Hierarchical autophagy of PRC2 components governs epigenetic stability at the CDKN2A/B locus during aging

Andre Yamada· Mar 24, 2026

Mechanism: In young cells, selective autophagy targets and degrades phosphorylated PRC2 components like EZH2/SUZ12, maintaining H3K27me3 at the CDKN2A/B locus. Readout: Readout: Aging leads to impaired selective autophagy, increasing nuclear EZH2/SUZ12, reducing H3K27me3, and elevating p16 levels, driving senescence.

Hypothesis

Aging-related epigenetic drift at the CDKN2A/B locus stems not only from loss of repressive marks but also from a breakdown in the hierarchical selectivity of autophagy that normally removes PRC2 complex components (EZH2, SUZ12, EED) and DNA methyltransferases (DNMT1, DNMT3A/B). When this selective flux falters, epigenetic regulators accumulate, reinforcing a feed‑forward loop that locks the locus in an open, transcriptionally active state.

Mechanistic Rationale

Autophagy exhibits substrate preference via cargo receptors such as SQSTM1/p62 and NBR1, which can recognize ubiquitinated proteins and specific post‑translational modifications.
Emerging data show that EZH2 and SUZ12 become phosphorylated and ubiquitinated under oxidative stress, creating potential autophagic "eat‑me" signals.
In young cells, active autophagy preferentially degrades these phosphorylated PRC2 subunits, limiting their nuclear residence and preserving H3K27me3 at CDKN2A/B promoters.
With age, mitochondrial dysfunction elevates ROS, impairing autophagosome formation and shifting cargo receptor occupancy toward bulk cytosol, thereby sparing PRC2 components.
The resulting increase in nuclear EZH2/SUZ12 paradoxically fails to restore repression because concurrent loss of H3K27me3 writers and gain of DNA methylation shift the chromatin state toward a senescent configuration.
This mirrors the seed idea that autophagy functions as a triage ritual: when the hierarchy collapses, the cell mistakenly spares the very regulators that should be removed first, accelerating senescence.

Testable Predictions

In fibroblasts from young donors, immunoprecipitation of SQSTM1/p62 will pull down phosphorylated EZH2 and SUZ12 more efficiently than total EZH2/SUZ12.
Pharmacological activation of autophagy (e.g., with spermidine) will reduce nuclear EZH2/SUZ12 levels and increase H3K27me3 at the CDKN2A/B promoter, whereas autophagy inhibition (chloroquine) will have the opposite effect.
CRISPR‑mediated mutation of the ubiquitin‑acceptor lysines on EZH2 will block its autophagic degradation and lead to premature p16^INK4a upregulation despite intact autophagic flux.
Senescent cells rescued by ectopic expression of a non‑degradable EZH2 mutant will retain high p16^INK4a expression even when autophagy is chemically induced, confirming that loss of PRC2 turnover, not autophagy per se, drives the epigenetic shift.

Experimental Approach

Isolate primary human fibroblasts from donors aged <30 and >70 years.
Measure autophagic flux (LC3‑II turnover with bafilomycin A1) and correlate with nuclear EZH2/SUZ12 levels by subcellular fractionation and western blot.
Perform SQSTM1/p62 pull‑down followed by mass spectrometry to quantify phosphorylated PRC2 subunits.
Treat cells with spermidine (100 µM) or chloroquine (20 µM) for 24 h, then assess CDKN2A/B promoter methylation (bisulfite sequencing) and H3K27me3 (ChIP‑qPCR).
Use lentiviral vectors to express wild‑type or lysine‑mutant EZH2 (K→R) in young fibroblasts; monitor p16^INK4a expression and senescence‑associated β‑galactosidase activity.
Statistical analysis: two‑way ANOVA with age and treatment as factors; significance set at p<0.05.

If the hypothesis holds, enhancing the selective removal of PRC2 components should delay or prevent age‑associated epigenetic drift at the CDKN2A/B locus, uncoupling autophagy flux from senescence outcomes.

Hypothesis

Mechanistic Rationale

Autophagy exhibits substrate preference via cargo receptors such as SQSTM1/p62 and NBR1, which can recognize ubiquitinated proteins and specific post‑translational modifications.

Emerging data show that EZH2 and SUZ12 become phosphorylated and ubiquitinated under oxidative stress, creating potential autophagic "eat‑me" signals.

In young cells, active autophagy preferentially degrades these phosphorylated PRC2 subunits, limiting their nuclear residence and preserving H3K27me3 at CDKN2A/B promoters.

With age, mitochondrial dysfunction elevates ROS, impairing autophagosome formation and shifting cargo receptor occupancy toward bulk cytosol, thereby sparing PRC2 components.

The resulting increase in nuclear EZH2/SUZ12 paradoxically fails to restore repression because concurrent loss of H3K27me3 writers and gain of DNA methylation shift the chromatin state toward a senescent configuration.

This mirrors the seed idea that autophagy functions as a triage ritual: when the hierarchy collapses, the cell mistakenly spares the very regulators that should be removed first, accelerating senescence.

Testable Predictions

In fibroblasts from young donors, immunoprecipitation of SQSTM1/p62 will pull down phosphorylated EZH2 and SUZ12 more efficiently than total EZH2/SUZ12.

Pharmacological activation of autophagy (e.g., with spermidine) will reduce nuclear EZH2/SUZ12 levels and increase H3K27me3 at the CDKN2A/B promoter, whereas autophagy inhibition (chloroquine) will have the opposite effect.

CRISPR‑mediated mutation of the ubiquitin‑acceptor lysines on EZH2 will block its autophagic degradation and lead to premature p16^INK4a upregulation despite intact autophagic flux.

Senescent cells rescued by ectopic expression of a non‑degradable EZH2 mutant will retain high p16^INK4a expression even when autophagy is chemically induced, confirming that loss of PRC2 turnover, not autophagy per se, drives the epigenetic shift.

Experimental Approach

Isolate primary human fibroblasts from donors aged <30 and >70 years.

Measure autophagic flux (LC3‑II turnover with bafilomycin A1) and correlate with nuclear EZH2/SUZ12 levels by subcellular fractionation and western blot.

Perform SQSTM1/p62 pull‑down followed by mass spectrometry to quantify phosphorylated PRC2 subunits.

Treat cells with spermidine (100 µM) or chloroquine (20 µM) for 24 h, then assess CDKN2A/B promoter methylation (bisulfite sequencing) and H3K27me3 (ChIP‑qPCR).

Use lentiviral vectors to express wild‑type or lysine‑mutant EZH2 (K→R) in young fibroblasts; monitor p16^INK4a expression and senescence‑associated β‑galactosidase activity.

Statistical analysis: two‑way ANOVA with age and treatment as factors; significance set at p<0.05.