In addition to RNA polymerase II (Pol2) pre-initiation complex (PIC) assembly and initiation, the transition from promoter-proximal paused Pol2 to productively elongating Pol2 is an essential step in gene transcription and an important process in the overall multicomponent orchestration of gene expression in metazoans (Core et al., 2008; Seila et al., 2008; Core and Adelman, 2019). After the recruitment of Pol2 to a promoter by its general transcription factors and assembly of a PIC (Nogales et al., 2017; Sainsbury et al., 2015), transcription initiation occurs concurrently with TFIIH phosphorylation of Ser5 of the Pol2 heptapeptide repeat C-terminal domain (CTD) (Cramer, 2019). In metazoan cells, Pol2 then transcribes approximately 30–60 bases downstream of the transcription start site (TSS) and then pauses because it is bound by negative elongation factor (NELF) and DRB-sensitivity-inducing factor (DSIF, Spt4 and Spt5 in Saccharomyces cerevisiae) (Jonkers and Lis, 2015; Adelman and Lis, 2012). Association of elongation factor P-TEFb, including its enzymatic subunits CDK9-cyclin T, results in phosphorylation of NELF, DSIF, and Ser2 of the Pol2 CTD, whereupon NELF dissociates and Pol2 is released and proceeds to productive elongation (Cramer, 2019; Jonkers and Lis, 2015; Adelman and Lis, 2012; Vos et al., 2018).

Histone acetylation is well known to contribute to a permissive chromatin state for Pol2 PIC assembly at active promoters, and there is recently published work concerning its function in facilitating transcriptional elongation as well. For example, the chromatin reader protein BRD4 is thought to recruit P-TEFb (CDK9-cyclin T) to promoters and serves as a Pol2 elongation factor dependent on its interactions with acetylated histone lysines through its bromodomains (Kanno et al., 2014). In addition, H3 acetylation mediated by the Drosophila CBP ortholog stimulates productive elongation past the +1 nucleosome (Boija et al., 2017). Recruitment of the yeast histone chaperone FACT by acetylated H3 has also been shown to stimulate elongation (Pathak et al., 2018).

The super-elongation complex (SEC) is a multi-subunit complex comprising P-TEFb (CDK9-cyclin T) along with AF4/FMR2 proteins AFF1/4, ELL family members ELL1/2/3, ELL-associated factors EAF1/2, and one or the other highly homologous proteins AF9 or ENL containing YEATS acetyl-lysine-binding domains (Luo et al., 2012). There are various forms of the SEC, including SEC-like complexes that contain different combinations of elongation factors suggesting diversity in their regulatory mechanisms (Luo et al., 2012). The central serine/threonine-kinase P-TEFb (CDK9-cyclin T), an AFF scaffold protein, and ENL or AF9 are consistent components of SEC complexes. ENL and AF9 have been functionally linked to SEC recruitment to acetylated chromatin via their YEATS domains (Li et al., 2014; Gates et al., 2017). The SEC then stimulates transcription elongation through phosphorylation by CDK9 of NELF, DSIF, and Ser2 of the Pol2 CTD repeat, as well as interactions with the PAF1 complex (He et al., 2011), which blocks NELF-binding to Pol2 (Cramer, 2019), and DOT1L, which deposits the active chromatin modification H3K79me in the first intron (Li et al., 2014; Huff et al., 2010). Importantly, AF9 and ENL YEATS domains bind to active chromatin marks H3K9ac and H4K15ac (Gates et al., 2017), and, to a lesser extent, H3K18/27ac (Li et al., 2014), and are essential for SEC-dependent activation of a luciferase reporter driven by the HIV-1 LTR (He et al., 2011). Despite these conclusions, the function of histone acetylation during the transition from promoter-proximal paused to productively elongating Pol2 remains incompletely understood.

We previously reported that CBP/p300 acetylation of H3K18 and K27 in the two to three nucleosomes spanning the TSS had very different effects on distinct steps in transcription from different human adenovirus 5 (HAdV-5) early promoters (Hsu et al., 2018). At the E3 promoter, loss of H3K18/27ac in the promoter region had little effect on PIC assembly, and the rate of E3 mRNA synthesis was only modestly reduced (less than twofold) compared to transcription activated by wt E1A, which induces H3K18/27ac at the early viral promoters. In contrast, PIC assembly at the E2early promoter was almost eliminated by loss of promoter H3K18/27ac, and E2 mRNA synthesis was undetectable at 12 hr post-infection (p.i.) (Hsu et al., 2018). For E4, loss of promoter H3K18/27ac had little effect on PIC assembly but caused a significant (approximately tenfold) decrease in E4 transcription at 12 hr p.i. (Hsu et al., 2018). This result was particularly striking as it suggested that E4 transcription is regulated by promoter H3K18/27 acetylation at a step in transcription subsequent to PIC assembly, possibly during release of promoter-proximal paused Pol2.

To investigate the function of H3K18/27ac in transcriptional elongation at E4, we mapped the association of transcriptionally active Pol2 on the Ad5 genome using Global Run-On sequencing (GRO-seq) (Core et al., 2008). We found defective paused Pol2 release at E4 in cells expressing an E1A mutant (‘E1A-DM’) with polyalanine substituted for two highly acidic regions of the E1A activation domain (AD) that each mediate an interaction with CBP/p300 (Hsu et al., 2018). ChIP-seq for BRD4 and SEC components CDK9, AF9, and ENL revealed decreased SEC recruitment to E4 by E1A-DM compared to wt E1A. Using the specific small molecule inhibitor of CBP/p300 acetyl transferase activity A-485 (Lasko et al., 2017), we determined that CBP/p300 histone acetyl transferase (HAT) activities are essential for maximal paused Pol2 release at the E4 promoter, but not at the E3 promoter.

We then extended our studies to the human genome, where we found that 2 hr of A-485 treatment resulted in hypoacetylation of total cell H3K18/27 to a new, extensively hypoacetylated steady state. This was associated with defective pause release at a subset of active genes (~7%) where promoter H3K18/27ac was decreased by the drug. Differences in the sensitivity of transcription from different promoters to H3K18/27ac correlated with differences in SEC component association with the genes after A-485 treatment. This was similar to what we had observed for the HAdV-5 E4 promoter during activation by the multi-site E1A mutant (DM-E1A) with mutations in the E1A-AD acidic peptides required for CBP/p300 binding to the E1A-AD. We also found that at a subset of enhancers with greatly decreased H3K18/27ac in response to A-485 treatment, H3K9ac is sufficient for BRD4 binding and stimulation of Pol2 pause release. Based on these results, we propose mechanisms of BRD4 and SEC recruitment by histone H3 acetylation during the transition from promoter-proximal paused to productively elongating Pol2 and report a homeostatic process that maintains promoter H3K18/27ac.

CBP/p300 acetylation of promoter histone H3K18 and K27 stimulates paused Pol2 release at the HAdV-5 E4 promoter, but is not required at the E3 promoter.

Transcription from HAdV-5 early promoters is activated by the first viral proteins expressed following infection, the E1A isoforms, primarily large E1A (Figure 1—figure supplement 1). While transcriptional activation from the viral early promoters is entirely dependent on the interaction of the large E1A isoform with the mediator complex, transcription of E4 is stimulated an additional tenfold through interactions between CBP/p300 and two highly acidic regions immediately flanking the E1A mediator-binding region (large E1A aa residues 133—138 and 189—200; Figure 1—figure supplement 1; Hsu et al., 2018). Separate Ad5 expression vectors were constructed that express the wt E1A from the wt E1A promoter/enhancer or DM-E1A with several mutations that convert these acidic peptides in wt E1A to polyAla (Figure 1—figure supplement 1b).

To analyze the effects of promoter H3K18/27ac on Pol2 elongation through the early Ad5 genes, we applied the GRO-seq method, which reveals the position and direction of transcribing Pol2 by BrU-labeling of 3′-ends of nascent RNA transcripts in isolated nuclei (Core et al., 2008). The nuclei were first washed with the non-ionic detergent sarkosyl to remove proteins from chromatin that block transcription elongation and prevent Pol2 initiation, so only actively transcribing RNA polymerases at the time the nuclei were isolated produce BrU-labeled RNA (Core et al., 2008). To avoid possible effects of cellular mutations in stable cell lines, we performed these studies in primary human bronchial-tracheal epithelial cells (HBTECs) derived from human adult lung transplant donors. These HBTECs are a cell culture model for the airway epithelial cells infected by HAdV-5 in humans.

We infected HBTECs with the wt E1A vector or the DM-E1A vector expressing mutant E1A with polyAla substitutions of the two highly acidic peptides flanking CR3 (Figure 1—figure supplement 1). Wt E1A binds CBP/p300 through these highly acidic peptides, inducing histone H3 acetylation at K18 and K27 by the CBP/p300 acetyl transferase domain at the viral E2early, E3, and E4 promoter regions (Hsu et al., 2018). In cells expressing DM-E1A, which does not interact in vivo with CBP/p300 through the E1A activation domain (E1A-AD) (Hsu et al., 2018), there was far less H3K18/27ac at these viral early promoter regions (Figure 1a). To express equal steady-state levels of the wt E1A and less stable DM-E1A proteins, infections were performed at a twentyfold higher multiplicity of infection for the DM-E1A vector than for the wt E1A vector (Hsu et al., 2018; Figure 1—figure supplement 1). Cells infected with the wt E1A vector were also coinfected with sufficient E1A deletion mutant dl312 to maintain the same number of viral DNA molecule templates for the viral early regions (~100 viral DNA molecules per nucleus) in cells expressing the same level of wt E1A and the DM-E1A protein (Hsu et al., 2018; Figure 1—figure supplement 1).

Figure 1 with 1 supplement see all

Download asset Open asset

Figure 1—source data 1

qRT-PCR for E2early, E3, and E4 pre-mRNA transcripts in cells treated with DMSO or A-485.

https://cdn.elifesciences.org/articles/63512/elife-63512-fig1-data1-v3.xlsx

Download elife-63512-fig1-data1-v3.xlsx

GRO-seq data at 12 hr p.i. with the vector expressing wt E1A revealed peaks of paused Pol2 with the expected orientation and location of promoter-proximal paused Pol2, ~40–60 bp downstream from the E3 and E4 TSSs (Figure 1a, highlighted, Figure 2a). At 12 hr p.i., very low GRO-seq signal was observed at the E2 early promoter or within the E2 gene body in wt E1A expressing cells compared to E3 and E4 (Figure 1a). This is probably because E2early transcription is delayed compared to E3 and E4 in these primary cells and increases by 18 hr p.i. (Hsu et al., 2018). The low GRO-seq signal in the E2early promoter region and gene body was decreased further in cells expressing DM-E1A compared to wt E1A (Figure 1a), supporting our previous conclusion that E2early transcription is regulated by H3K18/27ac in the promoter region because it is required for rapid PIC assembly (Hsu et al., 2018).

Figure 2

Download asset Open asset

To determine the degree of promoter-proximal pausing in the E2, E3, and E4 promoter regions where Pol2 association is detected by ChIP-seq at 12 hr p.i. (Hsu et al., 2018), we calculated the Pol2 pausing index (PI, Core et al., 2008). The PI equals the number of GRO-seq reads in the promoter-proximal region (TSS to +200 bp) divided by the total GRO-seq reads in the gene body (+201 to TTS). The GRO-seq reads in the promoter-proximal region reflect the amount of promoter-proximal paused Pol2 in the population of cells at the time the nuclei were isolated, while the GRO-seq reads in the gene body reflect the amount of elongating Pol2 subsequent to pause release. Therefore, an increase in PI indicates a reduced rate of promoter-proximal pause release.

After activation by DM-E1A, the PI at E4 increased 2.5-fold on average in three biological replicate GRO-seq experiments, a significant increase compared to the degree of pausing at E4 when activated by wt E1A (Figure 1b). Because the vectors expressing wt E1A and DM-E1A had different size deletions in E3 due to differences in their constructions (Figure 1a, bottom), the calculation of PI for E3 was based on the ~1.1 kb region of E3 common to both vectors (gray line above E3 in the Ad5 map, Figure 1a, bottom). In contrast to E4, there was no significant change in PI at E3 where promoter H3K18/27 acetylation had only a modest effect on transcription (Hsu et al., 2018; Figure 1b). Similar to E3, the low level of GRO-seq counts at the E2early promoter region showed an insignificant change in PI between wt E1A and DM E1A (Figure 1b), suggesting that H3K18/27ac acetylation does not affect pause release at the E2early promoter and primarily promotes Pol2 initiation. Therefore, loss of H3K18/27 promoter acetylation produced a smaller defect on promoter-proximal Pol2 pause release at the E2 and E3 promoters than at the E4 promoter.

CBP/p300 acetyl transferase activity is required for efficient Pol2 pause release and recruitment of AF9, ENL, and BRD4 at E4

A-485 is a potent and specific small molecule inhibitor of CBP/p300 acetyl transferase activity that competes with acetyl-CoA for binding to the acetyl transferase domain active site (Lasko et al., 2017). Decreased total cell H3K18ac after A-485 treatment in HBTECs was confirmed by western blot (Figure 3a). ChIP-seq for H3K9ac, H3K18ac, and H3K27ac in wt E1A vector-infected cells treated with A-485 for 2 hr demonstrated A-485 inhibition of H3K18/27ac, as expected (Lasko et al., 2017; Weinert et al., 2018), while H3K9ac was modestly affected at viral early promoters (Figure 3b). As a measure of the transcription rate of the early viral genes, we assayed pre-mRNA levels using qRT-PCR of intronic RNA isolated from HBTECs expressing wt E1A treated with 10 µM A-485 or control DMSO (dimethyl sulfoxide) vehicle alone for 2 hr. We observed significant decreases in E2early and E4 pre-mRNA after A-485 treatment, while E3 pre-mRNA was not decreased significantly (Figure 3c). This result confirms that E3 transcription is less dependent on promoter H3K18/27ac than E2early and E4 transcription (Hsu et al., 2018).

Figure 3

Download asset Open asset

Figure 3—source data 1

Average fold change in pause index for E2early, E3, and E4 in cells expressing wt E1A compared to DM E1A.

https://cdn.elifesciences.org/articles/63512/elife-63512-fig3-data1-v3.xlsx

Download elife-63512-fig3-data1-v3.xlsx

To determine if A-485-induced H3K18/27 hypoacetylation correlates with changes in elongation, we performed GRO-seq in DMSO or A-485-treated cells and calculated the pausing indices for transcription of the viral early genes (Figure 4a, b). There was little difference in PI with A-485 treatment at the E2early (DMSO (dimethyl sulfoxide) PI = 0.13, A-485 PI = 0.12) or E3 (DMSO (dimethyl sulfoxide) PI = 1.20, A-485 PI = 1.19) promoters, but a clear increase in PI was observed for E4 (DMSO (dimethyl sulfoxide) PI = 0.79, A-485 PI = 1.13) (Figure 4b). These results indicate that release of paused Pol2 in the E4 promoter-proximal region depends on CBP/p300 HAT activity and is consistent with a requirement for promoter H3K18/27 acetylation for maximal E4 elongation.

Figure 4

Download asset Open asset

Figure 4—source data 1

Pause indexes for E2early, E3, and E4 in cells treated with DMSO compared to A-485 treated cells.

https://cdn.elifesciences.org/articles/63512/elife-63512-fig4-data1-v3.xlsx

Download elife-63512-fig4-data1-v3.xlsx

To determine if inhibition of CBP/p300 HAT activity resulted in defective SEC recruitment at E4, we performed AF9, ENL, and BRD4 ChIP-seq in cells infected with the wt E1A vector after DMSO (dimethyl sulfoxide) or A-485 treatment. As with cells expressing DM-E1A, A-485 treatment greatly decreased AF9 association at the E2early, E3, and E4 promoter regions, and decreased total ENL ChIP-seq counts from the E2early TTS to the right end of the genome from 13,712,875 to 10,804,279 reads, an ~20% decrease (Figure 4a). We also observed decreased BRD4 throughout the transcribed early regions in A-485-treated cells. Taken together, these data show that maximal recruitment of SEC complexes to viral early promoters requires H3K18/27ac by the CBP/p300 acetyl transferase catalytic domain targeted to the promoter region by the interaction between CBP/p300 and the E1A-AD acidic regions (Figure 4c).

CBP/p300 HAT inhibition by A-485 affects H3 acetylation of cellular chromatin differently at promoters and enhancers

To determine if the effects of H3K18/27ac on Pol2 pause release at E4 is a general mechanism that also applies to transcription of cellular genes, we shifted our study to the human genome. First, we characterized the changes in H3K18/27ac in HBTECs after 2 hr treatment with 10 µM A-485. Western blotting demonstrated an extensive decrease in total cellular H3K18ac, which approached steady state by 2 hr after addition of A-485, as expected (Lasko et al., 2017; Weinert et al., 2018; Figure 3a). H3K18ac and H3K27ac enriched genomic regions were determined by separate ChIP-seq analyses with antibodies specific for either H3K18ac or H3K27ac. H3K18/27ac are enriched at both active promoters and enhancers. Enhancers can be divided into typical enhancers and super-enhancers, which are dense clusters of neighboring enhancers that often activate highly expressed cell identity genes (Whyte et al., 2013). We used the ranking of super-enhancers algorithm (Whyte et al., 2013; Lovén et al., 2013; Hnisz et al., 2013) with H3K27ac ChIP-seq data to determine super-enhancer regions in HBTECs (Zemke et al., 2019). H3K27ac peaks >2.5 kb from the nearest TSS but outside of super-enhancers were classified as typical enhancers. Comparing the average signals for H3K18ac and H3K27ac, we observed the expected decreases due to A-485 treatment at both typical and super-enhancers (Figure 5a). But, on average, A-485 caused greater decreases in both H3K18ac and H3K27ac at super-enhancers than typical enhancers (Figure 5a). This pattern was also true for H3K9ac, although the hypoacetylation by A-485 was to a much lesser extent (Figure 5—figure supplement 1).

Figure 5 with 3 supplements see all

Download asset Open asset

Figure 5—source data 1

Lists of all active genes, 2XPI, and 2XIn genes.

https://cdn.elifesciences.org/articles/63512/elife-63512-fig5-data1-v3.xlsx

Download elife-63512-fig5-data1-v3.xlsx

An unexpected result was the increase in the average level of H3K18ac and H3K27ac at all TSSs in cells treated with A-485 (Figure 5a). Furthermore, the percentage of total H3K27ac and H3K18ac peaks increased at promoters and decreased at enhancers after A-485 treatment, while the effects on H3K9ac peaks were minimal (Figure 5—figure supplement 2). These observations indicate that homeostatic mechanisms function to maintain H3K18/27ac at promoters when CBP/p300, the principal cellular acetyl transferases for these sites (Weinert et al., 2018; Horwitz et al., 2008; Ferrari et al., 2014), are extensively inhibited.

A-485-sensitive Pol2 pause release and SEC recruitment at genes with low promoter region H3 acetylation

We plotted the average H3K18ac, K27ac, and K9ac ChIP-seq counts ±5 kb from the TSS for all active genes, 2XPI genes, and 2XIn genes (Figure 6a). H3K18ac and K27ac at TSSs for 2XPI and 2XIn genes decreased in response to A-485, as expected for a specific competitive inhibitor of the CBP/p300 acetyl transferase. But this was in contrast to the surprising increase in H3K18 and K27 acetylation on average near the TSSs for all active genes in response to A-485. Thus H3K18/27 acetylation in the promoter regions of 2XPI and 2XIn genes was decreased by A-485 as it is throughout the vast majority of the genome (Figure 3a), whereas the average H3K18/27 acetylation in the promoter regions for all active genes was increased by treatment with A-485 (Figure 6a). In contrast, H3K9ac did not change at the TSS in the average plot for all active genes or 2XPI genes after A-485 treatment, but did decrease for 2XIn genes (Figure 6a). This suggests that H3K9ac may be sufficient for initiation but not elongation of genes with promoter H3K18/27 hypoacetylation.

Figure 6 with 1 supplement see all

Download asset Open asset

To determine if these decreases in H3 acetylation at 2XPI genes were correlated with decreased SEC component binding, we plotted metagene profiles of AF9 and ENL ChIP-seq counts for all active genes and 2XPI genes after DMSO- or A-485 treatment (Figure 6b). Remarkably, AF9 and ENL were extensively enriched at TSSs and throughout gene bodies of 2XPI genes compared to the average for all active genes. Further, after A-485 treatment, AF9 and ENL association with 2XPI genes fell to the average level for all genes, which did not change with A-485 treatment. Thus, genes with an increase in PI after A-485 treatment (2XPI genes) had a very high association of SEC complexes throughout their transcription units that was selectively sensitive to A-485 treatment. Similarly, we found enrichment of AF9 and ENL at TSSs and throughout gene bodies in 2XIn genes, which fell to the level of the average for all genes after A-485 treatment (Figure 6c, Figure 6—figure supplement 1). This decrease in elongation factor association with 2XIn genes is likely explained in part because this group of genes was selected for having decreased Pol2 in the gene body with which elongation factors associate (Figure 6—figure supplement 1). BRD4 association at the TSSs of all active genes was increased with A-485, while association near the TSS of 2XPI and 2XIn genes was reduced by A-485, but to a far less extent than the decrease in AF9 and ENL.

H3K9ac is sufficient for enhancer association with BRD4 that stimulates pause release at nearby genes

It was evident that BRD4 peaks were enriched at enhancers. Fifty-six percent of identified BRD4 peaks (see Materials and methods) were >2.5 kb from the nearest TSS in control cells treated with DMSO (dimethyl sulfoxide). Of these distal peaks, 84% overlapped with peaks of H3K27ac, indicating that these peaks were primarily at enhancers. We subsequently clustered all BRD4 peaks >2.5 kb from a TSS (i.e., within an enhancer) based on whether BRD4 association decreased to <50% of DMSO (dimethyl sulfoxide) control after A-485 treatment for 2 hr. There were 7554 peaks where BRD4 decreased to <50% of control after A-485 treatment and 6324 peaks where BRD4 association remained unaltered or was not reduced to <50% of control (referred to as A-485-sensitive and -resistant BRD4 enhancer peaks, respectively; Figure 7). At both A-485-resistant and -sensitive BRD4 enhancer peaks, H3K18/27ac decreased following A-485 treatment (Figure 7a). In contrast, H3K9ac decreased only at the A-485-sensitive BRD4 peaks but not at A-485-resistant BRD4 peaks. These observations indicate that enhancer H3K9ac correlates with BRD4 association and suggest that H3K9ac is sufficient for BRD4 association at enhancers.

Figure 7 with 4 supplements see all

Download asset Open asset

We next asked if BRD4 enhancer association correlated with the extent of Pol2 pause release in the promoter-proximal region of nearby genes. For example, COPS8 has enhancers downstream from the TSS in the first intron and beyond its TTS as indicated by H3K18/27ac peaks in control DMSO-treated cells (Figure 7b). H3K18/27ac, H3K9ac, and BRD4 association with these enhancer peaks was A-485 sensitive, but not their associations near the TSS (Figure 7b). A-485 caused only a modest decrease in the GRO-seq reads at the Pol2 pause site (~30%, Figure 7b) and, therefore, only an ~30% decrease in the amount of Pol2 that had initiated transcription at the COPS8 TSS in this population of cells compared to control DMSO-treated cells. But A-485 caused a much greater decrease in GRO-seq reads relative to control DMSO-treated cells downstream from the COPS8 promoter-proximal pause site, indicating that A-485 reduced the rate of Pol2 pause release for this gene (Figure 7b, Figure 7—figure supplement 1). Additionally, NDRG1 and PAG1 are examples of other genes with nearby enhancers that have A-485-sensitive BRD4 association (Figure 7—figure supplements 2 and 3). Similar to COPS8, A-485 treatment caused a decrease in release of promoter-proximal paused Pol2, but little decrease in Pol2 initiation at these genes (Figure 7—figure supplements 2 and 3). These results indicate that genes with A-485-sensitive BRD4 enhancer association are transcriptionally inhibited by A-485 during elongation.

In contrast to COPS8, NDRG1, and PAG1, PTPRF is a gene with A-485-resistant BRD4 association at nearby enhancer regions within its introns (Figure 7b, Figure 7—figure supplement 1). A-485 treatment reduced H3K18/27ac but not H3K9ac or BRD4 association at these enhancers. GRO-seq reads were reduced to approximately the same degree at the pause site and throughout the gene body, indicating regulation mainly at the initiation step. CSF3 is also an example of a gene where A-485 greatly inhibited Pol2 initiation (Figure 7—figure supplement 4). We tested if the correlation between BRD4 enhancer association and efficient Pol2 pause release in the promoter-proximal region was observed broadly. To examine this, we compared the change in PIs after A-485 treatment for genes near A-485-sensitive BRD4 enhancer peaks with that of all active genes. We observed a significant increase in PI distribution for genes near A-485-sensitive BRD4 acting at distal enhancers to stimulate transcription elongation of target genes (Kanno et al., 2014).

There is a substantial base of knowledge establishing a correlation between histone N-terminal tail lysine acetylation and transcriptional activity. However, understanding of the mechanisms underlying this correlation remains incomplete. In addition to regulating PIC assembly and Pol2 initiation, our results support a mechanism by which histone H3 N-terminal tail acetylation in the promoter-proximal region regulates Pol2 release from promoter-proximal pause sites in a subset (~4.3%) of active cellular promoters in primary human airway epithelial cells.

Sequencing data have been deposited in GEO under accession code GSE167094.

1. 1. Berk AJ

   (2021) NCBI Gene Expression Omnibus

   ID GSE167094. Promoter-specific changes in initiation, elongation and homeostasis of histone H3 acetylation during CBP/p300 Inhibition.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE167094

1. 1. Adelman K
   2. Lis JT

   (2012) Promoter-proximal pausing of RNA polymerase II: emerging roles in metazoans

   Nature Reviews Genetics 13:720–731.

   https://doi.org/10.1038/nrg3293

   • PubMed
   • Google Scholar
2. 1. Arsenian S
   2. Weinhold B
   3. Oelgeschläger M
   4. Rüther U
   5. Nordheim A

   (1998) Serum response factor is essential for mesoderm formation during mouse embryogenesis

   The EMBO Journal 17:6289–6299.

   https://doi.org/10.1093/emboj/17.21.6289

   • PubMed
   • Google Scholar
3. 1. Boija A
   2. Mahat DB
   3. Zare A
   4. Holmqvist PH
   5. Philip P
   6. Meyers DJ
   7. Cole PA
   8. Lis JT
   9. Stenberg P
   10. Mannervik M

   (2017) CBP regulates recruitment and release of Promoter-Proximal RNA polymerase II

   Molecular Cell 68:491–503.

   https://doi.org/10.1016/j.molcel.2017.09.031

   • PubMed
   • Google Scholar
4. 1. Bulger M
   2. Groudine M

   (2011) Functional and mechanistic diversity of distal transcription enhancers

   Cell 144:327–339.

   https://doi.org/10.1016/j.cell.2011.01.024

   • PubMed
   • Google Scholar
5. 1. Buratowski S

   (2009) Progression through the RNA polymerase II CTD cycle

   Molecular Cell 36:541–546.

   https://doi.org/10.1016/j.molcel.2009.10.019

   • Google Scholar
6. 1. Chen FX
   2. Smith ER
   3. Shilatifard A

   (2018) Born to run: control of transcription elongation by RNA polymerase II

   Nature Reviews Molecular Cell Biology 19:464–478.

   https://doi.org/10.1038/s41580-018-0010-5

   • PubMed
   • Google Scholar
7. 1. Core LJ
   2. Waterfall JJ
   3. Lis JT

   (2008) Nascent RNA sequencing reveals widespread pausing and divergent initiation at human promoters

   Science 322:1845–1848.

   https://doi.org/10.1126/science.1162228

   • PubMed
   • Google Scholar
8. 1. Core L
   2. Adelman K

   (2019) Promoter-proximal pausing of RNA polymerase II: a nexus of gene regulation

   Genes & Development 33:960–982.

   https://doi.org/10.1101/gad.325142.119

   • PubMed
   • Google Scholar
9. 1. Cramer P

   (2019) Organization and regulation of gene transcription

   Nature 573:45–54.

   https://doi.org/10.1038/s41586-019-1517-4

   • PubMed
   • Google Scholar
10. 1. Dhalluin C
    2. Carlson JE
    3. Zeng L
    4. He C
    5. Aggarwal AK
    6. Zhou MM

    (1999) Structure and ligand of a histone acetyltransferase bromodomain

    Nature 399:491–496.

    https://doi.org/10.1038/20974

    • PubMed
    • Google Scholar
11. 1. Eagen KP
    2. Hartl TA
    3. Kornberg RD

    (2015) Stable chromosome condensation revealed by chromosome conformation capture

    Cell 163:934–946.

    https://doi.org/10.1016/j.cell.2015.10.026

    • PubMed
    • Google Scholar
12. 1. Ferrari R
    2. Su T
    3. Li B
    4. Bonora G
    5. Oberai A
    6. Chan Y
    7. Sasidharan R
    8. Berk AJ
    9. Pellegrini M
    10. Kurdistani SK

    (2012) Reorganization of the host epigenome by a viral oncogene

    Genome Research 22:1212–1221.

    https://doi.org/10.1101/gr.132308.111

    • PubMed
    • Google Scholar
13. 1. Ferrari R
    2. Gou D
    3. Jawdekar G
    4. Johnson SA
    5. Nava M
    6. Su T
    7. Yousef AF
    8. Zemke NR
    9. Pellegrini M
    10. Kurdistani SK
    11. Berk AJ

    (2014) Adenovirus small E1A employs the lysine acetylases p300/CBP and tumor suppressor rb to repress select host genes and promote productive virus infection

    Cell Host & Microbe 16:663–676.

    https://doi.org/10.1016/j.chom.2014.10.004

    • PubMed
    • Google Scholar
14. 1. Filippakopoulos P
    2. Picaud S
    3. Mangos M
    4. Keates T
    5. Lambert JP
    6. Barsyte-Lovejoy D
    7. Felletar I
    8. Volkmer R
    9. Müller S
    10. Pawson T
    11. Gingras AC
    12. Arrowsmith CH
    13. Knapp S

    (2012) Histone recognition and large-scale structural analysis of the human bromodomain family

    Cell 149:214–231.

    https://doi.org/10.1016/j.cell.2012.02.013

    • PubMed
    • Google Scholar
15. 1. Franco CA
    2. Li Z

    (2009) SRF in angiogenesis: branching the vascular system

    Cell Adhesion & Migration 3:264–267.

    https://doi.org/10.4161/cam.3.3.8291

    • PubMed
    • Google Scholar
16. 1. Gao Y
    2. Chen L
    3. Han Y
    4. Wu F
    5. Yang WS
    6. Zhang Z
    7. Huo T
    8. Zhu Y
    9. Yu C
    10. Kim H
    11. Lee M
    12. Tang Z
    13. Phillips K
    14. He B
    15. Jung SY
    16. Song Y
    17. Zhu B
    18. Xu RM
    19. Feng Q

    (2020) Acetylation of histone H3K27 signals the transcriptional elongation for estrogen receptor alpha

    Communications Biology 3:165.

    https://doi.org/10.1038/s42003-020-0898-0

    • PubMed
    • Google Scholar
17. 1. Gates LA
    2. Shi J
    3. Rohira AD
    4. Feng Q
    5. Zhu B
    6. Bedford MT
    7. Sagum CA
    8. Jung SY
    9. Qin J
    10. Tsai MJ
    11. Tsai SY
    12. Li W
    13. Foulds CE
    14. O'Malley BW

    (2017) Acetylation on histone H3 lysine 9 mediates a switch from transcription initiation to elongation

    Journal of Biological Chemistry 292:14456–14472.

    https://doi.org/10.1074/jbc.M117.802074

    • PubMed
    • Google Scholar
18. 1. He N
    2. Chan CK
    3. Sobhian B
    4. Chou S
    5. Xue Y
    6. Liu M
    7. Alber T
    8. Benkirane M
    9. Zhou Q

    (2011) Human Polymerase-Associated factor complex (PAFc) connects the super elongation complex (SEC) to RNA polymerase II on chromatin

    PNAS 108:E636–E645.

    https://doi.org/10.1073/pnas.1107107108

    • Google Scholar
19. 1. Heinz S
    2. Benner C
    3. Spann N
    4. Bertolino E
    5. Lin YC
    6. Laslo P
    7. Cheng JX
    8. Murre C
    9. Singh H
    10. Glass CK

    (2010) Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities

    Molecular Cell 38:576–589.

    https://doi.org/10.1016/j.molcel.2010.05.004

    • PubMed
    • Google Scholar
20. 1. Hnisz D
    2. Abraham BJ
    3. Lee TI
    4. Lau A
    5. Saint-André V
    6. Sigova AA
    7. Hoke HA
    8. Young RA

    (2013) Super-enhancers in the control of cell identity and disease

    Cell 155:934–947.

    https://doi.org/10.1016/j.cell.2013.09.053

    • PubMed
    • Google Scholar
21. 1. Horwitz GA
    2. Zhang K
    3. McBrian MA
    4. Grunstein M
    5. Kurdistani SK
    6. Berk AJ

    (2008) Adenovirus small e1a alters global patterns of histone modification

    Science 321:1084–1085.

    https://doi.org/10.1126/science.1155544

    • PubMed
    • Google Scholar
22. 1. Hsu E
    2. Pennella MA
    3. Zemke NR
    4. Eng C
    5. Berk AJ

    (2018) Adenovirus E1A activation domain regulates H3 acetylation affecting varied steps in transcription at different viral promoters

    Journal of Virology 92:e00805-18.

    https://doi.org/10.1128/JVI.00805-18

    • PubMed
    • Google Scholar
23. 1. Huff JT
    2. Plocik AM
    3. Guthrie C
    4. Yamamoto KR

    (2010) Reciprocal intronic and exonic histone modification regions in humans

    Nature Structural & Molecular Biology 17:1495–1499.

    https://doi.org/10.1038/nsmb.1924

    • PubMed
    • Google Scholar
24. 1. Jang MK
    2. Mochizuki K
    3. Zhou M
    4. Jeong HS
    5. Brady JN
    6. Ozato K

    (2005) The bromodomain protein Brd4 is a positive regulatory component of P-TEFb and stimulates RNA polymerase II-dependent transcription

    Molecular Cell 19:523–534.

    https://doi.org/10.1016/j.molcel.2005.06.027

    • PubMed
    • Google Scholar
25. 1. Jin Q
    2. Yu LR
    3. Wang L
    4. Zhang Z
    5. Kasper LH
    6. Lee JE
    7. Wang C
    8. Brindle PK
    9. Dent SY
    10. Ge K

    (2011) Distinct roles of GCN5/PCAF-mediated H3K9ac and CBP/p300-mediated H3K18/27ac in nuclear receptor transactivation

    The EMBO Journal 30:249–262.

    https://doi.org/10.1038/emboj.2010.318

    • PubMed
    • Google Scholar
26. 1. Jonkers I
    2. Kwak H
    3. Lis JT

    (2014) Genome-wide dynamics of pol II elongation and its interplay with promoter proximal pausing, Chromatin, and exons

    eLife 3:e02407.

    https://doi.org/10.7554/eLife.02407

    • PubMed
    • Google Scholar
27. 1. Jonkers I
    2. Lis JT

    (2015) Getting up to speed with transcription elongation by RNA polymerase II

    Nature Reviews Molecular Cell Biology 16:167–177.

    https://doi.org/10.1038/nrm3953

    • PubMed
    • Google Scholar
28. 1. Juven-Gershon T
    2. Kadonaga JT

    (2010) Regulation of gene expression via the core promoter and the basal transcriptional machinery

    Developmental Biology 339:225–229.

    https://doi.org/10.1016/j.ydbio.2009.08.009

    • PubMed
    • Google Scholar
29. 1. Kanno T
    2. Kanno Y
    3. Siegel RM
    4. Jang MK
    5. Lenardo MJ
    6. Ozato K

    (2004) Selective recognition of acetylated histones by bromodomain proteins visualized in living cells

    Molecular Cell 13:33–43.

    https://doi.org/10.1016/S1097-2765(03)00482-9

    • PubMed
    • Google Scholar
30. 1. Kanno T
    2. Kanno Y
    3. LeRoy G
    4. Campos E
    5. Sun HW
    6. Brooks SR
    7. Vahedi G
    8. Heightman TD
    9. Garcia BA
    10. Reinberg D
    11. Siebenlist U
    12. O'Shea JJ
    13. Ozato K

    (2014) BRD4 assists elongation of both coding and enhancer RNAs by interacting with acetylated histones

    Nature Structural & Molecular Biology 21:1047–1057.

    https://doi.org/10.1038/nsmb.2912

    • PubMed
    • Google Scholar
31. 1. Knöll B
    2. Nordheim A

    (2009) Functional versatility of transcription factors in the nervous system: the SRF paradigm

    Trends in Neurosciences 32:432–442.

    https://doi.org/10.1016/j.tins.2009.05.004

    • PubMed
    • Google Scholar
32. 1. Lasko LM
    2. Jakob CG
    3. Edalji RP
    4. Qiu W
    5. Montgomery D
    6. Digiammarino EL
    7. Hansen TM
    8. Risi RM
    9. Frey R
    10. Manaves V
    11. Shaw B
    12. Algire M
    13. Hessler P
    14. Lam LT
    15. Uziel T
    16. Faivre E
    17. Ferguson D
    18. Buchanan FG
    19. Martin RL
    20. Torrent M
    21. Chiang GG
    22. Karukurichi K
    23. Langston JW
    24. Weinert BT
    25. Choudhary C
    26. de Vries P
    27. Van Drie JH
    28. McElligott D
    29. Kesicki E
    30. Marmorstein R
    31. Sun C
    32. Cole PA
    33. Rosenberg SH
    34. Michaelides MR
    35. Lai A
    36. Bromberg KD

    (2017) Discovery of a selective catalytic p300/CBP inhibitor that targets lineage-specific tumours

    Nature 550:128–132.

    https://doi.org/10.1038/nature24028

    • PubMed
    • Google Scholar
33. 1. Levine M

    (2010) Transcriptional enhancers in animal development and evolution

    Current Biology 20:R754–R763.

    https://doi.org/10.1016/j.cub.2010.06.070

    • PubMed
    • Google Scholar
34. 1. Li Q
    2. Price JP
    3. Byers SA
    4. Cheng D
    5. Peng J
    6. Price DH

    (2005) Analysis of the large inactive P-TEFb complex indicates that it contains one 7sk molecule, a dimer of HEXIM1 or HEXIM2, and two P-TEFb molecules containing Cdk9 phosphorylated at Threonine 186

    Journal of Biological Chemistry 280:28819–28826.

    https://doi.org/10.1074/jbc.M502712200

    • PubMed
    • Google Scholar
35. 1. Li Y
    2. Wen H
    3. Xi Y
    4. Tanaka K
    5. Wang H
    6. Peng D
    7. Ren Y
    8. Jin Q
    9. Dent SY
    10. Li W
    11. Li H
    12. Shi X

    (2014) AF9 YEATS domain links histone acetylation to DOT1L-mediated H3K79 methylation

    Cell 159:558–571.

    https://doi.org/10.1016/j.cell.2014.09.049

    • PubMed
    • Google Scholar
36. 1. Lovén J
    2. Hoke HA
    3. Lin CY
    4. Lau A
    5. Orlando DA
    6. Vakoc CR
    7. Bradner JE
    8. Lee TI
    9. Young RA

    (2013) Selective inhibition of tumor oncogenes by disruption of super-enhancers

    Cell 153:320–334.

    https://doi.org/10.1016/j.cell.2013.03.036

    • PubMed
    • Google Scholar
37. 1. Luo Z
    2. Lin C
    3. Shilatifard A

    (2012) The super elongation complex (SEC) family in transcriptional control

    Nature Reviews Molecular Cell Biology 13:543–547.

    https://doi.org/10.1038/nrm3417

    • PubMed
    • Google Scholar
38. 1. Martin RM
    2. Rino J
    3. Carvalho C
    4. Kirchhausen T
    5. Carmo-Fonseca M

    (2013) Live-Cell visualization of Pre-mRNA splicing with Single-Molecule sensitivity

    Cell Reports 4:1144–1155.

    https://doi.org/10.1016/j.celrep.2013.08.013

    • Google Scholar
39. 1. Nguyen VT
    2. Kiss T
    3. Michels AA
    4. Bensaude O

    (2001) 7SK small nuclear RNA binds to and inhibits the activity of CDK9/cyclin T complexes

    Nature 414:322–325.

    https://doi.org/10.1038/35104581

    • PubMed
    • Google Scholar
40. 1. Nogales E
    2. Louder RK
    3. He Y

    (2017) Structural insights into the eukaryotic transcription initiation machinery

    Annual Review of Biophysics 46:59–83.

    https://doi.org/10.1146/annurev-biophys-070816-033751

    • PubMed
    • Google Scholar
41. 1. Ou HD
    2. Phan S
    3. Deerinck TJ
    4. Thor A
    5. Ellisman MH
    6. O'Shea CC

    (2017) ChromEMT: visualizing 3D chromatin structure and compaction in interphase and mitotic cells

    Science 357:eaag0025.

    https://doi.org/10.1126/science.aag0025

    • PubMed
    • Google Scholar
42. 1. Ozato K
    2. Shin DM
    3. Chang TH
    4. Morse HC

    (2008) TRIM family proteins and their emerging roles in innate immunity

    Nature Reviews Immunology 8:849–860.

    https://doi.org/10.1038/nri2413

    • PubMed
    • Google Scholar
43. 1. Pathak R
    2. Singh P
    3. Ananthakrishnan S
    4. Adamczyk S
    5. Schimmel O
    6. Govind CK

    (2018) Acetylation-Dependent recruitment of the FACT complex and its role in regulating pol II occupancy Genome-Wide in Saccharomyces cerevisiae

    Genetics 209:743–756.

    https://doi.org/10.1534/genetics.118.300943

    • PubMed
    • Google Scholar
44. 1. Rahl PB
    2. Lin CY
    3. Seila AC
    4. Flynn RA
    5. McCuine S
    6. Burge CB
    7. Sharp PA
    8. Young RA

    (2010) c-Myc regulates transcriptional pause release

    Cell 141:432–445.

    https://doi.org/10.1016/j.cell.2010.03.030

    • PubMed
    • Google Scholar
45. 1. Sainsbury S
    2. Bernecky C
    3. Cramer P

    (2015) Structural basis of transcription initiation by RNA polymerase II

    Nature Reviews Molecular Cell Biology 16:129–143.

    https://doi.org/10.1038/nrm3952

    • PubMed
    • Google Scholar
46. 1. Schaukowitch K
    2. Joo JY
    3. Liu X
    4. Watts JK
    5. Martinez C
    6. Kim TK

    (2014) Enhancer RNA facilitates NELF release from immediate early genes

    Molecular Cell 56:29–42.

    https://doi.org/10.1016/j.molcel.2014.08.023

    • PubMed
    • Google Scholar
47. 1. Schwartz B
    2. Marks M
    3. Wittler L
    4. Werber M
    5. Währisch S
    6. Nordheim A
    7. Herrmann BG
    8. Grote P

    (2014) SRF is essential for mesodermal cell migration during elongation of the embryonic body Axis

    Mechanisms of Development 133:23–35.

    https://doi.org/10.1016/j.mod.2014.07.001

    • Google Scholar
48. 1. Seila AC
    2. Calabrese JM
    3. Levine SS
    4. Yeo GW
    5. Rahl PB
    6. Flynn RA
    7. Young RA
    8. Sharp PA

    (2008) Divergent transcription from active promoters

    Science 322:1849–1851.

    https://doi.org/10.1126/science.1162253

    • PubMed
    • Google Scholar
49. 1. Shahbazian MD
    2. Grunstein M

    (2007) Functions of site-specific histone acetylation and deacetylation

    Annual Review of Biochemistry 76:75–100.

    https://doi.org/10.1146/annurev.biochem.76.052705.162114

    • PubMed
    • Google Scholar
50. 1. Takahashi H
    2. Parmely TJ
    3. Sato S
    4. Tomomori-Sato C
    5. Banks CA
    6. Kong SE
    7. Szutorisz H
    8. Swanson SK
    9. Martin-Brown S
    10. Washburn MP
    11. Florens L
    12. Seidel CW
    13. Lin C
    14. Smith ER
    15. Shilatifard A
    16. Conaway RC
    17. Conaway JW

    (2011) Human mediator subunit MED26 functions as a docking site for transcription elongation factors

    Cell 146:92–104.

    https://doi.org/10.1016/j.cell.2011.06.005

    • PubMed
    • Google Scholar
51. 1. Vijayalingam S
    2. Chinnadurai G

    (2013) Adenovirus L-E1A activates transcription through mediator complex-dependent recruitment of the super elongation complex

    Journal of Virology 87:3425–3434.

    https://doi.org/10.1128/JVI.03046-12

    • PubMed
    • Google Scholar
52. 1. Vos SM
    2. Farnung L
    3. Urlaub H
    4. Cramer P

    (2018) Structure of paused transcription complex pol II-DSIF-NELF

    Nature 560:601–606.

    https://doi.org/10.1038/s41586-018-0442-2

    • PubMed
    • Google Scholar
53. 1. Weinert BT
    2. Narita T
    3. Satpathy S
    4. Srinivasan B
    5. Hansen BK
    6. Schölz C
    7. Hamilton WB
    8. Zucconi BE
    9. Wang WW
    10. Liu WR
    11. Brickman JM
    12. Kesicki EA
    13. Lai A
    14. Bromberg KD
    15. Cole PA
    16. Choudhary C

    (2018) Time-Resolved analysis reveals rapid dynamics and broad scope of the CBP/p300 acetylome

    Cell 174:231–244.

    https://doi.org/10.1016/j.cell.2018.04.033

    • Google Scholar
54. 1. Whyte WA
    2. Orlando DA
    3. Hnisz D
    4. Abraham BJ
    5. Lin CY
    6. Kagey MH
    7. Rahl PB
    8. Lee TI
    9. Young RA

    (2013) Master transcription factors and mediator establish super-enhancers at key cell identity genes

    Cell 153:307–319.

    https://doi.org/10.1016/j.cell.2013.03.035

    • PubMed
    • Google Scholar
55. 1. Yamada T
    2. Yamaguchi Y
    3. Inukai N
    4. Okamoto S
    5. Mura T
    6. Handa H

    (2006) P-TEFb-mediated phosphorylation of hSpt5 C-terminal repeats is critical for processive transcription elongation

    Molecular Cell 21:227–237.

    https://doi.org/10.1016/j.molcel.2005.11.024

    • PubMed
    • Google Scholar
56. 1. Yang Z
    2. Zhu Q
    3. Luo K
    4. Zhou Q

    (2001) The 7sk small nuclear RNA inhibits the CDK9/cyclin T1 kinase to control transcription

    Nature 414:317–322.

    https://doi.org/10.1038/35104575

    • PubMed
    • Google Scholar
57. 1. Yoon JH
    2. Choi WS
    3. Kim O
    4. Choi SS
    5. Lee EK
    6. Nam SW
    7. Lee JY
    8. Park WS

    (2015) NKX6.3 controls gastric differentiation and tumorigenesis

    Oncotarget 6:28425–28439.

    https://doi.org/10.18632/oncotarget.4952

    • PubMed
    • Google Scholar
58. 1. Zemke NR
    2. Gou D
    3. Berk AJ

    (2019) Dedifferentiation by adenovirus E1A due to inactivation of hippo pathway effectors YAP and TAZ

    Genes & Development 33:828–843.

    https://doi.org/10.1101/gad.324814.119

    • PubMed
    • Google Scholar

Author details

1. Emily Hsu

   Molecular Biology Institute, UCLA, Los Angeles, United States

   Present address

   Department of Biochemistry and Molecular Medicine and the Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Writing - original draft

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-5096-0745

2. Nathan R Zemke

   Molecular Biology Institute, UCLA, Los Angeles, United States

   Present address

   Department of Cellular and Molecular Medicine, University of California, San Diego School of Medicine, La Jolla, United States

   Contribution

   Formal analysis, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-6326-5925

3. Arnold J Berk

   1. Molecular Biology Institute, UCLA, Los Angeles, United States
   2. Department of Microbiology, UCLA, Los Angeles, United States

   Contribution

   Conceptualization, Funding acquisition, Investigation, Writing - original draft

   For correspondence

   arnold.berk@icloud.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-9379-6287

• 3,312

  views

• 399

  downloads

• 24

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.