Transcription occurs in stochastic bursts — periods of intense transcriptional activity, interrupted by extended periods of inactivity. The mechanism and purpose of transcriptional bursting are the major focus of our research. We address these problems theoretically, including the analysis of stochastic process models, and experimentally, by analyzing transcription and chromatin structure at the level of single gene molecules.
The specificity of transcriptional regulation – which genes are transcribed in response to a particular signal – is thought to rest entirely with the specificity of activator-gene recognition. Yet, for many eukaryotic activators the energetic differences between correct and incorrect sequence binding are remarkably small – too small to explain observed regulatory specificities. The specificity problem may be solved by ‘kinetic proofreading’, i.e., insertion of a free energy-transducing delay step between activator-promoter binding and transcription. The mechanism enhances specificity not by increasing the energetic difference between correct and incorrect associations but by exploiting the same difference twice, before the delay step and afterward, when free energy dissipation favors complex dissociation over formation and return of the suspended delay mechanism to its repressive state. A prime candidate among possible delay mechanisms for activator proofreading is ATP-dependent chromatin remodeling, which relieves genes from nucleosomal repression in an activator-dependent manner. Logical implications of this proposition, including the prediction that perturbation of chromatin remodeling affects the ability of activators to distinguish between correct and incorrect genes, are tested by employing a combination of life-cell multifocus fluorescence microscopy, single molecule-fluorescence in situ hybridization, nanopore sequencing, and reverse genetics in yeast. The outcomes will provide new insights into how near deterministic behavior of biological systems emerges from the random behavior of their molecular components.