Autophagy Doesn't Simply Decline With Age — It Reshapes Differently by Sex and Cell Type
New human data shows autophagy flux is sex- and cell-specific with age, and high flux in adults over 70 links to worse physical function.
Summary
Autophagy, the cell's recycling system, is widely believed to weaken with age. Researchers measured actual autophagy activity, not just gene expression, in skin cells, lab-grown neurons, and fresh blood cells from healthy adults aged 23 to 80. Gene activity for autophagy rose with age, but real autophagy activity did not follow. In skin cells, activity fell with age in men but held steady in women. In neurons, it rose with age in women only. In adults over 70, higher autophagy activity went with slower walking, lower fitness, and weaker physical performance. In a small pilot, 12 weeks of mild exercise lowered blood-cell autophagy activity while physical function improved. High autophagy in older adults may be a stress response, not a sign of healthier cells.
Detailed Summary
Disabled autophagy is a recognized hallmark of aging, and animal studies show that boosting it extends lifespan. Yet direct evidence in living humans is thin. Autophagy flux, the rate at which cells actually degrade material through lysosomes, cannot be measured in vivo. Common static readouts such as autophagosome counts, LC3 protein levels, and autophagy-gene expression cannot distinguish more autophagosome formation from blocked degradation. This leaves open whether autophagy truly declines with age in people, and whether it tracks health.
The team used the San Diego Nathan Shock Center healthy aging cohort, whose members were free of chronic disease and ranged in age from 23 to 80. They compared autophagy across subject-matched human cell types. These were low-passage dermal fibroblasts (N = 44 for flux assays), induced neurons (iNs) converted directly from those fibroblasts (N = 22), and freshly isolated peripheral blood mononuclear cells (PBMCs; N = 23, mostly lymphocytes). Flux was measured by blocking lysosomal degradation with chloroquine and quantifying the resulting LC3B accumulation. Bulk RNA-seq covered fibroblasts (young 23-33 years, N = 4; older 64-72, N = 5) and iNs.
Transcription told a misleading story. Fibroblasts from older donors showed 94 upregulated and 188 downregulated genes. Macroautophagy, autophagosome-component, and autophagosome-assembly gene modules were upregulated, though some genes (ATG5, ATG7, ATG4C) fell. iNs showed weaker, more variable changes. Static LC3B foci did not correlate with flux in any cell type, and gene expression did not predict functional activity, so direct flux measurement was essential.
Flux itself varied by sex and cell type. In fibroblasts, flux declined with age in males (p = 0.021) and stayed stable in females (age-by-sex interaction p = 0.075). In iNs, flux rose with age in females (R² = 0.61, p = 0.02), with a significant age-by-sex interaction (p = 0.001). Fibroblast and iN flux were uncorrelated within the same individuals (males r = -0.41; females r = -0.46). PBMC flux was not associated with age (p = 0.44) but grew more heterogeneous, trending higher after about age 70, with no sex effect.
Flux also related to function, and the direction depended on age. Across all participants, higher PBMC flux correlated with lower resting heart rate (r = -0.61, p = 0.002), while higher iN flux correlated with weaker grip strength (r = -0.51) and slower walking (r = -0.52). In those over 70, higher PBMC flux was associated with lower VO2max (r = -0.75, p = 0.026), slower walking speed (r = -0.78, p = 0.01), and longer chair-stand time (r = 0.73, p = 0.02). In a pilot study, 12 weeks of mild exercise lowered PBMC flux while physical function improved.
The authors conclude that autophagy does not decline uniformly with age. Elevated flux in late life may be a compensatory response to accumulating stress, not enhanced capacity. Caveats include small subgroups, cross-sectional design, cell-culture models, and a small pilot exercise study without a control group described in the available text.
Key Findings
- Autophagy-gene expression rose with age in fibroblasts (94 genes up, 188 down in older vs. young donors), but expression did not predict actual autophagy flux
- Fibroblast flux fell with age in males (p = 0.021) but stayed stable in females; the age-by-sex interaction trended toward significance (p = 0.075)
- iN flux rose with age in females (R² = 0.61, p = 0.02) but not in males, with a significant age-by-sex interaction (p = 0.001)
- Flux in fibroblasts and iNs from the same donors was uncorrelated (males r = -0.41, p = 0.25; females r = -0.46, p = 0.30), so autophagy is not coordinated across cell types
- PBMC flux was not associated with age (p = 0.44) but became more heterogeneous, trending higher after about age 70, with no sex differences
- In adults over 70, higher PBMC flux correlated with lower VO2max (r = -0.75, p = 0.026), slower walking speed (r = -0.78, p = 0.01), and longer chair-stand time (r = 0.73, p = 0.02)
- In a pilot study, 12 weeks of mild exercise lowered PBMC autophagy flux while physical function improved in older adults
Methodology
The study drew on the San Diego Nathan Shock Center healthy aging cohort (ages 23-80, self-reported free of chronic disease). It ran bulk RNA-seq with GSEA/GO analysis on fibroblasts and iNs and chloroquine-based flux assays in fibroblasts (N = 44), iNs (N = 22), and fresh PBMCs (N = 23, assessed by LC3B-II Western blot). Flux was related to treadmill VO2max, grip strength, walking speed, chair-stand time, SPPB, and actigraphy using Spearman correlations and multiple linear regression with age-by-sex interactions. A pilot intervention tested 12 weeks of mild exercise on PBMC flux in older adults.
Study Limitations
Subgroup sizes were small, especially in the 70+ group (N = 6-10 per cell type) and in sex-stratified analyses, and the design was cross-sectional and correlational. Fibroblasts and iNs are culture models, and the cohort was healthy and predominantly White, which limits generalizability. The exercise intervention was a pilot study, and conflict-of-interest disclosures were not included in the provided text.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: