Hypothesis: Inflammasome‑Mediated Neuronal Eviction Links Brain Aging to Immunosenescence

Core idea – Chronic low‑grade inflammation activates the NF‑κB/NLRP3 axis in microglia, which tags metabolically expensive, weakly connected neurons for complement‑dependent phagocytosis. This neuronal "eviction" mirrors synaptic pruning and runs in parallel with T‑cell exhaustion, suggesting that the aging brain sacrifices inefficient cells to preserve energy under rising inflammatory load.

Mechanistic rationale

1. Microglial inflammasome priming – Aged macrophages show lowered NLRP3 activation thresholds due to reduced TBK1/IKKε activity 1. Microglia share this signaling node; accumulated mitochondrial DNA from stressed neurons can engage cGAS‑STING, driving NF‑κB transcription of NLRP3 and pro‑IL‑1β 2.

2. Neuronal tagging – Neurons with high ATP demand and low synaptic activity release extracellular ATP and expose phosphatidylserine, marking them for microglial recognition. Complement component C1q binds these "stressed" neurons, a step amplified by NF‑κB‑driven C1q upregulation in microglia (see 3).

3. Phagocytic eviction – Activated NLRP3 inflammasome triggers caspase‑1–mediated release of IL‑1β and IL‑18, which potentiate microglial phagocytic activity. The result is selective removal of costly, under‑utilized neurons, preserving network efficiency at the expense of total cell count – a process akin to the seed idea that the brain evicts neurons for inefficiency.

4. Link to T‑cell exhaustion – The same IL‑1β‑rich milieu fuels systemic inflammation that drives NF‑κB in T cells, accelerating epigenetic commitment to exhaustion after ~2 weeks of chronic stimulation 4. Thus, neuronal eviction and T‑cell dysfunction share a common inflammatory trigger.

Testable predictions

• Prediction 1: In aged mice, microglial NLRP3 activation (measured by ASC speck formation) will spatially correlate with zones of neuronal loss that show low synaptic marker density (e.g., synaptophysin) and high metabolic stress markers (e.g., p‑AMPK).
• Prediction 2: Pharmacological enhancement of TBK1/IKKε activity (using a small‑molecule agonist) or NLRP3 inhibition (MCC950) will reduce complement C1q deposition on neurons, preserve cortical thickness, and improve performance on memory tasks.
• Prediction 3: These interventions will also lower serum IL‑1β, reduce PD‑1^hi TIM‑3^hi exhausted T‑cell frequencies, and improve viral clearance in a chronic infection model.
• Prediction 4: Neurons isolated from treated animals will exhibit reduced mitochondrial DNA release into the cytosol, indicating attenuated cGAS‑STING activation.

Experimental design

Subjects: 20‑month‑old C57BL/6 mice, split into four groups (n=10 each): vehicle, TBK1/IKKε agonist, NLRP3 inhibitor, combined treatment.

Interventions: Drug administration via osmotic pump for 4 weeks, starting at 18 months to capture pre‑exhaustion window.

Readouts:

• Iba1^+ ASC speck immunohistochemistry (confocal quantification).
• Neuronal density (NeuN^) and synaptic density (synaptophysin^) in prefrontal cortex and hippocampus.
• Complement C1q co‑localization with NeuN^.
• Serum cytokine panel (IL‑1β, IL‑6, TNF‑α) via ELISA.
• Flow cytometry of CD8^+ T cells for PD‑1, TIM‑3, TOX expression.
• Behavioral assays: Y‑maze spontaneous alternation, novel object recognition.
• Mitochondrial DNA in microglial cytosol (qPCR for mtDNA after subcellular fractionation).

Analysis: Two‑way ANOVA with post‑hoc Tukey; correlation analysis between microglial inflammasome score and neuronal loss.

Falsifiability: If TBK1/IKKε activation or NLRP3 blockade does not reduce neuronal C1q tagging, nor improve cognition or T‑cell phenotypes, the hypothesis that inflammasome‑driven neuronal eviction underlies brain aging‑immunosenescence linkage is refuted.