Multi-Tissue Metabolomic Signatures of Five Longevity Interventions Converge on Ergothioneine and Lipid Remodeling in Male UM-HET3 Mice

UM-HET3 metabolomics investigators

1. 01

   Five interventions profiled: rapamycin, acarbose, 17-alpha-estradiol, canagliflozin and caloric restriction, dosed from 4 to 12 months

2. 02

   Metabolomic profiles across seven tissues identified treated animals well before survival curves separate

3. 03

   Leave-one-intervention-out models trained on four interventions correctly classified mice receiving a fifth, unseen intervention

4. 04

   Ergothioneine increased consistently in brain, plasma and muscle across mechanistically distinct interventions

5. 05

   Shared lipid remodelling accompanied the ergothioneine signal

6. 06

   Suggests a candidate short-horizon efficacy readout for geroprotector screening, pending peer review

Study Design

Preprint

Sample Size

Not reported

Study Duration

Treatment from 4 to 12 months of age

Methodology

Untargeted multi-tissue metabolomics across seven tissues in male UM-HET3 mice treated from 4 to 12 months with one of five validated lifespan-extending interventions; feature-stabilized XGBoost classification pipeline; leave-one-intervention-out generalisation testing to assess shared versus intervention-specific metabolic signatures.

Limitations

PREPRINT — has not undergone peer review and must not be cited as established evidence. Male UM-HET3 mice only; no female cohort reported. The ergothioneine elevation is correlative — a downstream signature of intervention rather than demonstrated mediator — and no causal test of ergothioneine supplementation on lifespan is included. Machine-learning classification of treated versus untreated animals is not a validated surrogate for lifespan benefit. No human data.