Hypothesis

In the aged hippocampus, reduced m6A methylation of synaptic transcripts not only diminishes local translation of plasticity proteins (e.g., CAMKIIα, GluA1) but also triggers a compensatory increase in METTL14 activity that methylates the ubiquitin ligase UBR1. This methylation suppresses autophagy, leading to accumulation of damaged proteins and sequestration of translational machinery. The combined effect creates a proteostatic trap that locks synapses in an over‑consolidated, low‑plasticity state. Restoring autophagy should break this trap, re‑enable m6A‑independent translation, and rescue cognitive flexibility without restoring m6A levels.

Mechanistic Model

1. m6A deficit – Aging lowers global m6A in CA1/CA3/DG, decreasing ribosome recruitment to plasticity‑gene transcripts (loss of m6A‑dependent translation initiation) Global m6A RNA methylation levels decrease....
2. METTL14 shift – Contrary to the decline of METTL3 in some tissues, hippocampal METTL14 protein or activity remains stable or rises with age, promoting methylation of UBR1 (an E3 ubiquitin‑ligase) METTL14 methylation of UBR1 inhibits autophagy....
3. Autophagy block – Methyl‑UBR1 exhibits reduced affinity for LC3, impairing autophagosome formation and causing p62/SQSTM1 accumulation, which in turn binds and stalls ribosomes and eIF4E complexes.
4. Proteostatic feedback – Accumulated ubiquitinated proteins form insoluble aggregates that further sequester translation factors, creating a vicious cycle where translation stays low despite available mRNA.
5. Outcome – Synapses retain old protein composition, exhibit reduced capacity for activity‑dependent remodeling, and manifest as behavioral rigidity.

Predictions

• P1: Aged hippocampal neurons will show increased METTL14‑dependent UBR1 methylation concurrent with decreased autophagy flux (lower LC3‑II/I, higher p62).
• P2: Pharmacological or genetic induction of autophagy (e.g., rapamycin, spermidine, or ATG5 overexpression) will normalize p62 levels and restore synaptic protein synthesis rates without altering global m6A.
• P3: Rescue of translation will correlate with restored LTP magnitude and improved performance on reversal‑learning tasks.
• P4: Inhibiting autophagy in young mice will phenocopy the aged translation deficit and exacerbate m6A‑loss effects.

Experimental Design

Subjects: Young (3‑mo) and aged (20‑mo) C57BL/6J mice.

Groups (n=10 per age):

• Vehicle control
• Rapamycin (autophagy inducer) administered via chow for 4 weeks
• Spermidine (alternative inducer) via drinking water
• ATG5 hippocampal overexpression via AAV (aged only)
• Chloroquine (autophagy inhibitor) in young mice (to test P4)

Readouts:

• Western blot / quantitative proteomics for m6A (MeRIP‑seq), METTL3/14, methyl‑UBR1 (IP‑Western), LC3‑II/I, p62, CAMKIIα, GluA1, ribosomal puromycin incorporation (SUnSET assay).
• Autophagic flux assay (LC3‑II turnover with bafilomycin A1).
• Electrophysiology: field EPSPs and LTP in CA1.
• Behavior: Morris water maze reversal learning, novel object recognition.

Statistical plan: Two‑way ANOVA (age × treatment) with post‑hoc Tukey; significance set at p<0.05.

Potential Outcomes and Falsifiability

• If autophagy induction restores CAMKIIα/GluA1 synthesis, LTP, and reversal learning without changing m6A levels, the hypothesis is supported.
• If autophagy induction fails to rescue translation or behavior despite clearing p62, the proteostatic trap model is weakened, suggesting m6A loss alone is sufficient.
• If autophagy inhibition in young mice does not reproduce the aged translation deficit, the causal link between autophagy failure and translational silencing is questioned.

This framework directly links epitranscriptomic, ubiquitin‑ligase, and autophagy pathways, offering a testable, mechanistic alternative to the simple “over‑consolidation versus decay” narrative and indicating that re‑introducing controlled uncertainty (via autophagy‑mediated protein turnover) could reinstate synaptic adaptability in the aging brain.