Aging tissues accumulate dysfunctional cells not because they are actively pruned but because surveillance systems collapse. In the brain, neurons suffer oxidative damage and impaired protein clearance while microglial phagocytosis wanes; in the colon, DNA mismatch repair‑deficient (dMMR) clones evade immune elimination and expand, creating field cancerization. Both failures share a metabolic hinge: declining NAD+ levels reduce sirtuin activity and increase PARP consumption, leading to hyperacetylation of FOXO3 and NF‑κB p65. Hyperacetylated FOXO3 cannot drive autophagic gene expression in neurons, allowing damaged proteins to persist. Hyperacetylated p65 sustains a chronic inflammatory state that exhausts CD8+ T cells and compromises their cytotoxic granule release in the colonic mucosa. Consequently, the same NAD+ deficit simultaneously weakens neuronal quality control and immune surveillance, permitting the buildup of metabolically expensive, weakly connected neurons and hypermutated dMMR epithelial clones.

Testable prediction: Restoring NAD+ in aged mice will reinstate FOXO3‑dependent autophagy in neurons and restore CD8+ T cell cytotoxicity against dMMR clones, reducing both neuronal loss and clonal expansion. Experimentally, 24‑month‑old C57BL/6 mice receive nicotinamide riboside (NR) or vehicle for 12 weeks. Brain sections are stained for NeuN, LC3‑II, and 4‑HNE to quantify neuronal survival, autophagic flux, and oxidative damage. Colon tissues are immunostained for MLH1 loss (dMMR marker), CD8, granzyme B, and γH2AX to assess clone burden, T cell infiltration, cytotoxic activity, and DNA damage. Serum NAD+ metabolites are measured by LC‑MS. Expected outcomes: NR‑treated mice show increased NAD+, higher LC3‑II/NeuN ratios, lower 4‑HNE, and reduced neuronal loss versus controls. In the colon, NR mice display fewer MLH1‑negative crypts, elevated CD8+ granzyme B+ cells, and decreased γH2AX foci in dMMR lesions. If NAD+ restoration fails to improve either readout, the hypothesis is falsified. This links a single metabolic node to two seemingly disparate aging phenotypes, offering a unified intervention target.