Genome-wide dynamics of Pol II elongation and its interplay with promoter proximal pausing, chromatin, and exons
- Cornell University, United States
- University of Michigan, United States
Figures
Figure 1 with 2 supplements
Timed inhibition of pause escape (P-TEFb) or initiation (TFIIH) has similar effects on the gene body Pol II distribution, but opposite effect at the promoter-proximal region.
(A) Experimental set-up, 300 nM flavopiridol (FP) and 500 nM of triptolide (Trp) were used to block pause escape or transcription initiation in mES cells. Nuclei were isolated for GRO-seq at …
-
Figure 1—source data 1
Sequencing and alignment of GRO-seq replicates in the Trp and FP time courses.
The top tables depict the total reads sequenced, and the alignment to the spike-in controls, ribosomal, reference genomes for each replicate. One of the 25 min Trp replicates was of poor quality due to RNA degradation during the library preparation and wasn’t included in further analysis.
- https://doi.org/10.7554/eLife.02407.006
-
Figure 1—source data 2
Pearson correlation of GRO-seq replicates in the Trp and FP time courses.
Pearson correlations between the replicates for each time course after drug treatment in either the gene body or promoter regions.
- https://doi.org/10.7554/eLife.02407.007
-
Figure 1—source data 3
Spearman correlation of GRO-seq replicates in the Trp and FP time courses.
Spearman correlations between the replicates for each time course after drug treatment in either the gene body or promoter regions.
- https://doi.org/10.7554/eLife.02407.008
Figure 1—figure supplement 1
Inhibition of the P-TEFb kinase by Flavopiridol.
(A) Western blots of chromatin bound or unbound protein fractions of mES cells after 50 min of Trp, FP or DMSO (no Trp) treatment or no treatment (no FP), with antibodies against Ser2-, and …
Figure 1—figure supplement 2
Stability of Pol II elongation after the inhibition of initiation or pause escape.
(A) Composite profile of Pol II GRO-seq density from 230 to 240 kb downstream of the TSS in genes larger than 250 kb after treatment with Trp for 0, 12.5, 25 or 50 min shows that levels downstream …
Figure 2 with 1 supplement
Pol II distribution and change at the TSS region of active genes after pause escape (P-TEFb) or initiation (TFIIH) inhibition.
(A) Left panel shows a density plot of the log10 of reads of the no Trp dataset, in 10 bp windows ±2 kb around TSSs of active genes. The two right panels show the log10 difference in 10 bp windows …
-
Figure 2—source data 1
Change in promoter proximal and gene body read density, and pausing index after treatment with FP or Trp in time.
The gene body change was measured in the region from 1 to 3.5 kb downstream of the TSS for each timepoint, and was called significantly increased when the read density ratio was greater then 1 with a Fisher Exact p-value <0.05, or significantly decreased when the read density ratio was smaller then 1 with a Fisher Exact p-value <0.05. The same for the pause peak increase and decrease, but using a 250 bp region with maximal read density in a window of ±500 bp around the annotated TSS. The divergent peak was found by using a 250 bp region with maximal read density in a window of −1000 to +500 bp around the annotated TSS on the antisense strand, and change was defined similarly as the gene body change. Pausing indexes (PI) are the ratio between gene body and promoter proximal Pol II density, and the change in PI has therefore no Fisher Exact p-value associated with it. PI’s are increased or decreased when the ratio between untreated and treated PI’s is greater or smaller then 1, respectively.
- https://doi.org/10.7554/eLife.02407.011
Figure 2—figure supplement 1
Differential dynamics of paused Pol II upon the inhibitions of initiation or the pause escape.
(A) Scheme of selection process of active, long and well-resolved genes. (B) Log10 of GRO-seq read count in 10 bp bins ± 2 kb around the TSSs of active genes (n = 6380) before and after 50 min …
Figure 3 with 3 supplements
Genome-wide determination of Pol II elongation rates after FP and Trp treatment.
(A) Three representative genes used for measurement of transition points with the HMM. In green, the affected region after FP treatment as established by the HMM, the transition point being the …
-
Figure 3—source data 1
Full list of elongation rates derived after FP treatments between times spanning 5–12.5 min, 12.5–25 min, and 25–50 min.
- https://doi.org/10.7554/eLife.02407.013
Figure 3—figure supplement 1
Determination of elongation rates using the Hidden Markov Model of inhibition time-points.
(A) Scheme of the Hidden Markov Model (HMM), in which each hidden state follows a binomial distribution B of 20 observed states n of the read ratio. We assumed two hidden states as ‘inhibition …
Figure 3—figure supplement 2
Determination of elongation rates using the intron RNA-seq gradients.
(A) PolyA(−) RNA-seq pattern in genes with long introns display a ‘saw-tooth’ pattern as shown by Ameur et al. (2011) and represented here for a typical gene (AUTS2). (B) Intronic reads of poly(A)- …
Figure 3—figure supplement 3
Acceleration of Pol II and its accompanying CTD phosphorylation.
(A) Elongation rates of genes from which more than one elongation rate could be determined. p-values are calculated using Wilcox test. (B) Acceleration constant calculated in genes that have at …
Figure 4 with 1 supplement
Elongation rate explains transcriptional stability and the steady-state output.
(A) Models of Pol II elongation (left), and the expected steady-state GRO-seq density (D) and the inverse elongation rates (v−1) for each model (right). The termination model proposes that the …
Figure 4—figure supplement 1
Monte-Carlo modeling of the acceleration and the termination hypotheses.
(A) Simulation time-course of change in Pol II distribution after a block of Pol II entry into the gene body assuming an intrinsic acceleration model (top) and a termination model (bottom). See …
Figure 5 with 1 supplement
Genome-wide features that associate with the elongation rates.
(A) Profile of the 25 min FP GRO-seq data in mid elongation rate genes (n = 938), ordered by increasing rate (left panel). Right panels depict genomic features in the same gene order (but more …
-
Figure 5—source data 1
ChIP-seq datasets used in this study.
Complete table of all the ChIP-seq data and references taken into account in the correlation study between mid elongation rates (n = 938) and promoter or gene body occupancy of each of the listed factors. The references to the GEO datasets are shown in the “Major datasets” section.
- https://doi.org/10.7554/eLife.02407.019
Figure 5—figure supplement 1
Effects of exon density and CpG on elongation rates.
(A) Elongation rates plotted against exon density within the transition region for early elongation rates (5–12.5 min; n = 44). (B) As in (A) for late elongation rates (25-50 min; n = 199). Linear …
Figure 6 with 1 supplement
Variations in elongation rates are largely explained by exon density, H3K79me2 and CpG content.
(A) Most prominent correlations between mid elongation rates (n = 938) and various features of mid rate genes after transformation to a rank-order based z-score assuming a Gaussian distribution. (B) …
Figure 6—figure supplement 1
Comprehensive linear models of the elongation rates using multiple genomic features.
(A) Correlation plots of the most highly correlated genomic features and the elongation rates or the residuals of the elongation rates after fitting to exon density; exon density and H3K79me2; exon …
Author response image 1
LTR.
A. The decay rate plots of the three active genes identified by Wagschal et al. to be regulated at the level of termination of Pol II at the promoter or within the gene body in Hela cells. B. The …
Author response image 2
TFIIH sensitivity.
The genes assayed by ChIP-qPCR of Ser5 phosphorylated Pol II in Figure 5F of Tee et al. Cell 2014 were examined. In the top box, two Trp sensitive control genes taken from this figure are shown, and …
