Hypothesis

Tau propagation switches from functional connectivity (FC)‑driven to structural connectivity (SC)‑driven modes because neuronal burst firing triggers activity‑dependent release of tau‑laden extracellular vesicles that travel along existing synapses; as vesicle release wanes and aggregates accumulate, passive diffusion along white‑matter tracts becomes dominant.

Mechanistic Rationale

• Burst firing and vesicle release: In early AD, hub regions exhibit high‑frequency bursting that stimulates calcium‑dependent exocytosis of synaptic vesicles and exosomes containing tau seeds (4). This makes FC the primary conduit for tau spread, consistent with PINN models showing FC explains >80% of early nodal tau rates (1).
• Transition to SC dominance: Repeated vesicle release leads to perisynaptic accumulation of tau oligomers that impair vesicular release machinery and promote uptake by astrocytes and microglia, reducing activity‑dependent spread. Simultaneously, oligomers seed along axonal pathways, where they move via slow bulk flow or interstitial fluid convection guided by the brain’s structural connectome (2).
• Carrying capacity: Regional vulnerability reflects a balance between vesicle release capacity and clearance efficiency; low‑SUVR regions have high clearance, limiting logistic growth (5).

Testable Predictions

1. Pharmacological suppression of neuronal burst firing (e.g., low‑dose carbamazepine) in prodromal AD mouse models will reduce tau SUVR increases in FC‑rich subcortical areas by >30% after 3 months, without affecting later‑stage cortical tau accumulation driven by SC.
2. Enhancing extracellular vesicle clearance via overexpression of lysosomal‑associated membrane protein 2a (LAMP2a) will attenuate tau spread in both early and late stages, but the effect size will be larger in SC‑dominant cortical regions (>40% reduction) than in early FC‑dominant zones (<20%).
3. Simultaneous in‑vivo calcium imaging and PET tau tracing will show a positive correlation between burst firing frequency and early tau SUVR slope (r>0.6) that diminishes as SC‑weighted tau burden rises (partial correlation controlling for SC drops to <0.2).

Experimental Approach

• Use transgenic tau‑P301S mice crossed with a chemogenetic hM4Di receptor to inhibit bursting in entorhinal cortex via CNO administration. Longitudinal PET‑tau (AV‑1451) and structural/functional MRI will quantify SC and FC contributions using multi‑layer graph diffusion models (1).
• In a second cohort, AAV‑LAMP2a will be injected bilaterally into hippocampus; tau PET and exosome isolation from CSF will assess vesicle‑linked tau burden.
• Electrophysiological recordings (Neuropixels) will capture burst dynamics; calcium imaging (GCaMP) will corroborate activity patterns.
• Statistical validation: compare observed tau trajectories to model predictions using nested likelihood ratio tests; falsify hypothesis if burst suppression does not alter early FC‑driven tau spread or if LAMP2a overexpression fails to modify SC‑driven accumulation.

This hypothesis links soft‑matter vesicle dynamics to the observed shift in connectivity dominance, offering a concrete, falsifiable mechanism that bridges network diffusion models with neuronal physiology.