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

Abstract

Regulation of RNA Polymerase II (Pol2) elongation in the promoter proximal region is an important and ubiquitous control point for gene expression in metazoan cells. We report that transcription of the adenovirus 5 E4 region is regulated during the release of paused Pol2 into productive elongation by recruitment of the super elongation complex (SEC), dependent on promoter H3K18/27 acetylation by CBP/p300. We also establish that this is a general transcriptional regulatory mechanism for ~6% of genes expressed with FPKM>1 in primary human airway epithelial cells. We observed that a homeostatic mechanism maintains promoter, but not enhancer H3K18/27ac in response to extensive inhibition of CBP/p300 acetyl transferase activity by the highly specific small molecule inhibitor A-485. Further, our results suggest a function for BRD4 association at enhancers in regulating paused Pol2 release at nearby promoters. Taken together, our results uncover processes regulating transcriptional elongation by promoter region histone H3 acetylation and homeostatic maintenance of promoter, but not enhancer, H3K18/27ac in response to inhibition of CBP/p300 acetyl transferase activity.

Metabolic adaptations underpin resistance to histone acetyltransferase inhibition

“…In all, these data suggest that CBP/p300 HAT inhibition and concomitant histone hypoacetylation lead to eviction of acetyl-lysine binding proteins, deficient Pol II recruitment and activity, and a loss of enhancer-mediated leukemogenic transcription. These data are broadly consistent with results from other cell types 18,30,31 , but it remains unclear whether the changes we and others observe are sufficient to explain the anti-proliferative effects of CBP/p300 HAT inhibition. Comparison to lineage-matched insensitive controls have not been described and, therefore, the potential for these phenotypes to act as predictive biomarkers of response remains untested.…”

Theoretical and practical refinements of sans spike-in quantitative ChIP-seq with application to p300/CBP inhibition

“…However, using the correct physical scale will not only avoid seemingly inconsistent conclusions, it may also help bring ChIP-seq and RNA-seq, along with other orthogonal observations, into better alignment. [19] We speculate that the homeostatic mechanism[28] implied by the increased probability of finding H3K18ac fragments at TssA after A485 exposure is GCN5/KAT2A. Indeed, we speculate that this alternate and independent route to H3K18ac at TssA likely drives the missinterpretation in all the above-cited p300/CBP perturbation studies.…”

“…Indeed, this conclusion is reported in several papers. [16–19] In one study, the increase in H3K18ac signal at TssA was described as an ’increase in the average level of H3K18ac and H3K27ac.’[28] In another study, using spike-ins, it was ’observed that the H3K27ac mark was reduced at [TssA] whereas H3K18ac was slightly increased.’[17] Yet another recent report observed that ’surprisingly, only 226 [of 807] downregulated genes exhibited hypoacetylation of H3K27 under the condition of CBP/p300 HAT inhibition.’[19] Consistently, treating unscaled ChIP-seq data as though it were quantitative leads to the conclusion that the targeted acetylation may be unaffected or even increased in some genomic regions — counter to expectations and contrary to the indication of global loss in the isotherms (IP mass capture, Figure 3), which mirror what is seen by other global measures like western blots (Figure 3a of Reference 28 and SI-Fig. 4).…”

“…[16][17][18][19] In one study, the increase in H3K18ac signal at TssA was described as an 'increase in the average level of H3K18ac and H3K27ac.' [28] In another study, using spike-ins, it was 'observed that the H3K27ac mark was reduced at [TssA] whereas H3K18ac was slightly increased.' [17] Yet another recent report observed that 'surprisingly, only 226 [of 807] downregulated genes exhibited hypoacetylation of H3K27 under the condition of CBP/p300 HAT inhibition.'…”

Potential roles of super enhancers in inflammatory gene transcription

“…In general, enhancer regions can directly activate transcription via acting on the associated promoters and are crucial for cell-type-specific gene transcription [2,3]. Mechanistically, pause-release (i.e., productive elongation) of RNA polymerase II (pol II) has been considered as a major point of enhancer-mediated transcriptional activation, albeit a series of very recent studies has implicated the functional importance of enhancer in recruitment and transcriptional initiation of pol II [4][5][6]. Current evidence has shown that enhancer regions are associated with DNase I hypersensitivity (i.e., open chromatin), binding of transcription factors, recruitment of coregulators (e.g., Mediator, acetyltransferase CBP/ p300), epigenetic modifications [e.g., H3K4me1/2 (primed enhancers), H3K27ac (active enhancer)], and chromatin architecture, which contribute to the transcription of enhancer RNAs (eRNAs) [7,8] (Fig.…”