diagenode). Amplification was performed with a StepOnePlus Real-Time PCR System (Applied Biosystem) and was stopped manually at the exponential phase. Library after amplification was purified twice with Agencourt AMPure XP beads (A63880, Beckman) and sequenced on a Hiseq 2500 sequencer.

RNA-seq Data Analysis

Reads were aligned to the mouse reference genome mm9 using TopHat v2.0 with default settings. The expression level of RefSeq genes were counted by matching the aligned reads to the coordinates of the RefSeq exons. Reads falling into exons that belong to multiple transcripts of the same genes were counted one time, and the reads from all exons of the same gene were combined to represent the gene level expression. Raw counts were then normalized by trimmed mean of M value (TMM) method implemented in the Bioconductor package “edgeR”. The normalized counts were then scaled by the gene length and log2 transformed to represent the expression value of each RefSeq gene model.

ChIP-seq Data Analysis

Reads from ChIP-seq experiments were aligned to mouse genome assembly mm9 using Novoalign v3.02.07. Only reads aligned to a unique genome locus were reported. Peak calling was done by MACS v2.0 using the corresponding IgG control with the option of p = 0.001 and d = 200. All subsequent analysis was done using customized R scripts. Gene annotations of mm9 genome were downloaded from RefSeq database. For metagene profile and heatmap at gene body or promoter regions, these regions were divided into equal number of bins for each gene, and fold enrichment of read counts within each bin between ChIP sample and IgG control sample was calculated. These data matrices were used to generate metagene profiles by plotting the average signal in each bin for all the genes in NPC and neuron. They were also used to generate heat maps using Java TreeView. Promoters were defined as TSS+/−500bp, and proximal promoters were defined as 30bp upstream and 300bp downstream of TSS. The H3K4me3 and H3K27me3 promoter signal was calculated by the fold enrichment of read counts at the promoter region between the ChIP sample vs. the IgG control sample. Pausing index was calculated by the fold difference of read counts within the proximal promoter and gene body (TSS+300bp to transcription end).

Sequential ChIP real-time PCR

Around 16 million of FACS purified neurons at E15.5 were collected, fixed and fragmented as described in ChIP-seq. The sequential ChIP experiment was performed using a Re-ChIP-IT kit (Active Motif). Anti-Pol II-Ser5P (ab5131, Abcam) or anti-H3K27me3 (07-449, EMD Millipore) was used in the first immunoprecipitation, respectively. Then anti-H3K27me3 or anti-Pol II-Ser5P was added to precipitated chromatin in the second reaction. For control, a ChIP grade IgG (Abcam) was used replacing anti-Pol II-Ser5P or anti-H3K27me3 in the second reactions. DNAs eluted from the second immunoprecipitations were applied to real-time quantitative PCR analysis using a DyNAmo flash SYBR green qPCR kit (Thermo Fisher) on a StepOnePlus real-time PCR system (Applied Biosystems).

Statistical Analysis

Statistical analyses were performed with R v3.0.2. When two groups were compared, we first performed a normality test. Parametric data were compared using a Welch t test. Enrichment analysis of gene lists was performed using Fisher’s exact test. Statistical significance was set at p < 0.05. The data are presented as mean ± SEM.

Supplementary Material

1
2

Table S5 (Related to Figure 4): Paused genes in NPCs and neurons important for DNA damage response and repair

3

Table S2 (Related to Figure 4): NPC-specific paused genes

4

Table S3 (Related to Figure 4): Neuron-specific paused genes

5

Table S4 (Related to Figure 4): Paused genes shared by NPCs and neurons

6

Table S6 (Related to Figure 5): NPC-specific H3K4me3-marked genes

7

Table S7 (Related to Figure 5): NPC-specific bivalent genes

Highlights.

  • Global patterns of Pol II and bivalent marks were characterized in NPCs and neurons;

  • Pol II pausing is associated with cell type-specific functions in NPCs and neurons;

  • Bivalent marks prime neuronal specification genes for activation during differentiation;

  • Paused Pol II and H3K27me3 often co-exist on the promoter proximal regions.

In Brief.

Pol II pausing and chromatin bivalency are two mechanisms controlling gene transcription. Liu et al. reveal that Pol II pausing is associated with cell type-specific functions in mouse neural progenitor cells and their daughter neurons, while change in H3K4me3/H3K27me3 bivalency is associated with gene activation during mammalian neuronal differentiation.

Acknowledgments

We thank Donna Isbell and Cirila Arteaga for assistance with animal breeding and care; Lucy Brown and Jeremy Stark and their staff for helping with cell sorting; Jinhui Wang for performing next-generation sequencing; Jeremy Stark for helpful discussion of DNA damage response and repair-related genes. This work was supported by NIH grants NS075393 from NINDS to Q.L. and MH094599 from NIMH to Q.L. and G.P.P. In addition, research reported in this study included work performed in the Analytical Cytometry Core and Integrated Genomics Core supported by the National Cancer Institute under award number P30CA033572.

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.

Authors’ contributions

JCL and QL conceived of the study. JCL and HYZ conducted the experiments. XWW conducted bioinformatics analyses. JCL, XWW, GPP and QL analyzed the data. JCL, XWW, GPP and QL wrote the paper. All authors approved the final manuscript.

Accession Number

ChIP-seq data of Pol II, H3K4me3 and H3K27me3 as well as RNA-seq data comparing NPCs versus neurons can be accessed at Gene Expression Omnibus (GEO) with accession number GSE93011.

References

  1. Adelman K, Lis JT. Promoter-proximal pausing of RNA polymerase II: emerging roles in metazoans. Nature reviews Genetics. 2012;13:720–731. doi: 10.1038/nrg3293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. 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. doi: 10.1016/j.cell.2006.02.041. [DOI] [PubMed] [Google Scholar]
  3. Chen FX, Woodfin AR, Gardini A, Rickels RA, Marshall SA, Smith ER, Shiekhattar R, Shilatifard A. PAF1, a Molecular Regulator of Promoter-Proximal Pausing by RNA Polymerase II. Cell. 2015;162:1003–1015. doi: 10.1016/j.cell.2015.07.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Core LJ, Waterfall JJ, Gilchrist DA, Fargo DC, Kwak H, Adelman K, Lis JT. Defining the status of RNA polymerase at promoters. Cell reports. 2012;2:1025–1035. doi: 10.1016/j.celrep.2012.08.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Core LJ, Waterfall JJ, Lis JT. Nascent RNA sequencing reveals widespread pausing and divergent initiation at human promoters. Science (New York, NY. 2008;322:1845–1848. doi: 10.1126/science.1162228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cui K, Zang C, Roh TY, Schones DE, Childs RW, Peng W, Zhao K. Chromatin signatures in multipotent human hematopoietic stem cells indicate the fate of bivalent genes during differentiation. Cell stem cell. 2009;4:80–93. doi: 10.1016/j.stem.2008.11.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cui P, Liu W, Zhao Y, Lin Q, Zhang D, Ding F, Xin C, Zhang Z, Song S, Sun F, et al. Comparative analyses of H3K4 and H3K27 trimethylations between the mouse cerebrum and testis. Genomics, proteomics & bioinformatics. 2012;10:82–93. doi: 10.1016/j.gpb.2012.05.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Enderle D, Beisel C, Stadler MB, Gerstung M, Athri P, Paro R. Polycomb preferentially targets stalled promoters of coding and noncoding transcripts. Genome research. 2011;21:216–226. doi: 10.1101/gr.114348.110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Gaertner B, Zeitlinger J. RNA polymerase II pausing during development. Development (Cambridge, England) 2014;141:1179–1183. doi: 10.1242/dev.088492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Gilchrist DA, Fromm G, dos Santos G, Pham LN, McDaniel IE, Burkholder A, Fargo DC, Adelman K. Regulating the regulators: the pervasive effects of Pol II pausing on stimulus-responsive gene networks. Genes & development. 2012;26:933–944. doi: 10.1101/gad.187781.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gilmour DS, Lis JT. RNA polymerase II interacts with the promoter region of the noninduced hsp70 gene in Drosophila melanogaster cells. Molecular and cellular biology. 1986;6:3984–3989. doi: 10.1128/mcb.6.11.3984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gotz M, Huttner WB. The cell biology of neurogenesis. Nature reviews Molecular cell biology. 2005;6:777–788. doi: 10.1038/nrm1739. [DOI] [PubMed] [Google Scholar]
  13. Greig LC, Woodworth MB, Galazo MJ, Padmanabhan H, Macklis JD. Molecular logic of neocortical projection neuron specification, development and diversity. Nature reviews. 2013;14:755–769. doi: 10.1038/nrn3586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Guenther MG, Levine SS, Boyer LA, Jaenisch R, Young RA. A chromatin landmark and transcription initiation at most promoters in human cells. Cell. 2007;130:77–88. doi: 10.1016/j.cell.2007.05.042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hahn MA, Qiu R, Wu X, Li AX, Zhang H, Wang J, Jui J, Jin SG, Jiang Y, Pfeifer GP, et al. Dynamics of 5-hydroxymethylcytosine and chromatin marks in Mammalian neurogenesis. Cell reports. 2013;3:291–300. doi: 10.1016/j.celrep.2013.01.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hammoud SS, Nix DA, Zhang H, Purwar J, Carrell DT, Cairns BR. Distinctive chromatin in human sperm packages genes for embryo development. Nature. 2009;460:473–478. doi: 10.1038/nature08162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kinkley S, Helmuth J, Polansky JK, Dunkel I, Gasparoni G, Frohler S, Chen W, Walter J, Hamann A, Chung HR. reChIP-seq reveals widespread bivalency of H3K4me3 and H3K27me3 in CD4(+) memory T cells. Nature communications. 2016;7:12514. doi: 10.1038/ncomms12514. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kriegstein A, Alvarez-Buylla A. The glial nature of embryonic and adult neural stem cells. Annual review of neuroscience. 2009;32:149–184. doi: 10.1146/annurev.neuro.051508.135600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ku M, Koche RP, Rheinbay E, Mendenhall EM, Endoh M, Mikkelsen TS, Presser A, Nusbaum C, Xie X, Chi AS, et al. Genomewide analysis of PRC1 and PRC2 occupancy identifies two classes of bivalent domains. PLoS Genet. 2008;4:e1000242. doi: 10.1371/journal.pgen.1000242. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lagha M, Bothma JP, Levine M. Mechanisms of transcriptional precision in animal development. Trends in genetics : TIG. 2012;28:409–416. doi: 10.1016/j.tig.2012.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Levine M. Paused RNA polymerase II as a developmental checkpoint. Cell. 2011;145:502–511. doi: 10.1016/j.cell.2011.04.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Liu J, Wu X, Zhang H, Qiu R, Yoshikawa K, Lu Q. Prospective separation and transcriptome analyses of cortical projection neurons and interneurons based on lineage tracing by Tbr2 (Eomes)-GFP/Dcx-mRFP reporters. Dev Neurobiol. 2016a;76:587–599. doi: 10.1002/dneu.22332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Liu X, Wang C, Liu W, Li J, Li C, Kou X, Chen J, Zhao Y, Gao H, Wang H, et al. Distinct features of H3K4me3 and H3K27me3 chromatin domains in pre-implantation embryos. Nature. 2016b;537:558–562. doi: 10.1038/nature19362. [DOI] [PubMed] [Google Scholar]
  24. Lodato S, Arlotta P. Generating Neuronal Diversity in the Mammalian Cerebral Cortex. Annu Rev Cell Dev Biol. 2015;31:699–720. doi: 10.1146/annurev-cellbio-100814-125353. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lomvardas S, Maniatis T. Histone and DNA Modifications as Regulators of Neuronal Development and Function. Cold Spring Harb Perspect Biol. 2016;8:a024208. doi: 10.1101/cshperspect.a024208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Martynoga B, Drechsel D, Guillemot F. Molecular control of neurogenesis: a view from the mammalian cerebral cortex. Cold Spring Harb Perspect Biol. 2012;4:a008359. doi: 10.1101/cshperspect.a008359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Maxwell CS, Kruesi WS, Core LJ, Kurhanewicz N, Waters CT, Lewarch CL, Antoshechkin I, Lis JT, Meyer BJ, Baugh LR. Pol II docking and pausing at growth and stress genes in C. elegans. Cell reports. 2014;6:455–466. doi: 10.1016/j.celrep.2014.01.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Mikkelsen TS, Ku M, Jaffe DB, Issac B, Lieberman E, Giannoukos G, Alvarez P, Brockman W, Kim TK, Koche RP, et al. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature. 2007;448:553–560. doi: 10.1038/nature06008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Min IM, Waterfall JJ, Core LJ, Munroe RJ, Schimenti J, Lis JT. Regulating RNA polymerase pausing and transcription elongation in embryonic stem cells. Genes & development. 2011;25:742–754. doi: 10.1101/gad.2005511. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Muse GW, Gilchrist DA, Nechaev S, Shah R, Parker JS, Grissom SF, Zeitlinger J, Adelman K. RNA polymerase is poised for activation across the genome. Nature genetics. 2007;39:1507–1511. doi: 10.1038/ng.2007.21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Narciso L, Parlanti E, Racaniello M, Simonelli V, Cardinale A, Merlo D, Dogliotti E. The Response to Oxidative DNA Damage in Neurons: Mechanisms and Disease. Neural plasticity. 2016;2016:3619274. doi: 10.1155/2016/3619274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Nechaev S, Fargo DC, dos Santos G, Liu L, Gao Y, Adelman K. Global analysis of short RNAs reveals widespread promoter-proximal stalling and arrest of Pol II in Drosophila. Science (New York, NY. 2010;327:335–338. doi: 10.1126/science.1181421. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Pan L, Penney J, Tsai LH. Chromatin regulation of DNA damage repair and genome integrity in the central nervous system. J Mol Biol. 2014;426:3376–3388. doi: 10.1016/j.jmb.2014.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Rahl PB, Lin CY, Seila AC, Flynn RA, McCuine S, Burge CB, Sharp PA, Young RA. c-Myc regulates transcriptional pause release. Cell. 2010;141:432–445. doi: 10.1016/j.cell.2010.03.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rakic P. Evolution of the neocortex: a perspective from developmental biology. Nature reviews. 2009;10:724–735. doi: 10.1038/nrn2719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Reik W. Stability and flexibility of epigenetic gene regulation in mammalian development. Nature. 2007;447:425–432. doi: 10.1038/nature05918. [DOI] [PubMed] [Google Scholar]
  37. Rougvie AE, Lis JT. The RNA polymerase II molecule at the 5' end of the uninduced hsp70 gene of D. melanogaster is transcriptionally engaged. Cell. 1988;54:795–804. doi: 10.1016/s0092-8674(88)91087-2. [DOI] [PubMed] [Google Scholar]
  38. Sachs M, Onodera C, Blaschke K, Ebata KT, Song JS, Ramalho-Santos M. Bivalent chromatin marks developmental regulatory genes in the mouse embryonic germline in vivo. Cell reports. 2013;3:1777–1784. doi: 10.1016/j.celrep.2013.04.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Scheidegger A, Nechaev S. RNA polymerase II pausing as a context-dependent reader of the genome. Biochem Cell Biol. 2016;94:82–92. doi: 10.1139/bcb-2015-0045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Suzuki MM, Bird A. DNA methylation landscapes: provocative insights from epigenomics. Nature reviews Genetics. 2008;9:465–476. doi: 10.1038/nrg2341. [DOI] [PubMed] [Google Scholar]
  41. Taverna E, Gotz M, Huttner WB. The Cell Biology of Neurogenesis: Toward an Understanding of the Development and Evolution of the Neocortex. Annu Rev Cell Dev Biol. 2014;30:465–502. doi: 10.1146/annurev-cellbio-101011-155801. [DOI] [PubMed] [Google Scholar]
  42. Tee WW, Reinberg D. Chromatin features and the epigenetic regulation of pluripotency states in ESCs. Development (Cambridge, England) 2014;141:2376–2390. doi: 10.1242/dev.096982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Tie F, Banerjee R, Fu C, Stratton CA, Fang M, Harte PJ. Polycomb inhibits histone acetylation by CBP by binding directly to its catalytic domain. Proceedings of the National Academy of Sciences of the United States of America. 2016;113:E744–753. doi: 10.1073/pnas.1515465113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wang J, Zhang H, Young AG, Qiu R, Argalian S, Li X, Wu X, Lemke G, Lu Q. Transcriptome analysis of neural progenitor cells by a genetic dual reporter strategy. Stem cells (Dayton, Ohio) 2011;29:1589–1600. doi: 10.1002/stem.699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Zeitlinger J, Stark A, Kellis M, Hong JW, Nechaev S, Adelman K, Levine M, Young RA. RNA polymerase stalling at developmental control genes in the Drosophila melanogaster embryo. Nature genetics. 2007;39:1512–1516. doi: 10.1038/ng.2007.26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Zhou Q, Li T, Price DH. RNA polymerase II elongation control. Annual review of biochemistry. 2012;81:119–143. doi: 10.1146/annurev-biochem-052610-095910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Zhou VW, Goren A, Bernstein BE. Charting histone modifications and the functional organization of mammalian genomes. Nature reviews Genetics. 2011;12:7–18. doi: 10.1038/nrg2905. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS895588-supplement-1.doc (758KB, doc)
2

Table S5 (Related to Figure 4): Paused genes in NPCs and neurons important for DNA damage response and repair

NIHMS895588-supplement-2.xls (20.5KB, xls)
3

Table S2 (Related to Figure 4): NPC-specific paused genes

NIHMS895588-supplement-3.xls (2MB, xls)
4

Table S3 (Related to Figure 4): Neuron-specific paused genes

NIHMS895588-supplement-4.xls (469KB, xls)
5

Table S4 (Related to Figure 4): Paused genes shared by NPCs and neurons

NIHMS895588-supplement-5.xls (94KB, xls)
6

Table S6 (Related to Figure 5): NPC-specific H3K4me3-marked genes

NIHMS895588-supplement-6.xls (361KB, xls)
7

Table S7 (Related to Figure 5): NPC-specific bivalent genes

NIHMS895588-supplement-7.xls (350KB, xls)

RESOURCES