Human promoter directionality is determined by transcriptional initiation and the opposing activities of INTS11 and CDK9

RNA polymerase II (RNAPII) transcription initiates bidirectionally at many human protein-coding genes. Sense transcription usually dominates and leads to messenger RNA production, whereas antisense transcription rapidly terminates. The basis for this directionality is not fully understood. Here, we show that sense transcriptional initiation is more efficient than in the antisense direction, which establishes initial promoter directionality. After transcription begins, the opposing functions of the endonucleolytic subunit of Integrator, INTS11, and cyclin-dependent kinase 9 (CDK9) maintain directionality. Specifically, INTS11 terminates antisense transcription, whereas sense transcription is protected from INTS11-dependent attenuation by CDK9 activity. Strikingly, INTS11 attenuates transcription in both directions upon CDK9 inhibition, and the engineered recruitment of CDK9 desensitises transcription to INTS11. Therefore, the preferential initiation of sense transcription and the opposing activities of CDK9 and INTS11 explain mammalian promoter directionality.


INTRODUCTION
In humans, most protein-coding genes initiate transcription bidirectionally.Usually, only sense transcription efficiently elongates and leads to mRNA synthesis.Antisense transcription terminates within a few kilobases (kb), and the short non-coding (nc)RNA is degraded 1 .Similar asymmetry frequently occurs in unicellular eukaryotes (e.g., budding yeast) and at some plant promoters 2,3 .Thus, promoter directionality is a broadly observed phenomenon.Interestingly, bidirectionality is the ground state of promoters and directionality is acquired over evolutionary time 4 .This is proposed to be through a combination of DNA sequences and proteins that favour the directional initiation and elongation of transcription.The present explanation for the directionality of mammalian RNAPII promoters involves the arrangement of U1 snRNA binding sites and polyadenylation signals (PASs).U1 snRNA promotes RNAPII elongation through protein-coding genes by binding to RNA and preventing early termination [5][6][7] .RNA-bound U1 prevents early termination by inhibiting PASs and antagonising other attenuation mechanisms, which include the PP1 regulator, PNUTS, and the Restrictor complex 8,9 .In contrast, U1 binding sites are rarer in short antisense transcripts, which are often rich in PAS sequences that are proposed to promote transcriptional termination 10 .This model predicts that polyadenylation factors control a large fraction of antisense transcriptional termination.The multi-subunit Integrator complex also regulates promoter-proximal transcription 11- 15 .The Integrator complex comprises the backbone, arm/tail, phosphatase, and endonuclease modules 11 .Its endoribonuclease is INTS11, and its phosphatase activity is mediated by INTS6 and protein phosphatase 2A (PP2A).INTS11 endonuclease broadly affects promoterproximal transcriptional attenuation whereas INTS phosphatase is proposed to regulate the escape of RNAPII into elongation 16,17 .INTS6/PP2A phosphatase functionally antagonises CDK9, which is vital for RNAPII promoter escape and productive elongation 18 .By this model, CDK9 activity promotes elongation across protein-coding genes and INTS6/PP2A opposes it.
We tested the prediction that PAS factors control transcription directionality by terminating antisense transcription, but this is not usually the case.Instead, we find that promoter directionality is often conferred by preferential initiation in the sense direction and is thereafter maintained by INTS11 and CDK9.The termination of antisense transcription is constitutively INTS11-dependent, whereas sense transcription is hypersensitive to INTS11 only when CDK9 activity is simultaneously inhibited.We hypothesise that CDK9 activity protects sense transcription from attenuation by INTS11 and that reduced CDK9 activity in the antisense direction exposes RNAPII to INTS11-dependent termination.

Polyadenylation factor depletion does not increase or extend antisense transcription
The current explanation of mammalian promoter directionality invokes early PASdependent termination of antisense transcription 10 .This is partly based on the direct detection of polyadenylated antisense transcripts, which provides evidence that some of their transcriptional termination is PAS-dependent.However, there are non-polyadenylated antisense RNAs that might be attenuated in other ways 19 .To test the contribution of PASdependent termination toward antisense transcriptional termination and promoter directionality we tagged RBBP6 with the dTAG degron.Because RBBP6 is required to activate the PAS cleavage machinery 20,21 , its depletion should inhibit any PAS-dependent transcriptional termination.Three homozygous dTAG-RBBP6 clones were isolated (Supplemental Figure 1A) and tagged RBBP6 was efficiently depleted after exposure to the dTAGv-1 degrader (Figure 1A).To test the contribution of RBBP6 to nascent transcription, we used POINT (Polymerase Intact Nascent Transcript)-seq 22 , which maps full-length RNA extracted from immunoprecipitated RNAPII.
Figure 1B shows POINT-seq coverage over NEDD1 including the upstream antisense transcript.RBBP6 loss causes a termination defect beyond the PAS shown by the extended POINT-seq signal beyond the NEDD1 gene.However, RBBP6 does not affect the termination of upstream antisense RNA which is not extended when RBBP6 is depleted.Because RBBP6 loss induces strong termination defects beyond the PAS, we generalised its effects using genes separated from their neighbours by ≥10kb.Metaplot and heatmap analyses across these 1316 genes confirm that RBBP6 loss causes a general termination defect downstream of protein-coding genes but has little impact on antisense transcription (Figures 1C and D).
Although many antisense transcripts contain multiple AAUAAA sequences, most still terminate RBBP6-independently (Supplementary Figures 1B and C).Similarly, we recently showed that the PAS-dependent 5'3' exonucleolytic torpedo terminator, XRN2, does not affect antisense transcriptional termination 9 .Therefore, although some antisense transcripts are polyadenylated, most antisense transcription can terminate using PAS-independent mechanisms.Consequently, PAS-dependent termination cannot fully explain the promoterproximal attenuation of antisense transcription.

Integrator depletion increases and extends antisense transcription
Our data argue that PAS-independent termination mechanisms control a large fraction of antisense transcription.A major PAS-independent termination pathway is driven by the Integrator complex, which was first identified as the 3' end processing complex for snRNAs 23 .
Several reports show that Integrator terminates transcription from most promoters, including those that initiate antisense transcription, so it might affect promoter directionality 11,16,17 .To analyse this, we tagged its endonucleolytic subunit (INTS11) with a dTAG degron 24 , which enables rapid depletion of its catalytic activity (Figure 2A).We then performed POINT-seq on INTS11-dTAG cells depleted or not of INTS11 to assay the global impact of INTS11 on RNAPII transcription.Integrator is established to control the transcription of snRNAs, and this was detected in our POINT-seq, which demonstrates a clear extension of read density at RNU5A-1 and RNU5B-1 loci (Figure 2B).
As exemplified by PGK1 (Figure 2C), INTS11 loss does not affect the termination of protein-coding transcription but causes a strong upregulation of antisense transcription.Metaanalysis of the same gene set used for RBBP6 POINT-seq shows the generality of antisense transcriptional attenuation via INTS11 (Figure 2D).A heatmap analysis of transcription 3kb upstream and downstream of the same transcription start sites (TSSs) demonstrates the dominant impact of INTS11 on antisense vs. sense transcription at most of these promoters (Figure 2E).Although the transcription over protein-coding genes is less affected by INTS11 loss, an increase in promoter-proximal transcription is apparent in some cases (Supplemental Figure 2A).The most strongly affected genes are lowly expressed, which is consistent with recent findings (Supplemental Figure 2B) [12][13][14][15][16][17] .Overall, INTS11 frequently attenuates antisense transcription whereas a smaller fraction of sense transcription is affected.RNAPII.
The hypersensitivity of antisense transcripts to INTS11 might be due to the makeup of antisense RNAPII or some other promoter feature.To interrogate this further, we analysed two other promoter classes: those where protein-coding transcripts are initiated in both directions and those that initiate the bidirectional transcription of unstable enhancer (e)RNAs.
In the former case, both transcripts are extended and stable like most sense transcripts at directional promoters.In the latter case, both are short and unstable like most antisense transcripts at directional promoters.When both directions are protein-coding, INTS11 depletion causes modest reductions in transcription (Supplemental Figure 2C).Conversely, eRNA transcription is bidirectionally upregulated upon INTS11 elimination like the antisense transcripts in (Supplemental Figure 2D).We conclude that short ncRNAs are more strongly affected by INTS11 than protein-coding transcripts.At directional promoters this results in the attenuation of antisense transcription.

Transcription initiates more efficiently in the sense direction
The increased antisense POINT-seq signal following INTS11 loss is consistent with defects in transcriptional termination.However, it could result from increased transcriptional initiation.We were also interested in whether preferential sense initiation could contribute to mammalian promoter directionality as was proposed in budding yeast 4 .To precisely resolve directional aspects of initiation and assay any impact of INTS11, we devised a variant of POINT-seq called short (s)POINT.Briefly, sPOINT follows the POINT-seq protocol, but library preparation employs the selective amplification of 5' capped RNAs <150nts (Figure 3A, Supplemental Figure 3A, Experimental Procedures).The sPOINT signal over PGK1 exemplifies this and demonstrates a very restricted signal close to the TSS (Figure 3B). Figure 3C compares the meta profile of POINT-and sPOINT-seq on 684 well-expressed and well-spaced (≥10kb from neighbours) genes to highlight the full read coverage obtained by POINT-seq and the tightly restricted, TSS-proximal, sPOINT signal.Because most capped RNAPII-associated RNA <150nts are promoter-proximal, sPOINT effectively assays the promoter-proximal RNAPII pause.
The sPOINT-seq metaplot in Figure 3B shows a higher signal in the sense direction, suggesting that more efficient transcriptional initiation establishes promoter directionality.To assay this further and test the impact of INTS11, we performed sPOINT-seq in INTS11-dTAG cells treated or not with dTAGv-1 and plotted the coverage over the promoters of wellexpressed protein-coding genes (3060 promoters, Figure 3D).As sPOINT-seq maps the 5' and 3' ends of these reads, precise TSSs are mapped at single-nucleotide resolution and plotted in the lower meta profile in Figure 3D.This analysis once again shows a higher sPOINT signal in the sense direction.In addition, the lower TSS mapped plot also shows that sense transcription initiated in a more focused manner compared to in the antisense direction.Figure 3E (PGK1) and supplemental figure 3B (ACTB) exemplify these features on individual genes, and the metaplot in Figure 3F highlights the dispersed nature of antisense TSSs.Quantitation of the TSS-derived sense vs. antisense signal confirms a higher read count in the sense direction in untreated cells (Supplemental Figure 3C).The most noticeable (although still modest) effect of INTS11 loss on sPOINT-seq profiles is a mild signal reduction.This shows that INTS11 depletion does not generally enhance transcriptional initiation in either direction.
This slight reduction in sPOINT-seq signal could result from less transcriptional initiation or less efficient pausing if, for example, INTS11 loss allows more RNAPII to escape the promoter.
Although a lower resolution technique, RNAPII ChIP-seq confirmed these promoter characteristics and the mild impact of INTS11 (Supplemental Figures 3D and E).Overall, these data show that directionality might be established by more efficient initiation in the sense direction.

CDK9 inhibition sensitises sense transcription to INTS11
After transcription initiates, further control of directionality is evident because sense transcription goes further than antisense transcription.If INTS11 regulates this, an opposing force is required to promote elongation.INTS11 affects transcription very early, occupies promoters, and becomes less active as RNAPII moves into elongation [16][17][18]25,26 . Thereore, if INTS11 is counteracted to allow sense transcription, any responsible mechanism needs to act early.One of the first transcriptional checkpoints involves the phosphorylation of RNAPII and other factors by CDK9, which releases promoter-proximally paused RNAPII into elongation 27 .
During this process, the Integrator-associated phosphatase, PP2A, antagonizes CDK9 and presumably regulates the sensitivity of RNAPII to INTS11 18 .Because Integrator phosphatase has little effect on antisense transcription 16 , we hypothesised that INTS11 sensitivity and CDK9 activity are inversely correlated to maintain directionality after initiation.
Our hypothesis predicts that sense transcription will be attenuated via INTS11 when CDK9 is inactive.To test this genome-wide, we depleted INTS11 from INTS11-dTAG cells in the presence or absence of a specific CDK9 inhibitor (NVP-2 28 ) and performed POINT-seq.
As exemplified by TARDBP, the depletion of INTS11 alone caused an antisense transcriptional termination defect with a milder impact on the protein-coding sense direction (Figure 4A).As expected, NVP-2 treatment reduced transcription over the protein-coding gene body.Antisense transcription also displays some CDK9 sensitivity.Importantly, in NVP-2-treated cells, INTS11 loss increased transcription in both directions.This contrasts with the dominant antisense effect deriving from just depleting INTS11 (refer to Figure 2).As such, CDK9 activity prevents sense transcription from being attenuated by INTS11.This is a genome-wide trend as shown in the metaplots in Figures 4B and C. A heatmap of the effect of INTS11 loss after CDK9 inhibition shows the bidirectional upregulation of transcription at the same promoters assayed in Figure 2 (Figure 4D).Although this experiment employed longer INTS11 depletion than that in Figure 2 (2.5hr vs. 1.5hr to allow concurrent CDK9 inhibition), its antisense effects remain general, and the affected protein-coding genes strongly overlapped (Supplemental Figure 4A-C).Lastly, when CDK9 and INTS11 are both compromised, the POINT-seq signal remains higher in the sense vs. antisense direction (black line, Figure 4C).This is consistent with our sPOINT-based finding that initiation of sense transcription is generally more efficient.
Following CDK9 inhibition and INTS11 loss, the largest recovery of protein-coding transcription is often over the 5' end of genes.This presumably reflects poor elongation without CDK9 activity even when INTS11 is absent.RNAPII elongation is associated with phosphorylation of the C-terminal domain (CTD) of its largest subunit, RPB1, and most frequently occurs on Serine 5 or 2 (Ser5/2p) of its heptad repeat 29,30 .We assayed the effects of CDK9 inhibition and INTS11 depletion on these two modifications by western blotting (Figure 4E).As previously shown 28  We confirmed the CDK9-mediated suppression of INTS11 on four selected proteincoding genes using an alternative inhibitor, 5, 6-dichloro-1-β-D-ribofuranosylbenzimidazole (DRB) (Supplemental Figure 4D).Interestingly, inhibition of CDK7 with THZ2 sensitised the same selected protein-coding transcripts to INTS11 attenuation (Supplemental Figure 4E).
This may be because CDK7 is required to activate CDK9 31 , or because Integrator additionally targets RNAPII when CDK7 is inactive.This result reinforces the idea that the successful execution of early transcriptional checkpoints may overcome Integrator-mediated attenuation.

DISCUSSION
Most studies on mammalian promoter directionality have focused on understanding how antisense transcription terminates and how RNAPII elongation is enabled in the sense direction.However, we show that preferential sense transcriptional initiation significantly contributes toward the directionality of many mammalian promoters.More analyses are required to elucidate the mechanism, but we hypothesise that a unidirectional arrangement of at least some core promoter elements favours transcription in the protein-coding direction.
Consistently, the evolution of promoter elements explains how a bidirectional promoter "ground state" acquired directionality in yeast 4 .Consistent with our finding, focused transcription initiation is associated with highly expressed mammalian genes and promoters with clearly defined elements (e.g.TATA) 34 .Thus, the lower level and dispersed nature of antisense initiation may reflect the suboptimal orientation of promoter elements or the opportunistic transcription of open chromatin in promoter regions.
We hypothesise that INTS11 controls transcriptional attenuation via its endonuclease activity.An inactive mutant of INTS11 (E203Q) is widely used to interrogate its catalytic function 23 ; however, we discovered that it poorly associates with other Integrator components compared to wild-type INTS11 (Supplemental Figure 4G).Therefore, when employed in cells, it may not isolate the effects of INTS11 activity from those requiring an intact Integrator complex.Similarly, catalytic mutations in the highly related PAS endonuclease, CPSF73, disrupt its association with other cleavage and polyadenylation components 35 .In light of this, it is formally possible that non-catalytic consequences of INTS11 loss explain the upregulation of antisense transcription.
How CDK9 activity opposes INTS11 is unresolved, but INTS11 and SPT5 are adjacent in the RNAPII: Integrator structure 36 .As SPT5 is a prominent substrate of CDK9 [37][38][39] , its phosphorylation might evict INTS11 or prevent its association with the complex.Consistently, Integrator is enriched on promoters whereas phosphorylated SPT5 (SPT5p) is most prevalent during elongation over protein-coding gene bodies 25,38 .Furthermore, although antisense RNAPII occupancy is increased by INTS11 loss, published data suggest that this is not paralleled by increases in its Ser5p and Ser2p forms 16 .As these modifications require CDK7 and CDK9, this provides further support that these activities are less prevalent for antisense vs. sense transcription.
While INTS11 and CDK9 activities are the clearest on antisense and sense transcription, respectively, this is not binary.Some antisense transcription is NVP-2 sensitive, which indicates CDK9 activity on some RNAPII (Figure 4 and 40 ).CDK9-sensitive antisense transcription could undergo INTS11-independent attenuation, which might account for the polyadenylated antisense RNA observed previously 10,41 .Consistently, some antisense transcription is targeted by the poly(A) exosome targeting connection, which operates on cleaved and polyadenylated transcripts 42,43 .Reciprocally, promoter-proximal transcription from lowly expressed protein-coding transcription is INTS11-sensitive, which could reflect low CDK9 activity at those loci.We and others recently described the Restrictor complex as restraining antisense transcription 9,44,45 .Unlike Integrator, which associates with NELF-bound RNAPII 36 , Restrictor is proximal to the distinct PAF-bound RNAPII 9 .Thus, various complexes may target different forms of RNAPII before and after CDK9 activity.Even so, our data suggests that Integrator influences a greater volume of antisense RNAPII than the cleavage and polyadenylation pathway.
In addition to CDK9, multiple studies demonstrate a role for U1 snRNA in promoting elongation through protein-coding genes 5,6,8,46 .U1 promotes RNAPII elongation and shields transcription from early PAS-dependent termination and the Restrictor complex 5,9 .The U1mediated suppression of premature PAS-dependent termination inspired the original model of promoter directionality.As U1 sites are rarer in antisense RNAs, their termination was hypothesised to be driven by early PASs that would consequently be active.Our RBBP6 data (and other published data on PAS termination factors 9 ) argues that a large fraction of antisense transcriptional termination is PAS-independent.Nevertheless, the lack of U1 sites may limit RNAPII elongation and render it sensitive to Integrator (or Restrictor).Accordingly, it will be interesting to test whether U1 regulates the sensitivity of transcription to Integrator.
In conclusion, we provide a new model for mammalian promoter directionality and the subsequent control of bidirectional transcription (Figure 4G).sPOINT demonstrates more efficient sense transcription initiation, which establishes directionality.Early events dictate the decision to elongate or attenuate transcription.In the sense direction, CDK9 activity prevents attenuation by Integrator to favour productive elongation, whereas antisense transcription is hypersensitive to INTS11 and terminates early.As INTS11 occupies most promoters and CDK9 activity is near-universally required to achieve protein-coding transcription, this model can generally explain how directionality is initiated and maintained.Our new sPOINT-seq approach will be valuable in elucidating additional aspects of promoter-proximal transcriptional regulation.

POINT-seq and sPOINT-seq
For POINT-seq, we followed the protocol provided in 22 .The only modification was that we started with a confluent 10cm dish of cells and performed the immunoprecipitation with 6µg of anti-RNAPII.~2% cell volume of Drosophila S2 cells was included as a spike in control.
Libraries were prepared using the NEBNext Ultra™ II Directional RNA Library Prep Kit for Illumina (New England Biolabs).sPOINT was performed in the same manner with the following difference: a confluent 15cm dish of cells was used and 10µg of anti-RNAPII.In sPOINT, the immunoprecipitated and DNAse treated RNA was treated with Terminator™ 5-Phosphate-Dependent Exonuclease (lucigen) to remove any uncapped transcripts, and libraries were prepared with the SMARTer® smRNA-Seq Kit for Illumina (Takara Bio) to selectively capture transcripts <150nts.

ChIP-seq
For each experiment, 1x10cm dish of cells was used.Protein: DNA crosslinks were formed by adding Formaldehyde (1% v/v) to culture media for 10 mins then quenching with 125mM glycine.Cells were rinsed 2x with PBS, scraped off the dish, and pelleted in 10ml PBS at 500xg for 5 mins.We then employed the simple ChIP enzymatic kit (Cell Signalling Technologies) to fragment chromatin and purify RNAPII-bound DNA.We followed the kit protocol except for conjugating 5µg of anti-total RNAPII to sheep anti-mouse dynabeads (Life Technologies).Sequencing libraries were generated using the NEBNext® Ultra™ II DNA Library Prep Kit for Illumina.
Total/chromatin-associated RNA isolation and qRT-PCR For total RNA, a 24-well dish of cells was transfected with 100ng HIV reporter plasmid and, where co-transfected, 50ng TAT plasmid 52 using Jetprime (Polyplus) following the manufacturer's protocol.Media was refreshed 5 hours post-transfection and dTAGv-1 was added where appropriate.The next day, total RNA was isolated using Trizol (Thermo Fisher) following the manufacturer's protocol.For chromatin-associated RNA (for Supplemental Figure 4F), a 6 well dish of cells was transfected with 300ng HIV reporter plasmid and 150ng TAT. Pelleted cells were resuspended in 800µl hypotonic lysis buffer (HLB: 10 mM Tris-HCl (pH 7.5), 10 mM NaCl and 2.5 mM MgCl2, 0.5% NP40) and underlayered with 200µl HLB +10% sucrose.Nuclei were isolated by centrifugation for 5 mins at 500xg.These were resuspended in 100µl NUN1 (20 mM Tris-HCl (pH 7.9), 75 mM NaCl, 0.5 mM EDTA, and 50% Glycerol).After addition of 1ml of NUN2 (20 mM HEPES-KOH (pH 7.6), 300 mM NaCl, 0.2 mM EDTA, 7.5 mM MgCl2, 1% NP-40, 1 M Urea) chromatin was isolated by 10 min incubation on ice followed by 10 min centrifugation at 13000rpm.RNA was isolated from the chromatin pellet using Trizol following the manufacturers' instructions.For qRT-PCR of total and chromatin-associated RNA, RNA was DNase treated then 1µg was reverse transcribed using Protoscript II reverse transcriptase (New England Biolabs).qPCR was performed using LUNA SYBR green reagent (New England Biolabs) on a Qiagen Rotorgene instrument.Quantitative analysis used the ΔCT method.
Debris was pelleted at 13000rpm for 10 mins and supernatant was incubated with 20µl of antiflag magnetic beads (Sigma) for 2hrs at 4 o C. Beads were washed 6x with ice-cold chromatin digestion buffer and samples were eluted by boiling in protein loading buffer (100 mM Tris-Cl (pH 6.8) 4% (w/v) sodium dodecyl sulfate 0.2% (w/v) bromophenol blue 20% (v/v) glycerol) before polyacrylamide separation and western blotting.

qRT-PCR primers and gRNA target sites
See supplemental table 1.

BIOINFORMATICS POINT-Seq alignment and visualisation
A metagene list of genes with no overlapping regions within 10kb of any other expressed transcription unit (1316).Adapters were removed from raw reads using Trim Galore! and mapped to GRCh38 using HISAT2 using default parameters.Biological replicates were normalised and merged using SAMtools merge.Split strand metagene plots were produced using Drosophila spike-in normalised sense and antisense (scaled to -1) bigwig coverage files separately with further graphical processing performed in R. For heat maps, computematrix (DeepTools) was used to generate score files from the normalised bigwig files using the 10kb non-overlapping gene list.A log2 ratio (depletion/control) was applied to identify changes in reads.Plots were redrawn in R; parameters used for each heat map are detailed in figure legends.
sPOINT For sPOINT TSS metaplots showing full-length and 5' derived coverage, gene lists were determined by selecting principal protein-coding transcript isoforms from gencode v42 human annotation -specifically, those containing both "appris_principal_1" and "Ensembl_canonical" labels (15301 in total).The top 20% expressed (based on -dTAG) were used to generate meta profiles (3060 TSSs in total).The 684 genes used to exemplify the difference between POINT and sPOINT coverage are the top 50% of expressed genes (based on POINT-seq in INTS11-dTAG) cells, which are separated from neighbouring transcription usints by at least 10kb.

Antisense PAS analysis
Bed files from Figure 1C were edited to map the antisense transcript from each gene by fitting each gene's start and end point to 3kb upstream of their TSS.Sequences underlying these regions were obtained via BEDTools getFASTA (hg38) and consensus PAS (AATAAA) enumerated in R. PAS frequency per transcript was plotted using ggplot2.Heatmaps for transcripts containing > 1 or no PAS motifs were plotted with DeepTools.

ChIP-seq alignment and plotting
Adapters were removed from raw reads using Trim Galore! and mapped to GRCh38 using HISAT2 with default parameters.Reads were also mapped to Dm6 to identify spike-in signal.Reads with MAPQ score of ≤30 were removed with SAMtools 54 .Peaks were called using MACS2 in paired-end mode 55 .Metaplots were created from S2 scaled, merged replicate -log10 q-value bigwig coverage using DeepTools with further processing performed within R.
For Supplementary Figure 3E, plots used the same gene list described above for sPOINT promoter analyses (see sPOINT section above).C. Metaplot of POINT-seq data from dTAG-RBBP6 cells treated (RBBP6 dep) or not (Ctrl) (2hr) with dTAGv-1.This shows 1316 protein-coding genes selected as separated from any expressed transcription unit by ≥10kb.Signals above and below the x-axis are sense and antisense reads, respectively.The y-axis scale is RPKM.

Key Resources
TSS=transcription start site; TES=transcription end site (this marks the PAS position).
This is an average of two biological replicates.
D. Heatmap representation of the data in C, which displays signal as a log2 fold change (log2FC) in RBBP6 depleted versus un-depleted conditions.This is an average of two biological replicates.G. Model for promoter directionality depicting higher levels of focused transcriptional initiation in the sense direction together with opposing gradients of CDK9 and INTS11 activity that peak in sense and antisense directions, respectively.

If
CDK9 counteracts INTS11, its recruitment should prevent transcriptional attenuation by Integrator.To assay this, we employed a plasmid where transcription is driven by the human immunodeficiency virus (HIV) promoter.Transcription from the HIV promoter results in synthesis of the trans-activating response (TAR) element and promoter-proximal RNAPII pausing.Pause release requires the trans-activator of transcription (TAT), which promotes RNAPII elongation by recruiting CDK932 .INTS11 suppresses transcription from the HIV promoter when TAT is absent33 .To assay whether CDK9 affects this process, INTS11-dTAG cells were transfected with the HIV reporter with or without TAT before treatment or not with dTAGv-1.Transcription from the reporter was then analysed by qRT-PCR (Figure4F).INTS11 loss induces HIV transcription in the absence of TAT.However, TAT strongly stimulates transcription (~200 fold) and desensitises it to INTS11 loss (slightly less reporter RNA was recovered).Therefore, TAT-mediated CDK9 recruitment alleviates INTS11dependent attenuation of transcription.In line with endogenous protein-coding genes, CDK9 inhibition sensitises TAT-activated transcription to INTS11 (Supplemental Figure4F).Overall, our data strongly suggest that CDK9 activity counteracts transcriptional attenuation by INTS11.

FIGURE 3 :
FIGURE 3: Sense direction transcription initiation is more efficient and focused.

FIGURE 4 :
FIGURE 4: Some promoter directionality is lost when CDK9 is inhibited.