Hypothesis

Core claim: During natural sleep, neuronal AMPK undergoes brief, high‑amplitude pulses that selectively activate autophagy of stalled ribosomal complexes via phosphorylation of the autophagy receptor p62/SQSTM1 at Ser403. This “ribophagy” window prevents the buildup of translation‑derived proteotoxic species that would otherwise impair synaptic plasticity. In wakefulness, AMPK activity remains low or tonic, failing to trigger this selective pathway; chronic, unregulated AMPK activation (as seen in aging) drives non‑selective autophagy and synaptic loss.

Mechanistic rationale

1. Metabolic gate: Sleep reduces cerebral glucose utilization and raises the AMP:ATP ratio, providing the biochemical trigger for AMPK Thr172 phosphorylation.
2. Spatial specificity: Hippocampal neurons express predominantly AMPKα2, which couples to the scaffolding protein AXIN 1, facilitating rapid diffusion to perisynaptic ribosomes when activated.
3. Selective autophagy trigger: Phosphorylated p62/SQSTM1 acquires a high affinity for ubiquitin‑tagged 40S subunits and the translation‑stall marker eIF2α‑P, nucleating autophagosome formation around these cargos.
4. Temporal restriction: The pulse lasts ~5‑10 min per NREM cycle, sufficient to clear a defined ribosomal load without depleting essential synaptic proteins.
5. Pathological shift: In aged or sleep‑deprived brains, AMPK activation becomes prolonged, overwhelming the selective mechanism and engaging bulk autophagy that indiscriminately degrades postsynaptic density proteins (e.g., PSD‑95, GluA1), mirroring the synaptic loss reported with sustained AMPK activation.

Testable predictions

• Prediction 1: Pharmacological blockade of AMPK (Compound C) administered during the dark phase will reduce sleep‑associated phosphorylation of p62/SQSTM1 at Ser403 and decrease co‑localization of ubiquitin‑positive 40S subunits with LC3, without altering overall LC3‑II turnover.
• Prediction 2: Chemogenetic induction of AMPKα2 pulses in hippocampal neurons during wakefulness (using a light‑activated AMPK construct) will increase ribophagy flux, lower levels of puromycin‑labeled nascent chains stalled on ribosomes, and preserve long‑term potentiation, whereas constitutive AMPK activation will cause synaptic protein loss.
• Prediction 3: Aged mice subjected to sleep‑fragmentation will show blunted p62/SQSTM1 Ser403 phosphorylation peaks during NREM, concomitant accumulation of ubiquitinated ribosomal proteins in the hippocampus, and rescued ribophagy when AMPK pulses are restored via timed AICAR infusion.

Experimental approach

• Use fiber‑photometry AMPK activity reporters (e.g., AMPKAR) combined with electrophysiological sleep staging to capture hourly pAMPKα‑Thr172 oscillations in freely moving mice.
• Perform proximity‑labeling (BioID) of p62/SQSTM1 during sleep vs wake to identify ribosomal interactors.
• Measure ribophagy flux with mCherry‑GFP‑LC3‑RPS6 reporter and quantify stalled ribosomal complexes by puromycylation and polysome profiling.
• Assess synaptic integrity via Western blot for PSD‑95, GluA1 and electrophysiological LTP.

If the predictions hold, sleep’s “autopsy” function is redefined as a temporally gated, AMPK‑driven selective ribosome quality‑control pathway, distinct from the bulk clearance implicated in chronic AMPK pathology.

References

[1] https://pmc.ncbi.nlm.nih.gov/articles/PMC5945561/ [2] https://pmc.ncbi.nlm.nih.gov/articles/PMC6941559/ [3] https://doi.org/10.1038/s41419-019-1464-x [4] https://pmc.ncbi.nlm.nih.gov/articles/PMC6764723/ [5] https://pmc.ncbi.nlm.nih.gov/articles/PMC6420092/