Senescent cells do not merely secrete inflammatory SASP factors; they also load and release extracellular vesicles (EVs) enriched with molecular chaperones (e.g., HSP70, HSP90, small HSPs) that they acquire from neighboring cells through trogocytosis or endocytosis. These senescent‑cell‑derived EVs function as mobile chaperone depots, delivering proteostasis support to stressed or injured tissues. When senescent cells are cleared broadly, this chaperone‑EV supply is lost, compromising the ability of nearby cells to handle protein‑damage stress and contributing to the off‑target effects observed after senolytic treatment.

Mechanistic Rationale

• Senescent cells exhibit impaired intrinsic chaperone induction during stress [5] yet they reside in aging networks where longevity‑associated proteins are highly connected hubs enriched for chaperone pathways [4]. This suggests they may operate as signaling nodes that coordinate stress responses rather than as direct repair workers.
• Acute senescent cells promote wound healing by secreting PDGF‑AA and CCN1/CCN2 [1] and are cleared before causing harm [2]. Their early presence coincides with a surge in extracellular vesicle release observed in many wound‑healing models.
• Chronic accumulation of senescent cells drives pathology [3], but partial clearance (~30% of p16+ cells) extends healthspan without impairing immune function [7], indicating that a threshold of chaperone‑EV activity may be preserved at lower senescent‑cell loads.
• Broad senolytic ablation causes collateral damage such as reduced macrophage function, impaired immune memory, and beta‑cell loss [6], phenotypes that resemble defects in cellular chaperone capacity.

Combining these observations, we propose that senescent cells sequester chaperone proteins from their microenvironment—perhaps via surface receptors that bind exposed hydrophobic patches on misfolded proteins—and package them into EVs for export. This export buffers proteotoxic stress in adjacent stem, immune, or parenchymal cells, preserving tissue function. When senescent cells are removed, the chaperone‑EV flux drops, leaving neighboring cells vulnerable to stress‑induced dysfunction.

Testable Predictions

1. EV cargo enrichment: EVs isolated from p16+ senescent cells (in vivo or induced in culture) will show significantly higher levels of HSP70, HSP90, HSP27, and α‑B crystallin compared with EVs from non‑senescent counterparts, detectable by western blot or mass spectrometry.
2. Functional transfer: Adding senescent‑cell EVs to stressed primary fibroblasts or macrophages will increase their refolding capacity (e.g., luciferase reactivation assay) and reduce apoptosis, an effect blocked by chaperone inhibitors (e.g., VER‑155008).
3. EV‑release inhibition phenocopies senolysis: Treating aged mice with the neutral sphingomyelinase inhibitor GW4869 (to block EV secretion) will reproduce senolytic side‑effects such as diminished wound‑healing speed and impaired glucose tolerance, without reducing p16+ cell numbers.
4. Rescue by exogenous EVs: Administering purified senescent‑cell EVs to mice undergoing broad senolytic therapy (e.g., dasatinib + quercetin) will ameliorate off‑target phenotypes (immune memory loss, beta‑cell depletion) while retaining the beneficial clearance of pathogenic senescent cells.
5. Spatial correlation: In tissue sections, regions with high senescent‑cell density will display elevated EV‑associated chaperone signal (e.g., CD63‑HSP70 colocalization) that diminishes after senolytic treatment.

Experimental Approach

• Isolation and characterization: Sort p16+ cells from young and old mouse tissues using flow cytometry, isolate EVs by ultracentrifugation or size‑exclusion chromatography, quantify chaperone content via targeted proteomics.
• In vitro functional assays: Treat cultured myotubes or hepatocytes with tunicamycin or heat shock, then add senescent‑cell EVs; measure chaperone activity, protein aggregation (filter‑trap assay), and cell viability.
• In vivo validation: Use progeroid or naturally aged mice; administer GW4869 or senolytics; assess wound closure, glucose tolerance, and immune‑cell function; perform EV‑chaperone imaging via immunofluorescence or proximity ligation assay.
• Rescue experiments: Co‑inject senescent‑cell EVs with senolytics; monitor whether off‑target toxicity is alleviated while senescent‑cell burden (p16+ staining) remains reduced.
• Controls: Use EVs from non‑senescent cells, EV‑depleted supernatants, and EVs pre‑treated with protease to confirm cargo‑dependence.

Potential Implications

If validated, this hypothesis reframes senescent cells not merely as sources of harmful SASP but as dynamic contributors to tissue proteostasis. It suggests that therapeutic strategies should aim to modulate, rather than abolish, senescent‑cell‑derived EV chaperone transfer—perhaps by fine‑tuning senolytic dosing, timing, or by supplementing with chaperone‑rich EVs to preserve stress resilience during senescence‑targeted interventions.