Exon-intron architecture determines mRNA stability by dictating m6A deposition

Abstract

N6-methyladenosine (m6A), a widespread destabilizing mark on mRNA, is non-uniformly distributed across the transcriptome, yet the basis for its selective deposition is unknown. Here, we uncover that m6A deposition is not selective. Instead, m6A distribution is exclusion-based: m6A-consensus harboring sites are methylated by default, unless they are within a window of up to ∼200 nt from an exon-intron junction. A simple model, relying exclusively on presence of m6A motifs and exon-intron architecture allows high accuracy recapitulation of experimentally-measured m6A profiles and of all m6A hallmarks. We further establish that m6A serves as the long-sought mechanism underlying the strong association between exon-intron architecture and mRNA stability. Our findings establish a mechanism by which the memory of nuclear RNA splicing is covalently etched on an mRNA, in the form of m6A, and determines its cytoplasmic stability, with broad implications on the regulation, function, and evolution of both m6A and mRNA stability.

R2Dtool: Integration and visualization of isoform-resolved RNA features

“…1, right panel). The metaJunction plot shows a clear exclusion of m6A sites around 200nt away from the junctions, which is consistent with previous findings reporting a depletion of m6A from the vicinity of exon-exon junctions (Uzonyi et al, 2022) (Ke et al, 2017).…”

“…1H). This plot, produced with the command 'r2d plotMetaJunction', shows a clear exclusion of m6A sites in a window of 200nt around the exon-exon junctions, in agreement with previous reports (Uzonyi et al, 2022). Like the metatranscript plot, the metajunction plot displays the proportion of positive features at intervals around each splice-site, in increments of 1nt.…”

“…1H). The plot, produced with the command R2_metaJunction.R , shows a clear exclusion of m6A sites in a window of 200nt around the exon-exon junctions, in agreement with previous reports (Uzonyi et al, 2022). Similarly, R2_plotMetaCodon.R can be used to plot the distribution of RNA features in absolute distance around start and stop codons (also available at https://github.com/comprna/R2DTool).…”

“…R2Dtool exploits the isoform-resolved mapping of RNA features, such as those obtained from long-read sequencing, to enable simple, reproducible, and lossless integration, annotation, and visualization of isoform-specific RNA features. We illustrate R2Dtool’s capabilities with the analysis of isoform-resolved messenger RNA (mRNA) modification data, which can be accurately obtained from nanopore long-read sequencing data (Acera Mateos et al, 2022; Hendra et al, 2022) and its interpretation in the right mRNA isoform context has uncovered important properties and regulatory mechanisms (Gleeson et al, 2024; Uzonyi et al, 2022).…”