Back

A DNA deliverer-receiver mechanism for DNA recruitment in phase-separated transcriptional condensates

Blazquez, S.; Yamauchi, M.; Terakawa, T.

2026-03-11 biophysics
10.64898/2026.03.09.710715 bioRxiv
Show abstract

DNA transcription is a complex process involving numerous components that can assemble into phase-separated transcriptional condensates. However, whether condensates formed by multiple transcription factors behave through simple additive effects or instead exhibit non-additive, emergent properties remains unclear. Here, we use large-scale molecular dynamics simulations to investigate how three core transcription factors regulating pluripotency and early embryonic development--Nanog, Oct4, and Sox2--organize biomolecular condensates in the absence and presence of DNA. We find that condensate formation is primarily driven by intrinsically-disordered-region-mediated interactions of Nanog and Sox2, each of which individually promotes Oct4 phase separation; by contrast, when Nanog and Sox2 coexist in the absence of DNA, Oct4 is less efficiently incorporated into condensates. In the presence of DNA, condensates display a distinct spatial organization: Nanog and Sox2 form dense, well-mixed clusters, whereas Oct4 remains more dispersed in interstitial regions where DNA preferentially localizes, resulting in [~]20% higher DNA content in Oct4-containing condensates. Notably, phase separation reshapes DNA-protein interaction landscapes, altering the intramolecular regions that engage DNA. Together, these results support a synergistic DNA deliverer-receiver mechanism and suggest that non-additive, multi-component condensate organization constitutes an additional layer of gene expression regulation beyond canonical transcription factor-DNA binding.

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.