henhaobio
Sign in

bioRxivGenomicsPreprint

A rapid, sensitive, scalable method for Precision Run-On sequencing (PRO-seq)

Julius JuddiD, Luke WojenskiiD, Lauren M. WainmaniD, Nathaniel D. TippensiD, Edward J. Rice, Alexis DziubekiD, Geno J. VillafanoiD, Erin M. WissinkiD, Philip P. VersluisiD, Lina Bagepalli, Sagar R. Shah, Dig Bijay MahatiD, Jacob M. TomeiD, Charles G. DankoiD, John T. LisiD, Leighton J. CoreiD

PDFDOICode
In-depth full-text digest available · Jump to digest ↓

Abstract

Tracking active transcription with the nuclear run-on (NRO) assays has been instrumental in uncovering mechanisms of gene regulation. The coupling of NROs with high-throughput sequencing has facilitated the discovery of previously unannotated or undetectable RNA classes genome-wide. Precision run-on sequencing (PRO-seq) is a run-on variant that maps polymerase active sites with nucleotide or near-nucleotide resolution. One main drawback to this and many other nascent RNA detection methods is the somewhat intimidating multi-day workflow associated with creating the libraries suitable for high-throughput sequencing. Here, we present an improved PRO-seq protocol where many of the enzymatic steps are carried out while the biotinylated NRO RNA remains bound to streptavidin-coated magnetic beads. These adaptations reduce time, sample loss and RNA degradation, and we demonstrate that the resulting libraries are of the same quality as libraries generated using the original published protocol. The assay is also more sensitive which permits reproducible, high-quality libraries from 104-105 cells instead of 106-107. Altogether, the improved protocol is more tractable allows for nascent RNA profiling from small samples, such as rare samples or FACS sorted cell populations.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (19 May 2020) · CC BY 4.0

A precision run-on sequencing (PRO-seq) variant with one streptavidin bead binding and on-bead enzymatic steps yields adapter-ligated cDNA in about 10 h and libraries from 0.05–0.25 × 10^6 cells.

Precision run-on sequencing (PRO-seq) maps transcriptionally engaged RNA polymerases at near-nucleotide resolution, but the published protocol takes 4-5 working days, three bead-binding steps, five phenol:chloroform extractions and 0.5–2 × 10^7 cells, limiting adoption and work on rare samples. qPRO-seq cuts this to one bead binding with washes replacing organic extractions, adds dual unique molecular identifiers (UMIs), and avoids PAGE purification by titrating adapters.

Key results

  1. ShownqPRO-seq data from 10^6 K562 cells match original PRO-seq in quality. Compared with published, deeply sequenced K562 PRO-seq, the choice of protocol explains almost none of the variance, and promoter and enhancer profiles match (pause indices aside; see below).

    Two biological replicates per protocol. Principal component analysis attributed >99% of variance in gene-body and promoter signal to batch and/or sequencing depth, and less than 0.3% to protocol (PRO-seq vs qPRO-seq). Aggregate promoter and enhancer (dREG peaks >10 kb from a promoter) profiles showed no detectable difference; gene-body and promoter RPM were reproducible, and dREG found similar element sets. · 1. Introduction; Figure legends

  2. SuggestedLibraries can be made from 0.05 to 0.25 × 10^6 cells. Quality improves up to about 10^6 cells and changes little beyond that, but low-input data are sparser.

    The 250k-cell K562 qPRO-seq libraries had fewer unique reads and sampled polymerases at fewer positions. This gave spiky RPM tracks (ANLN, SRRM1) and inflated counts at highly expressed genes, though transcription was still seen at many enhancers. The authors report working down to 50k primary human cells but show no data. The abstract cites 10^4 to 10^5 cells vs 10^6 to 10^7 before. · Abstract; 1. Introduction; 4. Notes; Figure legends

  3. ShownLibrary preparation goes from three bead bindings plus five phenol:chloroform extractions and ethanol precipitations to a single bead binding, taking permeabilized cells to adapter-ligated cDNA within one day.

    Original protocol: 4-5 working days. qPRO-seq: about 10 h up to adapter-ligated cDNA. 3' adapter ligation to base-hydrolyzed total RNA was as efficient as standard ligation to purified nascent RNA (32P-labelled RNA, 1 h ligation). Ligation efficiency to synthetic biotinylated RNA in 15% PEG8000 was also tested. The text gives no numeric efficiencies. · 1. Introduction; Figure legends

  4. ShownA single MyOne C1 streptavidin enrichment after adapter ligation to bulk RNA still selects nascent RNA, with no sign of extra mature mRNA carry-over.

    Excess 32P-labelled non-biotinylated RNA was mixed with biotinylated RNA and the fractions measured by liquid scintillation counting; the biotinylated species was enriched. Genome-wide length-normalized exon-to-intron read ratios showed no detectable difference between PRO-seq and qPRO-seq. · 1. Introduction; Figure legends

  5. SuggestedqPRO-seq gives lower pause indices than original PRO-seq. The authors attribute this to C1 beads having less bias against long RNAs than M280 beads, so they capture more gene-body signal.

    Pause index distributions across all genes shifted lower in qPRO-seq. Pause index is pause-region signal divided by length-normalized gene-body signal, so more gene-body capture with unchanged pause-region capture lowers it. The text gives no summary statistic for the shift. · 1. Introduction; Figure legends

Methods, data and reproducibility

Modified PRO-seq: biotin-NTP run-on in permeabilized cells, base hydrolysis and 3' adapter ligation on total RNA, one MyOne C1 streptavidin enrichment, then on-bead kinase repair, RppH decapping and 5' adapter ligation, TRIzol elution, reverse transcription and indexed PCR with UMI adapters. Compared side by side with the original protocol in K562 cells and with published K562 data.

  • K562 qPRO-seq and original PRO-seq libraries made side by side from 10^6 cells each (GEO: raw reads, bigWig files, dREG peak files) · human (K562 is a human cell line) · n = two biological replicates per protocol · GSE150625
  • K562 low-input qPRO-seq library · human (K562 is a human cell line) · n = 0.25 x 10^6 cells · GSE150625
  • Published, deeply sequenced K562 PRO-seq data used as the reference in principal component analysis · human (K562 is a human cell line)
  • 32P-labelled nascent RNA and synthetic biotinylated RNA used to test ligation efficiency and bead stringency
Methods & tools
PRO-seq, dREG, Dynabeads MyOne Streptavidin C1 beads, T4 RNA ligase, RppH, T4 polynucleotide kinase, Maxima H Minus reverse transcriptase, Q5 High-Fidelity DNA polymerase, Norgen Total RNA Purification Kit, TRIzol LS, AMPure XP SPRI beads, Bioanalyzer
Software & environment
Reagents and kits are listed with vendor catalogue numbers. The protocol is also posted at https://dx.doi.org/10.17504/protocols.io.57dg9i6, which the authors say will be updated. No software versions are stated.
Reproducibility gaps
The text states no alignment parameters or software versions (these are left to the GitHub repositories). The Table 1 oligonucleotide sequences are not in the provided text.
Limitations (5)
  • Ligating adapters to bulk RNA and using fewer affinity purifications could, in principle, lower specificity for nascent RNA. The authors raise this risk and test it with C1 bead stringency and exon/intron read ratios. · stated by the authors
  • Low-input libraries (0.25 × 10^6 cells) have fewer unique reads and look spiky, which inflates per-million-normalized counts at highly expressed genes. The cell number needed depends on the cell type's transcriptional activity and genome size. · stated by the authors
  • On-bead reverse transcription removes all organic extraction but markedly lowers library yield and increases adapter dimer. It is advised only with abundant input (10^7 cells) when speed matters most. · stated by the authors
  • Pause indices come out lower than with original PRO-seq, which the authors attribute to C1 beads capturing gene-body (long) RNAs better than M280 beads. Pause indices from the two protocols are therefore not directly comparable. · noted in this summary
  • Validation rests on one cell line (K562): two biological replicates at 10^6 cells, plus a 250k-cell condition shown only as browser tracks and metaprofiles. No data below 2.5 × 10^5 cells are shown, so the abstract's lower bound of 10^4 cells and the 50k primary-cell claim are not backed by data in the text. · noted in this summary

Read this if you want nascent transcription maps from limited material (around 10^5 to 10^6 cells), or a PRO-seq protocol that does most of library preparation in one day, with bench notes and a side-by-side check in K562.

Citations
69

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISBarbieri '20Scott '22Cingaram '25Smith '21Vill '22Pastore '26
77% semantic matchGenomics
Elisa Barbieri, Connor Hill, Mathieu Quesnel-Vallières+3 more·Cell Reports·Nov 1, 2020· 34 citations