Dynamics and Heterogeneity of Gene Expression and Epigenetic Regulation at the Single-Cell Level

The ability of cells to establish and remember their gene expression states is a cornerstone of multicellular life. This thesis explores how gene expression states are regulated dynamically, and how these regulations differ in individual cells even under the same conditions. These properties underlie cellular state decisions and often determine the balance between different cell types in a multicellular system, but are typically inaccessible to conventional techniques that rely on static snapshots and population averaging. We address these issues in two separate contexts, one natural and one synthetic, using time-lapse imaging and other single-cell techniques. In the first context, we use embryonic stem cells (ES), which were shown to exist in a mixed population of at least two cellular states with distinct differentiation propensities, as a model to study natural dynamics of cellular states. These cells display rare, stochastic, and spontaneous transitions between the two states, as well as more frequent fluctuations in gene expression levels within each state. Our system enables us to further investigate how these dynamics are modulated under a cell signaling environment that enhances pluripotency, and the role DNA methylation plays in maintaining these states. In the second context, we investigate how chromatin regulators (CRs), part of a complex system that enables cells to modulate gene expression and epigenetic memory, operate dynamically in individual cells. We build a synthetic platform to measure the isolated effect of recruitment and de-recruitment of four individual CRs. In contrast to conventional transcription factor control, all CRs tested regulate gene expression in all-or-none events, controlling the probability of stochastic transitions between fully active and silent states rather than the strength of gene expression. The qualitative and quantitative responses of a cell population are determined by the set of event rates associated with each CR, as well as the duration of CR recruitment. These results provide a framework for understanding and engineering chromatin-based cellular states and their dynamics.