Surface condensation of a pioneer transcription factor on DNA

Abstract

In the last decade, extensive studies on the properties of non-membrane-bound compartments in the cellular cytoplasm have shown that concepts in phase separation drawn from physical chemistry can describe their formation and behaviour1–4. Current evidence also suggests that phase separation plays a role in the organization inside the cell nucleus5–8. However, the influence and role of DNA on the physical chemistry of phase separation is not well understood. Here, we are interested in the role of interactions between phase separating proteins and the DNA surface. The interaction of liquid phases with surfaces has been extensively studied in soft matter physics, in the context of macroscopic surfaces and non-biological liquids9–11. The conditions in the nucleus are different from those studied in conventional soft matter physics because DNA with a diameter of about 2 nm12 provides a microscopic surface, and liquid-like phases are complex mixtures of proteins subject to a myriad of biochemical modifications13. Transcriptional condensates, which are thought to serve as regulatory hubs in gene expression14–21, provide an accessible system to investigate the physics of condensates that emerge from DNA-protein and protein-protein interactions. These condensates are typically small22, and the mechanisms that determine their size are unknown. Whether they can be understood as phase separated compartments has been subject to debate23–26. Here, we use optical tweezers to directly observe the condensation of the pioneer transcription factor Klf427,28 on DNA in vitro. We demonstrate that Klf4 forms microphases that are enabled by interaction with the DNA surface. This sets their typical size and allows them to form below the saturation concentration for liquid-liquid phase separation. We combine experiment with theory to show that these microphases can be understood as forming by surface condensation on DNA via a switch-like transition similar to prewetting. Polymer surface mediated condensation reconciles several observations that were previously thought to be at odds with the idea of phase separation as an organizing principle in the nucleus.

Ligand Effects on Phase Separation of Multivalent Macromolecules

“…Morin et al, used the Brunauer-Emmett-Teller theory (45) for multilayer adsorption to explain their results. Taken together, it follows that preferential binding of multivalent ligands to scaffolds in dense phases, this work and that of Ghosh et al, (26) and Espinosa et al, (27), and adsorption of scaffolds to surfaces (46, 47) as shown by Morin et al, (44) can lead to similar effects in terms of lowering the saturation concentrations of scaffold macromolecules. Therefore, we propose that the stabilization of condensate formation via surface interactions derives from preferential adsorption of dense phases through spacer-surface interactions and / or enhancing of sticker-sticker crosslinks.…”

RNA polymerase II clusters form in line with liquid phase wetting of chromatin

“…The causal relevance of Pol II phosphorylation is supported by the application of chemical inhibitors of Pol II phosphorylation, which induces changes in cluster morphology and cluster number that are in line with predictions from our lattice simulations. In combination with previous work on Pol II liquid phase behavior 13,14 and studies showing condensation of transcription factors by wetting of DNA in vitro 23,24 , our findings in zebrafish, an embryonic model system, suggest that similar liquid phase wetting of chromatin might also occur in vivo.…”

“…11). This behavior is typical of liquid phase condensation on a surface, also called wetting 24 . To confirm that the simulations exhibit surface-mediated condensation, we demonstrate another key behavior: the size of clusters is controlled by the amount of available surface (Fig.…”

“…Formation of macromolecular clusters at genomic target regions has recently been described using a model of liquid phase condensation on polymers as microscopic surfaces 13,14,23,24 . To test whether such a model can reproduce the different cluster morphologies seen in our experiments, we implemented corresponding lattice kinetic Monte Carlo (LKMC) simulations 31,32 (for details see Supplementary Material).…”

Surface Densities Prewet a Near-Critical Membrane

“…While our focus is on membranes, prewetting can occur on other interesting biological surfaces. Phase separation has been proposed to play prominent roles in transcriptional regulation (47, 48), where DNA has been proposed to act as a one-dimensional ‘surface’ for prewetting (49).…”