H3K4me3 regulates RNA polymerase II promoter-proximal pause-release – Nature

mES cell culture and differentiation

E14 ES cells (129/Ola background) were maintained on 0.2% gelatin-coated plates in Glasgow minimum essential medium (GMEM, Sigma-Aldrich, G5154) containing 15% fetal bovine serum (Gibco, 26140079), supplemented with 1× penicillin–streptomycin (Thermo Fisher Scientific, 15140122), 2 mM GlutaMax (Thermo Fisher Scientific, 35050061), 50 µM β-mercaptoethanol (Thermo Fisher Scientific, 21985023), 0.1 mM non-essential amino acids (Thermo Fisher Scientific, 11140050), 1 mM sodium pyruvate (Thermo Fisher Scientific, 11360070) and Leukaemia inhibitory factor (LIF, 1,000 U ml⁻¹, Millipore), referred to as serum mES cell medium. Cells were passaged every 2 days by aspirating the medium, dissociating the cells with trypsin/EDTA solution (TE) briefly at room temperature before rinsing and dissociation in mES cell medium by pipetting. Cells were pelleted by centrifugation at 300g for 5 min. mES cell transfection was performed using Lipofectamine 3000 (Thermo Fisher Scientific, L3000-001) according to the manufacturer’s instructions. Cell counts were performed using the Countess II automated cell counter (Thermo Fisher Scientific, AMQAX1000) using 10 µl of cell suspension and 10 µl of Gibco Trypan Blue Solution (Gibco, 15250061) according to manufacturer’s instructions. For all-trans-retinoic acid (RA, Sigma-Aldrich, R2625-50MG) treatment, cells were induced with 1 μM RA and without LIF for the indicated time. During differentiation, RA medium was changed every 24 h. For SILAC experiments, cells were cultured in SILAC DMEM (Thermo Fisher Scientific, A33822) containing 15% dialysed FBS (Thermo Fisher Scientific, 26400044), to which either ¹³C₆¹⁵N₂l-lysine-2HCl (Thermo Fisher Scientific, 88209) and ¹³C₆¹⁵N₄l-arginine-HCl (Thermo Fisher Scientific, 89990) (heavy), or l-lysine (Sigma-Aldrich, L8662) and l-arginine (Sigma-Aldrich, L8094) containing only light isotopes (light) was added. All cell lines were subjected to STR authentification through ATCC and were tested for mycoplasma contamination.

Generation of mES cell knockin cell lines

For the generation of the auxin-inducible degradation system for DPY30 sgRNA targeting the stop-codon region were cloned into eSpCas9(1.1)-T2A-eGFP. Left and right homology arms, as well as the mAID-T2A-BFP middle part were ligated into a modified pUC19 vector backbone (a gift from S. Pollard) using the In-Fusion cloning kit (Takara, 638910). mES cells were co-transfected with sgRNA- and donor-vector using Lipofectamine 3000 and sorted 48 h later for GFP/BFP-double-positive cells. Homozygous clones were then transfected with pPB-hygro-OsTIR1-P2A-mCherry and pBase plasmids and selected with 100 µg ml⁻¹ hygromycin B (Thermo Fisher Scientific, 10687010). For the generation of the endogenous dTAG-inducible degradation system for RBBP5, sgRNA targeting the stop codon region were co-transfected into the cells and contained the following elements: left and right homology arms, as well as FKBP12(F36V), 2× HA tags, P2A and a neomycin-resistance gene. The transfected cells were selected with 100 µg ml⁻¹ Geneticin selective antibiotic (G418 Sulfate) (Thermo Fisher Scientific, 10131027), single-cell sorted to obtain clonal cell lines and screened for correct biallelic integration. All homozygous insertions and knock-ins were confirmed by Sanger sequencing and western blotting. A list of the oligos and the sequences of the sgRNAs is provided in Supplementary Table 1.

Generation of mES knockout cell lines

Cells were transfected with eSpCas9(1.1)-T2A-eGFP or eSpCas9(1.1)-T2A-mCherry vectors containing a sgRNA targeting the specific genomic locus, respectively, and cells were single-cell sorted 48 h after transfection. To generate the Kdm5a/b-dKO cells in the DPY30–mAID cell line, we first generated the Kdm5a-KO in the DPY30–mAID line (Kdm5a^−/−) by Cas9 (sgRNAs are listed in Supplementary Table 1), then we generated the Kdm5b knockout in the Kdm5a^−/− line (at passage three of the Kdm5a^−/− line). Subsequently, two clones with both Kdm5a and Kdm5b knockout (referred to as DPY30–mAID Kdm5a/b-dKO in this study) were picked up for the downstream analysis. All homozygous insertions and knockouts were confirmed by Sanger sequencing and western blotting. A list of the sgRNAs is provided in Supplementary Table 1. Further characterization of the dKO cell lines showed that they did not have detectable changes in proliferation and expressed normal levels of pluripotent genes; however, as expected, they showed an increase in H3K4me3 levels as compared with the wild-type cells.

Western blotting

Cells were lysed in RIPA buffer with Halt protease inhibitor (Thermo Fisher Scientific, 78429). Proteins that were separated by SDS–PAGE using acrylamide gels (BioRad gel system) were transferred onto nitrocellulose membranes (LI-COR, 926-31092). The membranes were blocked in 5% skimmed milk (Sigma-Aldrich) in PBS-T (0.1% Tween-20 in PBS) and incubated with the primary antibody of interest (Supplementary Table 2). As secondary antibodies, either IRDye 800CW goat anti-rabbit IgG (925-32211, LI-COR Bioscience, 1:15,000) or IRDye 800CW goat anti-mouse IgG (925-32210, LI-COR Bioscience, 1:15,000) was used. Proteins were imaged using Image Studio Lite (Odyssey CLx imager, Li-COR Biosciences). Immunoblotting source data are provided in Supplementary Fig. 4.

Flow cytometry

mES cells were dissociated with trypsin/EDTA, resuspended in culture medium, centrifuged and resuspended in PBS. For intracellular flow cytometry, 0.5 ml of cold fixation buffer (BioLegend, 420801) was added and then incubated at room temperature for 10 min. Subsequently, the cells were labelled with the unconjugated rabbit DPY30 antibodies (Bethyl Laboratories, A304-296A) and subsequently with a FITC-conjugated goat anti-rabbit IgG antibody. Flow cytometry was performed using the Beckman Coulter CytoFlex system. The gating strategy is shown for the E14 sample in Supplementary Fig. 5.

RNA extraction, cDNA synthesis and RT–qPCR analysis

Total RNA was extracted using the RNeasy Plus Mini Kit (Qiagen, 74134) according to the manufacturer’s protocol. One microgram of total RNA was subjected to reverse transcription using Transcriptor Universal cDNA Master (Sigma-Aldrich, 5893151001). qPCR with reverse transcription (RT–qPCR) reactions were set up in triplicate using PowerUp SYBR Green Master Mix (Thermo Fisher Scientific, A25778) and primers (listed in Supplementary Table 1). Relative quantitation was performed to a housekeeping gene using a ΔΔC_t method, as indicated in the corresponding figure legends. Statistical analysis was performed using GraphPad Prism v.7 (GraphPad).

3′-RNA Quant-seq and SLAM-seq

Cells (1 × 10⁶ per treatment condition) were resuspended in 350 μl of buffer RLT plus, and total RNA was extracted from cell pellets using the RNeasy Plus Mini kit (Qiagen, 74134). For SLAM-seq experiments²⁴, cells were incubated with 100 μM 4-thiouridine (4SU; Biosynth, NT06186) for 60 min before RNA isolation. RNA (1 μg) was treated with 10 mM iodoacetamide in a 50 μl reaction volume at 50 °C for 15 min with 50 mM NaH₂PO₄ (pH 8.0), and 50% (v/v) DMSO followed by addition of 1 μl of 1 M dithiothreitol (DTT) to stop the reaction. RNA was precipitated at −80 °C for 60 min with 1 μl of GlycoBlue (Thermo Fisher Scientific, AM9515), 5 μl of 3 M sodium acetate (pH 5.2) and 180 μl of ethanol (≥99.0%). RNA was pelleted at 4 °C (12,000g for 30 min), washed with 1 ml of 80% ethanol and centrifuged at 4 °C (12,000g for 30 min). The RNA pellet was dried at room temperature for 10 min and resuspended in 20 μl of nuclease-free H₂O. RNA yield and quality were assessed using the 2200 TapeStation (Agilent). Sequencing libraries were prepared using the QuantSeq 3′-mRNA Seq Library Prep Kit FWD for Illumina (Lexogen, SKU: 015.96) from 500 ng or 100 ng of total RNA spiked with ERCC RNA Spike-In Mix 1 (1:1,000, Thermo Fisher Scientific, 4456740). In brief, first-strand (oligo(dT)) cDNA synthesis was followed by RNA removal and second-strand synthesis by random priming. The double-stranded library was bead-purified to remove reaction components before PCR amplification with i7 single-index primers for 10 cycles. Amplified libraries were again bead-purified according to the manufacturer’s protocol, and the concentration was measured by Qubit assay. All of the samples were checked for fragment size distribution on the TapeStation before pooling for 75 bp or 45 bp single-end read sequencing on the Illumina NextSeq 550 platform.

ChIP–seq

ChIP experiments were performed according to a standard protocol. In brief, ES cells were cross-linked by the addition of 1% formaldehyde (Sigma-Aldrich, 252549-1L) in the dish for 10 min at room temperature before quenching with 0.125 M glycine. The fixed cells were washed with PBS and resuspended in SDS buffer (100 mM NaCl, 50 mM Tris-HCl pH 8.0, 5 mM EDTA, 0.5% SDS, 1× protease inhibitor cocktail from Roche). The resulting nuclei were precipitated, resuspended in the immunoprecipitation buffer at 1 ml per 16 million cells (SDS buffer and Triton dilution buffer (100 mM NaCl, 100 mM Tris-HCl pH 8.0, 5 mM EDTA, 5% Triton X-100) mixed at a 2:1 ratio with the addition of 1× protease inhibitor cocktail from Roche) and processed on the Bioruptor Plus Sonicator (Diagenode) to achieve an average fragment length of 200–300 bp. Chromatin concentrations were estimated using the Nanodrop according to manufacturer’s protocols. The immunoprecipitation reactions were set up in 1 ml of the immunoprecipitation buffer as indicated below and incubated overnight at 4 °C. The next day, BSA-blocked Protein G Dynabeads (Thermo Fisher Scientific, 10004D) were added to the reactions and incubated for 2 h at 4 °C. The beads were then washed three times with low-salt washing buffer (150 mM NaCl, 1% Triton X-100, 0.1% SDS, 2 mM EDTA, 20 mM Tris-HCl pH 8.0) and twice with high-salt washing buffer (500 mM NaCl, 1% Triton X-100, 0.1% SDS, 2 mM EDTA, 20 mM Tris-HCl pH 8.0). The samples were then reverse cross-linked overnight at 65 °C in the elution buffer (1% SDS, 0.1 M NaHCO₃) and purified using the QIAQuick PCR purification kit (Qiagen, 28506). A list of the antibodies used in this study is provided in Supplementary Table 2. Libraries for ChIP–seq were prepared using the NEBNext Ultra II DNA Library prep kit (NEB, E7645L), and AmpureXP beads (Beckman, A63881) were used for size selection. Libraries were quantified using the Qubit High Sensitivity DNA kit (Agilent, Q32854) and assessed on the TapeStation. Libraries were pooled as required, denatured and loaded onto the Illumina NextSeq 550 system with high-output kits (75 cycles). A list of all of the primers used for ChIP–qPCR is provided in Supplementary Table 1. For spiked-in ChIP–seq, 5% of the cross-linked Drosophila chromatin (homemade) with Spike-in Antibody (Active Motif, 61686) was added before the immunoprecipitation step according to the manufacturer’s instructions.

RNAPII IP followed by MS

To measure the difference between RNAPII interactome in the presence and absence of DPY30, 1 × 10⁸ DPY30–mAID cells were incubated with auxin for 8 h to induce DPY30 degradation (8 h samples), while 1 × 10⁸ DPY30–mAID cells treated with DMSO served as the control (0 h samples). Cells were collected and frozen on dry ice and kept at −80 °C until immunoprecipitation (IP). The cells were thawed at 37 °C for 30 s lysed in 1.6 ml of ice cold 50 mM EPPS pH 7.5, 150 mM NaCl, 1% Triton X-100 with cOmplete, EDTA-free Protease Inhibitor Cocktail (1 tablet per 20 ml of lysis buffer), 1:100 of Sigma-Aldrich phosphatase inhibitor 2 and 3 cocktails and 250 U μl⁻¹ of benzonase. Lysates were incubated on ice for 5 min to allow DNA digestion, centrifuged at 20,000g for 5 min to remove insoluble material and filtered through the AcroPrep 1.0 μm glass filter plate at 2,000g for 1 min. The concentration of protein was then estimated using the bicinchoninic acid assay. For each of the samples (DMSO- and auxin-treated cells) six immunoprecipitation reactions were performed: three with the anti-RNAPII antibody (Abcam, ab817, 8WG16) and three with IgG control (Invitrogen, 02-6102). Each reaction was performed with 1 mg of lysate at 3.3 g l⁻¹ lysate concentration and 5 μg of antibody bound to protein G Sepharose (Sigma-Aldrich, 17-0618-02). The incubation was performed at 4 °C with shaking at 1,100 rpm for 1 h. The beads were then transferred to the OF 1100 filter plate (Orochem Technologies) and washed five times with ice-cold 50 mM EPPS pH 7.5, 150 mM NaCl using vacuum manifold. Then, 18 μl of 10 mM EPPS pH 8.5 with 20 ng μl⁻¹ trypsin and 1 ng μl⁻¹ LysC was added to the beads in each well and digestion was performed for 2 h at 37 °C at 2,000 rpm. The partial digest was then collected into a 96-well PCR plate and left overnight at room temperature to complete digestion. Then, 4 μl of 22 g l⁻¹ 11 plex TMTPro tags were added to each sample. The samples were then pulled and 20 μl of the combined sample was set aside, and the rest was fractionated into six fractions using the High pH Reversed-Phase Peptide Fractionation Kit, as suggested by the manufacturer. The fractions were concatenated into four fractions (the first and fifth fractions, the second and sixth and so on were mixed) and evaporated in speed vac (0.5 μl of DMSO was added to each sample to prevent complete evaporation) and resuspended in 20 μl 0.1% TFA. For data acquisition, 4.5 μl of unfractionated sample and every fraction was analysed by using the nanoAcquity 2 μm particle size, 75 mm × 500 mm easyspray column in direct injection mode. The samples were separated using the following gradient at 300 nl min⁻¹ of buffer A (0.1% formic acid in water) and buffer B (0.1% formic acid in acetonitrile): 0–7% in 10 min, 7–30% in 92 min, 30–60% in 18 min, the column was then washed with 95% B for 10 min at 400 nl min⁻¹. The column was kept at 60 °C. Eluting peptides were analysed on the Orbitrap Fusion Lumos instrument using MS3 SPS with the settings recommended by the instrument manufacturer for TMT11 plex analysis with the following modifications: (1) CID NCE for MS2 was set at 32; (2) HCD NCE for MS3 was set at 45; (3) C series exclusion was disabled as TMTPro reagent was not enabled in C-series exclusion node. The cycle time was set at 3 s and the dynamic exclusion time was set at 15 s.

APEX2-based RNAPII proximity labelling and affinity enrichment of biotinylated proteins

CRISPR–Cas9 technology was used to target endogenous Rpb1 at the 5′ end with a cassette encoding a Flag affinity-tag and APEX2 (Flag–APEX2), resulting in RPB1 fused at its N terminus to Flag–APEX2. DPY30–mAID OsTIR1 E14 cells were co-transfected with espCas9 plasmid and a donor plasmid containing the puromycin-resistance selection gene, P2A self-cleavage site and Flag–APEX2 flanked by homology arms corresponding to the respective target genes. The sequences of the guide RNA and homology arms for targeting Rpb1 are provided in Supplementary Table 1. The APEX2-expressing cells were incubated with 4 mM biotin-phenol reagent (Iris Biotech, LS-3500.1000) for 2 h before the start of the labelling reaction. The cells were washed with PBS (with Ca⁺⁺ and Mg⁺⁺) and the labelling reaction was initiated by adding 1 mM H₂O₂ in PBS for 2 min at room temperature. The reaction was terminated by washing the cells three times with a quencher solution containing 10 mM sodium azide, 10 mM sodium ascorbate and 5 mM Trolox in PBS.

Cell fractionation for SILAC chromatin MS

Chromatin fractions were prepared as described previously⁵³ with some modifications. In brief, cells were lysed by swelling and mechanical force in buffer A (10 mM ammonium bicarbonate pH 8.0, 1.5 mM MgCl₂, 10 mM KCl, 10 mM sodium ascorbate, 5 mM Trolox, 10 mM sodium azide, 1× protease inhibitor cocktail and 0.2% NP40). Nuclei were then collected by centrifugation and chemically lysed in buffer C (20 mM ammonium bicarbonate pH 8.0, 420 mM NaCl₂, 20% (v/v) glycerol, 2 mM MgCl₂, 0.2 mM EDTA, 0.1% NP40, 10 mM sodium ascorbate, 5 mM Trolox, 10 mM sodium azide, 1× protease inhibitor cocktail and 0.5 mM DTT). Lysates were centrifuged at 20,800g for 45 min at 4 °C. The pellet contains the insoluble chromatin fraction and consists of DNA and proteins tightly bound to chromatin. To solubilize the chromatin pellet, 750 U Benzonase (Sigma-Aldrich) was added, followed by 10 min incubation on ice and 5 min of agitation at room temperature. Clarified lysate was collected and the protein concentration was quantified using Bio-Rad Bradford’s reagent. Approximately 4 mg lysates from SILAC heavy or light cells were mixed 1:1 and incubated with 50 μl Streptavidin magnetic beads (Pierce, 88817) at 4 °C on a rotating wheel overnight. The beads were washed four times with RIPA buffer.

Sample preparation for SILAC/MS

Eluates from biotin pull-down were transferred to fresh microfuge tubes. NuPAGE sample loading buffer was added to the beads and heated at 90 °C for 5 min. A magnetic rack was used to separate the beads from the proteins. The supernatant was then run on an SDS–PAGE gel (Bis-Tris, 4–12%) enough to get the sample into the gel. Gel sections were excised, washed, reduced with DTT, alkylated with iodoacetamide and digested overnight with trypsin at 37 °C (ref. ⁵⁴). Homemade C18 StageTips were prepared as described previously⁵⁵ and preconditioned with a 50 μl wash of methanol, 50 μl wash of 70% acetonitrile/0.1% trifluoroacetic acid and two 50 μl washes of 0.1% trifluoroacetic acid at 1,000g. Peptides were then loaded onto StageTips and washed with 50 μl of 0.1% formic acid and were eluted with 60 μl of 70% acetonitrile/0.1% formic acid. The samples were then vacuum centrifuged using the SpeedVac and reconstituted in 0.1% formic acid for LC–MS/MS and were analysed by microcapillary LC–MS/MS using the nanoAcquity system (Waters) with a 100 μm inner-diameter × 10 cm length C18 column (1.7 μm BEH130, Waters) configured with a 180 μm × 2 cm trap column coupled to a Q-Exactive Plus mass spectrometer (Thermo Fisher Scientific). Peptides were eluted at 300 nl min⁻¹ using a 4 h acetonitrile gradient (0.1% formic acid). The Q-Exactive Plus mass spectrometer was operated in automatic, data-dependent MS/MS acquisition mode with one MS full scan (380–1,600 m/z) at 70,000 mass resolution and up to ten concurrent MS/MS scans for the ten most intense peaks selected from each survey scan. Survey scans were acquired in profile mode and MS/MS scans were acquired in centroid mode at 17,500 resolutions with an isolation window of 1.5 amu and normalized collision energy of 27; AGC was set to 1 × 10⁶ for MS1 and 5 × 10⁴ and 50 ms max IT for MS2; charge exclusion of unassigned, +1 and greater than 6 was enabled with dynamic exclusion of 15 s.

TT_chem-seq and DRB/TT_chem-seq analysis

We performed TT_chem-seq as previously described³². In brief, cells in a 10 cm dish at 80% confluency were treated in biological duplicates at the specified time points. After the specified treatment, we supplemented the treatment medium with 1 mM 4SU and metabolically labelled the cells for 10 min. The cells were lysed in QIAzol (Qiagen, 79306) and total RNA was isolated according to the manufacturer’s instructions before the addition of 100 ng of RNA spike-in mix together with QIAzol. The RNA spike-in was extracted from Drosophila S2 cells using 4SU, metabolically labelling the cells for 20 min. The 100 μg RNA (in a total volume of 100 μl) was fragmented by addition of 20 μl of 1 M NaOH and left on ice for 20 min. Fragmentation was stopped by addition of 160 μl of 0.5 M Tris pH 6.8 and cleaned up twice using the Micro Bio-Spin P-30 Gel Columns (BioRad, 7326223) according to the manufacturer’s instructions. Biotinylation of 4SU-residues was performed in a total volume of 250 μl, containing 10 mM Tris-HCl pH 7.4, 1 mM EDTA and 5 mg of MTSEA biotin-XX linker (Biotium, BT90066) for 30 min at room temperature in the dark. RNA was then purified by phenol–chloroform extraction, denatured by 10 min incubation at 65 °C and added to 200 μl μMACS Streptavidin MicroBeads (Milentyl, 130-074-101). RNA was incubated with beads for 15 min at room temperature and beads were applied to a μColumn in the magnetic field of a μMACS magnetic separator. The beads were washed twice with pull-out wash buffer (100 mM Tris-HCl, pH 7.4, 10 mM EDTA, 1 M NaCl and 0.1% Tween-20). Biotinylated RNA was eluted twice by addition of 100 mM DTT and cleaned up using the RNeasy MinElute kit (Qiagen, 74204) using 1,050 μl ethanol (≥99%) per 200 μl reaction after addition of 700 μl of RLT buffer to precipitate RNA of less than 200 nucleotides. A total of 200 ng of the purified 4SU-labelled RNA was then used as input for the TruSeq Stranded Total RNA kit (Illumina, 20020596) for library preparation. The libraries were amplified according to the manufacturer’s instructions with modifications as previously described³². The library was amplified with 10 PCR cycles and quality-control checked on the TapeStation (Agilent) using the High Sensitivity DNA kit before pooling and paired-end sequencing on the NextSeq 550 (Illumina) system.

For DRB/TT_chem-seq, cells were incubated in 100 µM DRB (Sigma-Aldrich, D1916) for 3.5 h. The cells were then washed twice in PBS, and the prewarmed fresh DRB-free medium was added to restart transcription. The RNA was labelled in vivo with 1 mM 4SU for 10 min before the addition of QIAzol, which was used to stop the reaction at the desired time point.

mNET–seq

We performed mNET–seq with minor modifications to the original protocol²⁷. The Flag epitope tag was added to the N terminus of the first RNAPII subunit (RPB1) in DPY30–mAID degron cells (RPB1–APEX2 cells). Cells were seeded the day before the experiment to get 100 million cells per sample the next day. We randomly assigned flasks for each treatment and treated cells with DMSO only or auxin ligand, before extracting the chromatin-bound RNAPII. The cells were first washed with ice-cold DPBS, resuspended in 4 ml of ice-cold HLB + N buffer (10 mM Tris-HCl (pH 7.5), 10 mM NaCl, 2.5 mM MgCl₂, 0.5% (v/v) NP-40 and 1× proteinase inhibitor) and incubated on ice for 5 min, then cells were scraped to a 15 ml centrifugate tube. The cell suspension was then underlaid with 1 ml of HLB + NS buffer (10 mM Tris-HCl pH 7.5, 10 mM NaCl, 2.5 mM MgCl₂, 0.5% (v/v) NP-40, 10% (w/v) sucrose and 1× proteinase inhibitor) and centrifuged to pellet the nuclei at 400g at 4 °C. The supernatant and membrane debris were then removed, and the nuclei were resuspended in 125 μl of NUN1 lysis buffer (20 mM Tris-HCl pH 8.0, 75 mM NaCl, 0.5 mM EDTA, 50% (v/v) glycerol and 1× proteinase inhibitor) to which we added 1.2 ml NUN2 buffer (20 mM HEPES-KOH pH 7.6, 300 mM NaCl, 0.2 mM EDTA, 7.5 mM MgCl₂, 1% (v/v) NP-40, 1 M urea and 1× proteinase inhibitor) to precipitate the chromatin, and the sample was incubated on ice for 15 min with occasional vortexing. The lysates were then centrifuged at 16,000g for 10 min at 4 °C to pellet the chromatin. The chromatin pellets were then washed with 1× MNase buffer and then digested with 50 U MNase (NEB, M0247S) for 2 min at 37 °C with 1,400 rpm on a thermomixer. The digestion was stopped by adding 5 μl of 500 mM EGTA (to a final concentration of 25 mM; Thermo Fisher Scientific, 50-255-956) and transferred onto ice. The reactions were then centrifuged at 16,000g for 5 min at 4 °C and the supernatant was subsequently diluted with 1 ml of NET-2 buffer (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.05% (v/v) NP-40) per fraction and pooled per sample for the N-terminal Flag-RNAPII IP. We added 50 μl of anti-Flag M2 Affinity gel (Sigma-Aldrich, A2220) and incubated in the cold room for 1 h. This was followed by eight washes with NET-2 buffer and one wash with 500 µl of PNKT buffer containing 1× T4 polynucleotide kinase (PNK) buffer (NEB, M0201L) and 0.1% (v/v) Tween-20 (Thermo Fisher Scientific, BP337-100). The beads were incubated in 100 µl of PNK reaction mix containing 1× PNK buffer, 0.1% (v/v) Tween-20, 1 mM ATP and T4 PNK 3′ phosphatase minus (NEB, M0236L) at 37 °C for 10 min. We eluted the RNA by adding 350 μl of buffer RLT plus and 1 ng of fragmented Drosophila S2 mRNA spike-in to the extraction buffer. Next, short immunoprecipitated RNA fragments were size-selected (under 200 nucleotides) and purified from the eluates according to the manufacturer’s protocol after the purification of miRNA from animal cells using the RNeasy Plus Mini Kit and the RNeasy MinElute Cleanup Kit (Qiagen, 74204), and eluted in 14 μl of nuclease-free water. We checked 1 μl of the eluted RNA on the TapeStation for the RNA quality and proceeded to NGS library preparation using the NEBNext Multiplex Small RNA Library Prep kit (NEB, NC0477293). Both were performed according to the manufacturer’s protocol with low input material, with 14 PCR cycles for the library amplification step. We cleaned up and concentrated the DNA using the Monarch kit (NEB) and separated the library on a 6% TBE gel and performed size-selection by excising the smear between 147 and 307 nucleotides (according to the Quick-Load pBR322 DNA-MspI Digest ladder (NEB)). These libraries were quality-checked and quantified using the TapeStation. The samples were pooled and sequenced using paired-end sequencing on the NextSeq 550 (Illumina) system.

For promoter-proximal RNAPII half-life experiments determined by mNET–seq, chromatin was isolated from cells that were pretreated with DMSO or auxin and then incubated with 1 µM triptolide for 0, 5, 10, 20 or 40 min in the presence of DMSO/auxin and analysed using mNET–seq. Chromatin was digested with micrococcal nuclease (MNase, scissor) to release RNAPII engaged RNA from insoluble chromatin, and then immunoprecipitated using anti-Flag–RNAPII antibodies.

IP–MS data analysis

Data were analysed in Proteome Discoverer v.3.1. A database search was performed with the Sequest HT search engine using the Mouse UniProt database containing only reviewed entries and canonical isoforms (retrieved on 10 October 2019). Oxidation (M) was set as a variable modification, while TMT was set as fixed modification. A maximum of two missed cleavages were permitted. The precursor and fragment mass tolerances were 10 ppm and 0.6 Da, respectively. Peptide-spectrum matches (PSMs) were validated by percolator with a 0.01 posterior error probability threshold. Only PSMs with an isolation interference of <25% and at least 5 MS2 fragments matched to the peptide sequence selected for MS3 were considered. The quantification results of PSMs were combined into protein-level quantification using the MSstatsTMT R package⁵⁶. Only proteins with at least three peptides were reported. To identify interactors, we performed differential abundance analysis between the IP samples and their corresponding controls (that is, 0 h IP was compared to 0 h IgG control and 8 h IP was compared to 8 h IgG control). A protein was considered to be an interactor if in one or both comparisons its levels were statistically significantly different (Q ≤ 0.05, limma test, with P values adjusted by the Storey method) and at least twice higher in IP reactions than in the corresponding IG control (Supplementary Table 3).

SILAC/MS analyses

All SILAC/MS data were processed using the MaxQuant software (Max Planck Institute of Biochemistry; v.1.5.3.30). The default values were used for the first search tolerance and main search tolerance—20 ppm and 6 ppm, respectively. Labels were set to Arg10 and Lys8. MaxQuant was set up to search the reference mouse proteome database downloaded from UniProt on 9 January 2020. MaxQuant performed the search assuming trypsin digestion with up to two missed cleavages. Peptide, site and protein false-discovery rate (FDR) were all set to 1% with a minimum of one peptide needed for identification but two peptides needed to calculate a protein level ratio. The following modifications were used as variable modifications for identifications and included for protein quantification: oxidation of methionine, acetylation of the protein N terminus, ubiquitination of lysine, phosphorylation of serine, threonine and tyrosine residues, and carbamidomethyl on cystine. Intensity values measured in all replicates were log₂-transformed (Supplementary Table 4), P values were computed using Fisher’s tests and corrected using Benjamini–Hochberg FDR correction. All raw MS data files have been deposited to the ProteomeXchange Consortium (and PXD039176). The Gene Ontology term enrichment analysis was performed using Enrichr online tool (https://maayanlab.cloud/Enrichr/), STRING (https://string-db.org) and clusterProfiler⁵⁷.

ChIP–seq data analysis

The sequenced reads were demultiplexed using bcl2fastq (v.2.19.0.316), and basic quality control was performed on the resulting FASTQ files using FastQC (v.0.11.8). FASTQ reads were mapped to the GRCm38 (mm10) genome using Bowtie2 (v.2.4.1) using the standard settings. The resulting SAM files were converted to BAM files using the SAMtools (v.1.10) view command, after which the BAM files were sorted and indexed, and potential PCR duplicates were removed using the rmdup function. DeepTools (v.3.3.0) was used to generate occupancy heat maps, and the resulting normalized occupancy matrix was used as input for public R scripts to generate average profile plots and to calculate processivity indices. In brief, the BAM files were converted into BigWig files using the bamCoverage function (bamCoverage -p 8 –normalizeUsing RPGC –effectiveGenomeSize mm10 –centerReads -e –scaleFactor X –blackListFileName mm10.blacklist.bed). For comparison, quantitative ChIP–seq data using spike-in normalization were used, normalization to the Drosophila S2 spike-in was performed at this stage according to the manufacturer’s instructions (Active Motif, 61686; https://www.activemotif.com/catalog/1091/chip-normalization). The computeMatrix function was used to quantify the occupancy of reads across the specified intervals, and the plotProfile and plotHeatmap functions were used to plot the data. Reproducibility of replicates is shown in Supplementary Fig. 3.

Quant-seq/SLAM-seq analysis

Gene and 3′ untranslated region (UTR) annotations were obtained from the UCSC table browser (https://genome.ucsc.edu/cgi-bin/hgTables, mm10 vM14 3′ UTR). Adapters were trimmed from raw reads using cutadapt through the trim_galore wrapper tool with adapter overlaps set to 3 bp for trimming. For Quant-seq, concatenated fastq files were trimmed for adapter sequences, and masked for low-complexity or low-quality sequences using trim_galore, then mapped to the mm10 whole genome using HISAT v.2.2.1 with the default parameters. The number of reads mapped to the 3′ UTR of genes was determined using featureCounts. Raw reads were normalized to CPM. SLAM-seq analysis was performed as previously described⁵⁸ using the SlamDunk package⁵⁹. Trimmed reads were further processed with SlamDunk (v.0.3.4 16). The ‘Slamdunk all’ command was executed with the default parameters except ‘-rl 74 -t 8 fastq.gz -n 100 -m -mv 0.2 -o Slamdunk2’, running the full analysis procedure (slamdunk all) and aligning against the mouse genome (GRCm38), filtering for variants with a variant fraction of 0.2. Unless indicated otherwise, reads were filtered for having ≥2 T>C conversions. The remaining parameters were left as defaults.

Analysis of differential gene expression was restricted to genes with ≥10 reads in at least one condition. Differential gene expression calling was performed on raw read counts with ≥2 T>C conversions using DESeq2 with the default settings, and with size factors estimated on corresponding total mRNA reads for global normalization. Downstream analysis was restricted to genes that passed all internal filters for FDR estimation by DESeq2. Plots of differential gene expression were visualized using the ggplot2 package in R with significant genes (P value < 0.05, |log2FC| ≥ 1). Reproducibility of replicates is shown in Supplementary Fig. 3.

mNET–seq data analysis

Reads were demultiplexed using bcl2fastq, then trimmed for adapter content with cutadapt (-m 10 -e 0.05 –match-read-wildcards -n 1), and mapped with STAR to the GRCm38 (mm10) genome assembly. Further data processing was performed using the R/Bioconductor environment. Coverage tracks for further analysis were restricted to the last nucleotide incorporated by the RNAPII in the aligned mNET–seq reads as described⁶⁰. To calculate the half-life of paused RNAPII for each gene by mNET–seq, RNAPII density was calculated in a 300 bp window downstream of the TSS. RNAPII time-course measurements were fitted into an exponential decay model using the RNAdecay R package (https://bioconductor.org/packages/release/bioc/html/RNAdecay.html). We selected genes fulfilling the current criteria: (1) detectable RNAPII levels (reads per kilobase of transcript, per million mapped reads > 1), (2) highest RNAPII density under the no triptolide (0 min) condition and (3) low variance between replicates (σ < 0.05). Genes fitting the above criteria (n = 6,338) were used to calculate the RNAPII half-life. Reproducibility of replicates is shown in Supplementary Fig. 3.

TT_chem-seq data analysis

Paired-end reads were demultiplexed using bcl2fastq. TT_chem-seq raw data were processed essentially as described previously³². Raw reads were aligned to the mouse mm10 genome assembly using STAR. Mapped reads with a mapping quality score <10 were discarded with SAMtools. All further processing was performed using the R/Bioconductor framework. Antisense bias, sequencing depth and cross-contamination rates were calculated as described previously³². Reads were mapped to transcription units, which represent the union of all annotated UCSC RefSeq isoforms per gene. The number of transcribed bases per transcription unit was calculated as the sum of the coverage of evident (sequenced) fragment parts (read pairs only) for all fragments in addition to the sum of the coverage of the inner mate interval if not entirely overlapping a RefSeq annotated intron (UCSC RefSeq GRCm38). Computational analysis DRB/TT_chem-seq data were processed using a previously published protocol³². In brief, reads were aligned to human GRCm38 (mm10). Read depth coverage was normalized to account for differences between samples using a scale factor derived from a spike-in aligned and counted against Drosophila melanogaster. Biological replicate alignments were combined for the purpose of visualization and wave-peak analysis to increase read-depth coverage.

A set of non-overlapping protein-coding genes of >60 kb and <300 kb was selected for wave-peak analysis. A meta-gene profile was calculated by taking a trimmed mean of each base-pair coverage in the region around the TSS. This was further smoothened using a spline. Wave peaks were called at the maximum points on the spline, with the stipulation that the peak must advance with time before being subjected to manual review. Elongation rates (kb per min) were calculated by fitting a linear model to the wave-peak positions as a function of time. For elongation-rate analysis, the following criteria were used to filter genes: The 0 min timepoint DMSO control sample was required to show expression of the gene (mean expression of >100 rpm by TT-seq) and was required to have a wave peak called within 10 kb of the pausing peak region to remove artifacts. Genes showing an increase in transcription in the DMSO control sample for the time course were identified by requiring the wave peak in the 0 min sample to be less than the wave peak in the 10 min timepoint wave peak, and the wave peak in the 10 min sample to be less than the wave peak in the 20 min timepoint, and the wave peak in the 20 min sample to be less than the wave peak in the 30 min timepoint. This resulted in the identification of 855 genes, for which elongation rates were calculated for the samples by dividing the wave peak position by the timepoint. Reproducibility of replicates is shown in Supplementary Fig. 3.

Statistics and reproducibility

The statistical details of the experiments can be found in the figure legends and in the Methods. Western blotting in Figs. 1b,c,f–h, 3a and 5b,c,j was independently performed three times with similar results, and western blotting in Extended Data Figs. 8c,f,g and 9c,e was performed twice as biologically independent experiments.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.