Hypothesis

In aging neurons, METTL3 is progressively recruited to cytoplasmic stress granules via its interaction with the RNA‑binding protein TIA1. This sequestration reduces the nuclear‑cytoplasmic pool of active METTL3/14 complexes, leading to transcript‑specific loss of m6A on mRNAs that contain low‑affinity DRACH motifs and are normally retained in the nucleus for co‑transcriptional methylation (e.g., ubiquitin‑proteasome and autophagy genes). Synaptic genes, which harbor high‑affinity DRACH sites and are exported rapidly, retain or gain m6A through compensatory METTL14‑WTAP activity or alternative writer complexes, explaining the observed increase in m6A on synaptic transcripts.

Mechanistic Basis

1. Stress granule accumulation – Aging increases oxidative stress and reduces autophagy, promoting persistent stress granule formation. METTL3 contains an intrinsically disordered region that binds phosphorylated TIA1, a core stress granule scaffold.
2. Selective substrate vulnerability – Proteostasis transcripts often possess suboptimal DRACH contexts and rely on nuclear METTL3 for efficient methylation; when METTL3 is trapped in granules, these transcripts become hypomethylated. Synaptic transcripts contain consensus DRACH motifs with flanking enhancers that allow residual METTL14‑WTAP or METTL3‑independent m6A deposition (e.g., via ZC3H13).
3. Feedback on proteostasis – Loss of m6A on ubiquitin‑proteasome mRNAs diminishes their translation and stability, further impairing clearance of aggregated proteins and reinforcing stress granule persistence—a vicious loop.

Predictions

1. Biochemical – In aged human or mouse cortical neurons, a larger fraction of METTL3 will co‑localize with TIA1‑positive stress granules compared with young neurons.
2. Molecular – Targeted disruption of METTL3‑TIA1 interaction (e.g., via a peptide competitor) will restore nuclear METTL3 levels and increase m6A specifically on ubiquitin‑proteasome and autophagy transcripts without globally altering m6A on synaptic genes.
3. Functional – Restoring METTL3 nuclear activity in aged neurons will enhance proteasome activity, reduce ubiquitinated protein aggregates, and ameliorate age‑related synaptic dysfunction.
4. Falsifiability – If preventing METTL3 sequestration fails to increase m6A on proteostasis transcripts or does not rescue proteasome function, the hypothesis is falsified.

Experimental Design

• Model systems: Primary cortical neurons from young (3 mo) and aged (24 mo) mice; human iPSC‑derived neurons cultured with pro‑aging serum.
• Assays:
  • Proximity ligation assay (PLA) or immunofluorescence for METTL3/TIA1 colocalization.
  • Subcellular fractionation followed by Western blot for METTL3 in nuclear vs cytoplasmic fractions.
  • MeRIP‑seq or miCLIP to quantify m6A changes on specific gene sets (proteostasis vs synaptic) after interventions.
  • Proteasome activity assays (fluorogenic substrates) and filter‑trap assay for insoluble ubiquitin‑positive aggregates.
  • Electrophysiology (mEPSC frequency) to assess synaptic function.
• Interventions:
  • Expression of a TIA1‑binding deficient METTL3 mutant (ΔIDR).
  • Treatment with stress‑granule inhibitor (e.g., ISRIB) or overexpression of HSP70 to dissolve granules.
  • Controls: scrambled peptide, overexpression of WT METTL3.

Expected Outcomes and Interpretation

• If hypothesis is correct: Aged neurons show increased METTL3/TIA1 PLA signal; disruption of this interaction raises nuclear METTL3, restores m6A on ubiquitin‑proteasome transcripts (↑2‑fold), enhances proteasome activity (+30‑40 %), reduces aggregates, and improves synaptic transmission. No significant change in m6A levels on synaptic genes.
• If hypothesis is false: Manipulating METTL3‑TIA1 interaction does not alter m6A on proteostasis transcripts, proteasome activity remains unchanged, or both synaptic and proteostasis transcripts are affected indiscriminately, indicating that METTL3 sequestration is not the selective mechanism.

This framework directly links a cell‑biological stress response to the epitranscriptomic selectivity observed in aging brain, provides a clear, testable mechanism, and offers a route to intervene in age‑related proteostatic decline.