Image from freepik

Cellular reprogramming, critical for regenerative medicine and cloning, is hindered by epigenetic barriers that prevent somatic cells from regaining totipotency- the ability of a zygote and the first few cleavage-state blastomeres to give rise to any cell type. While histone modifications like H3K9me3 are known to impede reprogramming, a comprehensive “epigenetic barcode” for predicting outcomes remains elusive.

In a recent preprint article published by Ana Janeva et al., researchers from the Institute of Epigenetics and Stem Cells (Helmholtz Munich, Germany) and the Wellcome Trust/Cancer Research UK Gurdon Institute (University of Cambridge, UK) tackled this challenge through “Digital Reprogramming” by leveraging an extensive epigenetic reference dataset and artificial intelligence (AI) models to define predictors of embryonic outcomes following nuclear transfer (NT).

Using Xenopus laevis as an NT model system in concert with RNA-sequencing, the team defined a series of thousands of genes that were abnormally up- or down-regulated in specific tissue compartments in NT embryos but not in normally developing control embryos prepared through in vitro fertilisation.

Reasoning that these genes reflect a form of epigenetic memory that constrains the totipotent redifferentiation of somatic cell nuclei, the authors systematically characterised associated chromatin features via the use of convolutional neural network AI models, enabling the prediction of NT outcomes and defining a series of epigenetic markers of somatic memory gene expression. The representative target marker H3K27ac was then manipulated by pharmacologically inhibiting p300/CBP in embryos. An IonOpticks Aurora Ultimate™ 25×75 C18 UHPLC column was interfaced with a Q Exactive HF equipped with Ultimate 3000 RSLCnano pump to separate reconstituted in-gel tryptic digests during systematic characterisation of histone modifications in these embryos, revealing a 1.70-fold drop in H3K27ac levels in treated embryos relative to controls.

Building on these results, a series of memory-related genes influenced by H3K27ac modification in NT-derived embryos were validated, demonstrating that p300/CBP inhibition can partially normalise memory gene expression and improve developmental outcomes in cloned embryos.

These findings reinforce prior studies of induced pluripotent stem cells, extending current knowledge of p300/CBP and H3K27ac as key targets in cell fate reprogramming. Crucially, while more work will be needed to apply these insights in vivo, the predictive Digital Reprogramming pipeline offers an unprecedented tool that can serve as a foundation for further innovation and hypothesis generation, expediting the translation of epigenetic insights into actionable interventions in regenerative medicine.

Publication
bioRxiv

Authors

Ana Janeva, Christopher A. Penfold, Sara Llorente-Armijo, Huiwen Li, Tomas Zikmund, Marco Stock, Jerome Jullien, Tobias Straub, Ignasi Forne, Axel Imhof, Juan M. Vaquerizas, John B. Gurdon, & Eva Hörmanseder;

Title

‘Digital Reprogramming’ Decodes Epigenetic Barriers of Cell Fate Changes