Skip to main content

Thank you for visiting 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.

  • Article
  • Published:

Position-specific oxidation of miR-1 encodes cardiac hypertrophy


In pathophysiology, reactive oxygen species oxidize biomolecules that contribute to disease phenotypes1. One such modification, 8-oxoguanine2 (o8G), is abundant in RNA3 but its epitranscriptional role has not been investigated for microRNAs (miRNAs). Here we specifically sequence oxidized miRNAs in a rat model of the redox-associated condition cardiac hypertrophy4. We find that position-specific o8G modifications are generated in seed regions (positions 2–8) of selective miRNAs, and function to regulate other mRNAs through o8G•A base pairing. o8G is induced predominantly at position 7 of miR-1 (7o8G-miR-1) by treatment with an adrenergic agonist. Introducing 7o8G-miR-1 or 7U-miR-1 (in which G at position 7 is substituted with U) alone is sufficient to cause cardiac hypertrophy in mice, and the mRNA targets of o8G-miR-1 function in affected phenotypes; the specific inhibition of 7o8G-miR-1 in mouse cardiomyocytes was found to attenuate cardiac hypertrophy. o8G-miR-1 is also implicated in patients with cardiomyopathy. Our findings show that the position-specific oxidation of miRNAs could serve as an epitranscriptional mechanism to coordinate pathophysiological redox-mediated gene expression.

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

Access options

Buy this article

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

Fig. 1: Redox-dependent cardiac hypertrophy induces miRNA oxidation.
Fig. 2: o8G-miSeq for cardiac miRNAs.
Fig. 3: o8G-miR-1 redirects target repression via o8G•A base pairing.
Fig. 4: o8G-miR-1 generates cardiac hypertrophy.
Fig. 5: Transcriptome-wide target repression by o8G-miR-1 in cardiac hypertrophy.
Fig. 6: 7o8G-miR-1 is implicated in cardiomyopathy and its loss of function.

Similar content being viewed by others

Data availability

All raw sequencing data from o8G-miSeq (SRP189806, SRP189807, SRP189808, SRP226125), RNA-seq (SRP189813, SRP189117, SRP189812, SRP189811, SRP189809, SRP213998, SRP228274) and CLEAR-CLIP (SRP189810) were deposited to the Sequence Read Archive (SRA). All FASTQ files are available at, including sequencing data of o8G IP with spike-in. Source data are provided with this paper.

Code availability

The Python scripts used in this work are available at


  1. Burgoyne, J. R., Mongue-Din, H., Eaton, P. & Shah, A. M. Redox signaling in cardiac physiology and pathology. Circ. Res. 111, 1091–1106 (2012).

    Article  CAS  PubMed  Google Scholar 

  2. Kasai, H. & Nishimura, S. Hydroxylation of deoxyguanosine at the C-8 position by ascorbic acid and other reducing agents. Nucleic Acids Res. 12, 2137–2145 (1984).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Simms, C. L. & Zaher, H. S. Quality control of chemically damaged RNA. Cell. Mol. Life Sci. 73, 3639–3653 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Frey, N. & Olson, E. N. Cardiac hypertrophy: the good, the bad, and the ugly. Annu. Rev. Physiol. 65, 45–79 (2003).

    Article  CAS  PubMed  Google Scholar 

  5. Freudenthal, B. D. et al. Uncovering the polymerase-induced cytotoxicity of an oxidized nucleotide. Nature 517, 635–639 (2015).

    Article  ADS  CAS  PubMed  Google Scholar 

  6. Shibutani, S., Takeshita, M. & Grollman, A. P. Insertion of specific bases during DNA synthesis past the oxidation-damaged base 8-oxodG. Nature 349, 431–434 (1991).

    Article  ADS  CAS  PubMed  Google Scholar 

  7. Nunomura, A. et al. RNA oxidation is a prominent feature of vulnerable neurons in Alzheimer’s disease. J. Neurosci. 19, 1959–1964 (1999).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Chang, Y. et al. Messenger RNA oxidation occurs early in disease pathogenesis and promotes motor neuron degeneration in ALS. PLoS ONE 3, e2849 (2008).

    Article  ADS  PubMed  PubMed Central  Google Scholar 

  9. Seok, H., Ham, J., Jang, E. S. & Chi, S. W. MicroRNA target recognition: insights from transcriptome-wide non-canonical interactions. Mol. Cells 39, 375–381 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Wang, J. X. et al. Oxidative modification of miR-184 enables it to target Bcl-xL and Bcl-w. Mol. Cell 59, 50–61 (2015).

    Article  CAS  PubMed  Google Scholar 

  11. Carè, A. et al. MicroRNA-133 controls cardiac hypertrophy. Nat. Med. 13, 613–618 (2007).

    Article  PubMed  Google Scholar 

  12. Karakikes, I. et al. Therapeutic cardiac-targeted delivery of miR-1 reverses pressure overload-induced cardiac hypertrophy and attenuates pathological remodeling. J. Am. Heart Assoc. 2, e000078 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  13. Li, Q. et al. Attenuation of microRNA-1 derepresses the cytoskeleton regulatory protein twinfilin-1 to provoke cardiac hypertrophy. J. Cell Sci. 123, 2444–2452 (2010).

    Article  CAS  PubMed  Google Scholar 

  14. Seok, H. Y. et al. Loss of microRNA-155 protects the heart from pathological cardiac hypertrophy. Circ. Res. 114, 1585–1595 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Huang, Z. P. et al. MicroRNA-22 regulates cardiac hypertrophy and remodeling in response to stress. Circ. Res. 112, 1234–1243 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Zhao, Y. et al. Dysregulation of cardiogenesis, cardiac conduction, and cell cycle in mice lacking miRNA-1-2. Cell 129, 303–317 (2007).

    Article  CAS  PubMed  Google Scholar 

  17. Heidersbach, A. et al. microRNA-1 regulates sarcomere formation and suppresses smooth muscle gene expression in the mammalian heart. eLife 2, e01323 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  18. Zhao, Y., Samal, E. & Srivastava, D. Serum response factor regulates a muscle-specific microRNA that targets Hand2 during cardiogenesis. Nature 436, 214–220 (2005).

    Article  ADS  CAS  PubMed  Google Scholar 

  19. Yang, B. et al. The muscle-specific microRNA miR-1 regulates cardiac arrhythmogenic potential by targeting GJA1 and KCNJ2. Nat. Med. 13, 486–491 (2007).

    Article  CAS  PubMed  Google Scholar 

  20. Su, X. et al. Over-expression of microRNA-1 causes arrhythmia by disturbing intracellular trafficking system. Sci. Rep. 7, 46259 (2017).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  21. Pan, Z. et al. miR-1 exacerbates cardiac ischemia-reperfusion injury in mouse models. PLoS ONE 7, e50515 (2012).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  22. Alenko, A., Fleming, A. M. & Burrows, C. J. Reverse transcription past products of guanine oxidation in RNA leads to insertion of A and C opposite 8-oxo-7,8-dihydroguanine and A and G opposite 5-guanidinohydantoin and spiroiminodihydantoin diastereomers. Biochemistry 56, 5053–5064 (2017).

    Article  CAS  PubMed  Google Scholar 

  23. Chi, S. W., Zang, J. B., Mele, A. & Darnell, R. B. Argonaute HITS-CLIP decodes microRNA–mRNA interaction maps. Nature 460, 479–486 (2009).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  24. Moore, M. J. et al. miRNA-target chimeras reveal miRNA 3′-end pairing as a major determinant of Argonaute target specificity. Nat. Commun. 6, 8864 (2015).

    Article  ADS  CAS  PubMed  Google Scholar 

  25. Crackower, M. A. et al. Regulation of myocardial contractility and cell size by distinct PI3K–PTEN signaling pathways. Cell 110, 737–749 (2002).

    Article  CAS  PubMed  Google Scholar 

  26. Li, H. H. et al. Atrogin-1/muscle atrophy F-box inhibits calcineurin-dependent cardiac hypertrophy by participating in an SCF ubiquitin ligase complex. J. Clin. Invest. 114, 1058–1071 (2004).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Ying, X. et al. Novel protective role for ubiquitin-specific protease 18 in pathological cardiac remodeling. Hypertension 68, 1160–1170 (2016).

    Article  CAS  PubMed  Google Scholar 

  28. Spengler, R. M. et al. Elucidation of transcriptome-wide microRNA binding sites in human cardiac tissues by Ago2 HITS-CLIP. Nucleic Acids Res. 44, 7120–7131 (2016).

    PubMed  PubMed Central  Google Scholar 

  29. Park, S. et al. CLIPick: a sensitive peak caller for expression-based deconvolution of HITS-CLIP signals. Nucleic Acids Res. 46, 11153–11168 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Stark, K. et al. Genetic association study identifies HSPB7 as a risk gene for idiopathic dilated cardiomyopathy. PLoS Genet. 6, e1001167 (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  31. Ebert, M. S. & Sharp, P. A. MicroRNA sponges: progress and possibilities. RNA 16, 2043–2050 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Chi, S. W., Hannon, G. J. & Darnell, R. B. An alternative mode of microRNA target recognition. Nat. Struct. Mol. Biol. 19, 321–327 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Lee, H. S. et al. Abasic pivot substitution harnesses target specificity of RNA interference. Nat. Commun. 6, 10154 (2015).

    Article  ADS  CAS  PubMed  Google Scholar 

  34. Choi, Y. J., Gibala, K. S., Ayele, T., Deventer, K. V. & Resendiz, M. J. E. Biophysical properties, thermal stability and functional impact of 8-oxo-7,8-dihydroguanine on oligonucleotides of RNA-a study of duplex, hairpins and the aptamer for preQ1 as models. Nucleic Acids Res. 45, 2099–2111 (2017).

    Article  CAS  PubMed  Google Scholar 

  35. Kim, S. K., Lee, S. H., Kwon, O. S. & Moon, B. J. DNA·RNA heteroduplex containing 8-oxo-7,8-dihydroguanosine: base pairing, structures, and thermodynamic stability. J. Biochem. Mol. Biol. 37, 657–662 (2004).

    CAS  PubMed  Google Scholar 

Download references


We thank all members of the Chi laboratory for help and discussions. We thank in particular M.-H. Chung for sharing research experiences with H. Kasai and S. Nishimura, which inspired us to initiate this project; the late S.-D. Yoo for scientific discussions; the late Y.-S. Lee for initial help on this work; D.-Z. Wang for providing the pJG/ALPHA MHC vector; J. Han, H.-W. Lee and W. J. Park for providing AC16 cells and Y. W. Chung for H9c2 cells, which were used for preliminary studies; Samsung Medical Center for echocardiograms; J. Im and J. W. Park for initial help with the HPLC analysis to confirm the o8G immunoprecipitation; E. S. Cho for help in analysing sequence conservation of miR-1 oxo sites; D. H. Lee and W. Lee for initial help with the project; and the staff of Gyerim Experimental Animal Resource Center for animal care and technical assistance. This work was supported by Samsung Research Funding and Incubation Center of Samsung Electronics under project number SRFC-MA1801-10 and Korea University Grant.

Author information

Authors and Affiliations



H.S., H.L. and S.L. performed the major experiments. H.S., H.L., S.L., G.-W.D.K., S.H.A., E.-S.J. and S.W.C analysed the data. H.S., H.L., S.L., H.-S.L., J. Park, D.G. and S.E. performed the biochemical and molecular biology experiments. H.S. and H.L. conducted flow cytometry analyses. H.S. and H.L. performed the o8G-miSeq. H.S., H.L. and Yeojin Jeong conducted the experiments to demonstrate o8G>T mutation. H.S., H.L., S.L., Y.K.C., Yeahji Jeong and G.-W.D.K. performed the RNA-seq. J. Peak conducted the CLEAR-CLIP analysis. S.H.A., Y.K.C. and S.W.C. performed the bioinformatics analyses. H.S., H.L., S.L., H.-S.L. and J. Park conducted the luciferase reporter assays. H.S., S.L. and G.-W.D.K. performed the mouse experiments. H.S., E.-S.J. and S.W.C. conceived, designed and supervised the research. H.S., E.-S.J. and S.W.C. wrote the manuscript.

Corresponding author

Correspondence to Sung Wook Chi.

Ethics declarations

Competing interests

Korea University has filed a Korean patent application (10-2019-0156147) covering part of the work presented in this Article, listing H.S. and S.W.C. as inventors.

Additional information

Peer review information Nature thanks Marino Resendiz, Eva van Rooij and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Adrenergic cardiac hypertrophy depends on ROS.

a, b, ROS measurement in H9c2 cells after phenylephrine treatment in a time-dependent manner, detected by fluorescent ROS dye (DHE) and flow cytometry (10,000 cells); M1 (blue), cells with moderate ROS; M2 (red), cells with high level of ROS (a). The results were further analysed for cell size; log10(forward scatter), y axis; hypertrophy, cell size >3 (b). c, Same analysis as in a, b except subjecting to serum starvation and showing distribution of cell size in left panels; number of cells or counts (%), y axis (left); n = 3 biologically independent experiments. d, Morphology of H9c2 cells in the presence of phenylephrine and/or NAC treatment, quantified for their average cell size relative to NAC treatment (n ≥ 40 cells, ImageJ; top left graph); NT versus phenylephrine, P = 9.5 × 10−7; error bars, s.e.m. Immunofluorescence staining of cardiomyocytes, discriminated by staining of sarcomeric isoforms of myosin heavy chain (MF20, a marker of cardiomyocyte differentiation). Notably, differentiated cardiomyocyte lineage cells (MF20+) were quantified owing to different extent of fate heterogeneity in H9c2 cardiomyoblast cells; the same batch of H9c2 cells used in ac; scale bar, 50 μm. eh, Same analyses as in c except examining the effect of NAC treatment in H9c2 cells (e, f) and using AC16 cells (g, h); n = 3 biologically independent experiments; H9c2 cells in all of these figures (ah) were obtained from Korean Cell Line Bank. Of note, prevention of ISO induced cardiac hypertrophy by NAC treatment indicates that ROS production is probably upstream of the hypertrophic phenotype. i, j, Chronic intraperitoneal (IP) injection of 75 mg kg−1 ISO and PBS (used as control) was conducted for mice (n = 7 biologically independent mice) in every two days for 29 days (i) and their echocardiography results (n = 3 biologically independent mice) were obtained (j); IVS, interventricular septum; LVID, left ventricular internal diameter; LVPW, left ventricular posterior wall; EF, ejection fraction (%); FS, fractional shortening (%); LV, left ventricular; d, diastolic; s, systolic; *P = 0.024 (IVS;d, ISO versus Cont) and 0.027 (IVS;s, ISO versus Cont), respectively; data are mean ± s.e.m. k, Same experiments as performed in i, j, except treating NAC and using echocardiography at day 21; *P = 0.001, 0.004, 0.031 and 0.033, respectively; n ≥ 3 biologically independent mice; data are mean ± s.e.m. l, Measurement of ROS in lysate (100 μg proteins) from the ISO-induced hypertrophic mouse hearts by using fluorescent ROS dye, CM-H2DCFDA. *P = 0.046. All P values from two-sided t-test; *P < 0.05; n ≥ 3 biological independent samples; data are mean ± s.d. unless otherwise indicated.

Source data

Extended Data Fig. 2 Adrenergic cardiac hypertrophy oxidizes miRNAs.

a, Extent of ROS and alteration of cell size was monitored by using ROS fluorescence dye (DHE) in phenylephrine-treated H9c2 cells in time-dependent manner; scale bar, 50 μm. Notably, sustained elevation of ROS was also observed (up to 48 h) by flow cytometry (Extended Data Fig. 1a, b). b, Immunofluorescence staining of oxidized nucleic acids in ISO- and/or NAC-treated H9c2 cells by using o8G-specific antibody (top). Cytoplasmic pattern of o8G staining disappeared upon treating with RNase, confirming that o8G occurs in RNA (middle). Phenylephrine-treated H9c2 cells were also examined (bottom); scale bar, 50 μm. ce, Immunofluorescence staining of o8G and Ago2 in ISO- and/or NAC-treated rCMCs (c), phenylephrine- and/or NAC-treated H9c2 cells (d), and ISO- and/or NAC-treated H9c2 cells where their colocalization was quantified (e); coloured arrow, cytoplasmic puncta stained as both Ago2 and o8G (d); n ≥ 4; *P = 0.012 (e, right); Scale bars, 100 μm (c) or 50 μm (d, e). Every independent experiment (n ≥ 3) showed similar results (ae). f, Dot blot analyses of o8G for small RNA, extracted from phenylephrine- and/or NAC-treated H9c2 cells (top left) with quantification (relative to small RNA input in lower left panel; right); n = 3; *P = 0.009. g, Same analyses performed in phenylephrine- and/or NAC-treated rCMCs as in f. hk, Northwestern blotting of o8G for total RNA (20 μg), extracted from NAC-treated H9c2 cells (h), phenylephrine- and/or NAC-treated H9c2 cells (i) and ISO-injected mouse hearts (j) with quantification of normalized intensity (intensity ratio, o8G signal per amount of RNA; k) in 20 nt RNA (indicated with arrow or asterisk); n = 5 biological independent mice; Mock, a control set with injection of PBS; HW/BW, heart weight per body weight (j); 20 nt-o8G, synthesized 20-nt-long RNA with o8G used as a size marker; *P = 0.024 (k). l, miRNA was isolated from total RNA in mouse hearts based on their ~20 nt size (left; *), where existence of muscle-specific miR-1 was confirmed by qPCR with no detection of a brain-specific miRNA, miR-124 (right); M, synthesized 20-nt-long RNA used as a size marker; P = 4.24 × 10−5 (miR-1 versus miR-124). m, Dot blot analysis of o8G was performed for the gel-extracted miRNAs (40 ng), derived from ISO-treated hypertrophic mouse hearts (left) and the results were quantified as in k (right); error bars, s.d.; *P = 0.029; n = 3 biologically independent samples. All P values from two-sided t-test; *P < 0.05; n ≥ 3 biological independent samples; data are mean ± s.e.m. unless otherwise indicated.

Source data

Extended Data Fig. 3 Development of o8G-miSeq to identify oxidized miRNA and o8G position for cardiac hypertrophy.

a, Optimization of o8G IP by adjusting IP and wash condition, estimated by using synthesized RNA containing o8G as spike-in control (100 pg or 10 pg) followed by qPCR; relative enrichment, amount of o8G RNA per non-oxidized RNA. “Ab masking” denotes preincubation of o8G antibody with non-oxidized G; *P = 0.04, 1.6 × 10−6, 1.3 × 10−4 and 0.02, respectively; n ≥ 3; error bars, s.e.m. The detailed process and the composition of buffers are described in Supplementary Methods. bd, Sequencing results of cDNA, reverse transcribed from synthesized o8G RNA (Fastq files,; b). Sequencing reads with ≤2 mismatches (2MM) were analysed for the frequency of variation, found in the introduced position of o8G; based on raw counts (c, top); based on unique reads, to exclude artificial bias derived from PCR amplification (c, bottom; discriminated by degenerative barcode, details in Supplementary Methods). Numbers of variation in every position of synthesized RNA were also represented; G (d, top) versus o8G (d, bottom), position 7; read-count, y axis. Of note, ~50–60% G>T variation rate is in agreement with thermal denaturation transitions obtained from RNA:DNA duplexes34,35. e, o8G-miSeq results from H9c2 cells. o8G enrichment, log ratio of o8G IP normalized to the input read counts for miRNAs, was plotted depending on the miRNA abundance (input; e, left), number of G in sequences as heat map density (e, right). f, o8G enrichment (log2(IP/input)) was analysed depending on the level of miRNA (log2 value of read-count from miRNA-seq) in the absence (left) versus presence of phenylephrine treatment (right); miR-1b, red dot indicated by red arrow. g, The positional sequence variation of miR-1b is represented as a normalized mismatch (%) in o8G-miSeq results from H9c2 cells; error bars, s.e.m. The o8G in miR-1b was found to be enriched predominantly in seed region (positions 2, 3 and 7) based on G>T mutation rate. Notably, o8G at position 15 was also observed to be comparable to that of the other positions in seed. h, o8G-miSeq in phenylephrine-treated rCMCs with serum starvation. Relative o8G enrichment was plotted for miRNAs depending on the extent of phenylephrine-induced oxidation (log2(o8G IP) in phenylephrine). Of note, miR-1b showed the most marked phenylephrine-induced oxidation (relative o8G enrichment = 0.8) with a considerable level of o8G (log2(o8G-IP)>10, dotted red line). i, o8G-miSeq result of H9c2 cells was compared with previous results of oxidized miRNAs in H2O2-treated H9c2 cells10, which were identified by using o8G IP and microarray. The previous results were represented inside of the current o8G-miSeq data, highlighted as a yellow circle with indicated arrow in the plot of o8G enrichment (amount in o8G IP relative to abundance) and abundance of miRNA (log2(input)). Notably, except miR-184, the previously identified oxidized miRNAs were not so much enriched in the o8G-miSeq results, presumably due to disregarding background from different abundance of miRNA or lack of o8G IP optimization in the previous trial. The discrepancy also could be caused by heterogeneity of H9c2 cell lines, which behave differently depending on culture condition and batches. j, Relative o8G of miR-1 in ISO-treated mouse heart (chronic injection) was measured by o8G IP and qPCR; the amount of miR-1 in o8G IP was normalized by the amount of o8G-contained spike-in control; error bars, s.e.m.; *P = 0.002. miR-1 showed the most dramatic enhancement (fivefold). Of note, elevation of o8G-miR-1 in chronic injection of ISO implicated that o8G-miR-1 could persist during chronic stress for cardiac hypertrophy. k, l, Luciferase reporter assays with oxo sites of miR-1 as performed in Fig. 3e except for AC16 with phenylephrine (k) or H2O2 treatment (l); *P = 0.001, 6.1 × 10−3, 0.001, 0.025, 0.027, 0.034 and 0.012, respectively (k); *P = 2.3 × 10−4, 0.002, 0.001, 0.007, 0.018, 2.0 × 10−3 and 1.2 × 10−4, respectively (l). All P values from two-sided t-test; *P < 0.05; n = 3 biological independent samples; data are mean ± s.d. unless otherwise indicated.

Source data

Extended Data Fig. 4 Oxidized miR-1 silences new target sites via o8G•A base pairing.

a, dFP (dual fluorescent proteins) reporter with miR-1 seed sites in GFP (top) was validated to detect miR-1 dependent repression at the endogenous level of individual H9c2 cells; NT, non-targeting control; miR-1, synthesized miR-1 used as positive control; miR-1 inhibitor for blocking endogenous miR-1. b, Expression level of miR-1 in AC16 cells, H9c2 cells, rCMCs and mouse heart, measured by qPCR relatively to U6; error bars, s.e.m. Notably, heterogeneity of AC16 and H9c2 cells could result in low global level of miR-1 in total small RNAs, implicating the worth of examining miR-1 at individual cell level. c, Same dFP reporter analyses as performed in a except for 7oxo, 3oxo and 2oxo sites; 7U-miR-1, 3U-miR-1 or 2U-miR-1, used as positive control. Of note, every miR-1 oxo site (7oxo, 3oxo and 2oxo site) in the dFP reporters was endogenously suppressed with sensitivity to detect the repression mediated by the basal level of o8G-miR-1, confirmed by either transfecting miR-1 inhibitor or cognate miR-1 variants. d, Phenylephrine-dependent enlargement of H9c2 cells, measured by size (log10(FSC)) using flow cytometer. e, Distribution of GFP value (log10(GFP)) in dFP reporter, of which GFP contains miR-1 7oxo sites (left). Relative repression (log10(GFP/RFP)) was calculated by averaging the reporter fluorescence values (GFP-7oxo) with a similar range of control fluorescence values (RFP) in the cells (right). The relative repression was also examined in phenylephrine-treated H9c2 cells. Of note, flow cytometry in a, ce used a blue laser for the excitation. f, g, Quantification of dFP reporter in H9c2 cells by flow cytometry. Based on scatter plot of GFP versus RFP (f), GFP value (log10(GFP)) with 7oxo sites (RFP:GFP-site; red) was examined by averaging GFP values derived from similar range of control RFP values in the cells (g), further normalized as relative fold change (Fig. 3g), normalized by dFP reporter with no site (RFP:GFP-no site; black); 7o8G-miR-1, used as positive control (purple). Vectors with no fluorescence protein gene (control) were used to measure level of autofluorescence. h, Same analyses as in f except for considering NAC treatment. Of note, 10 times greater amount of reporter vector was used in h than in f, resulting in fewer transfected cells in f. ik, Flow cytometry analysis using switched dFP protein reporter, of which fluorescent proteins were interchanged in i, top, resulting in sensitive detection of the suppression of miR-1 7oxo site by the endogenous level of 7o8G-miR-1 (i, j); experiments with positive control, 7o8G-miR-1 (j). Repressive propensity of RFP with 7oxo site (GFP:RFP-site; i, middle) was shown in merge (i, right), relative to RFP with no site (GFP:RFP-no site; i, left), showing similar tendency of shifting; relative fold change, normalized RFP values denominated by dFP reporter with no site (RFP:GFP-no site; black). l, Same analysis as in ik with switched dFP reporter except using phenylephrine (left) and miR-1 inhibitor (right). Of note, phenylephrine-induced repression of miR-1 7oxo sites became more substantial when only a limited cell population, for which reporter values (RFP) were in the lowest 25%, was considered (middle). Observation of the restored reporter activity (RFP) by introducing miR-1 inhibitor (right) further confirmed that phenylephrine-dependent repression of miR-1 7oxo sites was mediated by miR-1. The flow cytometer used in fl was equipped with two separate lasers to maximize the excitation of both GFP and RFP, resulting in increased sensitivity of the detection. m, Luciferase reporter assays with miR-1 seed sites in the presence of 5 nM of control (cont; NT), 2o8G-miR-1, 3o8G-miR-1, 7o8G-miR-1 and miR-1; *P = 0.005, 0.4 × 10−3 and 0.1 × 10−4; error bars, s.e.m.; n = 6 biological independent samples. Notably, introduction of oxidized miR-1 (2o8G, 3o8G and 7o8G) could also suppress seed sites in luciferase reporter albeit less potent than unoxidized miR-1, presumably because of retained activity of o8G•C base pairing. All P values from two-sided t-test; *P < 0.05; n = 3 biological independent samples; data are mean ± s.d. unless otherwise indicated.

Source data

Extended Data Fig. 5 Oxidized miR-1 elicits hypertrophy of cardiomyocyte through o8G•A base pairing.

a, Effect of miR-1 expression on rCMCs in the presence of phenylephrine and/or NAC treatment; scale bar, 100 μm. b, Morphology of rCMCs after the transfection of oxidized miR-1 (2o8G, 3o8G or 7o8G); scale bar, 100 μm. Every independent experiment (n ≥ 3) showed similar results (a, b). c, Sequences of o8G-miR-1 mimics, of which o8G was substituted with U (2U-miR-1, 3U-miR-1 and 7U-miR-1), and its cognate sites (2oxo, 3oxo and 7oxo site). d, Luciferase reporter assays with different concentration of 2U-miR-1, 3U-miR-1 or 7U-miR-1 for its cognate 2oxo, 3oxo or 7oxo site (red), compared with miR-1 seed site (blue). Of note, U-miR-1 (2U, 3U or 7U) cannot repress seed site. e, rCMC with transfection of miR-1, 2U-miR-1, 3U-miR-1 or 7U-miR-1; scale bar, 100 μm. f, H9c2 cells with transfection of miR-1, oxidized miR-1 (2o8G, 3o8G or 7o8G) or U-miR-1 (2U, 3U or 7U). Phenylephrine treatment was used as a positive control inducing hypertrophy; scale bar, 50 μm. g, Time lapse images of H9c2 cells transfected with 7o8G-miR-1, 7U-miR-1 or miR-1; scale bar, 50 μm. Every independent experiment (n ≥ 3) showed similar results (eg). h, Distribution of H9c2 cell size after 7U-miR-1 transfection (flow cytometry, 10,000 cells, n = 3); NT, non-targeting miRNA control. P values from two-sided t-test, (d, h); *P < 0.05; n ≥ 3 biological independent samples; data are mean ± s.d. unless otherwise indicated.

Source data

Extended Data Fig. 6 7o8G-miR-1 induces cardiac hypertrophy in vivo.

a, b, In vivo injection of 7o8G-miR-1 as polyethylenimine (PEI) complex. Non-targeting control (NT) was injected together to check the delivery; “Cont”, 5×NT (100% NT); “7o8G”, 1×NT (1:4; 25% NT, 75% 7o8G-miR-1); 5 mg kg−1 (left); IV, intravenous tail injection. The delivery rates were estimated by qPCR (log2 ratio relative to the negative control (−), normalized to U6, n = 4 biological independent mice); error bars, s.d.; *P = 7.2 × 10−5 and 0.017, respectively (a, right). The mouse hearts were displayed in b. c, Immunofluorescence staining of interventricular septum from 7o8G-miR-1 delivered mouse hearts; wheat germ agglutinin (WGA), cell border, red; MF20, cardiomyocyte, green; DAPI, nucleus, blue; scale bar, 100 μm. Every independent experiment (n ≥ 3) showed similar results. d, e, Same experiments as in a, b except using 7U-miR-1 as mixture of NT (1:4, 5mg kg-1) and PEI (N/P ratio = 8); *P = 0.002 and 0.029, respectively (d, right); representative mouse hearts (e). f, Cardiomyocyte-specific expression vector (pJG/ALPHA MHC), which contains α-MHC (alpha myosin heavy chain) promoter to drive cardiomyocyte specific expression of 7U-miR-1 (α-MHC::7U-miR-1). g, Injection of α-MHC::7U-miR-1 plasmid (1.9 mg kg-1) as complex with PEI (N/P ratio = 8) into PBS- or NAC-administrated mice (100 mg kg-1, IP injection; time interval of ~8 h at the same day). Delivery of the plasmid to heart tissue was confirmed by qPCR of hGH poly (A) in the plasmid. h, α-MHC::7U-miR-1 delivered mouse hearts (left) and the quantification of their size (HW/TL; right); box plots with median line, first and third quartile; whiskers, minima and maxima (right); n ≥ 4 biological independent mice; *P = 0.022 (7U-miR-1 versus others). Of note, expression of 7U-miR-1 in cardiomyocyte promoted enlargement of hearts only in the absence of NAC treatment, implicating that basal level of ROS is required for 7o8G-miR-1-induced cardiac hypertrophy. i, Generation of cardiomyocyte-specific transgenic mouse with expression of 7U-miR-1 (7U-miR-1 TG); transgene cassette, generated from α-MHC::7U-miR-1 plasmid by treating BamHI (top). Genotyping results of two independent founders (F0; #7, #43; bottom left) and section of their hearts (bottom right) at the age of p73 (#7) and p56 (#43). Of note, both 7U-miR-1 TG showed marked increase in heart size (HW/TL = 13.23 and 10.08, relative to average size of HW/TL = ~5.0), especially evident in the size of right ventricle (rv); lv, left ventricle. jl, Generation of cardiomyocyte-specific transgenic mice of miR-1 (α-MHC::miR-1 TG) following the same process as used in i. Notably, there was no change in heart size of α-MHC::miR-1 TG; F0 (HW/TL = 4.0; j); F1, TG(−) versus TG(+) (k, left); HW/TL, n ≥ 4 (k, right). WGA staining of heart tissue from α-MHC::miR-1 TG (TG(−) versus TG(+)), showing no difference in cell size (l, left); 40 μm; quantification of relative cell size (n = 70), ImageJ (l, right). m, Survival rate (%) of α-MHC::7U-miR-1 TG (F1). n, Hearts of 7U-miR-1 TG (F1; TG(−) versus TG(+)), derived from 3 independent founders (#33, #37, #44; top); HW/TL (bottom). Expression of pri-miR-1 was examined by qPCR to infer the copy number of transgene; relative expression (“pri-miR-1”, bottom); *P = 4.8 × 10−5, 0.0009, 0.014, 0.009, 0.0007 and 0.002, respectively; n ≥ 3 biologically independent mice. o, Ventricle region of 7U-miR-1 TG heart (F1, #33, TG(−) versus TG(+); H&E staining); scale bar, 1 mm. p, Measurement of total miR-1 level, contributed by expressing 7U-miR-1 transgene (7U-miR-1 TG, F1, #33); TG(+) vs TG(−); qPCR results normalized by U6 abundance (miR-1/U6); n ≥ 9. Of note, expression of 7U-miR-1 transgene added negligible amount to endogenous miR-1 in hearts, implicating physiological relevance of 7U-miR-1 transgenic mouse. q, r, Immunofluorescence staining of cardiomyocytes in the interventricular septum of 7U-miR-1 TG (F1, #33, q) and their Masson’s trichrome staining (r); WGA, red; MF20, green; DAPI, blue; scale bar, 100 μm. s, H&E staining of heart section of 7U-miR-1 TG (F1, #37, TG(−) versus TG(+); top). Enlargement of cells in 7U-miR-1 TG was observed (bottom); rectangular part of the top panel is represented; H&E staining; scale bar, 100 μm. t, Hearts of 7U-miR-1 TG (F2; TG(−) versus TG(+)), derived from 3 independent founders (#33, #37, #44). Section of 7U-miR-1 TG heart (F2, #33, TG(−) versus TG(+); H&E staining) was also displayed (t, bottom left); scale bar, 1 mm. u, The heart size was quantified for 7U-miR-1 TG (F2; TG(−) versus. TG(+); #33 (n = 11 versus 6), #37 (n = 13 versus 12), #44 (n = 5 versus 7)); HW/TL; *P = 0.0001, 8.6 × 10−6 and 0.009, respectively. All P values from two-sided t-test; *P < 0.05; n ≥ 3 biological independent samples; data are mean ± s.e.m. unless otherwise indicated.

Source data

Extended Data Fig. 7 Transcriptome-wide analysis of 7o8G-miR-1-delivered mouse hearts.

a, Kolmogorov–Smirnov test results for Fig. 5a. There are significant downregulation (P = 5.4 × 10−23) of mRNAs with 7oxo site (6mer, positions 2–7, in 3′UTR) and significant upregulation (P = 1.6 × 10−35) of mRNAs with no site (neither miR-1 7oxo site nor seed site, 6mers) depending on 7o8G-miR-1 expression, relative to total (total mRNAs). But, mRNA with control site (“cont site”, mismatched site at position 6 of miR-1 in 3′UTRs but no 7oxo site in the transcript, previously reported to be unrecognized by miRNA32) showed no significance. b, CDF analysis for mRNAs with miR-1 7oxo site (7mer, position 2-8, in 3′UTR, n = 952, red line) depending on 7o8G-miR-1-mediated fold change (log2 ratio, x-axis), conducted for each replicate pairs (Exp A, B and C; Supplementary Table 2a) and also for combined results (Exp(A+B+C)). c, Kolmogorov–Smirnov test results for b, relative to total. d, Functional annotation results for downregulated DEG with miR-1 7oxo site in Fig. 5b, performed by DAVID ( using BIOCARTA and KEGG pathways (top left). The most significantly enriched pathway, “skeletal mouse hypertrophy regulated via AKT/mTOR pathway” (P = 0.02) in BIOCARTA were indicated with other components of hits in the table (red; bottom left) and pathway map (red star; right). GSK3B is highlighted with red dotted circle. Notably, activation of GSK3B has been reported to suppress cardiac hypertrophy25. eg, Concentrating on heart-related pathways in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database, the normalized enrichment score (NES) from gene set enrichment analysis (GSEA) elucidated “hypertrophic cardiomyopathy” (HCM, NES = 1.25; e, right), “arrhythmogenic right ventricular cardiomyopathy” (ARVC, NES = 1.20; f, right), and “dilated cardiomyopathy” (DCM, NES = 1.25; g, top) as pathways regulated by 7o8G-miR-1. Heat maps on the top indicate the mRNAs in the corresponding pathway with measured expression; red, downregulation; blue, upregulation; order of results from top, NT (n = 3), 7o8G-miR-1 (n = 3). Hits from downregulated DEG with miR-1 7oxo site in Fig. 5b were also displayed as red stars in HCM (e, left), ARVC (f, left) and DCM (only a distinct part from HCM and ARVC is displayed; g, bottom). ATP2A2 (SERCA2) is highlighted with red dotted circle and GJA1 (CX43) with blue dotted circle.

Source data

Extended Data Fig. 8 Transcriptome-wide analysis of o8G-miR-1 transfected H9c2 cells.

a, d, CDF analysis as performed in Fig. 5a except for 7o8G-miR-1-transfected H9c2 cells; all 7oxo sites (a), 7mer 7oxo site (positions 2–8; b), 6mer seed site (positions 2–7; c) and 7mer seed site (positions 2–8; d). Kolmogorov–Smirnov test (two sided) results were displayed in the table (b, c, d, top right table); mean. Control in c, d denotes mismatched site at position 6 of miR-1 in 3′UTRs but no seed site in the transcript. e, Gene Ontology analysis results of 7o8G-miR-1 targets identified in H9c2 cells (Fig. 5c); intensity of bubble colour, P value (derived from DAVID); bubble size, frequency of the Gene Ontology term. Highly similar Gene Ontology terms are linked by edges in the graph, where the line width indicates the degree of similarity. Of note, only Gene Ontology terms in biological process were used in this analysis. fh, CDF analysis as in ad except for 2o8G-miR-1 (f) or 3o8G-miR-1 (h) by using RNA-seq results (Supplementary Table 2c, d); Kolmogorov–Smirnov test (two sided) results in h. i, j, Gene Ontology analysis results as performed in e except for 2o8G-miR-1 (i) and 3o8G-miR-1 (j); P < 0.05 (derived from DAVID), n = 176 (2o8G-miR-1), n = 34 (3o8G-miR-1). 2o8G-miR-1 targets and 3o8G-miR-1 targets were selected as containing 6mer sites (position 2–9) in 3′UTR and showing log2 ratio <−0.8 in f, g. The biggest cluster was displayed as representative (i, j). k, Validation of putative 7o8G-miR-1 targets by qPCR in 7o8G-miR-1-transfected rCMC as performed in Fig. 5d, f. As control, GAPDH (for FBXO32 and PTEN) or ACTB (for USP18) was also measured for normalization. FBXO32, PTEN and USP18, of which 3′UTRs contain 7oxo sites (Supplementary Table 4c) were examined; t-test, two-sided; *P = 0.003, 5.4 × 10−4, 2.1 × 10−4 and 0.044, respectively; n ≥ 3 biological independent samples; data are mean ± s.e.m. Notably, PTEN also contains a miR-1 seed site in its 3′UTR. l, m, CDF analyses as conducted in Fig. 5g except for all o8G-miR-1 targets (2oxo, 3oxo or 7oxo sites; l) or only for 2o8G-miR-1 (m, left panel) or 3o8G-miR-1 targets (m, right). n, GSEA results performed as in Fig. 5h except using 2o8G-miR-1 (top) and 3o8G-miR-1 (bottom) dependent changes of putative cognate mRNA targets (6mers, 3′UTR) in H9c2 cells (Supplementary Table 2c, d) for the enrichment of ISO-dependent downregulation (log ratio < 0); NES, normalized enrichment score; P value, adjusted P value derived from GSEA.

Source data

Extended Data Fig. 9 Identification of 7o8G-miR-1 targets and their 7oxo sites during adrenergic cardiac hypertrophy and cardiomyopathy.

a, Experimental validation of Bicd2 as 7o8G-miR-1 target (7oxo site) in phenylephrine-treated rCMCs. Relative expression was measured by qPCR, normalized by Actb; *P = 0.024; n ≥ 3 biological independent samples; error bars, s.e.m. b, Same validation experiments as in a except using ISO-treated mouse hearts for Bicd2 (left, normalized by Actb) and Atp2a2 (right, normalized by Gapdh); *P = 0.005 and 0.06, respectively; n ≥ 4 biological independent samples; error bars, s.e.m. cj, miR-1 7oxo sites (red) in Ago bound regions (3′UTR) of patients with heart disease for ATP2A2 (c), GSK3B (d), BICD2 (e), GJA1 (f), GATA4 (g), DRAM1 (h), PTEN (i) and FBXO32 (j), represented as in Fig. 6d; identified in the Ago-mRNA peaks (red; sequences in Supplementary Tables 3 and 4) with significant peak height (P < 0.05 or 0.01, CLIPick29, black); with miR-1 seed site (orange). k, Luciferase reporter assays with different concentrations of 7o8G-miR-1 for 7oxo sites, derived from GATA4, ATP2A2 and BICD2. Half-maximal inhibitory concentration (IC50) was calculated. Of note, repression mediated by miR-1 7oxo site in GATA4 (IC50 = 1.2 nM) or ATP2A2 (IC50 = 0.7 nM) was efficient, albeit the maximum repression was only ~15–20% (left and middle). By contrast, miR-1 7oxo site in BICD2 exerted better maximum repression (52%) with low efficiency (IC50 = 11nM; right). ln, CLEAR-CLIP analysis for ISO induced hypertrophic mouse hearts, yielding miRNA-target chimaera reads (l). miR-1 containing chimaera reads were aligned and analysed for the frequency of miR-1 seed, 2oxo, 3oxo, and 7oxo sites depending on ISO treatment, examined by expanding the size of the sequences from the miR-1 ligated site (±0, 25, 50 or 100 nt; m); box plots with median line, first and third quartile; whiskers, minima and maxima (n). Although only limited number of chimaera reads were yielded for miR-1 (Supplementary Table 6a), global propensity of its target interaction could be analysed for their changes depending on ISO-treatment. o, Sequence of conserved miR-1 7oxo site in HSPB7, which overlapped with miR-1 chimaeras in ISO-treated mouse hearts and Ago-mRNA peaks of patients with cardiomyopathy (Fig. 6d). p, Validation of Hspb7 as miR-1 7oxo target by performing qPCR in 7o8G-miR-1 transfected rCMC (normalized by Gapdh, left) and ISO-injected mouse heart by qPCR (normalized by Actb, right).−, NT transfection (left) or mock (right); *P = 0.04 and 0.01, respectively; n = 3 (left) or 5 (right) biologically independent samples; error bars, s.e.m. q, Ago-bound miR-1 seed and 7oxo sites identified in the 3′UTR of GJA1 in patients with cardiomyopathy, as represented in cj; left. Activity of luciferase reporters, containing the indicated regions (left), were examined depending on the concentration of 7o8G-miR-1 (right). Of note, despite containing the miR-1 seed site, core-Δoxo completely lost repression by 7o8G-miR-1, implicating the importance of 7oxo site to execute miR-1 dependent repression of GJA1. All P values from two-sided t-test, *P < 0.05; n ≥ 3 biological independent samples; data are mean ± s.d. unless otherwise indicated.

Source data

Extended Data Fig. 10 Specific inhibition of 7o8G-miR-1 by competitive inhibitors, anti-7oxo.

a, Schematics of competitive miRNA inhibitors, which consist of tandem repeats of miRNA target sites (7mer, position 2–8) in RNA oligonucleotide; “Cont”, NT seed sites (2×); anti-seed, miR-1 seed sites (2×); Anti-7oxo, miR-1 7oxo sites (2×). Base pairing of miR-1 or 7o8G-miR-1 with cognate target site (seed or 7oxo site) was represented. b, c, Luciferase reporter assays used to validate activity and specificity of competitive miRNA inhibitors (anti-seed and anti-7oxo) in a. Luciferase reporter vector containing miR-1 seed sites or 7oxo sites was constructed and used to confirm derepression of the reporters by anti-seed (b) and by anti-7oxo (c); “Cont”, anti-NT (containing seed sites of NT); −, NT; *P = 1.87 × 10−11 and 5.08 × 10−12, respectively; n = 8 biologically independent samples (b); *P = 0.038; n ≥ 5 biologically independent samples (c); error bars, s.e.m. Of note, anti-7oxo significantly derepressed 7oxo site mediated target repression, albeit the extent was somewhat marginal. This is possibly because of containing only two repeats of target sites in the inhibitor. Notably, anti-7oxo showed no effect on seed site mediated target repression, confirming the specificity of anti-7oxo (c). d, CDF analysis to globally confirm derepression of 7oxo targets by anti-7oxo (9×); “7oxo site”, mRNAs with 7oxo site (7mer, position 2–8, in 3′UTR); “No site”, mRNAs with neither 7oxo site nor seed site (6mers, positions 2–8, in 3′UTR); “Cont”, transcripts with a mismatch site at position 6 of miR-1 (negative control; 7mer, positions 2–8, in 3′UTR) but without 7oxo (left) or seed site (right). Expressed transcripts (FPKM >1 in both samples, valid status from Cuffdiff, n = 7,958) from RNA-seq (Supplementary Table 6b) were used for the analysis. Notably, anti-7oxo (9×) significantly derepressed miR-1 7oxo sites (P = 0.038, relative to control; Kolmogorov–Smirnov test, two sided; left) but not miR-1 seed sites (P = 0.987, relative to control, Kolmogorov–Smirnov test, two sided; right). e, Time-lapse images of rCMC, transfected with either cont (anti-NT (4×); left) or anti-7oxo (4×; right) after phenylephrine treatment (with serum starvation); scale bar, 100 μm. Introduction of anti-7oxo to rCMCs attenuated phenylephrine-induced hypertrophy. f, Generation of cardiomyocyte-specific transgenic mice (TG) with expression of anti-7oxo (13×); transgene cassette, generated from α-MHC::anti-7oxo (13×) plasmid by BamHI. g, Genotyping results of four independent founders (F0; #65, #68, #69, #44). h, Mapping result of RNA-seq reads (Supplementary Table 6c) for the anti-7oxo transgene cassette, expressed in hearts of ISO-treated TG (Fo, #68). Notably, TG(+) showed sufficient expression of anti-7oxo (13×) sequences in heart tissue. i, CDF analyses of RNA-seq results (Supplementary Table 6c) to confirm anti-7oxo (13×) mediated derepression of 7o8G-miR-1 targets (7oxo sites) as analysed in d. Notably, significant derepression of 7o8G-miR-1 targets (as measured by log2 (TG(+)ISO/TG(−)ISO)) was observed (P = 0.021, relative to “Cont”, Kolmogorov–Smirnov test, two sided ; P = 4.1 × 10−5, relative to “no site”); by contrast there was no change for miR-1 seed targets (right). All P values from t-test, two-sided; *P < 0.05; n ≥ 3 biological independent samples; data are mean ± s.d. unless otherwise indicated.

Source data

Extended Data Fig. 11 Loss-of-function study of 7o8G-miR-1 in vivo by establishing an anti-7oxo transgenic mouse.

a, The ISO treatment schedule for anti-7oxo TG, cardiomyocyte-specific transgenic mice with expression of anti-7oxo (a specific sponge inhibitor against 7o8G-miR-1); intraperitoneal (IP) injection. b, Heart size of anti-7oxo TG(−) versus TG(+) in the presence of ISO administration (three independent F0 lines, #69, #68, #44; left); HW/BW; error bars, s.d.; *P = 0.014; n ≥ 3 (right). c, Genotyping results of F1, generated from two different F0 lines (#71, top; #77, bottom). d, Section of hearts from littermates of two different lines (#71, left; #77, right); TG(−) versus TG(+) under ISO treatment. e, Heart size in d was quantified as HW/TL (bottom) or HW/BW (top); box plots with median line, first and third quartile; whiskers, minima and maxima; x, mean; *P = 0.007 and 0.013. Of note, ISO-induced hypertrophy is prevented in anti-7oxo TG (F1). f, H&E stained tissue sections of interventricular septum (IS) of anti-7oxo TG (F0); n = 3; each experiment was repeated independently with similar results. g, Immunofluorescence staining of cardiomyocytes in IS of anti-7oxo TG (F0); wheat germ agglutinin (WGA) for refining the cell border; MF20 for cardiomyocytes; DAPI for nuclear staining; scale bar, 100 μm. h, Relative cell size of MF(+) in g was quantified (n = 200, ImageJ; *P = 3.2 × 10−59); box plots with median line, first and third quartile; whiskers, minima and maxima; “x”, mean. Notably, size of cardiomyocyte was maintained in anti-7oxo TG even in the presence of ISO treatment. i, A schematic model of the position-specific oxidation of miR-1; 7o8G-miR-1 targets identified in the adrenergic receptor (AR) pathway, red; others, blue. j, Quantification of o8G-miR-1 in ISO-injected mice in time-dependent manner. After IP injection of ISO, hearts were collected at different time points (0, 1, 5 and 7 days; top), of which small RNA was used to measure relative amount of o8G-miR-1 to o8G spike-in control by performing o8G IP and qPCR (miR-1; bottom). Notably, sustained oxidation of miR-1 was observed up to 7 days from ISO treatment; *P = 0.01, 0.03 and 0.01, respectively; n = 4 biologically independent mice; error bars, s.e.m. k, Sequence conservation of o8G-miR-1 target sites in 3′UTR. Conservation rates for miR-1 oxo sites (2oxo, 3oxo and 7oxo sites) including seed site were indicated in cumulative graph (left) and distribution (right) of seed sites of conserved miRNAs (conserved; blue line) and all heptamers (total; black line). Notably, sequences of miR-1 2oxo, 3oxo and 7oxo sites were evolutionally conserved in 3′UTR (11.3, 13.0 and 12.3%, respectively) as much as seed sites of conserved miRNAs in mammals—higher than median conservation rate (10.5%, 6mers) but lower than miR-1 seed site (14.5%). All P values from two-sided t-test, two-sided; n ≥ 3 biological independent samples; data show the mean unless otherwise indicated.

Source data

Supplementary information

Supplementary Information

Supplementary Methods: Detailed description of methods used in this study.

Reporting Summary

Supplementary Information

Supplementary Discussion: In-depth discussion about o8G-miR-1. More focused discussion about position-specific oxidation of miR-1 during cardiac hypertrophy.

Supplementary Tables

Supplementary Tables 1-9: Statistics of sequencing data, sequences of target sites and oligonucleotides for luciferase reporter construction and qPCR. Supplementary tables provide statistics of sequencing data processes (Tables 1, 2, 5 and 6), sequences of 7o8G-miR-1 target sites (Tables 3 and 4), oligonucleotide sequences used to construct luciferase reporter vectors (Table 7) and sequences of qPCR primers used to measure target mRNAs in rat (Table 8) and mouse (Table 9).

Video 1

Time-lapse video of 7o8G-miR-1 transfected rCMC. The movie starts from the transfection of 7o8G-miR-1 (entitled as “miR-1:7o8G” in video) and ends after 72 hours.

Video 2

Time-lapse video of 7U-miR-1 transfected rCMC. The movie starts from the transfection of 7U-miR-1 (entitled as “miR-1:7U” in video) and ends after 72 hours.

Source data

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Seok, H., Lee, H., Lee, S. et al. Position-specific oxidation of miR-1 encodes cardiac hypertrophy. Nature 584, 279–285 (2020).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


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