KDM5 Gates PRC2-Mediated H3K27me3 Redistribution at Glycolytic Loci During Aging

Hypothesis

KDM5 demethylases act as a chromatin‑based brake that prevents the age‑associated spread of H3K27me3 from peak to inter‑peak regions at glycolytic gene promoters. By binding to high‑H3K4me3, accessible chromatin through its JmjC catalytic domain and C5HC2 zinc finger, KDM5 maintains a locally permissive environment that limits PRC2 activity. When KDM5 function is compromised—either by loss of demethylase activity or disruption of its scaffolding interface—PRC2 gains access to these loci, leading to ectopic H3K27me3 deposition, transcriptional silencing of glycolytic enzymes, and the metabolic decline observed in aged Drosophila. This model directly links the divergent roles of KDM5 in cancer (where its overexpression promotes immune evasion) and aging (where its loss of function drives metabolic dysfunction) through a shared mechanism: regulation of the H3K4me3/H3K27me3 balance at bivalent domains.

Testable Predictions

1. Genetic perturbation – Knocking down KDM5 (via RNAi or CRISPR) in young adult flies will recapitulate the aged H3K27me3 drift profile, increasing inter‑peak signal by ~9% at glycolytic promoters without altering global H3K4me3 levels.

2. Pharmacological inhibition – Treatment with a selective KDM5 inhibitor (e.g., CPI-455) will produce the same epigenetic shift as genetic loss, and the effect will be rescued by concomitant PRC2 knock‑down, indicating epistatic relationship.

3. Chromatin context dependence – Mutating the C5HC2 zinc finger while preserving the JmjC domain will still cause H3K27me3 redistribution, confirming that KDM5’s scaffolding function, not its demethylase activity, is sufficient to block PRC2 spreading.

4. Metabolic read‑out – Flies exhibiting KDM5‑dependent H3K27me3 drift will show reduced glycolytic flux (measured by lactate production) and shortened lifespan; overexpression of glycolytic genes (e.g., Pfk, Gapdh) will suppress the longevity phenotype, mirroring PRC2 mutant rescue.

5. Cancer contrast – In Drosophila tumor models, forced KDM5 overexpression will decrease H3K27me3 at glycolytic loci, increase glucose uptake, and enhance resistance to ROS‑inducing chemotherapeutics, linking the aging mechanism to the oncogenic phenotype.

Experimental Approach

• Perform ChIP‑seq for H3K4me3 and H3K27me3 in young and aged wild‑type, KDM5 RNAi, and zinc‑finger mutant flies. Quantify peak‑to‑inter‑peak ratios at glycolytic promoters.

• Use ATAC‑seq to confirm that chromatin accessibility remains unchanged upon KDM5 loss, isolating the effect to histone modification balance.

• Measure glycolytic enzyme activity and metabolomics (LC‑MS) to connect epigenetic changes to metabolic output.

• Conduct lifespan assays under normal and high‑sugar diet conditions to test metabolic stress interaction.

• Parallel experiments in human cell lines (e.g., senescent fibroblasts) with KDM5 knockdown and EZH2 inhibition to assess conservation.

Falsification

If KDM5 loss does not increase inter‑peak H3K27me3 at glycolytic genes, or if the drift persists despite PRC2 depletion, the hypothesis is refuted. Likewise, if restoring glycolytic gene expression fails to extend lifespan in KDM5‑deficient flies, the proposed causal link between H3K27me3 drift and metabolic decline would be unsupported.

References

[1] https://elifesciences.org/articles/35368 [2] https://pmc.ncbi.nlm.nih.gov/articles/PMC12465163/ [3] https://pmc.ncbi.nlm.nih.gov/articles/PMC12676786/