RNA polymerase II stalling promotes nucleosome occlusion and pTEFb recruitment to drive immortalization by Epstein-Barr virus - PubMed (original) (raw)
RNA polymerase II stalling promotes nucleosome occlusion and pTEFb recruitment to drive immortalization by Epstein-Barr virus
Richard D Palermo et al. PLoS Pathog. 2011 Oct.
Abstract
Epstein-Barr virus (EBV) immortalizes resting B-cells and is a key etiologic agent in the development of numerous cancers. The essential EBV-encoded protein EBNA 2 activates the viral C promoter (Cp) producing a message of ~120 kb that is differentially spliced to encode all EBNAs required for immortalization. We have previously shown that EBNA 2-activated transcription is dependent on the activity of the RNA polymerase II (pol II) C-terminal domain (CTD) kinase pTEFb (CDK9/cyclin T1). We now demonstrate that Cp, in contrast to two shorter EBNA 2-activated viral genes (LMP 1 and 2A), displays high levels of promoter-proximally stalled pol II despite being constitutively active. Consistent with pol II stalling, we detect considerable pausing complex (NELF/DSIF) association with Cp. Significantly, we observe substantial Cp-specific pTEFb recruitment that stimulates high-level pol II CTD serine 2 phosphorylation at distal regions (up to +75 kb), promoting elongation. We reveal that Cp-specific pol II accumulation is directed by DNA sequences unfavourable for nucleosome assembly that increase TBP access and pol II recruitment. Stalled pol II then maintains Cp nucleosome depletion. Our data indicate that pTEFb is recruited to Cp by the bromodomain protein Brd4, with polymerase stalling facilitating stable association of pTEFb. The Brd4 inhibitor JQ1 and the pTEFb inhibitors DRB and Flavopiridol significantly reduce Cp, but not LMP1 transcript production indicating that Brd4 and pTEFb are required for Cp transcription. Taken together our data indicate that pol II stalling at Cp promotes transcription of essential immortalizing genes during EBV infection by (i) preventing promoter-proximal nucleosome assembly and ii) necessitating the recruitment of pTEFb thereby maintaining serine 2 CTD phosphorylation at distal regions.
Conflict of interest statement
The authors have declared that no competing interests exist.
Figures
Figure 1. High-level pol II accumulation at Cp and CTD phosphorylation at distal EBV genome regions.
(A) Diagram showing the locations of the amplicons generated by the indicated primer sets (Table S1) at the C promoter and around the circular EBV episome. Numbers indicate the 5′ end of the forward primer relative to the Cp transcription start site in the annotated B95-8 EBV sequence (NC_007605.1). The RBP-Jκ site (grey box) and TATA box (black box) are shown. (B) ChIP using anti-EBNA 2 antibodies. Percentage input signals, after subtraction of no antibody controls, are expressed relative to the highest signal obtained in all or the majority of experiments. Results show the mean +/− standard deviation of a minimum of three independent experiments carried out using at least 2 chromatin batches from Mutu I cells (open bars) and Mutu III cells (black bars). ChIP using anti-phospho serine 2 pol II CTD antibodies (C), anti-phospho serine 5 pol II CTD antibodies (D) and anti-pol II antibodies (E).
Figure 2. pTEFb is recruited to Cp and is required for serine 2 phosphorylation.
Results show the mean +/− standard deviation of a minimum of three independent experiments carried out using at least 2 chromatin batches. Mutu I signals (open bars) are compared to Mutu III signals (black bars) in ChIP using anti-CDK9 antibodies (A) or anti-cyclin T1 antibodies (B). (C) ChIP using anti-phospho serine 2 pol II CTD antibodies in Mutu III cells minus (open bars) or plus 500 µM DRB (black bars). (D) ChIP using anti-pol II antibodies in Mutu III cells −/+ DRB.
Figure 3. DSIF and NELF are recruited to Cp in Mutu III cells.
Results show the mean +/− standard deviation of a minimum of three independent experiments carried out using at least 2 chromatin batches. Mutu I signals (open bars) are compared to Mutu III signals (black bars). ChIP using anti-Spt5 antibodies to detect DSIF (A) or anti-NELF-A antibodies to detect NELF (B). Blue-edged graphs show zoomed-in sections to allow better visualization of downstream primer signals.
Figure 4. Pol II is not paused at the LMP gene loci and little pTEFb is recruited.
(A) Diagram of the LMP gene locus showing the locations of the amplicons generated by the indicated primer sets (1–8). Numbers indicate the 5′ end of the forward primer relative to the starts of the LMP1 or LMP2A mRNA sequences (bent arrows) in the B95-8 annotated EBV sequence (NC_007605). The RBP-Jκ sites (grey boxes) and PU.1 site (black box) are shown. The vertical arrow indicates the position of the LMP1 polyadenylation site. Results show the mean +/− standard deviation of a minimum of two independent experiments using Mutu I (open bars) and Mutu III cell chromatin (black bars). Percentage input signals, after subtraction of no antibody controls, are expressed for comparison purposes relative to the highest signal obtained using Cp-specific primers. ChIP using anti-EBNA 2 antibodies (B), anti-CDK9 antibodies (C), anti-pol II antibodies (D) and anti-NELF-A antibodies (E).
Figure 5. Pol II pausing at Cp prevents nucleosome assembly around the promoter.
ChIP using anti-histone H3 antibodies. Results show the mean +/− standard deviation of a minimum of three independent experiments carried out using at least 2 chromatin batches in Mutu I cells (open bars), Mutu III cells (grey bars) and the PER 253 B95-8 LCL (black bars). Primer sets probed promoter-proximal regions of Cp, LMP2Ap and LMP1p. Percentage input signals, after subtraction of no antibody controls are expressed relative to the highest signal obtained in all or the majority of experiments.
Figure 6. Pol II pausing at Cp is driven by DNA sequences that promote access by TBP.
(A) The probability of nucleosome occupancy (P occupancy) at regions upstream and downstream from the transcription start site (TSS) of Cp (thick black line), LMP1p (thick grey line) and LMP 2Ap (thin black line) predicted using tools available at
http://genie.weizmann.ac.il/software/nucleo\_prediction.html
. (B) ChIP using anti-TBP antibodies analysed using Cp and LMP gene primers (see Figure 4) in Mutu I cells (open bars) and Mutu III cells (black bars). Results show the mean +/− standard deviation of a minimum of three independent experiments carried out using at least 2 chromatin batches. Percentage input signals, after subtraction of no antibody controls, are expressed for comparison purposes relative to the highest signal obtained at Cp.
Figure 7. Brd4 is recruited to Cp through acetylated histones.
Results show the mean +/− standard deviation of a minimum of three independent experiments carried out using at least two chromatin batches from Mutu I cells (open bars) and Mutu III cells (black bars). ChIP using anti-acetyl Histone H3 antibodies (A), anti-acetyl Histone H4 antibodies (B), anti-p300 antibodies (C) and anti-Brd4 antibodies (D).
Figure 8. Brd4 recruits pTEFb to Cp.
Results show the mean +/− standard deviation of a minimum of three independent experiments carried out using at least 2 chromatin batches from Mutu III cells in the absence (open bars) or presence of DRB (black bars). ChIP using anti-EBNA 2 (A and G), anti-acetyl Histone H3 (B and H), anti-acetyl Histone H4 (C and I), anti-Brd 4 (D and J), anti-CDK9 (E) and anti-cyclin T1 antibodies (F). Cp analysis is shown in A–F and LMP1p/LMP 2A analysis in shown in G–J.
Figure 9. Brd4 is required for Cp transcription.
(A) Transcription of the Cp-initiated transcripts EBNA 2 and EBNA 1, but not the LMP1 transcript is inhibited when Brd4 binding to chromatin is blocked in the presence of the Brd4 inhibitor JQ1. Mutu III cells were treated with 50 nM JQ1 or DMSO (control) for 48 hrs and transcript levels determined using specific Q-PCR primers and actin as an endogenous control. Normalised cDNA levels are expressed relative to 48 hr control samples. ChIP using anti-Brd4 (B), anti-CDK9 (C) and anti-cyclin T1 antibodies (D) in Mutu III cells in the absence (black bars) or presence of 50 nM JQ1 for 48 hrs (open bars). Results show the mean +/− standard deviation of two independent experiments.
Figure 10. pTEFb inhibitors selectively reduce Cp transcription.
Mutu III cells were treated with the indicated concentrations of DRB (A) or Flavopiridol (B) for 24 hrs and transcript levels determined using the specific Q-PCR primers indicated and actin as an endogenous control. Normalised cDNA levels are expressed relative to 24 hr control samples. Results show the mean +/− standard deviation of two independent experiments.
References
- Klein E, Kis LL, Klein G. Epstein-Barr virus infection in humans: from harmless to life endangering virus-lymphocyte interactions. Oncogene. 2007;26:1297–1305. -PubMed
Publication types
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Other Literature Sources
Miscellaneous