Dissecting human gene regulation principles using single-cell measurements and statistical mechanics

The binding of transcription factors (TFs) to gene regulatory elements controls gene expression in human cells. However, the molecular details that link TF binding to gene expression levels remain unclear, preventing us from developing physics-based predictive models of gene regulation. We applied single-molecule footprinting in live human cells to measure the simultaneous occupancy of TFs, nucleosomes, and other regulatory proteins on engineered constructs with variable numbers of TF binding sites. Although TF binding events on nucleosome-free DNA are independent, TF binding on nucleosomal DNA does not fit a simple thermodynamic competition model between TFs and nucleosomes. Instead, the distribution of microstates fits a model where the activation domains of TFs recruit cofactors that destabilize nucleosomes, allowing TFs at nearby sites to bind more efficiently and driving apparent TF binding cooperativity. Once bound, each TF linearly contributes to the kinetic rate of promoter activation. Thus, TF strength can be decomposed mathematically into binding and activation terms, and each can be tuned independently. These thermodynamic and kinetic models quantitatively predict both the single-molecule microstates and gene expression dynamics. We are now expanding this approach to transcriptional repressors. This work provides a template for the quantitative dissection of distinct contributors to gene expression, including TF activation domains, concentration, binding affinity, binding site configuration and recruitment of chromatin regulators.