Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

ARM-seq: AlkB-facilitated RNA methylation sequencing reveals a complex landscape of modified tRNA fragments

Abstract

High-throughput RNA sequencing has accelerated discovery of the complex regulatory roles of small RNAs, but RNAs containing modified nucleosides may escape detection when those modifications interfere with reverse transcription during RNA-seq library preparation. Here we describe AlkB-facilitated RNA methylation sequencing (ARM-seq), which uses pretreatment with Escherichia coli AlkB to demethylate N1-methyladenosine (m1A), N3-methylcytidine (m3C) and N1-methylguanosine (m1G), all commonly found in tRNAs. Comparative methylation analysis using ARM-seq provides the first detailed, transcriptome-scale map of these modifications and reveals an abundance of previously undetected, methylated small RNAs derived from tRNAs. ARM-seq demonstrates that tRNA fragments accurately recapitulate the m1A modification state for well-characterized yeast tRNAs and generates new predictions for a large number of human tRNAs, including tRNA precursors and mitochondrial tRNAs. Thus, ARM-seq provides broad utility for identifying previously overlooked methyl-modified RNAs, can efficiently monitor methylation state and may reveal new roles for tRNA fragments as biomarkers or signaling molecules.

This is a preview of subscription content, access via your institution

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Prices may be subject to local taxes which are calculated during checkout

Figure 1: ARM-seq facilitates sequencing of m1A-, m3C- or m1G-modified RNAs.
Figure 2: ARM-seq reveals m1A-modified tRNA fragments in S. cerevisiae.
Figure 3: ARM-seq reveals methylated RNAs derived from human cytosolic tRNAs, tRNA precursors and mitochondrial tRNAs.

Accession codes

Primary accessions

Sequence Read Archive

References

  1. Lee, Y.S., Shibata, Y., Malhotra, A. & Dutta, A. A novel class of small RNAs: tRNA-derived RNA fragments (tRFs). Genes Dev. 23, 2639–2649 (2009).

    Article  CAS  Google Scholar 

  2. Haussecker, D. et al. Human tRNA-derived small RNAs in the global regulation of RNA silencing. RNA 16, 673–695 (2010).

    Article  CAS  Google Scholar 

  3. Cole, C. et al. Filtering of deep sequencing data reveals the existence of abundant Dicer-dependent small RNAs derived from tRNAs. RNA 15, 2147–2160 (2009).

    Article  CAS  Google Scholar 

  4. Hanada, T. et al. CLP1 links tRNA metabolism to progressive motor-neuron loss. Nature 495, 474–480 (2013).

    Article  CAS  Google Scholar 

  5. Maute, R.L. et al. tRNA-derived microRNA modulates proliferation and the DNA damage response and is down-regulated in B cell lymphoma. Proc. Natl. Acad. Sci. USA 110, 1404–1409 (2013).

    Article  CAS  Google Scholar 

  6. Mei, Y. et al. tRNA binds to cytochrome c and inhibits caspase activation. Mol. Cell 37, 668–678 (2010).

    Article  CAS  Google Scholar 

  7. Thompson, D.M. & Parker, R. Stressing out over tRNA cleavage. Cell 138, 215–219 (2009).

    Article  CAS  Google Scholar 

  8. Couvillion, M.T., Bounova, G., Purdom, E., Speed, T.P. & Collins, K. A Tetrahymena Piwi bound to mature tRNA 3′ fragments activates the exonuclease Xrn2 for RNA processing in the nucleus. Mol. Cell 48, 509–520 (2012).

    Article  CAS  Google Scholar 

  9. Selitsky, S.R. et al. Small tRNA-derived RNAs are increased and more abundant than microRNAs in chronic hepatitis B and C. Sci. Rep. 5, 7675 (2015).

    Article  Google Scholar 

  10. Motorin, Y., Muller, S., Behm-Ansmant, I. & Branlant, C. Identification of modified residues in RNAs by reverse transcription-based methods. Methods Enzymol. 425, 21–53 (2007).

    Article  CAS  Google Scholar 

  11. Phizicky, E.M. & Hopper, A.K. tRNA biology charges to the front. Genes Dev. 24, 1832–1860 (2010).

    Article  Google Scholar 

  12. Studte, P. et al. tRNA and protein methylase complexes mediate zymocin toxicity in yeast. Mol. Microbiol. 69, 1266–1277 (2008).

    Article  CAS  Google Scholar 

  13. Schaefer, M. et al. RNA methylation by Dnmt2 protects transfer RNAs against stress-induced cleavage. Genes Dev. 24, 1590–1595 (2010).

    Article  CAS  Google Scholar 

  14. Blanco, S. et al. Aberrant methylation of tRNAs links cellular stress to neuro-developmental disorders. EMBO J. 33, 2020–2039 (2014).

    Article  CAS  Google Scholar 

  15. Delaney, J.C. & Essigmann, J.M. Mutagenesis, genotoxicity, and repair of 1-methyladenine, 3-alkylcytosines, 1-methylguanine, and 3-methylthymine in alkB Escherichia coli. Proc. Natl. Acad. Sci. USA 101, 14051–14056 (2004).

    Article  CAS  Google Scholar 

  16. Aas, P.A. et al. Human and bacterial oxidative demethylases repair alkylation damage in both RNA and DNA. Nature 421, 859–863 (2003).

    Article  CAS  Google Scholar 

  17. Machnicka, M.A. et al. MODOMICS: a database of RNA modification pathways–2013 update. Nucleic Acids Res. 41, D262–D267 (2013).

    Article  CAS  Google Scholar 

  18. Falnes, P.Ø. Repair of 3-methylthymine and 1-methylguanine lesions by bacterial and human AlkB proteins. Nucleic Acids Res. 32, 6260–6267 (2004).

    Article  CAS  Google Scholar 

  19. Lowe, T.M. & Eddy, S.R. tRNAscan-SE: a program for improved detection of transfer RNA genes in genomic sequence. Nucleic Acids Res. 25, 955–964 (1997).

    Article  CAS  Google Scholar 

  20. Chan, P.P. & Lowe, T.M. GtRNAdb: a database of transfer RNA genes detected in genomic sequence. Nucleic Acids Res. 37, D93–D97 (2009).

    Article  CAS  Google Scholar 

  21. Hopper, A.K. Transfer RNA post-transcriptional processing, turnover, and subcellular dynamics in the yeast Saccharomyces cerevisiae. Genetics 194, 43–67 (2013).

    Article  CAS  Google Scholar 

  22. Anderson, J. et al. The essential Gcd10p-Gcd14p nuclear complex is required for 1-methyladenosine modification and maturation of initiator methionyl-tRNA. Genes Dev. 12, 3650–3662 (1998).

    Article  CAS  Google Scholar 

  23. Nishikura, K. & De Robertis, E.M. RNA processing in microinjected Xenopus oocytes. Sequential addition of base modifications in the spliced transfer RNA. J. Mol. Biol. 145, 405–420 (1981).

    Article  CAS  Google Scholar 

  24. Suzuki, T. & Suzuki, T. A complete landscape of post-transcriptional modifications in mammalian mitochondrial tRNAs. Nucleic Acids Res. 42, 7346–7357 (2014).

    Article  CAS  Google Scholar 

  25. Carlile, T.M. et al. Pseudouridine profiling reveals regulated mRNA pseudouridylation in yeast and human cells. Nature 515, 143–146 (2014).

    Article  CAS  Google Scholar 

  26. Schwartz, S. et al. Transcriptome-wide mapping reveals widespread dynamic-regulated pseudouridylation of ncRNA and mRNA. Cell 159, 148–162 (2014).

    Article  CAS  Google Scholar 

  27. Dominissini, D. et al. Topology of the human and mouse m6A RNA methylomes revealed by m6A-seq. Nature 485, 201–206 (2012).

    Article  CAS  Google Scholar 

  28. Schaefer, M., Pollex, T., Hanna, K. & Lyko, F. RNA cytosine methylation analysis by bisulfite sequencing. Nucleic Acids Res. 37, e12 (2009).

    Article  Google Scholar 

  29. Squires, J.E. et al. Widespread occurrence of 5-methylcytosine in human coding and non-coding RNA. Nucleic Acids Res. 40, 5023–5033 (2012).

    Article  CAS  Google Scholar 

  30. Yarham, J.W., Elson, J.L., Blakely, E.L., McFarland, R. & Taylor, R.W. Mitochondrial tRNA mutations and disease. Wiley Interdiscip. Rev. RNA 1, 304–324 (2010).

    Article  CAS  Google Scholar 

  31. Ryvkin, P. et al. HAMR: high-throughput annotation of modified ribonucleotides. RNA 19, 1684–1692 (2013).

    Article  CAS  Google Scholar 

  32. Findeiss, S., Langenberger, D., Stadler, P.F. & Hoffmann, S. Traces of post-transcriptional RNA modifications in deep sequencing data. Biol. Chem. 392, 305–313 (2011).

    Article  CAS  Google Scholar 

  33. Ebhardt, H.A. et al. Meta-analysis of small RNA-sequencing errors reveals ubiquitous post-transcriptional RNA modifications. Nucleic Acids Res. 37, 2461–2470 (2009).

    Article  CAS  Google Scholar 

  34. Kawaji, H. et al. Hidden layers of human small RNAs. BMC Genomics 9, 157 (2008).

    Article  Google Scholar 

  35. Ishimura, R. et al. Ribosome stalling induced by mutation of a CNS-specific tRNA causes neurodegeneration. Science 345, 455–459 (2014).

    Article  CAS  Google Scholar 

  36. D'Silva, S., Haider, S.J. & Phizicky, E.M. A domain of the actin binding protein Abp140 is the yeast methyltransferase responsible for 3-methylcytidine modification in the tRNA anti-codon loop. RNA 17, 1100–1110 (2011).

    Article  CAS  Google Scholar 

  37. Quartley, E. et al. Heterologous expression of L. major proteins in S. cerevisiae: a test of solubility, purity, and gene recoding. J. Struct. Funct. Genomics 10, 233–247 (2009).

    Article  CAS  Google Scholar 

  38. Langmead, B. & Salzberg, S.L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).

    Article  CAS  Google Scholar 

  39. Love, M.I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014).

    Article  Google Scholar 

  40. Griffiths-Jones, S., Grocock, R.J., van Dongen, S., Bateman, A. & Enright, A.J. miRBase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res. 34, D140–D144 (2006).

    Article  CAS  Google Scholar 

  41. RNAcentral Consortium. RNAcentral: an international database of ncRNA sequences. Nucleic Acids Res. 43, D123–D129 (2015).

Download references

Acknowledgements

We thank P. Chan for her assistance with technical edits and final figure improvements. This work was supported by the US National Institutes of Health (NIH) NHGRI grant 5R01HG006753 to T.M.L. E.Q. and E.M.P. were also supported by NIH grant GM052347 to E.M.P.

Author information

Authors and Affiliations

Authors

Contributions

E.M.P. and T.M.L. conceived the project; A.E.C., E.Q., A.D.H., E.H.-R., E.M.P. and T.M.L. designed and performed research; A.E.C. and A.D.H. contributed new analytic tools; A.E.C., E.M.P. and T.M.L. wrote the paper.

Corresponding authors

Correspondence to Eric M Phizicky or Todd M Lowe.

Ethics declarations

Competing interests

All authors are named on a provisional patent application describing this method, filed by the University of California.

Integrated supplementary information

Supplementary Figure 1 AlkB demethylates m1A and m3C

Primer extensions targeting the 3’ end of mature S. cerevisiae Thr-AGT tRNA show partial to complete demethylation of m1A58 and m3C32 as the ratio of AlkB enzyme to bulk RNA is varied from 0.12 – 2.0 (by mass). By contrast, bands corresponding to pauses or stops in reverse transcription at m5C48, D47 (dihydrouridine) and m2,2G26 modifications become more apparent when the blocks at m1A58 and m3C32 are removed. Similar results were observed with frozen AlkB used for ARM-seq (lanes 1-8), or freshly prepared enzyme (lanes 9-18). AlkB reaction buffer was prepared immediately prior to use (lanes 1-16); control reactions using 5 month old α-ketoglutarate and ferrous ammonium sulfate in the buffer showed no demethylation (lanes 17-18).

Supplementary Figure 2 ARM-seq results for S. cerevisiae tRNAs

ARM-seq log2 fold changes reported by DESeq2 (a, same as primary Figure 2d ), and read profiles for S. cerevisiae tRNAs expected to contain m1A58 (b) or unmodified A58 (c) based on documentation in Modomics. Panel c includes several tRNA subtypes (shown in grey) with more than three differences in nucleotide sequence relative to S. cerevisiae tRNAs in Modomics. S. cerevisiae tRNAs in isodecoder groups not currently represented in Modomics are shown in panel d. ARM-seq profiles for all significant responders (indicated by *) showed increases in 3′-fragments, where AlkB-sensitivity could be attributed to m1A58 modifications. Approximately 50% of reads for Ser-CGA could be attributed to a documented m3C32 modification; the remaining 3’-fragment reads revealed by ARM-seq were consistent with an undocumented m1A58 modification.

Supplementary Figure 3 ARM-seq reveals modified 5′ tRNA fragments

ARM-seq revealed modified 5′-fragments of several S. cerevisiae tRNAs, where the most prevalent AlkB substrate documented in Modomics is m1G at residue 9. These included both Pro-TGG subtypes and Val-AAC-2, where Modomics documents m1G9, as well as Pro-AGG, which is not represented in Modomics but contains G at residue 9. ARM-seq also revealed modified 5′-fragments of Phe-GAA tRNAs, where Modomics shows unmodified A9, and m2G10 (not a known AlkB substrate), suggesting the possible presence of an undocumented AlkB-sensitive 5’ modification (for example m1A14, found in mammalian Phe-GAA tRNAs). ARM-seq profiles for all significant responders also showed increases in 3-fragments, where AlkB-sensitivity could be attributed to m1A58 modifications. 3′-fragments of Pro-TGG tRNAs revealed by ARM-seq include the site of expected m1G modifications at residue 37 in the anticodon loop, consistent with demethylation of m1G37, but no cytosolic-type tRNAs showed significant ARM-seq responses that could be attributed exclusively to demethylation of m1G37.

Supplementary Figure 4 Primer extensions verify ARM-seq predictions for S. cerevisiae Lys-CTT, Leu-TAA and Pro-AGG tRNAs

Primer extensions corroborate unexpected ARM-seq results for Lys-CTT and Leu-TAA tRNAs, which were expected to contain m1A58 based on documentation in Modomics, and for Pro-AGG, which is not currently documented in Modomics. Primer extensions targeting S. cerevisiae Lys-CTT tRNA show no hard stop at the T-loop A58 residue, confirming ARM-seq results that also indicate unmodified A58. Primer extensions for Leu-TAA show a minimally detectable hard-stop at A58 that is removed by AlkB treatment, suggesting low-level m1A58 modification. Primer extensions for Pro-AGG tRNA show partial removal of a block in reverse transcription at A58 and increased read-through to higher molecular weight stops in AlkB-treated samples. Corresponding positive control primer extensions are shown for Thr-AGT tRNA. Negative control lanes show primer extensions performed without addition of RNA.

Supplementary Figure 5 ARM-seq provides evidence for m1A58 modifications in many human tRNAs and tRNA-derived small RNAs

ARM-seq increased the proportion of small RNA sequencing reads mapping to tRNAs by approximately 3.5-fold in two B-cell derived human cell lines (a), with the increased reads in each case corresponding primarily to 3′-fragments and half-molecules where m1A58 is the most prevalent “hard-stop” modification (b).

Supplementary Figure 6 ARM-seq predicts m1A58 modification state for human tRNAs

ARM-seq read profiles show significant increases consistent with m1A58 modifications for at least one subtype in 15 of 17 human isodecoder groups (88%) expected to contain m1A58 (a), and for 22 of 35 human isodecoder groups (63%) not currently represented in Modomics (b, shows the 22 significant responders). Isodecoder subtypes with the lowest P-values were in many cases major subtypes that also showed the highest read count – profiles for both are shown where these differ. Primer extensions performed with or without AlkB treatment confirmed m1A58 modifications predicted by ARM-seq for Pro and Cys tRNAs, which are not currently documented in Modomics for any mammal, and for Arg-ACG, where documentation is lacking for humans (c). ARM-seq produced profiles consistent with documentation for humans, Mus musculus, and Rattus norvegicus (the mammals with data for Asp/Glu tRNAs) showing un-modified A58 for several subtypes of Asp and Glu tRNAs (d), but also showed some response for Glu-TTC-4, suggesting unexpected m1A58 modification (notably, the response was just below 2-fold in GM05372 cells, and just over 2-fold in GM12878 cells; see Fig. 3a for Glu-TTC-4 comparison). Primer extensions targeting the 3′-end of Glu-CTC & Glu-TTC subtypes cannot distinguish between subtypes due to identical 3’ tRNA sequences, but did confirm the presence of m1A58 for some fraction of Glu-CTC/TTC tRNAs (c). Control lanes in (c) show extensions using a primer for S. cerevisiae Thr-AGT in combination with S. cerevisiae (Y), human (H) or no RNA (-).

Supplementary Figure 7 Overview of human ARM-seq results for mature tRNAs, tRNA precursors and other genomic features

Each set of panels shows log2 fold changes (x axis) and adjusted P-values (y-axis) reported by DESeq2 for GM05372 (a) and GM12878 cells (b). Dotted black lines show thresholds used to call significance. Reads for tRNA precursors (which mapped to genomic tRNA sequences, upper right panels in each set) show significant ARM-seq responses that are comparable to responses for mature tRNAs (upper left panels) indicating the presence of AlkB-sensitive modifications in human pre-tRNAs. Reads for human mitochondrial tRNAs also show significant ARM-seq responses, most of which were attributable to m1A9, m1G9, or m1G37 modifications rather than m1A58 (middle right panels in each set). Reads mapping to small subunit ribosomal RNAs were the only other genomic features that showed significant ARM-seq responses consistent with AlkB-sensitive modifications (see lower right panel for GM05372 cells).

Supplementary Figure 8 ARM-seq provides evidence for early m1A58 modification of many human pre-tRNAs and reveals methyl-modified mitochondrial tRNAs

ARM-seq read profiles revealed modified RNAs derived from tRNA precursor transcripts, where the presence of 3′-trailers or 5′-leader sequences are distinguishing features not found in mature tRNAs (a). Read profiles show a subset of pre-tRNAs identified as significant, providing evidence for modification of major and minor subtypes in a variety of isodecoder groups. Primer extensions performed with or without AlkB treatment confirmed the presence of an m1A58 modification in human Leu-CAA pre-tRNA (b). Primer extensions also show that AlkB treatment can demethylate m1G as well as m1A modifications in human mitochondrial tRNAs (b), confirming ARM-seq results showing significantly increased reads for human mitochondrial tRNAs expected to contain m1G or m1A based on modification patterns documented in Modomics (c). ARM-seq profiles for human mitochondrial tRNAs not represented in Modomics were often consistent with modifications documented for bovine mitochondrial tRNAs, and included significant responses for mito-Gln-TTG (where documentation for Bos taurus shows m1G37), mito-Glu-TTC (m1A9 & m1A58), mito-Ser-TGA (m3C32 & m1A58), mito-Thr-TGT (m1A9 & m3C32), and mito-Tyr-GTA (m1G9).

Supplementary Figure 9 Overview of ARM-seq read profiles for human pre-tRNAs

ARM-seq read profiles show increases in detection of RNAs derived from tRNA precursor transcripts for most acceptor types in both GM05372 (top panels) and GM12878 cell types (bottom panels). These reads mapped preferentially to tRNA genes rather than predicted mature tRNA sequences due to the presence of genomically-encoded 5′-leader or 3′-trailer sequences that are not found in mature tRNAs (demarcated by dashed yellow lines). The most pronounced increases overall correspond to 3′-fragments of pre-tRNAs that include the T-loop A58 residue. By contrast, isolated 3′-trailer fragments produced by 3′-end processing by RNaseZ, which were particularly abundant in GM05372 samples (top panels), showed a decrease in normalized abundance in ARM-seq indicating that these do not contain AlkB-sensitive modifications.

Supplementary Figure 10 ARM-seq profiles for human pre-tRNAs associated with abundant 3′ trailer fragments

Human pre-tRNAs that showed significant ARM-seq responses included several that also showed evidence for accumulated 3′-trailer fragments, which have been noted for their abundance in previous sequencing based studies. ARM-seq produced increases in full-length pre-tRNA reads for each of these, consistent with AlkB-sensitive modifications. By contrast, ARM-seq produced a decrease in relative read abundance for isolated 3′-trailer fragments, indicating that these by-products of pre-tRNA processing are not methyl-modified. Highly abundant 3′-trailer fragments derived from Ser-TGA-1-1 and other pre-tRNAs partially obscure less numerous reads from corresponding to modified full-length pre-tRNAs in GM12878 samples (left panels), and almost completely obscure reads for modified pre-tRNAs in GM05372 samples (right panels). Dashed yellow lines show boundaries of 5′-leader and 3′-trailer sequences.

Supplementary Figure 11 Read profiles for intron-containing human pre-tRNAs

Read profiles for intron-containing pre-tRNAs that showed significant ARM-seq responses (indicated by *) show increased reads primarily for 3′-exons and 3′-exon fragments, consistent with m1A58 modifications. Intronic sequences are indicated by dashed orange lines at conserved residues 37 & 38 to the left of the anticodon indicate intron-exon bounds. Dashed yellow lines demarcate boundaries between the body of each tRNA gene and 5′-leader and 3′-trailer sequences. Results are shown for GM05372 cells; similar results were observed in GM12878 cells.

Supplementary Figure 12 DESeq2 log2 fold changes for human pre-tRNAs

ARM-seq log2 fold changes reported by DESeq2 for pre-tRNAs in GM05372 (left) or GM12878 samples (right). Results for each cell line are abbreviated, showing only those specific subtypes with the lowest P-values or with the highest read counts within each isodecoder group; a single isodecoder subtype is shown where these converge. Dotted red lines indicate the two-fold threshold for identifying significant increases.

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–12 and Supplementary Data (PDF 11616 kb)

Supplementary Table 1

ARM-seq reads summarized by RNA class (XLSX 26 kb)

Supplementary Table 2

Yeast ARM-seq differential abundance analysis (XLSX 176 kb)

Supplementary Table 3

Human ARM-seq differential abundance analysis (XLSX 925 kb)

Supplementary Table 4

Abundant methylated RNAs (XLSX 1011 kb)

Supplementary Table 5

Sequences of oligonucleotides used for primer extension analyses (XLSX 9 kb)

Supplementary Software

UNIX command-line software for mapping and differential expression analysis of raw FASTQ sequencing read data for small RNAs derived from mature tRNAs, pre-tRNAs, and other genomic features (ZIP 91 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Cozen, A., Quartley, E., Holmes, A. et al. ARM-seq: AlkB-facilitated RNA methylation sequencing reveals a complex landscape of modified tRNA fragments. Nat Methods 12, 879–884 (2015). https://doi.org/10.1038/nmeth.3508

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nmeth.3508

This article is cited by

Search

Quick links

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