Hypothesis

Histone demethylases KDM5 and KDM6 regulate the transcriptional set‑point of Hsp70 and related chaperones, thereby determining whether age‑associated protein aggregates are diverted to the reversible JUNQ compartment or accumulate irreversibly in IPOD.

Mechanistic Model

1. Chromatin state at chaperone promoters – KDM5 removes repressive H3K4me2/me1 marks, while KDM6 removes H3K27me3; their combined activity creates a permissive H3K4me3/H3K27me3 low environment at Hsp70, Hsp110, and Hsp104 promoters.
2. Coupling to proteostasis flux – Elevated chaperone load promotes recognition of soluble, ubiquitinated substrates and their sequestration into JUNQ, where they remain refolding‑competent. When chaperone expression falls, the same substrates bypass JUNQ and nucleate into IPOD.
3. Feedback loop – Accumulated IPOD aggregates sequester Hsp104, further reducing chaperone availability and reinforcing a low‑KDM5/KDM6, high‑repressive chromatin state.

Testable Predictions

• Prediction 1: In aged yeast or mammalian cells, loss of KDM5 or KDM6 activity will decrease Hsp70 mRNA and protein levels, increase the IPOD/JUNQ ratio, and elevate insoluble ubiquitin‑positive aggregates.
• Prediction 2: Pharmacological inhibition of KDM5 demethylase (e.g., with CPI-455) will phenocopy the genetic loss, whereas overexpression of a demethylase‑dead KDM5 mutant that retains its PHD3 chromatin‑binding domain will still support Hsp70 transcription via scaffolding.
• Prediction 3: Restoring Hsp70 expression in a KDM5/KDM6‑deficient background will rescue JUNQ formation and reduce IPOD burden without altering demethylase activity.
• Prediction 4: Chromatin immunoprecipitation (ChIP) will show increased H3K27me3 and decreased H3K4me3 at Hsp70 promoters in cells with impaired KDM5/KDM6, correlating with aggregate phenotype.

Experimental Approach

1. Genetic manipulation – Create CRISPR‑knockout or inducible knockdown lines for KDM5A/B and KDM6A/B in human fibroblasts and in Saccharomyces cerevisiae. Include rescue constructs: wild‑type demethylase, catalytically dead mutant, and PHD3‑only fragment.
2. Readouts – Quantify Hsp70/Hsp110/Hsp104 mRNA (RT‑qPCR) and protein (Western blot). Assess aggregate distribution using fluorescently tagged ubiquitinated proteins (Ub‑GFP) and microscopy to score JUNQ (peripheral, Hsp70‑positive) versus IPOD (central, Hsp104‑positive) compartments.
3. Pharmacologic modulation – Treat cells with KDM5 inhibitor CPI-455 or KDM6 inhibitor GSK-J4 and measure the same readouts.
4. Chromatin analysis – Perform ChIP‑seq for H3K4me3, H3K27me3, KDM5, and KDM6 at chaperone loci.
5. Functional assay – Measure cell proliferation, senescence markers (β‑galactosidase, p16), and proteotoxic stress resistance (e.g., canavanine tolerance in yeast).

Potential Outcomes

• Supportive outcome: KDM5/KDM6 loss reduces chaperone expression, shifts aggregates from JUNQ to IPOD, and increases senescence; rescue of Hsp70 reverses the shift.
• Refutatory outcome: Altering KDM5/KDM6 levels changes global histone marks but does not affect chaperone transcription or aggregate compartmentalization, indicating that epigenetic control of proteostasis operates via alternative regulators.

This framework directly links the "last attempt at order" hypothesis to an epigenetically tunable chaperone gateway, offering a clear, falsifiable path to determine whether modulating KDM5/KDM6 can convert pathological aggregation into a protective, reversible state.