Written in development and lost in ageing: the grammar of cellular identity.
Abstract
Ageing has been described through many theoretical frameworks, yet its mechanistic basis remains unresolved. Here we propose that ageing reflects progressive decanalization of mammalian cell identity, driven by erosion of a regulatory grammar written during development. Extending Waddington's epigenetic landscape, development canalizes cell fates through globally coordinated chromatin programs that establish identity constraints across multiple timescales. Subsequently, these constraints can only be maintained through local mechanisms with limited fidelity. The asymmetry between high-fidelity writing and imperfect maintenance generates predictable drift that accumulates with biological time. Antagonistic chromatin pathways, particularly polycomb repressive complex 2 (PRC2)-mediated repression opposed by H3K4/36 methylation and RNA polymerase II binding, enforce identity boundaries through mutual constraint. With age, this balance erodes. Consistent with this framework, the vast majority of age-associated DNA methylation gain in somatic mitotic tissues occurs at PRC2-bound low-methylated regions, indicating that these domains function as conserved coordinates of slow-layer drift. This model explains cross-tissue ageing signatures, developmental timing-to-lifespan correlations and robustness of pan-mammalian epigenetic ageing clocks. It provides testable predictions for interventions that stabilize architecture to preserve cellular identity and function.
The following analysis is generated by AI and may contain errors. It is separate from the authors’ abstract and is not peer review.
Research summary
The paper proposes that aging is driven by the progressive decanalization of cell identity due to the erosion of a developmental regulatory grammar, specifically predicting that age-associated DNA methylation gains occur at PRC2-bound regions. This is relevant because it offers a mechanistic framework for the root cause of aging (loss of epigenetic identity) rather than treating symptoms, though it is primarily a theoretical model with limited direct experimental validation in this abstract.
Author of this paper? Send a correction.