Article Figures & Data
Figures
- Figure S1. Validation of m6A-IPs.
(A) Enrichment of m6A mRNA after m6A-IP shown by m6A mRNA dot blot of input RNA, flow through/supernatant, and RNA immunoprecipitated (IP) with m6A antibody (m6A Ab) or IgG antibody (IgG Ab) as control. (B) Enrichment of m6A mRNA after m6A-IP shown by qPCR on Nde1 and Drd1, previously shown to be unmethylated or methylated, respectively. (C) Immunoblot for Mettl3 in human DCM samples showing a trend to increased Mettl3 expression. + denotes a positive control sample (Hela cell lysate). (D) mRNA expression of m6A machinery in healthy and DCM cardiac samples analyzed by qPCR. P < 0.05 by one-tailed t test (n = 3 per group). (E) Enrichment of DCM-specific GDPG and GJD3 or control-specific ZNF324, HNRPHR3, and HSPA1A m6A mRNA after m6A-IP shown by qPCR (n = 3, **P < 0.005, ***P < 0.0005 by one-way ANOVA). (F) Gene ontology (GO) term enrichment analysis for the 1,595 DCM-specific methylated mRNAs for molecular functions.
- Figure 1. The m6A mRNA methylome differs between healthy and failing human heart tissue.
(A) Percentage of m6A in mRNA from DCM and control samples measured by m6A ELISA (n = 6 control and n = 10 DCM, *P = 0.0476 by t test). (B) mRNA expression of m6A machinery in healthy and DCM cardiac samples analyzed by normalized read counts from RNA-seq from myocardial biopsies (n = 33 DCM and n = 24 control). (C) Venn diagram showing overlap of healthy and DCM m6A mRNA methylome analyzed by sequencing of m6A immunoprecipitated mRNA. (D) Gene Ontology (GO) term enrichment analysis for biological processes on the 1,595 DCM-specific methylated mRNAs (high m6A) and mRNAs not enriched in IP (low m6A). The width of the bars represents the significance (–log10 (adjusted P-value, Fisher’s exact test)) of the respective GO term enrichment. (E) Volcano plot of sequencing of m6A immunoprecipitated mRNA data. Significant enriched transcripts are shown in green. Transcripts involved in transcription (GO term: transcription and DNA-templated are shown in red). Examples of enriched transcripts in DCM hearts are indicated. (F) IGV plots of sequencing reads after m6A-IP for DCM-specific methylated GJD3 transcript and control-specific ZNF324B transcript.
- Figure S2. Gain and loss of function of Mettl3 in in vitro and in vivo lead to changes cell size and gene expression changes.
(A) Left: immunofluorescence of neonatal (top) and adult rat cardiomyocytes (bottom); actin antibody (red), Mettl3 (green), and nuclei (blue). Right: immunofluorescence of neonatal (top) and adult rat cardiomyocytes (bottom); actin antibody (red), Fto (green), and nuclei (blue). Bar graph: 20 μM. (B) Knockdown of Mettl3 and Fto in NRVMs demonstrated by immunoblot. (C) Relative m6A content (n-fold change) in mRNA in NRVMs after Mettl3 or Fto knockdown by siRNA measured by m6A ELISA (n = 3, P < 0.05 by one-way ANOVA). (D) Cell surface of NRVMs with Mettl3 and Fto knockdown compared with siScr with and without PE for 24 h stimulation (n = 3, ****P < 0.0001 by one way ANOVA). (E) mRNA expression of Nppa and Nppb in NRVMs after Mettl3 or Fto KD compared with siScr with and without PE stimulation (n = 3, **P < 0.005, ***P < 0.0005 by one-way ANOVA). (F) Overexpression of Mettl3 and enzymatically inactive mutant in NRVMs in neonatal cardiomycytes by adenoviral infection; actin antibody (red), flag (green), and nuclei (blue). Bar graph: 20 μM. (G) Cardiac overexpression of Mettl3 in mice demonstrated by immunoblot (top) and immunofluorescence (bottom); actin antibody (red), Mettl3 antibody (green), and nuclei (blue). Bar graph: 40 μM. (H) Percentage of m6A in mRNA in NRVMs after Mettl3 overexpression in vivo measured by m6A ELISA (n = 3). *P < 0.05 by t test. (I) mRNA expression of Nppa and Nppb of either control virus or Mettl3-injected mouse hearts after either sham or TAC surgery in 10-wk-old mice (n = 6–14,*P < 0.05, **P < 0.005).
- Figure 2. m6A affects cell growth and cardiac function.
(A) Percentage of m6A in mRNA in NRVMs after Mettl3 or Mettl3 enzymatically inactive mutant overexpression measured by m6A ELISA (n = 7 for control and n = 8 for Mettl3 WT and Mettl3 mutant). ****P < 0.0001 by one-way ANOVA followed by Bonferroni's post hoc comparisons. (B) Cell surface of NRVMs overexpressing Mettl3 or inactive mutant compared with control virus with and without PE stimulation (n = 3 independent biological experiments with >69 cells for each group *P < 0.0047, ****P < 0.0001 by one-way ANOVA). (C) Gross morphology of either control virus or Mettl3-injected mouse hearts after either sham or TAC surgery in 10-wk-old mice. (D) Ratio of LV weight to tibia length of control and Mettl3 OE mice (n = 10–12, ***P < 0.001, ****P < 0.0001 by one-way ANOVA). (E) Immunofluorescence of heart sections WGA (green), sarcomeric actin (red), and nuclei blue. Bar graph: 20 μM. (F) Cell surface area measurement from WGA staining (n = 4 animals per group and 200–300 cells in total, *P < 0.05, ****P < 0.0001 by one-way ANOVA). (G) Masson trichrome staining on paraffin heart sections. Bar graph: 80 μM. (H) Quantification of fibrotic area from Masson trichrome staining (n = 4–10, *P < 0.05 by one-way ANOVA). (I) mRNA expression of Col1a1 analyzed by qPCR (n = 6–14, ****P < 0.0001 by one-way ANOVA).
- Figure S3. Decrease of m6a levels in murine hearts after TAC surgery is associated with differential regulation of translation.
(A) Percentage of m6A in mRNA in mice hearts 2 d after TAC or sham surgery (n = 5–6, *P < 0.05 by t test). (B) Venn diagram showing overlap between m6A-containing transcripts between sham- and TAC-operated mice 2 d after surgery. (C) RT-PCR and immunoblots for Mettl3 in mice after sham or 2 d post TAC surgery. n = 3–4, *P < 0.05 by t test). (D) Enrichment of Gene Ontology (GO) terms in the group of m6A-enriched genes (sham conditions) for biological processes. Colors depict the log fold change (logFC) of individual genes within the GO category and numbers behind the bars correspond to the number of genes within the GO category. (E) Cumulative fraction of mRNAs relative to their fold change of RNA-seq and mRNAs relative to their fold change of Ribo-seq between all transcripts and methylated transcripts 2 d after TAC surgery (Kolmogorov–Smirnov test P < 0.01). (F) GO term enrichment analysis for the TAC-specific methylated mRNAs for biological processes and cellular compartments. (G) Overexpression of Mettl3 in HL-1 cells shown by immunoblotting. (H) List of top translationally regulated transcripts by Mettl3 overexpression in HL-1 cells. (I) Enrichment of m6A-mRNA after m6A-IP shown by qPCR of Arhgef3 and Myl2 after different time points post TAC surgery. <0.05 by t test. n = 3.
- Figure 3. m6A affects translation efficiency.
(A) Scatter blot of fold change of RNA-seq and Ribo-seq (sham/TAC) from murine hearts 2 d after TAC surgery. All transcripts (gray), enriched in m6A-IP (blue). (B) Box plots of fold change (sham/TAC) of Ribo-seq and RNA-seq from murine hearts 2 d after TAC surgery for methylated transcripts in sham-operated mice (sham only m6A) or post TAC surgery (TAC only m6A). *P < 0.05 by t test. (C) Volcano blot of Ribo-seq data from HL-1 cells overexpressing Mettl3. (D) GO term enrichment analysis (biological process) for Mettl3 target mRNAs, which are identified as regulated in Ribo-seq. The width of the bars represents the significance (–log10 (adjusted P-value, Fisher’s exact test)) of the respective GO term enrichment. Colors depict the log fold change (logFC) of individual genes within the GO category. (E) Cumulative fraction of mRNAs relative to their fold change of Ribo-seq and RNA-seq (sham/TAC) between all transcripts and Mettl3-regulated transcripts 2 d after TAC surgery (Kolmogorov–Smirnov test, P < 0.001).
- Figure 4. m6A affects expression of target genes by regulating transcript stability.
(A) Polysome profile of control and Mettl3-overexpressing cardiomyocytes (HL-1 cells). (B) RT-PCR (left panel) and Ribo-seq counts (right panel) for Myl2 and Arhgef3 in HL-1 cells (*P < 0.05 after EdgeR analysis for the Ribo-seq quantification, n = 3). (C) Percentage of total transcript abundance of Arhgef3 and Myl2 in mRNPs, 80s, and polyribosomes (*P < 0.05 by t test; n = 2 independent experiments). (D) Enrichment of m6A-mRNA after m6A-IP shown by qPCR of Arhgef3 and Myl2 after Mettl3 overexpression. *P < 0.01 by t test; n = 3 independent experiments. (E) mRNA stability of Mettl3 targets Arhgef3, Polr3d, and Myl2 analyzed by qPCR upon actinomycin D treatment. (F, G) Immunoblots and (G) quantification for Arhgef3 and Myl2 in HL-1 cells after confirming the Mettl3 overexpression. P < 0.05 by t test; n = 3 independent experiments. (H, I) Immunoblots and (I) quantification for Arhgef3 and Myl2 in mice after sham or TAC surgery cells after Mettl3 overexpression. *P < 0.05 by one-way ANOVA (n = 3 for each group; SE, short exposure; LE, long exposure).
Supplementary Materials
