Transient senescent cells release extracellular vesicles (EVs) that donate functional mitochondria to neighboring cells, boosting local ATP production and supporting tissue repair. This EV‑mediated mitochondrial transfer represents a mechanistic basis for their chaperone‑like roles in wound healing and tumor surveillance. In contrast, chronically accumulated senescent cells load EVs with damaged mitochondrial DNA, oxidized proteins, and a heightened SASP repertoire, converting the same vesicle pathway into a driver of inflammation and immunosuppression. We hypothesize that the functional output of senescent‑cell‑derived EVs switches from reparative to pathogenic as senescence persists, and that modulating EV release will have opposite effects in acute versus chronic tissue contexts.

Testable predictions

1. EVs isolated from transiently senescent human fibroblasts (induced by low‑dose ionizing radiation for 48 h) will contain higher ratios of intact mitochondrial DNA to oxidative damage markers and will increase oxygen consumption rate in recipient keratinocytes by ≥30 % compared with EVs from proliferating cells.
2. EVs from chronically senescent fibroblasts (induced by oncogenic RAS expression for 10 days) will carry elevated mtDNA lesions, oxidized proteins, and SASP factors (IL‑6, CXCL8) and will decrease recipient cell proliferation while increasing ROS production.
3. Pharmacological inhibition of EV release (using GW4869 to block neutral sphingomyelinase) in young mice with excisional wounds will delay closure by ~20 % and reduce granulation tissue vascularization, whereas the same inhibition in aged or diabetic mice with full‑thickness wounds will accelerate closure by ~15 % and lower histological fibrosis scores.
4. Genetic ablation of Rab27a specifically in p16^high^ senescent cells (via p16‑3MR‑Cre;Rab27a^fl/fl^) will phenocopy the drug results, confirming that the effect is senescent‑cell‑specific.

Experimental approach

• Generate transient and chronic senescent fibroblast populations using established protocols[3][4]. Isolate EVs by ultracentrifugation, quantify mitochondrial content via qPCR for mtDNA and assess damage with qPCR for lesions and immunoblot for oxidized proteins.
• Co‑culture EVs with human epidermal keratinocytes or endothelial cells; measure Seahorse OCR/ECAR, ROS (DCFDA), and proliferation (EdU).
• In vivo, use 8‑week‑old and 20‑month‑old C57BL/6 mice, and a db/db diabetic wound model. Apply GW4869 or vehicle topically daily; monitor wound area via planimetry, histology (Masson’s trichrome, CD31), and immune infiltrates (flow cytometry for Ly6G^+, F4/80^+).
• Parallel tumor‑suppression assay: transplant pre‑malignant keratinocytes into senescent‑cell‑rich or -depleted skin; monitor tumor latency and burden.

If EV‑mediated mitochondrial transfer is beneficial only when senescent cells are transient, then blocking EV secretion should impair acute wound healing while improving outcomes in chronic, inflammation‑driven ulcers. Conversely, if chronic senescent‑cell EVs are neutral or beneficial, the hypothesis would be falsified. This framework directly links the "chaperone" concept to a concrete vesicle‑based mechanism and offers a clear, falsifiable route to refine senolytic strategies.