Epigenetic Noise at HOX Loci Drives Positional Memory Loss and Differentiation Defects in Aging MSCs
Hypothesis
Aging MSCs accumulate stochastic methylation changes at HOX gene loci that act as epigenetic noise, eroding the HOX code and collapsing positional memory. This noise disrupts topologically associating domain (TAD) boundaries surrounding HOX clusters, leading to ectopic enhancer contacts and transcriptional dysregulation that uncouples lineage‑specific differentiation signals from positional identity.
Mechanistic model: In young MSCs, precise methylation patterns maintain insulated HOX TADs, allowing collinear expression that maps anteroposterior position to mesenchymal potency. With age, random gain/loss of methylation at CpG islands within HOX promoters and intergenic regions increases variance (epigenetic drift). This alters CTCF/cohesin binding, weakening TAD insulation. Consequently, enhancers intended for one HOX gene aberrantly activate neighboring loci, producing mixed HOX signatures and a loss of coherent positional cues. The resulting transcriptional noise drives MSCs toward a senescent, differentiation‑refractory state.
Testable predictions:
- Single‑cell multi‑omics (scATAC‑seq + scRNA‑seq) of MSCs across passages will show increased methylation variance at HOX loci correlating with loss of HOX collinearity and elevated chromatin accessibility at ectopic enhancer‑promoter loops.
- Targeted restoration of methylation patterns at key HOX loci (e.g., HOXA9, HOXC8) using CRISPR‑dCas9‑DNMT3A or demethylation with dCas9‑TET1 will rescue TAD boundary integrity (measured by Hi‑C) and restore adipogenic/osteogenic/chondrogenic differentiation in aged MSCs.
- Conversely, inducing stochastic methylation noise at HOX loci in young MSCs (via low‑dose DNMT1 inhibitor or CRISPR‑based methyltransferase recruitment) will prematurely erode positional identity and impair differentiation, even without replicative senescence.
Falsifiability: If rescuing HOX methylation does not improve differentiation, or if inducing methylation noise fails to phenocopy aging‑associated loss of lineage competence, the hypothesis that HOX code erosion via epigenetic noise drives MSCs senescence would be refuted.
Broader impact: Linking epigenetic noise at developmental regulators to functional decline provides a mechanistic bridge between epigenetic clocks and tissue‑specific aging, suggesting that HOX‑focused epigenome editing could revitalize aged mesenchymal stromal cells for regenerative therapies.
References