Single-molecule localization-integrated trajectory analysis for the dynamic interaction of chromatin structure and transcription machinery.

Abstract

The localization and dynamics of individual proteins in a living cell are essential information for investigating protein functions. Live-cell single-molecule imaging has provided direct information on dynamics involved in biological processes. We have identified the characteristic movement of transcription-related factors in elongation sites and directly measured the viscoelastic properties of phase-separated nucleolus proteins using single-molecule trajectory analysis in living cells. However, the functional dynamics are difficult to identify because of the highly heterogeneous distribution and mobility of individual molecules. Here, we developed localization-integrated single-molecule trajectory analysis with machine-learning-based trajectory classification for simultaneous single-molecule imaging. We applied this method to the dynamics of the largest subunit of RNA polymerase II (Pol II) in a crowded chromatin structure visualized by histone H3.1 labeled with spontaneously blinking protein in a living cell nucleus. We identified distinct Pol II mobility features by calculating various mobility indicators, including diffusion coefficient, spring coefficient, and confinement radius, then applied machine-learning methods, including dimensionality reduction and clustering. We also developed a series of analysis methods to quantify the spatial characteristics of the point distribution adjacent to the trajectory. These methods reveal that the spatial probability of surrounding histone distribution depends on the mobility state of Pol II. The simultaneous imaging technique and the new analysis method developed in this study directly link the physical properties of the transcription machinery with the biological function of chromatin.