Biology Atlas

Map / Concepts

concept · intermediate · partly written

Transcription by RNA polymerase II: initiation, pausing, elongation, termination

How human cells copy a gene into RNA, how fast and how accurately, and which of those numbers were seen and which were fitted.

Also called: Pol II, transcription, promoter-proximal pausing, RNA polymerase II

In short

RNA polymerase II (Pol II) is the enzyme that copies protein-coding genes into RNA. It starts at the promoter, usually halts for a while shortly after the start (promoter-proximal pausing), is released to run along the gene, and lets go past the end. Where Pol II sits is measured well. How fast it moves and how long it waits are fitted from snapshots and drug time-courses, and the fits disagree. 12

Reported, not yet replicated

The cycle and the factors named more

In the model most studies use, Pol II pauses about 30 to 60 nucleotides past the start, held by DSIF (DRB sensitivity-inducing factor) and NELF (negative elongation factor). The kinase P-TEFb (positive transcription elongation factor b, which contains CDK9, cyclin-dependent kinase 9) phosphorylates the polymerase tail and both factors, NELF leaves and elongation starts. These expansions are the ones the opened sources print; they were not checked against HGNC (the gene-naming authority) in this run. The names DSIF and NELF record the assay that found them, not what the factors are. The 30 to 60 nt position is a background statement in Steurer 2018, not measured there. In mouse embryonic stem cells the CDK9-blocking drug flavopiridol (and triptolide) stopped entry into productive elongation at over 95% of genes, while polymerases already in gene bodies kept going, so release from the pause is near-universal for expressed genes (mouse, not human). 234

Reported, not yet replicated

Initiation and bursting: numbers more

Human K562 cells, 4-thiouridine labelling (TT-seq) plus nascent-RNA sequencing (mNET-seq) fitted to a kinetic model: a median of 1 productive initiation per cell per minute for protein-coding genes, 0.3 for long intergenic non-coding RNAs. One endogenous gene, TFF1 (trefoil factor 1) in MCF7 cells, imaged live with 24 loops in the 3' UTR: 1.5 +/- 0.5 transcripts per burst, one burst about every 66 +/- 7 min, active periods 16.0 +/- 0.5 min. The model fit in the same paper gives initiation of 0.5 +/- 0.02 per minute while active. A HIV-1 reporter in HeLa cells fires polymerases in convoys, one every few seconds for about a minute. A reporter gene and an endogenous gene are different promoters with different definitions of a burst; they do not conflict. 567

Reported, not yet replicated

Elongation speed more

Human cells, drug-chase or induction followed by run-on sequencing: median 2.1 kb/min over 140 oestrogen-induced genes in MCF-7 cells (middle 90% of genes 0.37 to 3.57), 2.80 over 26 TNF-induced genes in AC16 cells (Danko 2013); about 1.7 in HEK293 cells (Fong 2014); about 3.5 with a range of about 2 to 6 in HeLa (Fuchs 2014). Imaging of beta-globin reporter in U2-OS cells gave 2.6 +/- 0.16 kb/min, slowed to 1.44 by camptothecin (Coulon 2014). Mouse embryonic stem cells: 0.5, 1.8 and 2.4 kb/min in early, middle and late windows of genes (Jonkers 2014). Danko says published estimates differ up to 6-fold and that no study traced this to a cause. Part is biology (speed rises along a gene, and TNF targets ran about 50% faster than oestrogen targets), part is method. 489103

Reported, not yet replicated

How long is the pause? Three answers that are not one quantity more

Human MRC-5 fibroblasts, live-cell photobleaching of tagged Pol II: about 42 s, with 23% of nuclear Pol II in the paused compartment (Steurer 2018). HEK293, time-varying run-on: release rate 0.139 +/- 0.007 per minute, about 7 min (Zhang 2021). Mouse embryonic stem cells, run-on signal after triptolide: mean half-life 6.9 min (Jonkers 2014; mouse). Human, under hypertonic stress: most paused Pol II turned over in under 2 min (Erickson 2018). Drosophila: over 15 min at genes with inefficient release (Henriques 2013, abstract). K562 model: median effective pause 1 min, defined as the gap between productive initiations (Gressel 2019). A review spans about 2 to 30 min and calls it a controversy (Core and Adelman 2019). They differ in what they call a pause: triptolide designs assume initiation stops at once, the imaging design assumes fitted compartments map onto biochemical states, and Gressel's figure is a gap, not a dwell. They are not averaged here. 2113121351

Contested

Transcription error rate more

Circular-consensus sequencing re-reads each RNA many times so that reverse-transcriptase errors cancel. Human primary fibroblasts, untreated, from a healthy 42-year-old female: 4.7e-6 per base (about 1 error in 213,000 bases). The same paper reports yeast 2.9e-6, worm 4.0e-6, fly 5.69e-6 and mouse cells 4.9e-6. 14

Reported, not yet replicated

Termination and readthrough more

Where termination happens is inferred from where labelled signal ends, not seen. After 1 hour in 80 mM KCl, one neuroblastoma line made downstream-of-gene transcripts at over 10% of 20,314 protein-coding genes, often over 45 kb long; their function is not shown. In HeLa mNET-seq, knockdown of cleavage factors raised pausing of Pol II near transcription start sites about 2.3 to 3.6 fold (an effect on start-site pausing, not a cut to gene-end pausing). The torpedo and allosteric models of termination remain unresolved in the sources opened. 15161

Reported, not yet replicated

What is measured and what is assumed more

Directly measured: where engaged Pol II sits (read 3' ends from run-on or NET-seq), the amount of newly made RNA, fluorescence at one tagged gene, and (with repeated reads) per-base errors. Assumed or modelled: a pause duration (from a drug chase that assumes an instant block, or from fitted compartments), an elongation rate (uniform speed within a window and a clean induction), the paused fraction of initiations ('1 in 100 initiations makes an mRNA' is Monte Carlo output) and where termination occurs. No assay watches one human polymerase at one site and reads out its fate (our reading of the methods opened). The nascent-RNA assays are in assays/nascent-rna-sequencing. 24171

Our inference

Plain answer

Pol II copies a gene in four stages: start at the promoter, pause shortly after, run along the gene at about 1 to 4 kb/min in most human studies, and release RNA past the gene end. Where Pol II sits is mapped well; how long it waits and how fast it runs depend on drugs and models, and published estimates differ about ten-fold for pausing. 4211

Reported, not yet replicated

Why it matters

Every RNA-seq number is production minus decay, so knowing how transcription itself is timed and measured is the first link in the chain from gene to protein. 1

Reported, not yet replicated

Still open here

Not yet written: Equation, Real numbers.

Linked from

Sources

  1. Core L, Adelman K. Promoter-proximal pausing of RNA polymerase II: a nexus of gene regulation.. Genes Dev (2019). paper; opened 2026-11-17 all records citing this
  2. Steurer B, Janssens RC, Geverts B, et al.. Live-cell analysis of endogenous GFP-RPB1 uncovers rapid turnover of initiating and promoter-paused RNA Polymerase II.. Proc Natl Acad Sci U S A (2018). paper; opened 2026-11-17 all records citing this
  3. Jonkers I, Kwak H, Lis JT. Genome-wide dynamics of Pol II elongation and its interplay with promoter proximal pausing, chromatin, and exons.. Elife (2014). paper; opened 2026-11-17 all records citing this
  4. Danko CG, Hah N, Luo X, et al.. Signaling pathways differentially affect RNA polymerase II initiation, pausing, and elongation rate in cells.. Mol Cell (2013). paper; opened 2026-11-17 all records citing this
  5. Gressel S, Schwalb B, Cramer P. The pause-initiation limit restricts transcription activation in human cells.. Nat Commun (2019). paper; opened 2026-11-17 all records citing this
  6. Rodriguez J, Ren G, Day CR, et al.. Intrinsic Dynamics of a Human Gene Reveal the Basis of Expression Heterogeneity.. Cell (2019). paper; opened 2026-11-17 all records citing this
  7. Tantale K, Mueller F, Kozulic-Pirher A, et al.. A single-molecule view of transcription reveals convoys of RNA polymerases and multi-scale bursting.. Nat Commun (2016). paper; opened 2026-11-17 all records citing this
  8. Fong N, Kim H, Zhou Y, et al.. Pre-mRNA splicing is facilitated by an optimal RNA polymerase II elongation rate.. Genes Dev (2014). paper; opened 2026-11-17 all records citing this
  9. Fuchs G, Voichek Y, Benjamin S, et al.. 4sUDRB-seq: measuring genomewide transcriptional elongation rates and initiation frequencies within cells.. Genome Biol (2014). paper; opened 2026-11-17 all records citing this
  10. Coulon A, Ferguson ML, de Turris V, et al.. Kinetic competition during the transcription cycle results in stochastic RNA processing.. Elife (2014). paper; opened 2026-11-17 all records citing this
  11. Zhang J, Cavallaro M, Hebenstreit D. Timing RNA polymerase pausing with TV-PRO-seq.. Cell Rep Methods (2021). paper; opened 2026-11-17 all records citing this
  12. Erickson B, Sheridan RM, Cortazar M, et al.. Dynamic turnover of paused Pol II complexes at human promoters.. Genes Dev (2018). paper; opened 2026-11-17 all records citing this
  13. Henriques T, Gilchrist DA, Nechaev S, et al.. Stable pausing by RNA polymerase II provides an opportunity to target and integrate regulatory signals.. Mol Cell (2013). paper; opened 2026-11-17 all records citing this
  14. Chung C, Verheijen BM, Navapanich Z, et al.. Evolutionary conservation of the fidelity of transcription.. Nat Commun (2023). paper; opened 2026-11-17 all records citing this
  15. Vilborg A, Passarelli MC, Yario TA, et al.. Widespread Inducible Transcription Downstream of Human Genes.. Mol Cell (2015). paper; opened 2026-11-17 all records citing this
  16. Nojima T, Gomes T, Grosso ARF, et al.. Mammalian NET-Seq Reveals Genome-wide Nascent Transcription Coupled to RNA Processing.. Cell (2015). paper; opened 2026-11-17 all records citing this
  17. Hunter S, Sigauke RF, Stanley JT, et al.. Protocol variations in run-on transcription dataset preparation produce detectable signatures in sequencing libraries.. BMC Genomics (2022). paper; opened 2026-11-17 all records citing this
  18. Mahat DB, Kwak H, Booth GT, et al.. Base-pair-resolution genome-wide mapping of active RNA polymerases using precision nuclear run-on (PRO-seq).. Nat Protoc (2016). paper; opened 2026-11-17 all records citing this
Where it sits on the map
History of this record (7)

This record has been changed 7 times, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.

Status
partly written
Last updated
2026-11-19
Record
content/concepts/transcription-by-rna-polymerase-ii.json