UCSF

The phenotypic behavior of a cell is driven by its molecular microstate, which emerges as a function of its epigenetic, transcriptomic, and proteomic states. Variability in these states can arise due to stochastic processes, leading to extensive microstate and functional variability in genetically clonal populations. It is of fundamental interest in cell biology to understand the number of states a cell can occupy, as well as the laws governing state transitions and the functional correlates of state variability. In this work, we develop an ensemble of molecular biology, computational, and statistical tools to measure and quantify morphological heterogeneity in both fixed and live populations of isogenic wild-type (WT) mouse embryonic fibroblasts (MEFs). We discover that WT MEFs occupy a continuum of states in morphology space, with increased density in specific subspaces and that this increased occupancy is due to a non-uniform energy landscape underlying WT MEF morphology space. Finally, we show that drug induction can induce topological changes to the energy landscape and that these changes correspond directly to functional divergence at the cellular level.

This work produces new insight into the topological and dynamical properties of morphological heterogeneity and provides a body of evidence suggesting that variability in these parameters correlate directly to functional heterogeneity within an isogenic population. Furthermore, this work positions morphological heterogeneity as an important and consequential form of non-genetic heterogeneity, setting the groundwork for future studies of the relationships between microstate, morphological, and functional variability.