Hypothesis

Senescent glia induce age‑related synaptic decline by tagging mitochondria for phagocytosis via a complement‑independent, TREM2‑dependent pathway.

Rationale

• Synaptic ultrastructure mismatch (presynaptic bouton‑mitochondria‑synapse scaling) correlates with working‑memory decline, yet total neuron loss is minimal[1].
• Aged primates show synaptic loss without corresponding cognitive impairment, implying functional defects at surviving synapses[2].
• Senescent microglia and astrocytes secrete IL‑6/TNF‑α, driving neuroinflammation and exacerbating synaptic pathology[3].
• Senolytic clearance of senescent glia improves cognition and BDNF levels, indicating that removing these cells restores synaptic health[3].
• Epigenetic marks define senescent glia, and partial reprogramming can reset their state[4][5].

Novel Mechanism

We propose that senescent glia up‑regulate the phagocytic receptor TREM2 and its adaptor DAP12, which recognizes altered lipid patterns on stressed mitochondria (e.g., exposed cardiolipin) in presynaptic terminals. This triggers a complement‑independent engulfment of mitochondria, leading to local ATP depletion, impaired vesicle cycling, and synaptic failure. Unlike developmental complement‑mediated pruning (C1q/C3), this pathway does not require opsonization by classical complement components; instead, mitochondrial "eat‑me" signals directly engage TREM2 on senescent glia.

Testable Predictions

1. Spatial correlation – In aged mouse hippocampus, TREM2‑positive microglia will be enriched around synapses showing reduced presynaptic mitochondrial density (visualized by mito‑GFP and synaptophysin staining) compared with young controls[1][2].
2. Phospholipid exposure – Isolated mitochondria from aged presynaptic boutons will display increased surface cardiolipin, detectable by annexin V‑like probes, and this signal will correlate with TREM2 binding affinity in vitro.
3. Functional rescue – Genetic deletion of TREM2 specifically in microglia (CX3CR1‑CreER;Trem2^fl/fl) or pharmacological blockade with an anti‑TREM2 antibody will preserve mitochondrial‑synapse coupling, improve working‑memory performance in the Y‑maze, and reduce senescent‑glia SASP markers without altering overall microglial numbers.
4. Senolytic synergy – Combining senolytics (dasatinib+quercetin) with TREM2 inhibition will produce additive improvements in synaptic plasticity markers (BDNF, p‑CREB) and cognitive outcomes beyond either treatment alone[3].
5. Epigenetic link – Senescent glia exhibiting heightened TREM2 expression will carry a distinct DNA methylation signature at the Trem2 promoter, reversible by partial reprogramming (OSK induction)[4][5], linking epigenetic drift to phagocytic activation.

Experimental Approach

• Use aged (20‑month) C57BL/6J mice; perform immuno‑EM to quantify mitochondrial‑synapse apposition and TREM2+ glial contacts.
• Apply proximity ligation assay (PLA) to detect TREM2–cardiolipin interactions in situ.
• Administer anti‑TREM2 antibody (or use conditional knockout) and assess behavior (Y‑maze, novel object recognition) and electrophysiology (LTP).
• Measure SASP cytokines (IL‑6, TNF‑α) and synaptic proteins (PSD‑95, synaptophysin) via ELISA and Western blot.
• Validate epigenetic changes via bisulfite sequencing of Trem2 promoter in FACS‑sorted senescent glia (p16^INK4a^+).

Falsifiability

If aged brains show no enrichment of TREM2+ glia at mitochondria‑depleted synapses, or if TREM2 loss fails to rescue mitochondrial‑synapse coupling and cognition, the hypothesis would be refuted. Conversely, confirmation would support a model where synaptic decline in aging results from maladaptive phagocytic signaling by senescent glia rather than wholesale neuronal culling or complement‑driven pruning.