Theses Doctoral

Spatial Organization and Segregation of Cells in Breast Cancer

Devanny, Alexander

The aim of this thesis is to establish a simple physical framework that captures and predicts key aspects of the spatial organization of cells in models of breast cancer, while also probing the downstream consequences of particular tumor composition and cell organization within tumors on disease progression. An in vitro model of tumor heterogeneity and a complementary minimal computational model of cell sorting were used to accomplish these goals. We evaluated the tendencies of cells to sort and segregate, the factors driving the sorting process, and the mechanisms of invasion that cells exhibited as a result of different composition and cellular organization.

Chapter 1 presents background information on breast cancer progression, the origins and consequences of heterogeneity in tumors and their local microenvironment, existing theoretical and computational approaches to explain cell segregation in tissues, and commonly employed experimental models of cancer invasion.

Chapter 2 explores cell sorting of healthy and cancerous breast cells in an in vitro tumor model. This work was motivated by previous observations that mixing genetically distinct breast cancer cells results in cell sorting and the formation of sharp boundaries between cell types, analogous to the segregation of cells during embryonic development. We examined cell segregation among six different breast cell lines and found that more invasive breast cancer cells tended to sort to the outside of mixed cell-type aggregates, such that more aggressive cells were poised to invade the surrounding extracellular matrix. The particular sorting among all binary sets of breast cells studied was found to follow predictions of the differential adhesion hypothesis, which predicts cell sorting to be dependent on a combination of available adhesion proteins and actomyosin contractility. Differential adhesion was found to be a useful lens for not only rationalizing cell sorting tendencies but also directing the assembly of cells. In fact, we showed that through use of a simple contractility inhibiting agent, invasive cell types could be made to sort inside mixed-type aggregates, reducing subsequent invasion.

In Chapter 3, we further probed the applicability of differential adhesion frameworks for explaining cell segregation in cancer by employing a Cellular Potts model. Experimentally observed sorting patterns were replicated using a minimal model and varying only two parameters across simulated cell types – one that governs cell morphology and one that governs cell adhesion, thus validating the differential adhesion hypothesis as a useful minimal model to rationalize sorting in this system. Less invasive cell types were found to have more fluid-like character that drives the sorting process and leads to their positioning in the interior of simulated aggregates, surrounded by more invasive cells. We observed evidence of non-equilibrium behavior in certain less adhesive cell types, as well as the capacity for more adhesive cell types to enhance motility and fluidize those that otherwise demonstrate non-equilibrium, slow dynamics. Chapter 4 summarizes these findings and suggests future studies.

Throughout this work, we show the value of applying existing views of cell sorting for rationalizing cell segregation and tumor organization in breast cancer. We find that cells sort in a predictable manner that relates to aspects of their adhesive character as captured by differential adhesion, which is shown experimentally to depend on co-regulation of adhesion protein function and actomyosin contractility. These same properties also dictate cell invasive strategy and efficiency, making this a critical area of study to enhance understanding of cancer invasion and metastasis. The drivers of spatial organization of cells in tumors and the consequences of particular organization remain an underexplored topic in breast cancer research. We argue that continued study in this area can yield improved understanding of the impacts of tumor heterogeneity on cancer progression.


  • thumnail for Devanny_columbia_0054D_16852.pdf Devanny_columbia_0054D_16852.pdf application/pdf 3.43 MB Download File

More About This Work

Academic Units
Thesis Advisors
Kaufman, Laura J.
Ph.D., Columbia University
Published Here
October 6, 2021