Hypothesis: During slow-wave sleep, AMPK activation phosphorylates astrocytic aquaporin-4 (AQP4) at serine residues, increasing its water channel activity and biasing glymphatic flow toward clearance of phosphorylated tau species while sparing amyloid-beta. This selective clearance depends on the conformational state of tau that exposes phosphorylation-dependent epitopes, which are preferentially dragged into perivascular CSF conduits when AQP4-mediated convective flux is heightened. Disruption of this AMPK‑AQP4 axis—via genetic knock‑in of non‑phosphorylatable AQP4 or pharmacological AMPK inhibition—should decouple glymphatic influx from autophagic tau degradation, leading to tau accumulation despite normal amyloid-beta clearance. We propose to test this in mice expressing human P301S tau by measuring CSF tau isoforms, AQP4 phosphorylation (p‑S321), and glymphatic influx (using intrathecal fluorescent tracer) under three conditions: (1) normal sleep, (2) sleep deprivation, and (3) sleep deprivation rescued by AMPK activator (AICAR) or astrocyte‑specific AQP4‑S321D phosphomimetic. Predictions: (a) Normal sleep shows increased AQP4‑pS321, elevated CSF tau clearance, and unchanged amyloid-beta levels; (b) Sleep deprivation reduces AQP4‑pS321, diminishes tau clearance but leaves amyloid-beta flux unchanged; (c) AMPK activation or AQP4‑S321D restores tau clearance without affecting amyloid-beta. Failure to observe these patterns would falsify the hypothesis that AMPK‑driven AQP4 phosphorylation mediates a tau‑selective glymphatic window during slow-wave sleep.
Supporting e





























