Gut Senescent Cells as Modulators of the Microbiota‑Gut‑Brain Axis: A Threshold Model for Mental Health
Hypothesis
Transient senescence in gut epithelial cells sustains a homeostatic microbiome by releasing SASP factors that stimulate mucosal barrier integrity and microbial metabolite production; when senescent burden exceeds a tissue‑specific threshold, these same signals become pro‑inflammatory and disrupt microbiota‑gut‑brain signaling, worsening anxiety‑ and depression‑like phenotypes.
Mechanistic Rationale
- Beneficial SASP in low‑burden senescence – Senescent epithelial cells secrete PDGF‑AA, TGF‑β, and indole‑3‑propionic acid (IPA) precursors that promote tight‑junction protein expression and stimulate commensal production of short‑chain fatty acids (SCFAs) [1][2]. These metabolites activate vagal afferents via free‑fatty‑acid receptors (FFAR2/3) and modulate HPA‑axis activity, dampening stress responses [5].
- Threshold shift with accumulation – Persistent senescence amplifies SASP components such as IL‑6, IL‑1β, and MMP‑9, which increase intestinal permeability, facilitate translocation of lipopolysaccharide (LPS), and drive low‑grade systemic inflammation [3][4]. LPS engages TLR4 on enterochromaffin cells, altering serotonin synthesis and vagal signaling toward a pro‑anxiety state.
- Microbiota feedback – Dysbiotic shifts induced by leaky gut reduce SCFA‑producing taxa, diminishing FFAR2/3 signaling and removing a brake on neuroimmune activation, creating a vicious loop that links gut senescence to mood pathology.
Testable Predictions
- Prediction 1: In aged mice (24‑mo), gut‑targeted senolytic treatment (e.g., navitoclax conjugated to a gut‑restricted polymer) will decrease p16^INK4a^‑positive epithelial cells by ~50% but will increase serum LPS, fecal calprotectin, and anxiety‑like behavior in the elevated plus maze and forced swim test compared with vehicle controls.
- Prediction 2: The senolytic‑induced behavioral changes will correlate with reduced fecal IPA and SCFA concentrations and attenuated vagal nerve firing recorded via electrophysiology.
- Prediction 3: Co‑administration of IPA (10 mg/kg/day) or a SCFA mixture (acetate, propionate, butyrate) will rescue the anxiety‑like phenotype and normalize microbiota composition without affecting senescent cell clearance.
- Prediction 4: Germ‑free mice receiving senolytic‑treated aged mouse microbiota will exhibit heightened stress‑induced corticosterone responses, confirming that the microbiota mediates the behavioral effect.
Experimental Design (outline)
- Groups: young (3‑mo) vehicle, aged vehicle, aged gut‑senolytic, aged senolytic + IPA, aged senolytic + SCFA mix (n=12 per group).
- Interventions: oral gut‑restricted navitoclax (50 mg/kg, 3×/week for 4 weeks); IPA or SCFA mix in drinking water.
- Readouts: immunohistochemistry for p16^INK4a^ and SASP cytokines in colonic crypts; 16S rRNA sequencing; fecal metabolomics (IPA, SCFAs); serum LPS, cytokines; vagal afferent firing (ex vivo); behavioral assays; HPA‑axis activity (cortisol AUC after restraint stress).
Falsifiability
If gut‑targeted senolysis does not increase permeability, inflammatory markers, or anxiety‑like behavior, or if metabolite supplementation fails to rescue the phenotype, the hypothesis that low‑burden gut senescence supports microbiota‑gut‑brain homeostasis would be refuted. Conversely, confirmation would support a threshold model where senescence is chaperone‑like until it overwhelms tissue integrity, directly linking the "witness" analogy to neuropsychiatric outcomes.