Hypothesis

Rapamycin’s lifespan‑extending effects arise primarily from reactivation of the Hedgehog (Hh)‑GLI transcriptional program in aged stromal cells, which restores morphogen gradients required for tissue repair and suppresses senescence‑associated secretory phenotype (SASP). While rapamycin does mimic caloric‑restriction‑like metabolic stress (e.g., ULK1‑dependent autophagy), the data show that DNA‑damage reduction, antioxidant upregulation, and senescence reversal occur even when autophagy is blocked, indicating a parallel rejuvenation route. We propose that mTORC1 inhibition in aged fibroblasts and mesenchymal stromal cells relieves the ciliary trafficking block on Smoothened (SMO), permitting GLI2/GLI1 accumulation in the cilium and nuclear translocation of active GLI transcriptional complexes. This restores Hh‑driven expression of regenerative genes (e.g., Ptch1, Gli1, Foxf1) and antioxidant enzymes (Cat, Sod2, Prdx3), thereby lowering oxidative DNA lesions and p21‑mediated senescence independent of bulk protein synthesis reduction.

Mechanistic Model

1. Baseline aging – Elevated mTORC1 activity phosphorylates and retains 14‑3‑3 proteins on SMO, preventing its ciliary entry; GLI2 is sequestered at the tip of the cilium in a repressor conformation, leading to low Gli1 transcription.
2. Rapamycin treatment – mTORC1 inhibition reduces 14‑3‑3–SMO interaction, freeing SMO to traffic to the primary cilium via the IFT‑B complex. Ciliary SMO then suppresses SUFU‑mediated GLI processing, allowing full‑length GLI2/GLI1 to act as transcriptional activators.
3. Downstream outcomes – Activated GLI induces:
   • Antioxidant genes (Cat, Sod2, Prdx3) → reduced ROS → lower DNA lesion load.
   • Pro‑regenerative chemokines (Cxcl12, Tgfb1) that modulate SASP toward a tissue‑supportive phenotype.
   • Cell‑cycle regulators (Cyclin D1, Myc) that preserve proliferative capacity in quiescent stromal niches.
4. Regeneration trade‑off – In tissues requiring acute proliferative bursts (satellite cells, axons), mTORC1‑driven translation is indispensable; thus rapamycin impairs regeneration there, consistent with the observed tissue‑specific dichotomy.

Testable Predictions

• Prediction 1: In aged human gingival fibroblasts, rapamycin will increase ciliary SMO density and nuclear GLI1 levels; this increase will be abolished by pharmacological inhibition of IFT‑B (e.g., ciliobrevin D) or by SMO antagonists (vismodegib), and will correlate with reduced SA‑β‑Gal and γH2AX foci.
• Prediction 2: Genetic ablation of Gli1 in fibroblasts will block rapamycin‑mediated antioxidant upregulation and senescence rescue, despite normal mTORC1 inhibition and autophagy induction.
• Prediction 3: In a murine models of aged muscle, rapamycin will improve stromal‑cell‑derived Hh signaling (measured by Gli1 expression in PDGFRα+ fibroblasts) while impairing satellite‑cell proliferation; rescuing Hh signaling downstream of SMO (e.g., with a SAG analog) should ameliorate the regenerative defect without affecting rapamycin’s systemic lifespan extension.
• Prediction 4: Rapamycin‑treated old mice treated with a cilia‑disrupting mutation (IFT88 hypomorph) will show diminished lifespan benefits compared with wild‑type controls, indicating that cilia‑dependent Hh reactivation contributes to longevity.

Falsifiability

If any of the following occurs, the hypothesis is weakened:

• Rapamycin still extends lifespan and reduces senescence in fibroblasts lacking functional primary cilia (IFT88 KO) or with constitutively active SMO that is cilium‑independent.
• Gli1 knockdown does not affect rapamycin‑induced antioxidant gene expression or senescence markers.
• Pharmacological activation of Hh signaling (SAG) fails to mimic rapamycin’s effects on DNA damage and senescence in aged stromal cells.

Conclusion

By positioning mTORC1 as a regulator of ciliary Hedgehog signaling, we reframe rapamycin’s anti‑aging action as a genuine restoration of developmental morphogen gradients that promote stromal repair and antioxidant defenses, rather than a mere imitation of famine. This mechanistic shift predicts specific, experimentally accessible nodes (SMO ciliary trafficking, GLI1 transcriptional output, primary‑cilium integrity) whose manipulation will either corroborate or refute the model.