# Hypothesis
Age‑associated shifts in the gut microbiome increase production of the bacterial D‑amino acid D‑serine, which translocates across an age‑compromised blood‑brain barrier (BBB) and drives NMDA‑receptor‑mediated calcium excitotoxicity in cortical neurons. This excitotoxic cascade disrupts cortical metabolic coupling with peripheral tissues, amplifies neuroinflammation, and accelerates cognitive decline.

## Rationale
- Multi‑organ metabolomics shows accumulation of gut‑derived microbial catabolites (e.g., TMAO, indole‑3‑acetic acid) in the aged brain, indicating BBB leakage【1](https://pmc.ncbi.nlm.nih.gov/articles/PMC12968584/)】.
- Aging weakens metabolite correlations between cortex and serum, reflecting a loss of systemic metabolic coupling【1](https://pmc.ncbi.nlm.nih.gov/articles/PMC12968584/)】.
- Certain gut bacteria (e.g., *Enterococcus*, *Lactobacillus*) possess D‑serine synthases and D‑amino acid aminotransferases that can release D‑serine into the lumen【2](https://pmc.ncbi.nlm.nih.gov/articles/PMC11972123/)】.
- D‑serine is a potent co‑agonist at the NMDA receptor glycine site; elevated extracellular D‑serine enhances neuronal calcium influx and can trigger excitotoxic pathways【3](https://journals.sagepub.com/doi/10.1177/11795549251384582)】.
- NMDA‑receptor overactivation is linked to neuroinflammatory cytokine release (IL‑6, IL‑1β, TNF‑α) and oxidative stress, hallmarks of inflammaging【1](https://pmc.ncbi.nlm.nih.gov/articles/PMC12968584/)】.

## Mechanistic Insight
We propose that age‑related dysbiosis enriches bacterial strains with heightened D‑serine synthesis while simultaneously depleting taxa that consume D‑serine (e.g., via D‑amino acid oxidase activity). The resulting luminal D‑serine surplus is absorbed through upregulated intestinal peptide transporters (PEPT1/2) that have increased permeability in aging gut epithelium. Circulating D‑serine then crosses the BBB via elevated expression of the neutral amino acid transporter SLC1A4/ASC‑1, which is known to be upregulated in response to inflammatory cues. Once in the brain interstitial fluid, D‑serine binds NMDA receptors, promoting prolonged channel opening, calcium overload, activation of calpains and caspases, and subsequent release of damage‑associated molecular patterns (DAMPs). DAMPs activate microglia, amplifying IL‑1β, IL‑6, and TNF‑α production, which further tightens a positive feedback loop that exacerbates BBB permeability and cortical metabolic decoupling.

## Testable Predictions
1. **Metabolite profiling** – Aged mice and humans will show significantly higher fecal, serum, and cerebrospinal fluid (CSF) D‑serine levels compared with young counterparts; levels will correlate positively with BBB permeability markers (e.g., serum‑to‑CSF albumin ratio) and negatively with cortex‑serum metabolite correlation coefficients.
2. **Microbiota causality** – Germ‑free mice colonized with a defined consortium of high D‑serine‑producing bacteria will exhibit accelerated cognitive decline, increased cortical calcium signaling biomarkers (p‑CAMKII, p‑CREB), and disrupted cortical‑periphery metabolic coupling, whereas colonization with D‑serine‑deficient mutants will not.
3. **Transport blockade** – Pharmacological inhibition of SLC1A4/ASC‑1 (using α‑methyl‑serine) or genetic knockdown in brain endothelial cells will reduce brain D‑serine accumulation, attenuate NMDA‑receptor‑mediated calcium fluxes, lower neuroinflammatory cytokine levels, and restore cortical‑serum metabolite correlations in aged animals.
4. **Receptor antagonism** – Chronic low‑dose NMDA‑receptor glycine‑site antagonism (e.g., with 7‑chlorokynurenic acid) will rescue cognitive performance and metabolic coupling without altering peripheral D‑serine levels, indicating a central mechanism of action.

## Falsifiability
If any of the following observations hold, the hypothesis is falsified:
- No significant elevation of D‑serine in aged CSF despite confirmed gut microbiome shifts.
- Blocking D‑serine transport or NMDA‑receptor glycine site fails to improve cortical‑periphery metabolic coupling or cognitive outcomes in aged models.
- Microbiome manipulations that alter D‑serine production do not change brain D‑serine levels or neuroinflammatory readouts.

## Experimental Outline (brief)
- **Cohorts**: Young (3 mo) vs. aged (18‑24 mo) mice; parallel human aging cohort with metagenomic stool, serum, CSF sampling.
- **Measurements**: Targeted LC‑MS/MS for D‑serine; BBB permeability via Evans blue or fluorescent albumin; NMDA‑receptor activity via phosphorylated GluN1 subunits; cytokine panels; untargeted metabolomics for cortex‑serum coupling analysis.
- **Interventions**: Antibiotic depletion, fecal microbiota transfer from aged donors, colonization with engineered *E. coli* overexpressing or knocking‑down D‑serine synthase; SLC1A4 antisense oligonucleotides; NMDA‑receptor glycine‑site antagonist.
- **Analysis**: Mixed‑effects models linking microbial D‑serine production rates to CSF D‑serine, BBB metrics, cortical calcium biomarkers, and metabolic coupling scores; mediation analysis to test whether D‑serine mediates the microbiome‑cognition relationship.