Contributions of cluster shape and intercellular adhesion to epithelial discohesion and emergent dynamics in collective migration

As a physical system, a cell interacts with its environment through physical and chemical processes. The cell can change these interactions through modification of its environment or its own composition. This dissertation presents the overarching hypothesis that both biochemical regulation of intercellular adhesion and physical interaction between cells are required to account for the emergence of cluster migration and collective dynamics observed in epithelial cells. Collective migration is defined as the displacement of a group of cells with transient or permanent cell-cell contacts. One mode, cluster migration, plays an important role during embryonic development and in cancer metastasis. Despite its importance, collective migration is a slow process and hard to visualize, and therefore it has not been thoroughly studied in three dimensions (3D). Based on known information about cluster migration from 2D studies of epithelial sheets and 3D single cell migration, this dissertation presents theoretical and experimental techniques to assess the independent contribution of physical and biochemical factors to 3D cluster migration. It first develops two computational models that explore the interaction between cells and the ECM and epithelial discohesion. These discrete mechanistic models reveal the need to account for intracellular regulation of adherens junctions in space and time within a cluster. Consequently, a differential algebraic model is developed that accounts for cross-reactivity of three pathways in a regulatory biochemical network: Wnt/β-catenin signaling, protein N-glycosylation, and E-cadherin adhesion. The model is tested by matching predictions to Wnt/β-catenin inhibition in MDCK cells. The model is then incorporated into a self-propelled particle (SPP) model, creating the first SPP model for study of adhesive mammalian cellular systems. MDCK cell clusters with fluorescent nuclei are grown, seeded, and tracked in 3D collagen gels using confocal microscopy. They provide data on individual cell dynamics within clusters. Borrowed from the field of complex systems, normalized velocity is used to quantify the order of both in vitro and simulated clusters. An analysis of sensitivity of cluster dynamics on factors describing physical and biochemical processes provides new quantitative insights into mechanisms underlying collective cell migration and explains temporal and spatial heterogeneity of cluster behavior