Hypothesis

Inflammasome‑driven metabolic reprogramming of microglia converts activity‑dependent synaptic pruning into indiscriminate synapse loss in the aging brain.

Mechanistic Basis

Healthy synaptic pruning depends on microglia recognizing phosphatidylserine (PS) exposed on inactive synapses, an "eat me" signal that correlates inversely with synaptic activity [1]. In aging, chronic low‑grade inflammation activates the NLRP3 inflammasome in microglia, leading to IL‑1β release that disrupts mitochondrial function and shifts microglial metabolism toward aerobic glycolysis [5,6]. This metabolic shift raises extracellular adenosine triphosphate (ATP) and reactive oxygen species (ROS) in the microglial phagocytic synapse.

We propose that elevated ATP acts as a dominant "find‑me" signal that overrides PS‑dependent recognition, causing microglia to engulf synapses irrespective of their activity state. Simultaneously, ROS oxidize PS residues, reducing their availability for recognition and further biasing phagocytosis toward synapses with high metabolic stress rather than low activity. Senescent microglia, which accumulate with age, sustain NLRP3 activation and secrete IL‑6, creating a feedback loop that amplifies inflammatory signaling and solidifies the metabolic phenotype [6].

Testable Predictions

1. Metabolic rescue – Pharmacological inhibition of microglial glycolysis (e.g., with 2‑deoxy‑D‑glucose) in aged mice will restore the inverse relationship between synaptic activity and PS exposure, decreasing synapse loss despite ongoing NLRP3 activation.
2. ATP blockade – Preventing extracellular ATP signaling (using apyrase or P2Y12 antagonists) will normalize microglial phagocytic selectivity and rescue synaptic density in aged animals.
3. ROS scavenging – Microglia‑targeted expression of catalase will preserve PS labeling on synapses and reduce indiscriminate engulfment, correlating with improved performance on hippocampal‑dependent memory tasks.
4. NLRP3 dependence – Genetic deletion of NLRP3 specifically in microglia will attenuate the metabolic shift (lower glycolysis, ATP release) and protect synapses, even in the presence of systemic inflammation.

Experimental Design

• Use aged (18‑month) C57BL/6 mice and microglia‑specific Cre lines (Cx3cr1‑CreER) to deliver CRISPR‑Cas9 knockdown of Hk2 (glycolysis) or Nrp3.
• Measure extracellular ATP in hippocampal slices with luciferase assay, PS exposure on dendritic spines via Annexin V‑fluorescence, and microglial phagocytic activity using pHrodo‑labeled synaptosomes.
• Quantify synapse number (synaptophysin/PSD‑95 immunostaining) and correlate with behavioral outcomes (Morris water maze, novel object recognition).
• Include young (3‑month) controls and aged mice treated with systemic NLRP3 inhibitor (MCC950) to differentiate cell‑autonomous versus systemic effects.

Potential Outcomes and Interpretation

If glycolysis inhibition or ATP scavenging restores activity‑dependent pruning and preserves synapses, the hypothesis gains support: inflammatory metabolic reprogramming, not mere synaptic "inefficiency", drives pathological loss. Conversely, if manipulations fail to alter the activity‑PS relationship or synapse numbers, the hypothesis would be falsified, suggesting that other mechanisms (e.g., complement tagging or lysosomal dysfunction) dominate age‑related synapse loss.

References

[1] Synaptic pruning of murine adult‑born neurons by microglia – https://rupress.org/jem/article/219/4/e20202304/213073/Synaptic-pruning-of-murine-adult-born-neurons-by [2] Age‑related downregulation of CX3CR1 with sustained microglial activation – https://pmc.ncbi.nlm.nih.gov/articles/PMC12692065/ [3] Senolytics may affect inflammation‑related cognitive decline – https://lifespan.io/news/senolytics-may-affect-inflammation-related-cognitive-decline/ [4] Synaptic pruning and schizophrenia – https://psychscenehub.com/psychinsights/synaptic-pruning-and-schizophrenia-2/ [5] NLRP3 inflammasome activation links systemic inflammation to functional decline – https://doi.org/10.1016/j.cmet.2013.09.010 [6] Microglial processing of tau drives IL‑6 elevation – https://pmc.ncbi.nlm.nih.gov/articles/PMC12641016/