Hypothesis

In normal aging, synaptic pruning follows a use-it-or-lose-it logic that conserves energy by removing weak connections. This process depends on microglial complement tagging (C1q rise, CD47 fall) and requires sufficient mitochondrial ATP to power phagocytic machinery. We hypothesize that age‑related mitochondrial decline reduces the energy available for microglia, causing a shift from selective, efficient pruning to stochastic, excessive removal of synapses regardless of activity. Consequently, the brain loses its ability to refine circuits, leading to cognitive decline that resembles neurodegenerative pathology despite absent neurodegeneration.

Mechanistic Basis

1. Energy‑Dependent Phagocytosis – Microglial engulfment of tagged synapses consumes ATP for actin remodeling and lysosomal acidification. Studies show microglial phagocytic capacity drops when cellular ATP falls below ~70 % of young adult levels [3].
2. Mitochondrial Decline in Aging – Aging neurons and glia exhibit decreased oxidative phosphorylation and increased ROS, limiting ATP supply [4].
3. Complement Tagging Remains Intact – C1q upregulation and CD47 downregulation persist, continuing to flag low‑activity synapses for removal [1][2].
4. Outcome – When ATP is insufficient, microglia cannot selectively execute the “eat me” signal; instead, they exhibit frustrated phagocytosis or release inflammatory mediators that destabilize nearby synapses, causing non‑selective loss.

Testable Predictions

• Prediction 1: In aged mice, microglial ATP levels will correlate positively with the specificity of synapse elimination (i.e., higher ATP → pruning enriched for low‑activity synapses; lower ATP → pruning independent of activity).
• Prediction 2: Pharmacologically boosting microglial mitochondrial function (e.g., with SS‑31 peptide) in aged animals will restore selective pruning and preserve cognitive performance without altering overall microglial numbers.
• Prediction 3: Genetic reduction of neuronal activity in specific circuits will protect those synapses from loss in young adults but not in aged animals with mitochondrial impairment, indicating loss of activity‑dependence.
• Prediction 4: Post‑mortem human tissue will show a negative correlation between microglial mitochondrial markers (e.g., COX‑IV, ATP5A) and the variance of synapse density across activity‑matched cortical layers.

Experimental Approach

1. Measure Microglial Energy State – Isolate microglia from young (3 mo) and old (24 mo) mice; quantify ATP, mitochondrial membrane potential, and ROS using flow cytometry.
2. Assess Pruning Selectivity – Use viral‑mediated GFP labeling of defined excitatory synapses combined with two‑photon imaging to track elimination of high‑ vs low‑activity spines over weeks. Correlate spine fate with microglial ATP levels from the same animals (via ex vivo assay after imaging).
3. Rescue Experiments – Treat aged mice with SS‑31 or a NAD⁺ booster (NR) for 4 weeks; repeat pruning selectivity assays and behavioral tests (Morris water maze, novel object recognition).
4. Human Validation – Obtain prefrontal cortex samples from cognitively normal elderly and Alzheimer’s donors; immunostain for microglial COX‑IV, C1q, and synaptophysin; compute synapse density variance across layers.

Potential Impact

If confirmed, this hypothesis reframes cognitive aging as a failure of energy‑dependent synaptic refinement rather than outright neurodegeneration. It suggests that interventions targeting microglial metabolism could preserve circuit precision and delay cognitive decline, offering a therapeutic avenue distinct from anti‑amyloid or anti‑tau strategies.

References

[1] https://doi.org/10.1523/jneurosci.1333-13.2013 [2] https://www.bu.edu/kilachandcenter/cognitive-decline-in-old-age-may-be-linked-to-increased-pruning-of-brain-cell-connections/ [3] https://doi.org/10.1101/2024.06.24.600214 [4] https://blog.cirm.ca.gov/2016/05/02/an-inside-look-reveals-the-adult-brain-prunes-its-own-branches/