IPA predicts mortality independently of epigenetic age, with effect modification by renal function and sex via PXR‑FXR signaling

Hypothesis

Baseline serum indole‑3‑propionic acid (IPA) concentration predicts all‑cause mortality in older adults after adjusting for established aging biomarkers (e.g., GrimAge acceleration, frailty index). This association is stronger in individuals with preserved renal function (eGFR ≥ 60 mL/min/1.73 m²) and is attenuated in males versus females due to sex‑differential hepatic pregnane X receptor (PXR) expression that modulates IPA‑FXR crosstalk.

Rationale

Animal work shows IPA extends lifespan in Drosophila and improves musculoskeletal health in aged mice, yet these effects are genotype‑ and sex‑specific[1][2]. Human data link low IPA to cognitive impairment, Alzheimer’s disease, obesity, and metabolic syndrome[3], but lack mortality hazard ratios. Mechanistically, IPA reduces oxidative stress, mitochondrial dysfunction, and DNA damage[4] and activates neuronal PXR to mediate cognitive benefits of intermittent fasting[2]. Human intervention reveals that polyphenol‑rich diets raise serum IPA only when renal function is normal, and IPA changes correlate with CRP (β = 0.32, p = 0.010)[5], highlighting kidney clearance as a key modulator not yet examined in aging contexts.

We propose that IPA’s systemic effects extend beyond direct antioxidant activity: hepatic PXR activation by IPA induces fibroblast growth factor 19 (FGF19) signaling, which suppresses hepatic bile acid synthesis via FXR inhibition, altering the bile acid pool and reducing senescence‑associated secretory phenotype (SASP) in peripheral tissues. Sex‑specific PXR expression (higher in females) and renal clearance together determine the circulating IPA‑FXR axis strength, thereby influencing mortality risk.

Predictions

1. In a prospective cohort, each 1‑SD increase in baseline serum IPA will be associated with a ~15 % lower hazard of death after adjusting for age, sex, comorbidities, GrimAge acceleration, and frailty.
2. The hazard ratio will be significantly stronger (interaction p < 0.01) in participants with eGFR ≥ 60 mL/min/1.73 m² compared with those with reduced eGFR.
3. Females will show a steeper mortality‑IPA slope than males, reflecting higher hepatic PXR activity.
4. Adding baseline IPA to a model containing GrimAge will improve the C‑index for 5‑year mortality prediction by ≥0.02.
5. Mediation analysis will indicate that ~30 % of IPA’s mortality effect is mediated through changes in the secondary bile acid deoxycholic acid (DCA) concentration, a downstream readout of PXR‑FXR crosstalk.

Study Design

• Population: 5,000 community‑dwelling adults aged ≥ 65 years from existing aging cohorts (e.g., Framingham, ARIC) with stored baseline serum and DNA.
• Measurements: Baseline serum IPA (LC‑MS/MS), serum creatinine/eGFR, liver PXR mRNA (proxy via peripheral blood mononuclear cell expression if feasible), GrimAge acceleration, frailty index, CRP, IL‑6, and a targeted bile acid panel (including DCA).
• Follow‑up: Minimum 5 years for vital status ascertainment.
• Analysis: Cox proportional hazards models with IPA as a continuous predictor; test IPA × eGFR and IPA × sex interaction terms; assess incremental predictive value via net reclassification improvement (NRI) and integrated discrimination improvement (IDI); conduct causal mediation analysis using the bile acid panel as mediator.

Potential Confounders & Mitigation

Dietary polyphenol intake, antibiotic use, and hepatic dysfunction could affect IPA levels. We will adjust for food frequency questionnaire–derived polyphenol scores, recent antibiotic exposure (< 3 months), and liver enzymes (ALT, AST). Sensitivity analyses will exclude participants with eGFR < 30 mL/min/1.73 m² or overt liver disease.

Expected Outcomes & Falsifiability

If the hypothesis is false, baseline IPA will show no independent association with mortality after covariate adjustment, or the association will not differ by eGFR or sex, and IPA will not improve prediction beyond GrimAge. Conversely, confirmation of the predicted interactions and mediation would elevate IPA from a mechanistic curiosity to a clinically actionable biomarker of biological age, guiding targeted interventions (e.g., renal‑preserving lifestyles, PXR modulators).