Agingreverses cancer’s bivalent promoter logic: KDM5/6 activity loss shifts HiBiv to LoBiv states, eroding stem cell plasticity
Hypothesis
In aging, the balance of H3K4me3 and H3K27me3 at bivalent promoters flips relative to cancer, leading to a loss of stem‑like plasticity.
Mechanistic premise
- Cancer maintains a subclass of bivalent promoters (CSBPs) with low H3K27me3/H3K4me3 ratio (LoBiv) that preserves expression of stemness genes such as SOX9 [1, 2].
- This state depends on high H3K4me3 driven by MLL2 recruitment at CpG‑rich promoters and restrained KDM5 activity [4].
- In contrast, aged stem cells show global gain of H3K27me3 and loss of H3K4me3 at loci governing quiescence exit [5].
- We hypothesize that aging drives the opposite epigenetic shift: increased KDM5 activity and/or decreased KDM6 activity converts LoBiv/CSBP‑like promoters into HiBiv configurations (high H3K27me3/H3K4me3), thereby locking cells into a differentiated, non‑plastic state.
Testable predictions
- In aged murine muscle or intestinal stem cells, ChIP‑seq will reveal a significant increase in the H3K27me3/H3K4me3 ratio at promoters that are LoBiv in young cancer‑derived CSBPs (e.g., Sox9, Nanog, Oct4) [3].
- Genetic knockdown of Kdm5b in aged stem cells will reduce H3K27me3 accumulation at these promoters, restoring a LoBiv‑like chromatin signature and improving regenerative capacity after injury.
- Pharmacological inhibition of Kdm6 (e.g., GSK‑J4) in young stem cells will phenocopy the aged state, increasing HiBiv promoters and decreasing clonogenic potential, whereas KDM5 inhibition in aged cells will rescue plasticity.
- The CpG‑MLL2 protective axis will be weakened in aging: bisulfite sequencing will show increased DNA methylation at CpG islands of formerly LoBiv promoters, correlating with reduced MLL2 binding (ChIP‑qPCR) and elevated KDM5 occupancy.
Experimental design
- Isolate young (2‑month) and aged (24‑month) murine muscle satellite cells and intestinal crypt stem cells.
- Perform CUT&RUN for H3K4me3, H3K27me3, MLL2, KDM5B, and KDM6A/B; integrate with RNA‑seq to link chromatin states to transcription.
- Use CRISPRi to attenuate Kdm5b or Kdm6a in aged cells; assess colony‑forming units, EdU incorporation, and in vivo transplantation efficacy.
- Validate causality by rescuing the aged phenotype with a demethylase‑dead KDM5B mutant versus wild‑type overexpression.
If the data show that aging elevates the H3K27me3/H3K4me3 ratio at formerly LoBiv promoters and that reversing KDM5/6 activity restores a cancer‑like LoBiv state and stemness, the hypothesis will be supported. Conversely, absence of these shifts would falsify the model, indicating that aging employs distinct epigenetic mechanisms unrelated to the cancer CSBP paradigm.