Controlling Pol II Pausing
Published: Last Edited:
Disclaimer: This essay has been submitted by a student. This is not an example of the work written by our professional essay writers. You can view samples of our professional work here.
Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of UK Essays.
Traditional studies on eukaryotic gene regulation were mostly based on yeast, which emphasized Pol II recruitment as the main regulatory step within the transcription cycle, with little regulation following PIC formation. However, during the past three decades, more and more evidence in metazoan systems revealed that much of the regulation occurs during early elongation, through controlling Pol II pausing and releasing paused Pol II into productive elongation (reviewed in Adelman and Lis, 2012; Gaertner and Zeitlinger, 2014).
Pol II pausing is widespread in metazoans
Pol II pausing was first discovered on Drosophila heat shock genes (Hsp genes) by the Lis lab (Giardina et al., 1992; Gilmour and Lis, 1986; Rasmussen and Lis, 1993; Rougvie and Lis, 1988), that transcriptionally engaged Pol II accumulates downstream of Hsp promoters, and associates with 20-60 nt of nascent transcript. They termed promoter-proximal Pol II as “paused” (Rougvie and Lis, 1990). Later Pol II pausing was found at a few other Drosophila genes, human genes MYC and FOS, and HIV long terminal repeat (reviewed in Adelman and Lis, 2012). More recently, genome-wide studies using ChIP-chip on Drosophila S2 cells and early embryos (Lee et al., 2008; Muse et al., 2007; Zeitlinger et al., 2007) and human primary lung fibroblasts and ESCs (Guenther et al., 2007; Kim et al., 2005) revealed that Pol II pausing is widespread, affecting 10-40% of all genes. Moreover, Pol II pausing is not only restricted to genes transcribed at low levels and poised for later activation, it also occurs at highly active genes (reviewed in Gaertner and Zeitlinger, 2014).
People have used different techniques to detect paused Pol II. First, Pol II ChIP-seq and ChIP-chip assays reveal high levels of Pol II 20-60 nt downstream of the TSS of many genes, as mentioned above. Second, permanganate footprinting experiments detect the open transcription bubble of transcriptionally engaged Pol II at pause sites (Gilmour and Fan, 2009). Third, short, capped RNA sequencing (scRNA-seq) detected 25-60 nt transcripts generated from paused Pol II in Drosophila (Nechaev et al., 2010). Last, global or precision nuclear run-on sequencing (GRO-seq, PRO-seq) maps transcriptionally engaged Pol II that can resume RNA synthesis (Core et al., 2012; Core et al., 2008; Kwak et al., 2013). Combining different assays, which all point to a similar conclusion, these experiments suggest that poised Pol II near the TSS primarily exists in paused form, with very little in the PIC (not imitating RNA synthesis), arrested (stably engaged in elongation but backtracked), or terminating (unstably elongating and in the process of dissociating from DNA) states (reviewed in Adelman and Lis, 2012; Gaertner and Zeitlinger, 2014).
Mechanisms of Pol II pausing
After Pol II initiates transcription, it escapes the promoter and is phosphorylated by the GTF TFIIH on the serine 5 residue of the largest subunit CTD (Ser5P), forming an early elongation complex. Two key pause-inducing complexes, DRB-sensitivity-inducing factor (DSIF) and negative elongation factor (NELF), bind Pol II following promoter escape and establish the pausing. The release of Pol II into productive elongation depends on positive transcription elongation factor b (P-TEFb), which phosphorylates NELF, DSIF, and serine 2 residue on the Pol II CTD (Ser2P) through its cyclin-dependent kinase Cdk9 subunit. This leads to dissociation of NELF from paused Pol II, DSIF conversion from a negative elongation factor to a positive one, and maturation of paused Pol II into elongation. Pol II carrying Ser2 can serve as a platform for the binding of RNA-processing factors, chromatin modifiers and remodelers, which all contribute to productive elongation. P-TEFb can be recruited to promoters by bromodomain-containing 4 (Brd4) which can bind acetylated histone tails, DNA-binding activators such as NF-κB, c-Myc and VP16, co-activators such as Mediator, etc. Due to its crucial role in pause release, targeting P-TEFb to paused promoters can be an important step subject to regulation, and the targeting rate may determine the transcription levels as well as Pol II pausing levels (reviewed in Adelman and Lis, 2012; Gaertner and Zeitlinger, 2014; Peterlin and Price, 2006).
The intrinsic properties of the core promoter play an important role in determining Pol II pausing levels. When placed in front of a reporter gene, the 200 bp promoter region of the Drosophila tup gene was found sufficient to establish paused Pol II, reminiscent of the highly paused Pol II profile at the endogenous tup locus (Lagha et al., 2013). Moreover, during Drosophila early embryogenesis and muscle development, genes with paused Pol II were found to possess a special promoter sequence component that is different from that of non-paused or maternal genes: they are strongly enriched for initiator (Inr), downstream promoter element (DPE), motif ten element (MTE), pause button (PB), and GAGA motifs (Chen et al., 2013; Gaertner et al., 2012; Hendrix et al., 2008), the previous three are characteristic for promoters with a focused initiation pattern within a very narrow window (Rach et al., 2009). The high G-C content of these promoters might contribute to the regulation of Pol II pausing (Hendrix et al., 2008; Nechaev et al., 2010), possibly by conferring a high energy barrier for the elongating Pol II to melt the DNA double helix (reviewed in Levine, 2011). The intrinsic affinity for nucleosomes of these paused promoters may also contribute to Pol II pausing, as they highly favor the formation of a promoter nucleosome in the absence of Pol II (Gaertner et al., 2012; Gilchrist et al., 2010). GAF was found to bind many paused promoters (Lee et al., 2008), and mutation of GAGA motifs led to reduction of paused Pol II at the hsp70 promoter (Lee et al., 1992). GAF may help establish paused Pol II by antagonizing the promoter nucleosome (reviewed in Gaertner and Zeitlinger, 2014; Levine, 2011), and recruiting NELF (Li et al., 2013).
Functions of Pol II pausing
GRO-seq performed on human cells revealed that ~30% of human genes are paused, and few of them are transcriptionally inactive (<1%; Core et al., 2008). Moreover, even during activation, Drosophila Hsp genes continue to exhibit pausing (reviewed in Adelman and Lis, 2012). In Drosophila, paused Pol II was found enriched near many developmental control genes, suggesting an important role of Pol II pausing in transcription regulation during development (Muse et al., 2007; Zeitlinger et al., 2007). This evidence suggests that Pol II pausing is not simply inactivating genes, but should be considered a way of fine-tuning the expression of active genes, or poising genes for future activation (reviewed in Adelman and Lis, 2012).
Rapid induction: Traditional studies on paused Drosophila Hsp genes revealed that they can be induced very rapidly, leading to the assumption that Pol II pausing can prime genes for rapid activation upon induction. However, not all rapidly inducible genes have paused Pol II, nor are all paused Pol II genes highly inducible (reviewed in Adelman and Lis, 2012). A recent study on Drosophila S2 cells and murine ESCs showed that for the stimulus-responsive gene networks, paused Pol II is more enriched at genes encoding the pathway components and regulators, rather than the downstream inducible targets, and that pausing largely functions in maintaining the basal transcription level of pathway key factors (Gilchrist et al., 2012). Furthermore, the vast majority of Drosophila pre-CB genes are non-paused (Chen et al., 2013), but they are induced as rapidly as, or perhaps even faster than paused genes. A recent GRO-seq study on early Drosophila embryos also indicated that paused genes during 2-2.5 hr have higher transcription levels during later embryogenesis than those not paused, but they have similar activation timing (Saunders et al., 2013). Therefore rapid induction may not be the major function of paused Pol II.
Establishing a permissive chromatin state: As mentioned above, the DNA sequence of highly paused promoters intrinsically favor high promoter nucleosome formation (Gaertner et al., 2012; Gilchrist et al., 2010). In order to recruit Pol II to these promoters, nucleosome removal needs to occur. The promoter regions of Drosophila Hsp genes were shown to be depleted of nucleosomes even without induction, and mutating the promoter-proximal sequences not only reduced Pol II pausing levels, but also disrupted heat shock factor (HSF) binding and subsequent gene activation following heat shock, suggesting that pausing may facilitate an open chromatin structure at promoters and hence the binding of TF and subsequent Pol II (reviewed in Adelman and Lis, 2012). A genome-wide study of Drosophila S2 cells (Gilchrist et al., 2010) further demonstrated the relationship between paused Pol II and nucleosome-depleted promoters: strongly paused genes globally tend to have low promoter nucleosome occupancy, which is dependent on promoter paused Pol II, because NELF RNAi significantly decreases Pol II pausing at strongly paused genes, with a concurrent increase of promoter nucleosome occupancy. Therefore, paused Pol II seems to compete with nucleosomes for promoter binding, to antagonize chromatin repression and establish a permissive state for future or further rounds of activation. Consistent with this, genes with diminished Pol II pausing after NELF knockdown typically have decreased expression (Gilchrist et al., 2012; Gilchrist et al., 2008). The permissive state established by Pol II may have a significant impact during development. Paused Pol II was found gradually acquired at many initially closed promoters during Drosophila embryogenesis in a stage-specific but not tissue-specific manner, suggesting that Pol II pausing may open promoters to prime their activation by tissue-specific TFs (Gaertner et al., 2012). Polycomb group (PcG) proteins bind many developmental control genes and prevent their inaccurate activation in a spatio-temporal specific manner. The combination of Pol II pausing and PcG repression (“balanced state”) could convey more dynamic transcription regulation during development (Gaertner et al., 2012). Besides this, the permissive chromatin state established by Pol II pausing might also help maintain the basal transcription level of genes and prepare them for bursts of activation under induction (Gilchrist et al., 2012), or for synchronous activation (see below).
Synchronous activation: In Drosophila early embryos, paused genes were found to express more synchronously within their expression domain than non-paused genes (Boettiger and Levine, 2009). Promoter swapping experiments (Lagha et al., 2013) showed that this synchrony depends on the core promoter, rather than the enhancer. When analyzing the expression of transgenes containing different promoters with a spectrum of intrinsic pausing levels, the stronger promoter-Pol II they bear in vivo, the more synchronous the transgene activation is. Moreover, by replacing the promoter of sna (highly paused) with that of sog (moderately paused) or ths (non-paused), the sna BAC transgene displayed an increasing spectrum of stochastic activation and defects in rescuing mesoderm invagination, indicating that transcriptional synchrony determined by Pol II pausing is essential for coordinating tissue morphogenesis (Lagha et al., 2013). One possibility of how Pol II pausing achieves this is by reducing the stochasticity caused by stepwise assembly of the PIC, to prepare all Pol II in a transcriptionally engaged status for synchronous activation in response to signal. Another possibility is that it may space out transcripts to prevent transcription bursts, also allow longer periods between re-initiation, so that the transcription machinery is less crowded and has more time for post-transcriptional processing (reviewed in Gaertner and Zeitlinger, 2014).
Checkpoint of early elongation: Pol II carrying Ser5P serves as a platform to interact with the 5’ capping enzyme, which adds 5’-methyl cap when nascent RNA is ~20-30 nt in length (reviewed in Adelman and Lis, 2012). Consistent with this, the majority of RNA associated with paused Pol II is capped (Nechaev et al., 2010). DSIF has also been shown to interact with the capping machinery. Therefore, pausing may function as a checkpoint for RNA processing before proceeding into productive elongation (reviewed in Adelman and Lis, 2012).
Regulatory signal integration: Some activators primarily stimulate GTFs and Pol II recruitment, or establish paused Pol II, such as Sp1 (Blau et al., 1996) and GAF (Lee et al., 2008). Some activators predominantly promote elongation by bringing P-TEFb to promoters, such as c-Myc and HIV TAT, while others stimulate both recruitment and release of paused Pol II, such as NF-κB and VP16 (reviewed in Adelman and Lis, 2012; Blau et al., 1996). Thus, Pol II pausing represents an additional layer of transcriptional regulation, possibly by allowing activators with different activation abilities to work together to achieve combinatorial control of expression levels.
To summarize, although Pol II pausing has emerged as an important step in transcriptional regulation, it remains unknown how Pol II is initially recruited to promoters, and what TFs trigger Pol II release through P-TEFb. Particularly, during Drosophila ZGA, whether the two waves of zygotic genes are regulated via different mechanisms is unknown.
Zelda is the key activator of Drosophila zygotic genome
A long-standing question in the embryogenesis field has been what factors trigger ZGA. In 2008, our lab identified Zelda (Zld; Zinc-finger early Drosophila activator; also known as Vfl) to be the key activator of the early Drosophila zygotic genome (Liang et al., 2008).
zld mRNA is maternally loaded from the nurse cells in the ovary, and is evenly distributed in the embryo excluding PGCs until cellularization, when maternally deposited zld is replaced by zygotically transcribed zld (Liang et al., 2008; Staudt et al., 2006). Zld protein can be detected in embryo nuclei as early as nc 2, whose level greatly increases and peaks at nc 10, coinciding with ZGA, then gradually decreases afterwards (Harrison et al., 2010; Kanodia et al., 2012; Nien et al., 2011). It is present ubiquitously in early embryos, then becomes restricted to the central nervous system (CNS), the peripheral nervous system (PNS), the brain, and imaginal disc primordia later in embryogenesis (Giannios and Tsitilou, 2013; Pearson et al., 2012; Staudt et al., 2006).
Cite This Essay
To export a reference to this article please select a referencing stye below: