Thiol-linked alkylation of RNA to assess expression dynamics

Abstract

Gene expression profiling by high-throughput sequencing reveals qualitative and quantitative changes in RNA species at steady state but obscures the intracellular dynamics of RNA transcription, processing and decay. We developed thiol(SH)-linked alkylation for the metabolic sequencing of RNA (SLAM seq), an orthogonal-chemistry-based RNA sequencing technology that detects 4-thiouridine (s4U) incorporation in RNA species at single-nucleotide resolution. In combination with well-established metabolic RNA labeling protocols and coupled to standard, low-input, high-throughput RNA sequencing methods, SLAM seq enabled rapid access to RNA-polymerase-II-dependent gene expression dynamics in the context of total RNA. We validated the method in mouse embryonic stem cells by showing that the RNA-polymerase-II-dependent transcriptional output scaled with Oct4/Sox2/Nanog-defined enhancer activity, and we provide quantitative and mechanistic evidence for transcript-specific RNA turnover mediated by post-transcriptional gene regulatory pathways initiated by microRNAs and N6-methyladenosine. SLAM seq facilitates the dissection of fundamental mechanisms that control gene expression in an accessible, cost-effective and scalable manner.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Figure 1: Detection of s4U by chemical derivatization and sequencing.
Figure 2: Thiol-linked alkylation for the metabolic sequencing of RNA (SLAM seq).
Figure 3: Quantitative description of the polyadenylated transcriptional output in mESCs.
Figure 4: Global and transcript-specific mRNA stability in mESCs.
Figure 5: Molecular determinants of mRNA stability in mESCs.

Accession codes

Primary accessions

Gene Expression Omnibus

References

  1. 1

    Schwanhäusser, B. et al. Global quantification of mammalian gene expression control. Nature 473, 337–342 (2011).

    PubMed  Article  CAS  PubMed Central  Google Scholar 

  2. 2

    Rabani, M. et al. High-resolution sequencing and modeling identifies distinct dynamic RNA regulatory strategies. Cell 159, 1698–1710 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  3. 3

    Yao, S.Y.M., Ng, A.M.L., Cass, C.E., Baldwin, S.A. & Young, J.D. Nucleobase transport by human equilibrative nucleoside transporter 1 (hENT1). J. Biol. Chem. 286, 32552–32562 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  4. 4

    Cleary, M.D., Meiering, C.D., Jan, E., Guymon, R. & Boothroyd, J.C. Biosynthetic labeling of RNA with uracil phosphoribosyltransferase allows cell-specific microarray analysis of mRNA synthesis and decay. Nat. Biotechnol. 23, 232–237 (2005).

    CAS  PubMed  Article  Google Scholar 

  5. 5

    Dölken, L. et al. High-resolution gene expression profiling for simultaneous kinetic parameter analysis of RNA synthesis and decay. RNA 14, 1959–1972 (2008).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  6. 6

    Miller, C. et al. Dynamic transcriptome analysis measures rates of mRNA synthesis and decay in yeast. Mol. Syst. Biol. 7, 458 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  7. 7

    Duffy, E.E. et al. Tracking distinct RNA populations using efficient and reversible covalent chemistry. Mol. Cell 59, 858–866 (2015).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  8. 8

    Schwalb, B. et al. TT-seq maps the human transient transcriptome. Science 352, 1225–1228 (2016).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  9. 9

    Miller, M.R., Robinson, K.J., Cleary, M.D. & Doe, C.Q. TU-tagging: cell-type-specific RNA isolation from intact complex tissues. Nat. Methods 6, 439–441 (2009).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  10. 10

    Rabani, M. et al. Metabolic labeling of RNA uncovers principles of RNA production and degradation dynamics in mammalian cells. Nat. Biotechnol. 29, 436–442 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  11. 11

    Gay, L. et al. Mouse TU tagging: a chemical/genetic intersectional method for purifying cell-type-specific nascent RNA. Genes Dev. 27, 98–115 (2013).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  12. 12

    Sun, M. et al. Comparative dynamic transcriptome analysis (cDTA) reveals mutual feedback between mRNA synthesis and degradation. Genome Res. 22, 1350–1359 (2012).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  13. 13

    Li, X., Xiong, X. & Yi, C. Epitranscriptome sequencing technologies: decoding RNA modifications. Nat. Methods 14, 23–31 (2016).

    PubMed  Article  CAS  Google Scholar 

  14. 14

    Heiss, M. & Kellner, S. Detection of nucleic acid modifications by chemical reagents. RNA Biol. 0, 1–9 (2016).

    Google Scholar 

  15. 15

    Shugart, L. Effect of selective chemical modification of 4-thiouridine of phenylalanine transfer ribonucleic acid on enzyme recognition. Arch. Biochem. Biophys. 148, 488–495 (1972).

    CAS  PubMed  Article  Google Scholar 

  16. 16

    Ong, C.-T. & Corces, V.G. Enhancer function: new insights into the regulation of tissue-specific gene expression. Nat. Rev. Genet. 12, 283–293 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  17. 17

    Young, R.A. Control of the embryonic stem cell state. Cell 144, 940–954 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  18. 18

    Whyte, W.A. et al. Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell 153, 307–319 (2013).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  19. 19

    Pott, S. & Lieb, J.D. What are super-enhancers? Nat. Genet. 47, 8–12 (2015).

    CAS  PubMed  Article  Google Scholar 

  20. 20

    Min, I.M. et al. Regulating RNA polymerase pausing and transcription elongation in embryonic stem cells. Genes Dev. 25, 742–754 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  21. 21

    Tani, H. & Akimitsu, N. Genome-wide technology for determining RNA stability in mammalian cells: historical perspective and recent advantages based on modified nucleotide labeling. RNA Biol. 9, 1233–1238 (2012).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  22. 22

    Bartel, D.P. MicroRNAs: target recognition and regulatory functions. Cell 136, 215–233 (2009).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  23. 23

    Huntzinger, E. & Izaurralde, E. Gene silencing by microRNAs: contributions of translational repression and mRNA decay. Nat. Rev. Genet. 12, 99–110 (2011).

    CAS  Article  PubMed  Google Scholar 

  24. 24

    Ameres, S.L. & Zamore, P.D. Diversifying microRNA sequence and function. Nat. Rev. Mol. Cell Biol. 14, 475–488 (2013).

    CAS  PubMed  Article  Google Scholar 

  25. 25

    Mullokandov, G. et al. High-throughput assessment of microRNA activity and function using microRNA sensor and decoy libraries. Nat. Methods 9, 840–846 (2012).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  26. 26

    Meyer, K.D. & Jaffrey, S.R. The dynamic epitranscriptome: N6-methyladenosine and gene expression control. Nat. Rev. Mol. Cell Biol. 15, 313–326 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  27. 27

    Yue, Y., Liu, J. & He, C. RNA N6-methyladenosine methylation in post-transcriptional gene expression regulation. Genes Dev. 29, 1343–1355 (2015).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  28. 28

    Batista, P.J. et al. m(6)A RNA modification controls cell fate transition in mammalian embryonic stem cells. Cell Stem Cell 15, 707–719 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  29. 29

    Geula, S. et al. Stem cells. m6A mRNA methylation facilitates resolution of naïve pluripotency toward differentiation. Science 347, 1002–1006 (2015).

    CAS  PubMed  Article  Google Scholar 

  30. 30

    Fu, Y., Dominissini, D., Rechavi, G. & He, C. Gene expression regulation mediated through reversible m6A RNA methylation. Nat. Rev. Genet. 15, 293–306 (2014).

    CAS  PubMed  Article  Google Scholar 

  31. 31

    Wang, X. et al. Structural basis of N(6)-adenosine methylation by the METTL3-METTL14 complex. Nature 534, 575–578 (2016).

    CAS  PubMed  Article  Google Scholar 

  32. 32

    Ke, S. et al. m(6)A mRNA modifications are deposited in nascent pre-mRNA and are not required for splicing but do specify cytoplasmic turnover. Genes Dev. 31, 990–1006 (2017).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  33. 33

    Limbach, P.A., Crain, P.F. & McCloskey, J.A. Summary: the modified nucleosides of RNA. Nucleic Acids Res. 22, 2183–2196 (1994).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  34. 34

    Burger, K. et al. 4-thiouridine inhibits rRNA synthesis and causes a nucleolar stress response. RNA Biol. 10, 1623–1630 (2013).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  35. 35

    Hafner, M. et al. Transcriptome-wide identification of RNA-binding protein and microRNA target sites by PAR-CLIP. Cell 141, 129–141 (2010).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  36. 36

    Herzog, V.A et al. Thiol-linked alkylation for the metabolic sequencing of RNA (SLAMseq). Protoc. Exch. https://doi.org/10.1038/protex.2017.105 (2017).

  37. 37

    Spitzer, J. et al. PAR-CLIP (photoactivatable ribonucleoside-enhanced crosslinking and immunoprecipitation): a step-by-step protocol to the transcriptome-wide identification of binding sites of RNA-binding proteins. Methods Enzymol. 539, 113–161 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  38. 38

    Nilsen, T.W. RNA sequencing by primer extension. Cold Spring Harb. Protoc. 2013, 1182–1185 (2013).

    PubMed  Google Scholar 

  39. 39

    Elling, U. et al. Forward and reverse genetics through derivation of haploid mouse embryonic stem cells. Cell Stem Cell 9, 563–574 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  40. 40

    Mohn, F., Sienski, G., Handler, D. & Brennecke, J. The rhino-deadlock-cutoff complex licenses noncanonical transcription of dual-strand piRNA clusters in Drosophila. Cell 157, 1364–1379 (2014).

    CAS  Article  PubMed  Google Scholar 

  41. 41

    Reimão-Pinto, M.M. et al. Uridylation of RNA hairpins by tailor confines the emergence of microRNAs in Drosophila. Mol. Cell 59, 203–216 (2015).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  42. 42

    Hodgkins, A. et al. WGE: a CRISPR database for genome engineering. Bioinformatics 31, 3078–3080 (2015).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  43. 43

    Sanjana, N.E., Shalem, O. & Zhang, F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods 11, 783–784 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  44. 44

    Han, B.W., Hung, J.-H., Weng, Z., Zamore, P.D. & Ameres, S.L. The 3′-to-5′ exoribonuclease Nibbler shapes the 3′ ends of microRNAs bound to Drosophila Argonaute1. Curr. Biol. 21, 1878–1887 (2011).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  45. 45

    Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).

    CAS  Article  Google Scholar 

  46. 46

    Liao, Y., Smyth, G.K. & Shi, W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics 30, 923–930 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  47. 47

    Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760 (2009).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  48. 48

    Trapnell, C., Pachter, L. & Salzberg, S.L. TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25, 1105–1111 (2009).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  49. 49

    Sedlazeck, F.J., Rescheneder, P. & von Haeseler, A. NextGenMap: fast and accurate read mapping in highly polymorphic genomes. Bioinformatics 29, 2790–2791 (2013).

    CAS  PubMed  Article  Google Scholar 

  50. 50

    Koboldt, D.C. et al. VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res. 22, 568–576 (2012).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  51. 51

    Quinlan, A.R. BEDTools: The swiss-army tool for genome feature analysis 27, 11.12.1–11.12.34 (John Wiley & Sons, 2002).

  52. 52

    Ramírez, F. et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 44, W1, W160 (2016).

    Article  CAS  Google Scholar 

  53. 53

    Thomas, P.D. et al. PANTHER: a library of protein families and subfamilies indexed by function. Genome Res. 13, 2129–2141 (2003).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  54. 54

    Carbon, S. et al. AmiGO: online access to ontology and annotation data. Bioinformatics 25, 288–289 (2009).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  55. 55

    Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. USA 102, 15545–15550 (2005).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  56. 56

    Mootha, V.K. et al. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat. Genet. 34, 267–273 (2003).

    CAS  PubMed  Article  Google Scholar 

  57. 57

    Friedman, R.C., Farh, K.K.-H., Burge, C.B. & Bartel, D.P. Most mammalian mRNAs are conserved targets of microRNAs. Genome Res. 19, 92–105 (2009).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

Download references

Acknowledgements

We thank J. Jude (Research Institute of Molecular Pathology, IMP) for generously providing a modified version of pLenti-CRISPR-v2-GFP, G. Krssakova and K. Mechtler (IMP, Vienna) for high-performance liquid chromatography analysis, IMP/Institute of Molecular Biotechnology (IMBA) Biooptics facility for FACS support, and all laboratory members for support and discussions. Mass spectrometry was performed at the Vienna Biocenter Core Facilities (VBCF) Metabolomics unit (http://www.vbcf.ac.at), funded by the City of Vienna through the Vienna Business Agency. High-throughput sequencing was performed at the VBCF NGS Unit (http://www.vbcf.ac.at). This work was supported by grants from the European Research Council to S.L.A. (ERC-StG-338252) and J.Z. (ERC-StG-336860) and the Austrian Science Fund to S.L.A (Y-733-B22 START, W-1207-B09 and SFB F43-22) and A.v.H. (W-1207-B09). The IMP is generously supported by Boehringer Ingelheim.

Author information

Affiliations

Authors

Contributions

V.A.H. and S.L.A. conceived the approach and wrote the paper. V.A.H., B.R. and S.L.A. developed the methods, performed the experiments and analyzed the data. W.W. performed initial s4U-alkylation experiments. T.N., P.R., V.A.H., J.Z., A.v.H. and S.L.A. developed SLAM-DUNK. P.B., T.R.B., V.A.H. and S.L.A. performed mRNA 3′-end annotation.

Corresponding author

Correspondence to Stefan L Ameres.

Ethics declarations

Competing interests

V.A.H., B.R. and S.L.A. declare competing financial interest. A patent application related to this work has been filed.

Integrated supplementary information

Supplementary Figure 1 4-thiouracil-derivatization by thiol-linked alkylation.

(a) 4-thiouracil (s4Uracil) reacts with the thiol-reactive compound iodoacetamide (IAA), attaching a carboxyamidomethyl-group to the thiol-group in s4Uracil because of a nucleophilic substitution (SN2) reaction. Absorption maxima of educt (4-thiouracil; s4Uracil; λmax 335 nm) and product (carboxyamidomethylated 4-thiouracil; *s4Uracil; λmax 297 nm) are indicated. (b) Absorption spectra of 4-thiouracil (s4Uracil) in the absence and presence of the indicated concentration of iodoacetamide (IAA). 1 mM s4U was incubated with the indicated concentration of IAA for 1 h at 37˚C in the presence of 50 mM sodium phosphate buffer (pH 8.0) and 10% DMSO. Data represents mean ± SD of independent experiments (untreated n=13; 1mM & 10 mM IAA n=3). (c) Quantification of absorption at 335 nm as shown in (B). P-values (two-tailed Student’s t-test) are indicated. (d) Absorption spectra of 1 mM 4-thiouracil (s4Uracil) in the absence and presence of 10 mM iodoacetamide (IAA) after incubation at the indicated temperature for 5 min in the presence of 50 mM sodium phosphate buffer (pH 8.0) and 10% DMSO. Data represents mean ± SD of independent experiments (untreated n=9; 37˚C, 50˚C, 80˚C, n=3). (e) Quantification of absorption at 335 nm as shown in (d). P-values (two-tailed Student’s t-test) are indicated. (f) Absorption spectra of 1 mM 4-thiouracil (s4Uracil) in the absence and presence of 10 mM iodoacetamide (IAA) after incubation at 37˚C for the indicated time in the presence of 50 mM sodium phosphate buffer (pH 8.0) and 10% DMSO. Data represents mean ± SD of independent experiments (untreated n=15; 5 min, 15 min, 75 min n=3). (g) Quantification of absorption at 335 nm as shown in (f). P-values (two-tailed Student’s t-test) are indicated. (h) Absorption spectra of 1 mM 4-thiouracil (s4Uracil) in the absence and presence of 10 mM iodoacetamide (IAA) after incubation at 50˚C for 2 min in the presence of 50 mM sodium phosphate buffer (pH 8.0) and the indicated amount of DMSO. Data represents mean ± SD of independent experiments (untreated n=13; 10%, 30 %, 50% DMSO n=3). (i) Quantification of absorption at 335 nm as shown in (H). P-values (two-tailed Student’s t-test) are indicated. (j) Absorption spectra of 1 mM 4-thiouracil (s4Uracil) in the absence and presence of 10 mM iodoacetamide (IAA) after incubation at 50˚C for 5 min in the presence of 50 mM sodium phosphate buffer with the indicated pH and the 10 % DMSO. Data represents mean ± SD of independent experiments (untreated n=25; pH6, 7, 8 n=3). (k) Quantification of absorption at 335 nm as shown in (j). P-values (two-tailed Student’s t-test) are indicated. (l) Absorption spectra of 1 mM 4-thiouracil (s4Uracil) in the absence and presence of 10 mM iodoacetamide (IAA) after incubation at 50˚C for 15 min in the presence of 50 mM sodium phosphate buffer (pH 8.0) and 50 % DMSO (optimal reaction [rxn] conditions). Data represent mean ± SD of independent experiments (untreated n=13; optimal reaction conditions n=3). (m) Quantification of absorption at 335 nm as shown in (j). P-values (two-tailed Student’s t-test) are indicated. Points represent individual measurements. Mean (center line) ± SD (whiskers) is indicated.

Supplementary Figure 2 4-thiouridine-derivatization by thiol-linked alkylation.

(a) 4-thiouridine (s4U) reacts with the thiol-reactive compound iodoacetamide (IAA), attaching a carboxyamidomethyl-group to the thiol-group in s4U because of a nucleophilic substitution (SN2) reaction. (b) Analysis of s4U-alkylation by mass spectrometry. 40 nmol 4-thiouridine were incubated with the indicated concentration of iodoacetamide in standard reaction buffer (50 mM NaPO4 (pH 8), 50 % DMSO) at 50°C for 15 minutes. The reaction was stopped with 1% acetic acid and analyzed by mass spectrometry. Normalized signal intensity of representative examples is shown. (c) Quantification of one (5 mM; 250 mM) or two (0 mM, 1 mM, 10 mM, 100 mM) independent experiments with two technical replicates shown in (b). Fraction alkylated s4U at indicated IAA concentrations represent relative normalized signal intensities at peak retention times of s4U and alkylated s4U. Points represent individual measurements. Mean (center line) ± SD (whiskers) is indicated.

Supplementary Figure 3 Alkylation of 4-thiouridine-containing RNA does not affect reverse transcription processivity.

(a) To determine the effect of s4U-alkylation on reverse transcriptase-processivity we employed a synthetic 76 nt long RNA that contains 4-thiouridine (s4U) incorporation at a single position (p9) within the sequence of the Drosophila small RNA dme-let-7, flanked by 5′ and 3′ adapter sequences. Reverse transcription was assayed before and after treatment with iodoacetamide (IAA) using commercially available reverse transcriptases by following the extension of a 5′ 32P –labeled DNA oligonucleotide, reverse and complement in sequence to the 3′ adapter sequence. (b) Reactions as described in (a) were analyzed by polyacrylamide gel electrophoresis followed by phosphorimaging. Primer extension results of s4U-containing and non-containing RNA in the presence and absence of IAA-treatment, conducted with the reverse transcriptase Superscript II (SSII), Superscript III (SSIII) or Quant-seq RT (QS) are depicted. The sequence of the RNA component excluding adapter sequences are shown; the position of the s4U residue is indicated in red. RNA sequencing lanes were generated by addition of the indicated ddNTPs to the reverse transcription reaction. PR, 5′ 32P –labeled DNA primer; bg, background stop signal; *p9, termination signal at position 9; FL, full length product. (c) Quantification of three independent experiments shown in (b). Ratio of drop off signal (+ vs – IAA treatment) at p9 after normalization to preceding background drop off signal was determined for control and s4U-containing RNA employing the indicated reverse transcriptase. Points represent individual experiments. Mean (center line) ± SD (whiskers) is indicated. P-Value (two-tailed Student’s t-test) is shown. N.s., not significant (p>0.05).

Supplementary Figure 4 Alkylation enables the quantitative identification of s4U-incorporations in RNA at single nucleotide resolution.

(a) RNA with or without 4-thiouridine (s4U) incorporation at a single position (p9) was treated with iodoacetamide (IAA) and subjected to reverse transcription and gel-extraction of full-length product followed by PCR amplification and high-throughput (HTP) sequencing. (b) Conversion rates for each position of a control RNA (left panels) and a s4U-containing RNA (right panels) in the presence or absence of iodoacetamide (IAA) treatment employing the indicated reverse transcriptase are shown. Points represent individual experiments (n=3). Average conversion rates (center line) ± SD (whiskers) are shown. Number of sequenced reads in each replicate (r1-r3) are indicated. Nucleotide identity occurrence at p9 is shown. (c) Conversion rates for the indicated conversions in the presence or absence of iodoacetamide (IAA) treatment employing Superscript II (SSII), Superscript III (SSIII), or Quant-seq reverse transcriptase (QS). Conversion rates were averaged across positions with the same nucleotide identity for both, s4U-containing and non-containing RNA oligonucleotides. Number of positions is indicated. P-Values (two-tailed Student’s t-test) are indicated. N.s., not significant (p>0.05).

Supplementary Figure 5 Effect of s4U treatment on mESC viability and metabolic RNA labeling.

(a) Viability of mESCs cultured in the presence of the indicated concentration of 4-thiouridine (s4U) for 12 h (left) or 24 h (right) relative to untreated conditions is shown. Note, that s4U-containing media was exchanged every 3h in the course of the labeling experiment. Final concentration used in subsequent experiments (100 μM) is indicated by triangle and dotted line. Mean ± SD of three independent cell cultures are shown. (b) Quantification of s4U-incorporation into total RNA after s4U-metabolic labeling incubated for the indicated time in a pulse, or following media replacement after 24 h pulse labeling followed by uridine-chase for the indicated times. s4U-incorporation was determined by HPLC analysis following digestion and dephosphorylation of total RNA to single nucleosides. Representative absorbance spectrum of background-subtracted s4U signal intensities normalized to 24 h pulse labeling time point is shown. Column retention time (min) relative to s4U-adsorption maxima are shown. (c) Substitution rate of s4U compared to unmodified uridine determined by HPLC as shown in (b) and previously described (Spitzer et al., 2014). s4U incorporation in total RNA across all time points of a s4U-metablic pulse and chase labeling experiment in mESCs. Values represent mean ± SD of three independent cell cultures. Maximum incorporation rates after 24 h labeling are indicated. (Note, that median s4U incorporation frequencies for mRNA [Fig. 2c] are higher than estimated by HPLC analysis of single nucleoside- digested total RNA, most certainly because stable RNA polymerase I and III transcripts, such as rRNA and tRNA, are strongly overrepresented in total RNA but depleted from RNA polymerase II-specific Quant-seq libraries. See also Supplementary Fig. 9.)Spitzer, J. et al. PAR-CLIP (Photoactivatable Ribonucleoside-Enhanced Crosslinking and Immunoprecipitation): a step-by-step protocol to the transcriptome-wide identification of binding sites of RNA-binding proteins. Meth Enzymol 539, 113–161 (2014).

Supplementary Figure 6 Validation of mRNA 3′ end sequencing using Quant-seq.

(a) Quant-seq uses total RNA as input, hence no prior poly(A) enrichment or rRNA depletion is required. Library generation is initiated by oligo-dT-anchor priming. Primer already contains Illumina-compatible linker sequences (green). After first-strand synthesis, the RNA is removed and second-strand synthesis is initiated by random priming and a DNA polymerase. The random primer also contains Illumina-compatible linker sequences (shown in blue). No purification is required between first and second strand synthesis. The insert size is optimized for shorter reads (SR50 or SR100). Second strand synthesis is followed by a magnetic bead-based purification step. The library is then amplified, introducing the sequences required for cluster generation (shown in red and purple). External barcodes (BC) for multiplexing are introduced during the PCR amplification step. (b) Normalized coverage of poly(A)-selected RNA-seq (top three panels) and Quant-seq (bottom three panels) generated from mESCs and analyzed across 24.440 Entrez genes. Coverage is reported for 500 bp regions surrounding the 5′ end or the 3′ end of all genes. Gene body coverage was normalized to gene length (including introns). (c) Gene expression values determined by Quant-seq correlate with RNA-seq-derived, transcript-length-normalized expression in mESCs. Length normalized RNA-seq signal (in TPM) is compared to gene-tag values (counts per million, CPM) for 17.821 genes. Average of three independent cell culture replicates is shown. Spearman correlation coefficient (rS) and associated p-Value is indicated. (d) Quant-seq produces highly-reproducible gene expression values. Comparison of 20.818 transcripts from three independent biological Quant-seq replicates (r#1-r#3) generated from total RNA of mESCs are shown. Spearman correlation coefficient (rS) and associated p-Value are indicated.

Supplementary Figure 7 Expanded mRNA 3′ end annotation in mESCs based on Quant-seq data.

(a) Nucleotide profile (left) and frequency of polyA-site motif (PAS, right) for RefSeq-annotated mRNA 3′ ends. Location of mRNA 3′ end-associated motifs, i.e. U-rich upstream sequence element (USE), poly-A-signal motif (PAS), Fip1-binding site (Fip1) and U-rich downstream sequence element (DSE) are indicated. (b) Nucleotide profile (left) and frequency of polyA-site motif (PAS, right) for RefSeq-overlapping mRNA 3′ ends determined by Quant-seq (see Online Methods for details) of transcripts expressed at >5cpm. (c) Nucleotide profile (left) and frequency of polyA-site motif (PAS, right) for de-novo annotated mRNA 3′ ends as determined by Quant-seq of transcripts expressed at >5cpm. Note, that de-novo annotated 3′ ends are supported by both nucleotide profile and PAS-motif (compare with (a)). (d) Overview of expanded 3′ end annotation in mESCs. Distribution of Quant-seq supported UTRs among the displayed types of alternative isoforms. For single UTRs, the number of isoforms overlapping with RefSeq annotations (+/- 5 nts) is indicated. The nucleotide profile and frequency of polyA-site motif for each UTR type is shown. (e) Genome browser plot of single-UTR example gene. Normalized sequencing tracks (in reads per million, RPM) for RNA-seq (black), Cap-seq (brown) and Quant-seq (green) are shown. (f) Meta-analysis of RNA-seq signal upstream and downstream of Quant-seq-supported mRNA 3′ ends. 6382 genes with single-UTR were analyzed. Relative RNA-seq signal (in %) reports for each position the coverage relative to the total coverage within a 200 nt window (left). Quantification of cumulative RNA-seq signal within 100 bp upstream (us) or downstream (ds) of the annotated 3′ end is shown (right). P-value (Mann-Whitney test) is indicated. (g and j) Genome browser plot of tandem-UTR example gene. Normalized sequencing tracks for RNA-seq (black), Cap-seq (brown) and Quant-seq (green) (in RPM) are shown. Example for a gene with two (g) or three (j) UTRs is shown. (h) Meta-analysis of RNA-seq signal upstream and downstream of Quant-seq-supported mRNA 3′ ends. 824 genes with two annotated UTRs were analyzed. Rel. RNA-seq signal (%) reporting for each position the coverage relative to the total coverage of the 300 nt window analyzed (left). Quantification of cumulative RNA-seq signal within 100 bp upstream (us) of the proximal 3′ end, in-between the two 3′ ends (m) or 100 bp downstream (ds) of the distal 3′ end is shown (right). P-value (Mann-Whitney test) is indicated. (i) Analysis as shown in (h) but on centered 3′ ends. Quantification of cumulative RNA-seq signal within 100 bp upstream (us) or 100 bp downstream (ds) of the 3′ ends is shown (right). P-value (Mann-Whitney test) is indicated. (k) Meta-analysis of RNA-seq signal upstream and downstream of Quant-seq-supported mRNA 3′ ends. 92 genes with three annotated UTRs were analyzed around centered sites. Rel. RNA seq signal (%) reporting for each position the coverage relative to the total coverage of the 600 nt window analyzed (left). Quantification of cumulative RNA-seq signal within 100 bp upstream (us), in-between (m1 and m2), and 100 bp downstream (ds) of 3′ ends is shown (right). P-value (Mann-Whitney test) is indicated. (l) Analysis as shown in (k) but on centered 3′ ends. Quantification of cumulative RNA-seq signal within 100 bp upstream (us) or 100 bp downstream (ds) of 3′ ends is shown (right). P-value (Mann-Whitney test) is indicated.

Supplementary Figure 8 SLAMseq robustly differentiates s4U-labeled and unlabeled transcripts in the context of total RNA.

(a) Steady-state gene expression determined by Quant-seq generated from total RNA of mESCs subjected to s4U-metabolic labeling for 24 h in three independent cell cultures (r#1-r#3). All reads (sum of T>C containing and non-T>C containing) report steady-state gene expression. Spearman correlation coefficient (rS) and associated p-Values are indicated. (b) Relative abundance of background-subtracted T>C conversion-containing reads in Quant-seq libraries generated from total RNA of mESCs subjected to s4U-metabolic labeling for 24 h in three independent cell cultures (r#1-r#3). Spearman correlation coefficient (rS) and associated p-Values are indicated. (c) Relative abundance of labeled (T>C reads) and all reads (sum of T>C containing and non-T>C containing) shows uniform s4U labeling of transcripts in mESCs subjected to s4U-metabolic labeling for 24 h. Average of three independent cell cultures is shown. Spearman correlation coefficient (rS) and associated p-Values are indicated. (d) Treatment of unlabeled total RNA does not perturb gene expression values. Relative abundance of transcripts in Quant-seq libraries generated from total RNA of untreated mESCs before and after treatment with iodoacetamide. Average of three independent cell cultures is shown. Spearman correlation coefficient (rS) and associated p-Values are indicated. (e) Meta-analysis of three independent SLAMseq libraries generated from mESCs subjected to s4U-metabolic labeling for 24 h. Relative coverage across 250 nt counting windows at the 3′ end of transcripts, transcript U-content, as well as raw and U-content-normalized (norm.) conversion rates (Conv.) are shown for all counting windows expressed at steady-state >5 cpm (n=8408). (f) Conversion rates in counting window-mapping reads of Quant-seq libraries, prepared from mESCs before (no s4U) and after metabolic labeling for 24 h (+s4U) in three independent cell cultures are shown for all counting windows expressed at steady-state >5 cpm (n=8408). Dashed line represents expected background sequencing error rate. Outlieres are not shown. p-Value (Mann-Whitney test) is indicated.

Supplementary Figure 9 SLAMseq robustly identifies s4U incorporation events in poly-adenylated RNA.

4-thiouridine incorporation events in total RNA and polyA+ enriched RNA prepared from mESCs subjected to s4U labeling for 6 h was determined by LC-MS/MS (black) and compared to T>C conversions determined by SLAMseq mRNA 3′ end sequencing (red). Based on this analysis, SLAMseq recovers ~90 % of s4U incorporation events in polyadenylated RNA. Data represent mean (center line) ± SD (whiskers) of two independent cell cultures and two independent measurements (LC-MS/MS) or three independent cell cultures (Sequencing).

Supplementary Figure 10 Quantitative description of the polyadenylated transcriptional output in mESCs.

(a) Experimental setup for determining transcriptional mRNA output by SLAMseq, coupled to short (45 min) s4U-pulse labeling using 100 μM s4U. (b) Model of genes involved in maintenance of stem cell state (adapted from Young et al., 2011). (c) Steady state expression analysis does not quantitatively report enhancer activity. Steady state abundance (top) and transcriptional output, as measured in number of T>C conversion containing reads (bottom, see Fig. 3c), for expressed genes (steady-state >5cpm) without adjacent enhancer (no, n=4994), proximal to canonical Oct4/Sox2/Nanog enhancer (OSN, n=2029) or proximal to strong enhancers (SE, n=156). Outliers are not shown. P-values determined by Mann-Whitney test are indicated. (d) Transcriptional-output measurement by SLAMseq correlates with global run-on sequencing (GRO-seq). GRO-seq data was normalized to gene length, omitting signal derived from the first kilobase of each transcript to exclude polymerase-stalling events (Min et al., 2011) and compared to T>C conversion containing reads after SLAMseq of mESCs subjected for s4U metabolic labeling for 45 min (n=6205). Spearman correlation coefficient (rS) and associated p-Value are indicated.Young, R. A. Control of the Embryonic Stem Cell State. Cell 144, 940–954 (2011).Min, I. M. et al. Regulating RNA polymerase pausing and transcription elongation in embryonic stem cells. Genes Dev. 25, 742–754 (2011)

Supplementary Figure 11 Global and transcript-specific mRNA stability in mESCs.

(a) Experimental setup for profiling mRNA stability by SLAMseq. mESCs were subjected to metabolic labeling with s4U (100 μM final conc.) in 3h intervals for a total of 24 h, followed by a chase with unlabeled uridine (10 mM final conc.) for 0, 0.5, 1, 3, 6, 12, and 24 h, followed by total RNA preparation and SLAMseq. (b) Representative genome browser plots of the indicated genes represent SLAMseq data prepared from mESCs subjected to s4U-pulse/chase labeling as shown in (a). Sequencing data mapping to defined counting windows in libraries prepared from the indicated timepoint of the chase are shown. All mapped reads (steady-state, in RPM) are shown in black; T>C conversion-containing reads (labeled, in RPM) are indicated in red. (c) Global analysis of mRNA stability in mESCs. Correlation of steady-state gene expression (all reads) or T>C conversion containing reads at the indicated time with steady-state expression at time 0 of the chase. (d) Relationship between transcript-specific mRNA half-life and its physiological function. Genes associated with the GO terms “Regulation of Transcription” (GO:0006357), “Cell cycle” (GO:0007049), “Translation” (GO:0006412) and “Extracellular Matrix” (GO:0031012) was tested for enrichment among transcripts ranked according to stability determined by SLAMseq by GSEA (Subramanian et al., 2005). Normalized enrichment score (NES) and associated p-Value is shown. (e) Messenger RNA half-life measured by SLAMseq correlates with mRNA stabilities determined by transcriptional inhibition. mESCs were treated with Actinomycin D (ActD) to inhibit transcription, followed by Quant-seq analysis of gene expression. Half-life was determined from global transcription inhibition data by normalizing gene expression to the 50 most stable transcripts. Half-lives for 6665 transcripts are shown. Spearman correlation coefficient (rS) and associated p-Value are indicated.Subramanian, A. et al. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 102, 15545–15550 (2005).

Supplementary Figure 12 Global relationships between transcriptional output, mRNA stability, and steady-state gene expression in mESCs as determined by SLAMseq.

(a) Correlation between the rate of transcriptional output (Trx. Output) and steady-state abundance for 6442 genes in mESCs. Spearman correlation coefficient (r) and associated p-Value are indicated. (b) Correlation between half-life and steady-state abundance for 6442 genes in mESCs. Spearman correlation coefficient (r) and associated p-Value are indicated. (c) Correlation between half-life and the rate of transcriptional output for 6442 genes in mESCs. Spearman correlation coefficient is indicated. (d) Correlation between the rate of transcriptional output and the rate of RNA decay for 6442 genes in mESCs. Pearson’s correlation coefficient and associated p-Value are indicated.

Supplementary Figure 13 Impact of microRNAs on mRNA stability in mESCs.

(a) Schematic representation of microRNA (miRNA)-mediated gene regulation. Target site-types with increasing (top to bottom) repressive function are shown. (b) Abundance of microRNAs (miRNAs) in mESCs. Relative abundance was determined by small RNA sequencing. Members of the miR-291-3p/294-3p/295-3p/302-3p-family are highlighted in red. The miRNAs miR-292a-3p and miR-291b-3p (highlighted in rose) share a 6mer seed with the miR-291-3p/294-3p/295-3p/302-3p-family. Together, these miRNAs (referred to as “miR-291a-family”) represent more than half of all miRNAs in mESCs. (c) Schematic representation of the Xpo5 gene locus in the mouse genome. Zoom-in shows a region in the first coding exon of Xpo5, which was targeted by CRISPR/Cas9 genome editing. Disruptive frameshift mutations in three independent xpo5ko clones are shown. (d) Western blot analysis for Exportin-5 (Xpo5) and Actin in wild-type (wt) mESCs and three independent xpo5ko clones. n/a, alternative clones not used in this study. (e) Northern hybridization assay for miR-290-5p and U6 snRNA in wild-type mESCs and three independent xpo5ko clones. n/a, alternative clones not used in this study. (f) MicroRNA-abundance determined by spike-in-controlled small RNA sequencing in xpo5ko mESCs relative to wild-type cells. Ratio for all miRNAs (black), miR-291a-family (red), miR-290a-family (rose), miR-293-family (light grey) and miR-291-family (dark grey) are shown. Data represent mean (center line) ± SD (whiskers) of three independent xpo5ko clones shown in (c) and (d). (g) Abundance of miR-families in wildtype mESCs as determined by small RNA sequencing. Highlighted are the most abundant miR-families, which are all transcribed from the miR-290-295 cluster: Members of the miR-291-3p/294-3p/295-3p/302-3p and miR-292a-3p/467a-5p family (“miR-291a”, in red), miR-290a-5p/292a-5p-family (“miR-290a”, in rose), miR-293-3p-family (“miR-293”, in light grey) and miR-291-5p-family (“miR-291”, in dark grey). (h) Cumulative distribution of ranked mRNA stabilities. Plotted are distributions for transcripts harboring no predicted target sites for any miRNA belonging to the ten most highly expressed miR-families (black, n=2188); at least one predicted target site for miRNAs belonging to the miR-291a-family (red, n=1450); at least one predicted target site for miRNAs belonging to the miR-290a-family, but no site for miR-291a-family miRNAs (rose, n=1326); at least one predicted target site for miR-293, but no site for miR-291a-family or miR-290a-family miRNAs (light grey, n=117); at least one predicted target site for miR-291 miRNAs, but no predicted sites for miR-291a-family, miR-290a-family or miR-293 family miRNAs (dark grey, n=708). P-value was determined by KS-test. (i) Cumulative distribution of mRNA stability changes in xpo5ko relative to wt mESCs. Plotted are distributions for transcripts that have predicted miRNA target sites as described in (b). P-value was determined by KS-test. N.s., not significant (p>0.05).

Supplementary Figure 14 Impact of m6A modification on mRNA stability in mESCs.

(a) Schematic representation of adenosine (A) conversion into N6-methyladenosine (m6A) by the methyltransferase complex Mettl3/14. (b) Schematic representation of the Mettl3 gene locus in the mouse genome. Zoom in shows a region in the first coding exon of Mettl3, which was targeted by CRISPR/Cas9 genome editing. Disruptive frameshift mutations in three independent mettl3ko clones are shown. (c) Western blot analysis for Mettl3 and Actin in wild-type (wt) mESCs and three independent mettl3ko clones. n/a, alternative clones not used in this study. (d) Cumulative distribution of ranked mRNA stabilities as in Fig. 5d, but Mettl3-dependent m6A targets were defined based on m6A-CLIP data set (Ke et al., 2017). Plotted are distributions for transcripts that do (red, n=3864) or do not (black, n=2801) contain a m6A mark. P-value was determined by KS-test. (e) Cumulative distribution of ranked mRNA stabilities harboring no m6A site (black, n=2801) or at least one m6A site exclusively in the 5′ UTR (grey, n=25), the coding sequence (CDS, green, n=829) or the 3′ UTR (red, n=1550) defined by m6A-CLIP (Ke et al., 2017). (f) Cumulative distribution of mRNA stability changes in mettl3ko relative to wt mESCs. Plotted are distributions for transcripts harboring m6A sites as described in (b). P-value was determined by KS-test. N.s., not significant (p>0.05).Ke, S. et al. m6A mRNA modifications are deposited in nascent pre-mRNA and are not required for splicing but do specify cytoplasmic turnover. Genes Dev. 31, 990–1006 (2017).

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–14 and Supplementary Table 3. (PDF 3168 kb)

Life Sciences Reporting Summary (PDF 129 kb)

Supplementary Table 1

Transcriptional output for 7179 genes in mES cells. (XLS 589 kb)

Table is attached as Excel file and reports the steady-state abundance and background-subtracted T>C conversion containing reads, as determined by SLAM-seq mRNA 3′ end sequencing of mES cells subjected to s4U metabolic labeling for 45 min.

Supplementary Table 2

High-confidence half-life data for 6665 transcripts in mES cells. (XLS 1335 kb)

Table is attached as Excel file and reports for each transcript the chromosome, start and end of the counting windows identifying transcript 3′ ends, Name of the associated transcript, length of the counting window, half-life (h), decay rate (k; h-1), standard error of half-life and decay rate (error of single exponential fit), and accuracy of fit (rsquare).

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Herzog, V., Reichholf, B., Neumann, T. et al. Thiol-linked alkylation of RNA to assess expression dynamics. Nat Methods 14, 1198–1204 (2017). https://doi.org/10.1038/nmeth.4435

Download citation

Further reading

Search

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing