Rapamycin extends lifespan by imposing a fasting‑like state on striatal D1‑medium spiny neurons, which in turn activates autophagy‑dependent clearance of presynaptic α‑synuclein and preserves habit learning before dopaminergic loss.

Hypothesis Chronic low‑dose rapamycin reduces mTORC1 activity specifically in D1‑receptor expressing medium spiny neurons (D1‑MSNs). This suppression triggers a hormetic stress response that up‑regulates TFEB‑driven lysosomal biogenesis and autophagic flux, leading to enhanced clearance of extracellular α‑synuclein that accumulates at corticostriatal synapses during normal aging. By preventing synaptic α‑synuclein buildup, rapamycin maintains the balance of long‑term potentiation and depression (LTP/LTD) in the direct pathway, thereby safeguarding habit‑formation learning before dopaminergic neurodegeneration begins.

Testable predictions

1. In aged (18‑month) wild‑type mice, rapamycin (14 ppm in chow) will decrease phospho‑S6K (a read‑out of mTORC1) in FACS‑sorted D1‑MSNs but not in D2‑MSNs Neuroprotective mechanism FKBP12.
2. The same treatment will increase LC3‑II/I ratio and lysosomal Lamp1 levels in synaptosomal fractions from the dorsolateral striatum, indicating heightened autophagic activity at presynaptic terminals.
3. Extracellular α‑synuclein measured by ELISA in striatal synaptosomes will be reduced by ~30 % relative to vehicle‑treated controls.
4. Behavioral assessment on a habit‑learning task (e.g., lever pressing after outcome devaluation) will show that rapamycin‑treated aged mice retain goal‑to‑habit transition performance comparable to young (3‑month) mice, whereas vehicle‑treated aged mice exhibit a deficit.
5. The effects in predictions 1‑4 will be absent in mice with D1‑MSN‑specific Raptor knockout (mTORC1 already inactive) or with D1‑MSN‑specific Atg5 knockdown (autophagy blocked), confirming that rapamycin’s benefit requires mTORC1 inhibition coupled with functional autophagy in D1‑MSNs.

Falsification If rapamycin fails to lower phospho‑S6K in D1‑MSNs, does not enhance autophagic markers, does not reduce synaptic α‑synuclein, or does not rescue habit learning despite adequate brain exposure, the hypothesis is falsified. Conversely, if rapamycin improves habit learning without changing mTORC1 activity or autophagy in D1‑MSNs, the proposed mechanistic link is insufficient and alternative mechanisms must be sought.

Broader implication This view reframes rapamycin’s anti‑aging action as a circuit‑specific stress mimic that preserves synaptic health by engaging evolutionarily conserved famine‑response programs, rather than as a global repair agent. It suggests that targeting the metabolic state of defined neuronal populations may uncouple lifespan extension from the quality of later life Aging basal ganglia reconfiguration D1‑MSN mTORC1 and dyskinesia D1‑MSN mTORC1 and FKBP12.