Back

Self-organization of Drosophila chromatin architecture in a cell-free system

Jayakrishnan, M.; Kars, G.; Campos-Sparr, A.; Karpinska, M. A.; Zunjarrao, S.; Maziak, N.; Margulies, C. E.; Vaquerizas, J. M.; Gambetta, M. C.; Oudelaar, M.; Becker, P. B. B.

2026-03-20 genomics
10.64898/2026.03.17.712299 bioRxiv
Show abstract

Metazoan genomes are organized by folding of the nucleosome fiber into loops and domains that support long-range regulatory interactions. Although cohesin-mediated loop extrusion and architectural DNA-binding proteins are central to current models of genome organization, how these mechanisms integrate to generate higher-order structure remains incompletely understood. Early Drosophila melanogaster embryogenesis provides a unique window into the emergence of chromatin architecture, as rapid syncytial nuclear divisions occur largely in the absence of transcription. However, probing the mechanisms underlying this primordial folding in vivo is technically challenging. Here, we establish an in vitro system that reconstitutes complex chromatin using extracts from syncytial embryos. Nucleosome mapping and Micro-C analyses reveal that long-range interactions, including loops and topologically associating domains (TADs), emerge spontaneously from soluble extract components. While some structures resemble those observed in early embryos, others represent latent interaction potentials that are constrained in vivo. Focusing on the eve locus, we find that TAD formation is incompatible with a simple loop extrusion model and instead requires direct pairing of boundary elements mediated by the insulator protein Suppressor-of-hairy-wing Su(Hw). Together, our work demonstrates that key features of 3D genome organization can be reconstituted in a cell-free system and provides a tractable platform for mechanistic dissection of chromatin folding in Drosophila.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.