Hypothesis

Persistent theta‑band activity in medial entorhinal cortex (MEC) layer II stellate cells promotes activity‑dependent calcium influx that, owing to their intrinsic lack of calbindin and aFGF, overwhelms cytosolic buffering and triggers localized tau phosphorylation. Phosphorylated tau is then packaged into activity‑released exosomes and secreted preferentially toward afferent hippocampal inputs, seeding retrograde tau pathology that follows the gradient of synaptic usage.

Mechanistic Rationale

• Selective vulnerability: Layer II stellate cells (Reelin+, Wfs1+) lack calbindin and aFGF, removing key calcium‑buffering and neuroprotective mechanisms [1]. This makes them prone to calcium‑dependent kinase activation (e.g., CDK5, GSK‑3β) during sustained firing.
• Activity‑linked tau modification: Non‑aggregated tau suppresses neuronal firing and alters grid‑cell firing patterns [1]. We propose the converse: heightened firing drives tau phosphorylation via calcium‑activated kinases, creating a feed‑forward loop where early tau pathology further disrupts grid oscillations.
• Vesicular release pathway: Active neurons release tau‑laden exosomes and microvesicles in an activity‑dependent manner [2]. In MEC stellate cells, theta‑rhythmic bursts would synchronize exosome release toward dendritic terminals that project to CA1 and subiculum.
• Retrograde bias: Hippocampal injection of tau fibrils induces upstream EC pathology [4], suggesting that afferent terminals are efficient uptake sites. If stellate cells secrete tau preferentially at synapses, hippocampal neurons would internalize it and transport it retrogradely via endocytic pathways, explaining the observed retrograde dominance.
• Proteostatic collapse: Age‑related decline in autophagy/lysosomal function disproportionately affects calbindin‑deficient neurons [5], reducing clearance of exosomal tau and accelerating tangle formation in the EC before spreading.

Testable Predictions

1. Chemogenetic silencing of MEC layer II stellate cells will reduce extracellular tau levels in hippocampal interstitial fluid and slow the emergence of p‑tau in CA1, despite unchanged expression of tau transgenes.
2. Optogenetic enhancement of theta‑range bursting in these cells will increase exosomal tau release (measured by CSF exosome tau ELISA) and accelerate retrograde tau spread to EC Layer II.
3. Genetic rescue of calbindin specifically in Reelin+ stellate cells will attenuate activity‑dependent tau phosphorylation and protect against tangle formation, without altering overall neuronal excitability.
4. Blockade of exosome release (e.g., via GW4869 or neuronal‑specific knockout of Rab27a) will diminish hippocampal tau seeding after EC‑specific tau expression, while somatic tau accumulation in stellate cells remains unchanged.
5. Pharmacological inhibition of CDK5/GSK‑3β during periods of high theta activity will prevent activity‑linked tau phosphorylation and disrupt the feed‑forward loop.

Experimental Approach

• Use AAV‑DIO‑hM4Di or AAV‑DIO‑ChR2 in Reelin‑Cre mice to inhibit or excite stellate cells while monitoring local field theta power and hippocampal tau via microdialysis and ELISA.
• Employ AAV‑synaptophysin‑GFP‑exosome reporter to visualize activity‑dependent exosome release from stellate cell terminals in hippocampal slices.
• Cross Reelin‑Cre;Calb1‑flox mice with tau‑P301S lines to assess the impact of calbindin rescue on pathology progression.
• Apply exosome release inhibitors or kinase inhibitors via osmotic pumps and quantify p‑tau spread using immunohistochemistry and PET‑compatible tau ligands.
• Perform single‑cell proteomics on FACS‑sorted stellate cells from young vs. aged animals to identify dysregulation of autophagy/lysosomal pathways linked to calcium load.

If these predictions hold, the hypothesis would establish a direct mechanistic bridge between grid cell network dynamics, intrinsic calcium‑handling deficits, and the directional spread of tau, offering novel targets for early‑stage intervention.