How the embryonal epigenome organizes itself

Enlarge

The development of an embryo is one of the most fundamental processes in biology. Early in this process, it is determined which cells will give rise to which tissues – controlled by epigenetic marks such as DNA methylation. Researchers at LMU have now shown that, rather surprisingly, this highly complex process is governed by physical laws.

An interdisciplinary team led by Professor Steffen Rulands has demonstrated for the first time that the patterns of DNA methylation in the early embryo can be described by universal dynamic rules. Rulands is a researcher at LMU’s Faculty of Physics and a member of the ORIGINS and BioSysteM Clusters of Excellence. His results combine modern genome research with concepts from statistical physics. “Our work shows that physical principles play a key role in the organization of the embryonal genome,” says Rulands. “This opens up whole new opportunities for understanding complex biological processes with the methods of physics.”

Simple rules instead of complex networks

Regarding the background: Cellular identity emerges from the finely tuned interplay of numerous biochemical processes. Chemical modifications to DNA and histones determine which genes are active and which are switched off. This interplay has long been regarded as extremely complex and hard to distill into general principles.

The new study reveals that universal physical principles underlie these complex biological processes. At the heart of the mechanism is a dynamic feedback loop: The enzymes that add DNA methyl groups simultaneously change the spatial structure of chromatin – and this altered structure, in turn, dictates where further methylation occurs. This gives rise to nanoscale structures, which form by means of phase separation – that is to say, through a physical process in which different molecular states within the cell nucleus segregate from one another to form stable domains.

Combination of state-of-the-art methods

To decipher these processes, the researchers combined single-cell multi-omics, high-resolution microscopy, and theoretical models from non-equilibrium physics. The data were provided by comprehensive measurements of DNA methylation in cell cultures and mouse embryos.

They discovered that DNA methylation in the early embryo follows self-similar scaling behavior: The patterns repeat across multiple orders of magnitude and can be described by a small number of fundamental principles. At the same time, these processes operate with surprising independence from the local genomic context, with the same rules applying in both active and inactive domains. Crucially, physical effects – such as chromatin compaction and the interactions among the enzymes involved – largely govern the dynamics. A particularly revealing discovery was that epigenetic changes at certain genes can be detected one to two days before their actual silencing – an early indication that these processes actively prepare the subsequent gene activity.

The results furnish new mechanistic explanations for a key step in embryonal development: the transition from initially identical cells to different cell types. This so-called symmetry breaking is the foundation for the formation of complex tissues and organs.

“What’s particularly exciting is that we can infer spatial and temporal processes in the cell nucleus directly from linear DNA sequence data. This allows us to observe and theoretically describe the self-organization of the genome”, says Rulands.

Significance for medicine and biology

In the long term, these findings could help scientists gain a deeper understanding of developmental processes – in the differentiation of stem cells, for instance, or in pathogenesis associated with epigenetic alterations. In particular, the early predictability of gene silencing opens promising new avenues for regenerative medicine and cancer research.

Moreover, the study highlights how closely biology and physics are intertwined: Even highly complex processes of life often follow universal principles that can be described using the tools of physics.

Olmeda F., Lohoff T., Kafetzopoulos I. et al.: Scaling and self-similarity in the formation of the embryonic epigenome. Nature Physics 2026