Imaging dynamic and selective low-complexity domain interactions that control gene transcription

Abstract

INTRODUCTION. DNA binding transcription factors (TFs) are quintessential regulators of eukaryotic gene expression. Early studies of TFs revealed their well-structured DNA binding domains (DBDs) and identified functionally critical activation domains (ADs) required for transcription. It later became evident that many ADs contain intrinsically disordered low-complexity sequence domains (LCDs), but how LCDs activate transcription has remained unclear. Although it is known that transcriptional activation by LCDs requires selective interaction with binding partners, it has been challenging to directly measure selective LCD-LCD recognition in vivo and unravel its mechanism of action. RATIONALE. Traditional biochemical reconstitution and genetics studies have identified most of the molecular players central to transcription regulation. However, the mechanism by which weak, dynamic protein-protein interactions drive gene activation in living cells has remained unknown. Advances in live-cell single-molecule imaging have opened a new frontier for studying transcription in vivo. In this study, we used synthetic LacO (Lac operator) arrays as well as endogenous GGAA microsatellite loci to study LCD-LCD interactions of TFs such as EWS/FLI1, TAF15, and Sp1 in live cells. To probe the dynamic behavior of TF LCDs at target genomic loci, we have combined CRISPR-Cas9 genome editing, mutagenesis, gene activation, cell transformation assays, and various high-resolution imaging approaches including fluorescence correlation spectroscopy, fluorescence recovery after photobleaching, lattice light-sheet microscopy, three-dimensional DNA fluorescence in situ hybridization, and live-cell single-particle tracking. RESULTS. Live-cell single-molecule imaging revealed that TF LCDs interact to form local high-concentration hubs at both synthetic DNA arrays and endogenous genomic loci. TF LCD hubs stabilize DNA binding, recruit RNA polymerase II (RNA Pol II), and activate transcription. LCD-LCD interactions within hubs are highly dynamic (seconds to minutes), selective for binding partners, and differentially sensitive to disruption by hexanediols. These findings suggest that under physiological conditions, rapid, reversible, and selective multivalent LCD-LCD interactions occur between TFs and the RNA Pol II machinery to activate transcription. We observed formation of functional TF LCD hubs at a wide range of intranuclear TF concentrations. Although we detected apparent liquid-liquid phase separation with gross overexpression of LCDs, transcriptionally competent TF LCD hubs were observed at physiological TF levels at endogenous chromosomal loci in the absence of detectable phase separation. In addition, mutagenesis, gene expression, and cell transformation assays in Ewing's sarcoma cells revealed a functional link between LCD-LCD interactions, transactivation capacity, and oncogenic potential. CONCLUSION. The use of various imaging methods in live cells powerfully complements in vitro studies and provides new insights into the nature of LCD interactions and their role in gene regulation. We propose that transactivation domains function by forming local high-concentration hubs of TFs via dynamic, multivalent, and specific LCD-LCD interactions. It also seems likely that weak, dynamic, and transient contacts between TFs play a role in disease-causing dysregulation of gene expression (i.e., EWS/FLI1 in Ewing's sarcoma), suggesting that LCD-LCD interactions may represent a new class of viable drug targets. Although we examined a small subset of TF LCDs, the principles uncovered regarding the dynamics and mechanisms driving LCD-LCD interactions may be applicable to other classes of proteins and biomolecular interactions occurring in many cell types.

Files

Additional details