Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jan 12;168(1-2):135-149.e22.
doi: 10.1016/j.cell.2016.12.020. Epub 2017 Jan 12.

RNA Binding to CBP Stimulates Histone Acetylation and Transcription

Daniel A Bose  1 , Greg Donahue  1 , Danny Reinberg  2 , Ramin Shiekhattar  3 , Roberto Bonasio  1 , Shelley L Berger  4
Affiliations

RNA Binding to CBP Stimulates Histone Acetylation and Transcription

Daniel A Bose et al. Cell. .

Abstract

CBP/p300 are transcription co-activators whose binding is a signature of enhancers, cis-regulatory elements that control patterns of gene expression in multicellular organisms. Active enhancers produce bi-directional enhancer RNAs (eRNAs) and display CBP/p300-dependent histone acetylation. Here, we demonstrate that CBP binds directly to RNAs in vivo and in vitro. RNAs bound to CBP in vivo include a large number of eRNAs. Using steady-state histone acetyltransferase (HAT) assays, we show that an RNA binding region in the HAT domain of CBP-a regulatory motif unique to CBP/p300-allows RNA to stimulate CBP's HAT activity. At enhancers where CBP interacts with eRNAs, stimulation manifests in RNA-dependent changes in the histone acetylation mediated by CBP, such as H3K27ac, and by corresponding changes in gene expression. By interacting directly with CBP, eRNAs contribute to the unique chromatin structure at active enhancers, which, in turn, is required for regulation of target genes.

Keywords: CBP/p300; RNA-binding; acetyltransferase; chromatin modification; eRNA; enhancers; epigenetic enzymes; gene regulation; histone acetylation; non-coding RNA.

PubMed Disclaimer

Figures

Figure 1
Figure 1. CBP interacts with RNA in vivo
A) Native RNA-IP of CBP. Top, RNA immunprecipitated with CBP. Bottom, CBP western blot. B) PAR-CLIP protocol. 4-Thiouridine (4-SU). C) CBP PAR-CLIP required 4-SU: top, autoradiography; bottom, CBP western blot.. D) Quantification of CBP PAR-CLIP. Error bars represent mean +/− s.e.m; n=4. E) CBP PAR-CLIP signal was sensitive to RNAse. 1× RNAse cocktail contained: RNAse A (0.01mU/ul) + RNase T1 (0.4mU/ul). F) Quantification of RNase titration. Error bars represent mean +/− s.e.m; n=4; P-values from two-tailed Student’s t-test: *P< 0.05; **P< 0.01; ***P< 0.001.
Figure 2
Figure 2. CBP-RNAs arise at regions of CBP enrichment
A) Autoradiography of CBP PAR-CLIP and nls-GFP (nGFP) and yeast Gal4 DNA binding domain (Gal4-DBD). Membrane excised for sequencing (Dashed box). B) UCSC genome browser view of CBP-RNA upstream of the Tet2 promoter. Top; GROseq signal (purple=sense, grey=antisense); CBP-RNA (PARalyzer) orange bars; 2 replicates PAR-CLIP reads (orange/purple); CBP peaks (blue bars); 2 replicates CBP reads (light/dark blue). C) Close-up of box from G). Sense (s) and antisense (as). D) Distribution of CBP and CBP-RNA by genome region. E-G) Read density at CBP-RNAs and control RNAs. CBP peaks were randomly downsampled to match the size of the background dataset (5581); F) CBP ChIPseq signal (GSE54453); G) DNaseI hypersensitivity (ENCODE, GSE37074); H) GROseq (GSM1524922). P-values from Mann-Whitney U-test.
Figure 3
Figure 3. CBP binds to eRNAs and directs gene expression from CBP-eRNA enhancers
A) Heatmaps of RNA bound CBP peaks (<3.5kb) show +/− 2.5kb window from centre of CBP peak, reads were binned over 50bp. B) UCSC genome browser views of CBP-eRNAs proximal to Klf6, Sp3, Med13l, YY1, Ccnd1 and Tet2. Colours as in Figure 2B and H3K27ac (green). Sense (s) and antisense (as). C-D) Reads at CBP-RNAs: C) H3K4me3; D) H3K4me1. P-values from Mann-Whitney U-test. E-F) Venn diagrams show CBP ChIPseq peaks with H3K4me1 intersecting H3K27ac for: E) CBP-RNAs; F) CBP-eRNAs. P-values from permutation test with random regions restricted to TSS-40kb. G-H) RT-qPCR of G) CBP-eRNAs (s=sense, as=antisense); and H) Nearest gene. Data show log2 fold change between control adenoviral GFP (Adv-GFP, green) or knockdown adenoviral Cre (Adv-Cre, purple); n=3.
Figure 4
Figure 4. In vitro reconstitution of CBP RNA binding
A) CBP domains and RNA binding prediction (BindN). Non-binding (green) and binding residues (red). Magnified sequence shows predicted RBR in CBP-HATwt. B-C) In vitro pull down of B) eRNA-Klf6s and C) eRNA-Med13ls. s=sense, as=antisense strand. Replicate images in Figure S4C-D. D) Quantification of RNA-pulldown data in B-C. n=3. E) In vitro pull down of eRNA-Klf6s. RNA Input and protein fractions in Figure S4E. F) Quantification of RNA-pulldown data in B-C. n=3. G-H) RNA EMSA of eRNAs using CBP-HATwt. G) eRNA-Mdm2s; H) eRNA-Med13ls. I-J) Competition binding RNA EMSAs. Binding of 2nM 32-P radiolabelled eRNA-Mdm2 to CBP-HATwt (2000nM) was competed with: I) 0–20nM unlabelled eRNA-Mdm2; J) 1nM, 10nM and 20nM un-labeled eRNA-Mdm2 (RNA), dsDNA or ssDNA with the same sequence. K) RNA EMSA using: K) CBP-HATdelta-loop and eRNA-Mdm2; L) CBP-HATmutant-loop and eRNA-Med13l. RNA was titrated with 0–8000nM CBP-HAT. (*) CBP-HATwt (2000nM). M) RBR mediates RNA binding to FL-CBP in vivo. PAR-CLIP for GFP-tagged CBPwt, CBPdelta-loop or CBPmutant-loop in MEFs was followed by RT-qPCR. Control lncRNA Malat-1 was not identified by PARalyzer v1.1. P-values from two-tailed Student’s t-test: *P< 0.05; **P< 0.01; ***P< 0.001.
Figure 5
Figure 5. CBP acetyltransferase activity is stimulated by RNA binding
A) RNA stimulated CBP HAT activity in filter binding assays. 5nM CBP-HATwt was titrated with 0–40nM eRNA-Ccnd1 and eRNA-Klf6 or control (dsDNA/ssDNA with same sequence). Data shows fold change in rate (Vmax/[E], (s−1)) from 0nM RNA. Shaded regions show mean +/− s.e.m (RNA n=3, control n=1). s=sense, as=antisense strand RNA. B) Stimulation of CBP HAT activity required the RBR. 1nM CBP-HATwt or CBP-HATdelta-loop was titrated with 0–40nM eRNA-Mdm2. Shaded regions show mean +/− s.e.m (n=4). C-E) Western blot HAT assay using recombinant nucleosome substrate. 1nM CBP-HATwt was titrated with 0–10nM eRNA-YY1as. RNA stimulated: C) H3K27ac (H3 normalized) and; D) H4K5ac (H4 normalized). E) Western blot for H3K27ac, H4K5ac and H3/H4. Coomassie (bottom panel) shows individual histones. F-G) Steady state filter binding assay. Michaelis-Menten plots for: F) CBP-HATwt; G) CBP-HATmutant-loop. Reactions contained 0nM (green) or 10nM (orange) eRNA-Mdm2. Concentrations of CBP-HAT domain and acetyl-CoA were 10nM and 100uM respectively. H3-1-21 peptide from 0–200uM (n=4). Derived kinetic parameters for Km and Kcat in Figure S5K. H-I) Specificity constant (Kcat/Km(H3-1-21)) for reaction with 0nM or 10nM eRNA-Mdm2. H) CBP-HATwt; I) CBP-HATmutant-loop. J) Mechanism for stimulation of CBP-HAT activity by RNA binding.
Figure 6
Figure 6. Localized eRNA binding can stimulate acetyltransferase activity of CBP in vivo
A) UCSC genome browser of YY1 enhancer and promoter. Colours as in Figure 2B. B-C) Depletion of PAR-CLIP eRNA using antisense olignucleotide (ASO) targeting: B) eRNA-YY1as (purple); and C) eRNA-Ccnd1s (blue); GFP-control (green). RT-qPCR shows fold-change in eRNA and associated mRNA at target (top) or control gene (bottom). Error bars represent mean +/− s.e.m; n=4. D-F) ChIP-qPCR following depletion of eRNA-YY1 (purple) or GFP-control (green). Foldchange (IP/H3) for: D) H3K27ac; E) H3K18ac and F) CBP (IP/Input) at YY1 and control gene Ccnd1 (bottom). Error bars represent mean +/− s.e.m; n=4. G-I) ChIP-qPCR following depletion of eRNA-Ccnd1 (blue) or GFP-control (green). Foldchange (IP/H3) for: G) H3K27ac; H) H3K18ac and I) CBP (IP/Input) at Ccnd1 and control gene YY1 (bottom). Error bars represent mean +/− s.e.m; n=4. P-values from two-tailed Student’s t-test: *P< 0.05; **P< 0.01; ***P< 0.001.
Figure 7
Figure 7. Model for RNA stimulation of CBP HAT activity during enhancer activation
i) At inactive enhancers, CBP activity is limited by the activation loop, which occupies the active site. ii) During activation, recruitment of PolII by bound TFs results in eRNA transcription. iii) eRNAs bind to the CBP HAT domain RBR, displacing the activation loop, stimulating the HAT activity of CBP and increasing H3K27ac at the enhancer and associated promoter.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Bailey TL, Elkan C. Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol. 1994;2:28–36. - PubMed
    1. Barlev NA, Liu L, Chehab NH, Mansfield K, Harris KG, Halazonetis TD, Berger SL. Acetylation of p53 activates transcription through recruitment of coactivators/histone acetyltransferases. Mol Cell. 2001;8:1243–1254. - PubMed
    1. Bedford DC, Kasper LH, Fukuyama T, Brindle PK. Target gene context influences the transcriptional requirement for the KAT3 family of CBP and p300 histone acetyltransferases. Epigenetics. 2010;5:9–15. - PMC - PubMed
    1. Beltran M, Yates CM, Skalska L, Dawson M, Reis FP, Viiri K, Fisher CL, Sibley CR, Foster BM, Bartke T, et al. The interaction of PRC2 with RNA or chromatin is mutually antagonistic. Genome Res. 2016;26:896–907. - PMC - PubMed
    1. Berndsen CE, Denu JM. Assays for mechanistic investigations of protein/histone acetyltransferases. Methods. 2005;36:321–331. - PubMed