IDH1-dependent m6A methylation defines transcriptomic heterogeneity in glioma

Abstract

Gliomas are biologically heterogeneous brain tumors with marked differences in clinical behavior based on the IDH1 mutation status. While epigenetic dysregulation is well characterized, the contribution of RNA modifications, particularly N6-methyladenosine (m6A), remains underexplored. Using direct RNA nanopore sequencing of patient-derived gliomas, we generated the first isoform-resolved m6A maps across IDH1-mutant and wild-type tumors. IDH1-mutant gliomas exhibited globally elevated m6A methylation, along with increased expression of methyltransferases (METTL3, METTL14) and stabilizing readers (YTHDF3). In contrast, wild-type glioblastomas showed enhanced expression of m6A erasers (ALKBH5, FTO) and RNA decay factors (YTHDF2). These subtype-specific differences in m6A architecture impacted transcript stability, isoform usage, and gene expression. Isoform-level analyses revealed stronger prognostic associations than gene-level parameter, including for IGF2BP2-202, PUF60-202, and GLUL-203. Our study establishes m6A as a critical, subtype-specific layer of RNA regulation in glioma with clinical and therapeutic implications.

Keywords: IDH1; N6-methyladenosine; glioma; isoforms.

Figure 1.. m6A modified sites, transcripts, and genes in IDH1 mutant and wild-type gliomas.

(a-c) Pie charts (left) showing the prevalence of m6A modified sites (green), isoforms (orange), and genes (red) in each glioma subtype (AA, Astrocytoma (green); OO, Oligodendroglioma (red); GBM, Glioblastoma (blue)). Stacked bars and accompanying pie charts (right) depict the biotype distribution of m6A-modified isoforms, including protein-coding and less abundant RNA biotypes. (d-f) Proportion of m6A sites mapped to 5’UTR, CDS and 3’UTR regions within (d) protein-coding, (e) retained intron, and (f) nonsense-mediated decay (NMD) isoforms across glioma subtypes (AA: green, OO: red, GBM: blue). (g-i) Distribution of single-site or multi-site (>1) m6A-modified isoforms in (g) AA, (h) OO, and (i) GBM. (j) Correlation plot showing: (top) number of m6A-modified isoforms vs. number of sites per patient (probability >0.9); and (bottom) total isoforms detected vs number of m6A-modified isoforms. (k) Metagene plots showing m6A site distribution across isoform regions in IDH1 mutant (IDH1mt) and wild-type (IDHwt) glioma across pathways including RNA splicing, DNA Damage, mRNA Processing, Cell Death, and Protein Folding.

Figure 2.. Transcript-level analysis of hyper- and hypomethylated m6A isoforms.

(a) Venn Diagrams showing the overlap of m6A modified genes (left) and isoforms (right) across glioma subtypes (AA, OO, GBM). A total of 226 commonly modified isoforms were used for downstream analysis. (b) Metagene plots showing isoform-region distribution of m6A sites in hypermethylated isoforms for each subtype. (c-e) Scatter plots integrating m6A methylation (x-axis, log2 fold change of weighted mod ratio) and gene expression (y-axis, log2 fold change) for common m6A-modified isoforms in each comparison: (c) AA.vs GBM, (d) OO vs GBM, and (e) AA vs OO. Dashed lines indicate threshold for significance. (f) Ratio of m6A site density in the CDS vs 3’UTR for hypermethylated protein-coding transcripts stratified by gene expression direction (up, down, unchanged). (g-i) Number of m6A sites in the 5’UTR region of hypermethylated retained intron isoforms, stratified by direction of gene expression, for each comparison.

Figure 3.. Functional functional enrichment of m6a-modified upregulated genes in IDH1 mutant and wild-type gliomas.

(a) Volcano plots showing differentially expressed genes (DEGs) from each pairwise comparison: AA vs. GBM (top), OO vs. GBM (middle), and AA vs. OO (bottom). Significant DEGs (Log2 FC > 0.58 or < −0.58) are highlighted in red. (b) Stacked plot quantifying the number of upregulated genes in IDH1 mutant (mut) and IDH1 wild-type (wt) gliomas that are also m6A modified (light green). (c) Heatmap displaying the weighted m6A modification ratio per isoform in upregulated, m6A-modified genes from the mutant (n=68) and wild-type (n=17) group. (d) Heatmap of top enriched gene ontology (GO) terms among m6A modified upregulated genes. The y-axis denotes the enriched functional pathway, and the associated genes identified in each ontology term. Pathway enrichment is plotted by patient (rows) and associated gene expression (counts per million, columns), stratified by IDH1 status and key molecular features (EGFR, TERT, ATRX, TP53, MGMT).

Figure 4.. Differential isoform usage between Astrocytoma (AA) and Glioblastoma (GBM).

(a) Stacked bar plot quantifying the number of high usage isoforms in AA vs GBM and AA vs OO. (b) Pie charts depicting the RNA biotype distribution among high usage isoforms in AA (n = 77, top panel) and GBM (n = 71, bottom panel). (c) Scatter plots showing the relationship between difference in isoform usage (difference in isoform fraction; dIF, y-axis) and gene level log2 fold change (x-axis) in AA vs GBM. Significant DEGs are highlighted in red. (d) Chromosomal distribution of upregulated genes in GBM (blue) and AA (green) linked to high or low isoform usage. (e) Consequence enrichment analysis identifying predicted structural consequences of isoform switches between AA and GBM. Y-axis: consequence, x-axis: fraction of genes having that consequence (95% confidence interval, CI). Dot size correlates with the number of genes with that consequence. (f) Alternative splicing event enrichment analysis in AA vs GBM. (g) Differential isoform expression (top, Log2 FC, top panel; gray) and usage (dIF, bottom panel; blue) in selected non-m6A modified upregulated genes in AA or GBM. (h) Differential isoform expression of m6A-modified transcripts with significant dIF usage in AA or GBM. Dotted line indicates a threshold for significance. (i, j) Isoform plots of genes depicted in (g) and (h).

Figure 5.. m6A regulators and transcriptomics consequences across glioma subtypes.

Heatmap of differentially expressed (a) m6A regulators (readers, writers, erasers) and (b) RNA decay factors across three pairwise comparisons: AA vs GBM, OO vs, GBM, and AA vs. OO. (c) Correlation matrix of m6A regulator expression, m6A site/transcript abundance, and clinical outcomes (OS, PFS). (d) Forest plot of hazard ratios (HR, COX regression analysis) and 95% confidence intervals (CI) for isoforms of m6A regulators. Significant associations (p < 0.05%) are highlighted in red. (e) Density distribution comparing transcript lengths of m6A-modified RNAs in IDH1 mutant (yellow) and wild-type (purple) gliomas. (f) Gene expression correlation plots: IGF2BP2 with oncogenes BST2 (top, left) and IFITM3 (top, right); YTHDC3 with ELAVL1 (bottom, left) and EIF3A (bottom, right). (g) Correlation analysis (Pearson correlation, two-tailed test) between the m6A reader proteins and selected T Cell markers (CD4, IFNGR1) in IDH mutant (left) and wild-type (right) gliomas. (h) Correlation plots showing association of hypermethylated transcripts (PUF60–202, GLUL-203, RHOB-201) with overall survival in IDH1 mutant and IDH1 wild type gliomas.