It is essential for cells to control which genes are transcribed into RNA. In eukaryotes, two major control points are recruitment of RNA polymerase II (Pol II) into a paused state, and subsequent pause release toward transcription. Pol II recruitment and pause release occur in association with macromolecular clusters, which were proposed to be formed by a liquid–liquid phase separation mechanism. How such a phase separation mechanism relates to the interaction of Pol II with DNA during recruitment and transcription, however, remains poorly understood. Here, we use live and super‐resolution microscopy in zebrafish embryos to reveal Pol II clusters with a large variety of shapes, which can be explained by a theoretical model in which regulatory chromatin regions provide surfaces for liquid‐phase condensation at concentrations that are too low for canonical liquid–liquid phase separation. Model simulations and chemical perturbation experiments indicate that recruited Pol II contributes to the formation of these surface‐associated condensates, whereas elongating Pol II is excluded from these condensates and thereby drives their unfolding.

Fluorescence microscopy, a central tool of biological research, is subject to inherent trade-offs in experiment design. For instance, image acquisition speed can only be increased in exchange for a lowered signal quality, or for an increased rate of photo-damage to the specimen. Computational denoising can recover some loss of signal, extending the trade-off margin for high-speed imaging. Recently proposed denoising on the basis of neural networks shows exceptional performance but raises concerns of errors typical of neural networks. Here, we present a work-flow that supports an empirically optimized reduction of exposure times, as well as per-image quality control to exclude images with reconstruction errors. We implement this work-flow on the basis of the denoising tool Noise2Void and assess the molecular state and three-dimensional shape of RNA Polymerase II (Pol II) clusters in live zebrafish embryos. Image acquisition speed could be tripled, achieving 2-second time resolution and 350-nanometer lateral image resolution. The obtained data reveal stereotyped events of approximately 10 seconds duration: initially, the molecular mark for initiated Pol II increases, then the mark for active Pol II increases, and finally Pol II clusters take on a stretched and unfolded shape. An independent analysis based on fixed sample images reproduces this sequence of events, and suggests that they are related to the transient association of genes with Pol II clusters. Our work-flow consists of procedures that can be implemented on commercial fluorescence microscopes without any hardware or software modification, and should therefore be transferable to many other applications.

Fluorescence microscopy, a central tool of biological research, is subject to inherent trade-offs in experiment design. For instance, image acquisition speed can only be increased in exchange for a lowered signal quality, or for an increased rate of photo-damage to the specimen. Computational denoising can recover some loss of signal, extending the trade-off margin for high-speed imaging. Recently proposed denoising on the basis of neural networks shows exceptional performance but raises concerns of errors typical of neural networks. Here, we present a work flow that supports an empirically optimized reduction of exposure times, as well as per-image quality control to exclude images with reconstruction errors. We implement this work flow on the basis of the denoising tool Noise2Void and assess the molecular state and three-dimensional shape of RNA Polymerase II (Pol II) clusters in live zebrafish embryos. Image acquisition speed could be tripled, achieving 2-second time resolution and 350-nanometer lateral image resolution. The obtained data reveal stereotyped events of approximately 10 seconds duration: initially, the molecular mark for initiated Pol II increases, then the mark for active Pol II increases, and finally Pol II clusters take on a stretched and unfolded shape. An independent analysis based on fixed sample images reproduces this sequence of events, and suggests that they are related to the transient association of genes with Pol II clusters. Our work flow consists of procedures that can be implemented on commercial fluorescence microscopes without any hardware or software modification, and should therefore be transferable to many other applications.

It is essential for cells to control which genes are transcribed into RNA. In eukaryotes, two major control points are recruitment of RNA polymerase II (Pol II) into a paused state and subsequent pause release to begin transcript elongation. Pol II associates with macromolecular clusters during recruitment, but it remains unclear how Pol II recruitment and pause release might affect these clusters. Here, we show that clusters exhibit morphologies that are in line with wetting of chromatin by a liquid phase enriched in recruited Pol II. Applying instantaneous structured illumination microscopy and stimulated emission double depletion microscopy to pluripotent zebrafish embryos, we find recruited Pol II associated with large clusters, and elongating Pol II with dispersed clusters. A lattice kinetic Monte Carlo model representing recruited Pol II as a liquid phase reproduced the observed cluster morphologies. In this model, chromatin is a copolymer chain containing regions that attract or repel recruited Pol II, supporting droplet formation by wetting or droplet dispersal, respectively.