Hypothesis

Mechanistic Basis

• Short-chain fatty acids (SCFAs) such as acetate, propionate, and butyrate produced by gut microbiota can cross the blood‑brain barrier and bind GPR41/43 receptors on microglia, shifting them toward an anti‑inflammatory phenotype and lowering TSPO‑PET signal in the fronto‑insular cortex.
• SCFAs also act as histone deacetylase (HDAC) inhibitors, increasing transcription of BDNF and KLOTHO in astrocytes and neurons.
• Elevated BDNF supports synaptic plasticity and cholinergic input to Von Economo neurons (VENs), while KLOTHO enhances antioxidant defenses, together promoting VEN survival despite age‑related stress.
• Superagers maintain higher VEN densities in anterior cingulate cortex [https://www.news-medical.net/news/20250810/Special-von-Economo-neurons-may-hold-key-to-super-sharp-memory-in-802b-superagers.aspx]
• A resilient VEN network improves the integration of interoceptive signals in the anterior insula, which can be read out as higher accuracy on heartbeat‑tracking tasks [https://pmc.ncbi.nlm.nih.gov/articles/PMC3140770/]

Predictions

1. In aged mice, administering sodium butyrate in drinking water will raise circulating butyrate, reduce TSPO‑PET binding in fronto‑insula, increase VEN density (immunohistochemistry for TOMM40), and improve performance on an interoceptive conditioning assay.
2. Fecal microbiota transplantation from young donors will replicate these effects, whereas colonization with a low‑SCFA producer consortium will not.
3. Pharmacological blockade of GPR41/43 will abolish the protective effects of SCFA supplementation, confirming microglial mediation.
4. In humans, plasma butyrate concentration will correlate positively with anterior cingulate cortical thickness (MRI) and negatively with TSPO‑PET binding, and both variables will mediate the relationship between fecal microbial diversity and interoceptive accuracy (heartbeat detection).

Experimental Design (Mouse)

Groups: young control, aged control, aged + butyrate (200 mM in water), aged + FMT from young donors, aged + FMT + GPR41/43 antagonist (e.g., GLPG0974), aged + scramble control.

Outcomes (measured at 8 weeks):

• TSPO‑PET imaging of fronto‑insula (baseline and post‑treatment).
• Immunohistochemical count of VENs in layer V of anterior cingulate (TOMM40‑positive, large bipolar cells).
• Western blot for BDNF and KLOTHO in frontal cortex homogenates.
• Interoceptive assay: mice trained to lick for a reward when they detect an internal state change induced by mild intra‑gastric distillation; accuracy measured as hit rate minus false alarms.
• Plasma SCFA quantification by GC‑MS.

Analysis: One‑way ANOVA with Tukey post‑hoc; mediation analysis to test whether TSPO signal mediates the link between butyrate levels and VEN density.

Human Translation

Recruit 30 superagers (≥80 y, high memory scores) and 30 age‑matched controls. Collect stool for 16S rRNA sequencing, plasma for SCFA metabolites, structural MRI for cortical thickness, TSPO‑PET for microglial activation, and administer a heartbeat‑tracking task. Use structural equation modeling to test whether SCFA levels mediate the effect of microbiota diversity on cortical thickness, TSPO binding, and finally interoceptive performance.

Falsifiability

If chronic SCFA elevation fails to lower TSPO signal, increase VEN density, or improve interoceptive accuracy in either mice or humans, the hypothesis would be refuted. Conversely, demonstrating that microglial HDAC inhibition alone (without SCFA changes) rescues VENs would support a downstream mechanism but would not invalidate the upstream microbial link.