Longevity & AgingResearch PaperPaywall

Boosting This Cellular Cleanup Pathway Clears Senescent Cells and Fights Aging

A decline in chaperone-mediated autophagy lets senescent cells accumulate with age — and restoring it in mice reduces this burden.

Wednesday, October 7, 2026 2 views
Published in Nat Aging
A fluorescence microscopy image showing enlarged, flattened senescent cells stained blue alongside smaller active macrophages in pink, on a dark laboratory slide background

Summary

Senescent cells pile up as we age, partly because the immune system stops removing them. This mouse and cell study shows why. Chaperone-mediated autophagy (CMA) is a cellular recycling pathway that declines with age. Without it, senescent cells release a more harmful mix of secreted signals that spreads senescence to neighbors. These signals also shut down CMA in macrophages, the immune cells that engulf senescent cells. Macrophages lacking CMA build up a 'do not eat me' receptor and clear senescent cells poorly. Mice with CMA-deficient macrophages carried more senescent cells and healed wounds more slowly. An oral CMA activator given to aged mice for five months kept senescent cell burden closer to young levels and eased lung fibrosis in a mouse model. The work is preclinical, but it names CMA as a promising target.

Detailed Summary

Senescent cells accumulate with age and drive disease, yet young bodies clear them efficiently. This study from Ana Maria Cuervo's group at Albert Einstein College of Medicine asks whether the age-related decline of chaperone-mediated autophagy (CMA) explains that failure. CMA selectively sends proteins carrying a KFERQ-like motif to lysosomes through the LAMP2A receptor. The question links two hallmarks of aging, disabled autophagy and cellular senescence.

In primary ear fibroblasts, cells from 4-month-old mice upregulated CMA seven days after palbociclib-induced senescence. Cells from 23-month-old mice had lower baseline CMA and failed to raise it (n=6 mice). Lamp2a knockdown in NIH3T3 fibroblasts produced cells with partial senescence-like features, including SA-β-gal, p21, γH2A.X and nuclear invaginations, but normal proliferation until senescence was induced. In principal-component analysis (PC1 57.7%) these cells sat between control and senescent cells. CMA loss alone therefore does not cause senescence, but it changes how cells respond to it.

Proteomics showed that blocking CMA in young cells reproduced nearly half of the age-related proteome changes. After palbociclib, CMA-knockout young cells and aged cells showed 45% concordance, with about 400 proteins changed versus young controls and only about 200 differing between the two groups. Lysosomal-inhibitor proteomics found 64 proteins no longer degraded in aged cells, over half of them CMA substrates. Metabolomics showed that both groups failed to engage fatty acid β-oxidation and shifted TCA cycle, nucleotide and amino acid pathways.

CMA-deficient cells also secreted a different mix of proteins. The ER-TurboID approach, which separates conventional from unconventional secretion, found 161 conventionally secreted proteins, with 71 increased and 47 decreased under CMA blockade. Of 340 unconventionally secreted proteins, 130 were reduced. Their medium induced paracrine senescence in healthy fibroblasts even before senescence was triggered. Bone marrow macrophages from aged mice of both sexes had less CMA, and the senescent secretome lowered macrophage CMA further. Macrophages lacking CMA (ML2A−/−) showed reduced phagocytosis and efferocytosis because SIRPα was no longer degraded. Higher SIRPα raised SHP-1 and strengthened the CD47 'do not eat me' signal. CMA-deficient senescent fibroblasts also killed these macrophages within 48 hours.

In vivo, 23-month-old male ML2A−/− mice had more SA-β-gal+ cells in gonadal fat, liver and lung than controls, with weaker differences in females. Nuclear HMGB1 fell in most tissues of both sexes. SHP-1, CD47 and liver TUNEL+ cells rose, and a PCA of 16 markers separated the groups. In 12-month-old mice, ML2A−/− animals had larger wounds at day 15 with more p21+ cells (ANOVA p=0.0117).

The CMA activator CA77.1 (30 mg/kg orally, 18-month-old mice, 5 months) prevented the age-related rise in SA-β-gal+ cells in fat, liver and lung in both sexes. It blunted p21, Tnf, Il1α and p16 upregulation and modestly reduced fibrosis, mainly in fat. Treated mice clustered nearer young animals. The compound also restored phagocytosis in macrophages from 23-month-old mice to the level of those from 6-month-old mice, and it lessened disease severity in a mouse pulmonary fibrosis model. All work is in mice and cultured cells, and the effects varied by tissue and sex.

Key Findings

  • Ear fibroblasts from 23-month-old mice had lower baseline CMA and failed to upregulate it after palbociclib-induced senescence, unlike 4-month-old cells (n=6 mice; p=0.0008 and 0.0041 for the comparisons shown)
  • Blocking CMA in young cells reproduced almost half of age-related proteome changes, with 45% concordance with aged cells after senescence induction (about 400 proteins changed vs young controls)
  • 64 proteins lost lysosomal degradation in aged fibroblasts, and over half were CMA substrates
  • The secretome of CMA-deficient cells induced paracrine senescence in healthy fibroblasts even before senescence induction (increased SA-β-gal+ cells and p21)
  • Macrophages lacking CMA accumulated SIRPα and SHP-1 (SHP-1 p=0.0012) and showed reduced phagocytosis (p=0.0307) and efferocytosis (p<0.0001)
  • 23-month-old male mice with CMA-deficient macrophages had more SA-β-gal+ cells in fat, liver and lung, and 12-month-old mice showed delayed wound closure (ANOVA p=0.0117) with more p21+ cells
  • Oral CA77.1 (30 mg/kg, 5 months, from age 18 months) prevented the age-related rise in SA-β-gal+ cells in both sexes and restored aged macrophage phagocytosis to 6-month-old levels

Methodology

The study combined primary mouse ear fibroblasts (4 vs 23 months), NIH3T3 Lamp2a-knockdown cells, bone marrow-derived macrophages, and LysM-Cre macrophage-specific LAMP2A knockout (ML2A−/−) mice, using a KFERQ-Dendra reporter mouse to measure CMA. Senescence was induced with palbociclib, etoposide, paraquat or BIBR1532. Readouts included quantitative proteomics with and without lysosomal inhibitors, metabolomics, ER-TurboID secretome profiling, co-culture clearance assays, flow cytometry, liver single-cell RNA-seq, and wound healing. Statistics were mainly two-way ANOVA, unpaired two-sided t-tests, and PCA, with n of 3 to 13 per experiment and 2 to 13 mice per group in vivo.

Study Limitations

All evidence comes from mice and cultured cells, so translation to humans is unproven. Effects varied by tissue and sex, with weaker SA-β-gal differences in females, and CMA activation only partly restored macrophage phagocytosis. The macrophage-specific knockout does not separate every immune mechanism, and conflict-of-interest details were not visible in the supplied portion of the full text.

Enjoyed this summary?

Get the latest longevity research delivered to your inbox every week.

Enter your email to subscribe: