A Center-Based Model Reveals that Cell Shape Variability Drives Cell Sorting in Spheroids

Abstract

We present a center-based, agent-based model of multicellular spheroids, which are 3D spherical in vitro cell aggregates that recapitulate key features of solid tumours. The model is implemented within the CompuTiX framework, in which each cell is an agent whose motion is governed by the balance of cell-cell and cell-ECM forces. We use the model to investigate cell sorting in a mixed spheroid initialised with two intermingled, quiescent cell types. To this end, we test several biomechanical hypotheses for what drives segregation. We find that the hypotheses of differential adhesion, high heterotypic interfacial tension and added random cell motility each fail to produce the sorted final state. Instead, we introduced variation in cell shape by modelling each cell’s volume as oscillating sinusoidally between cell nucleus volume and full-cell volume, with a random phase. We found that this latter hypothesis of cell shape variability causes the cells to cluster by type, reproducing the experimentally observed segregation both qualitatively and quantitatively. We conclude that the model is a promising tool for studying spheroid dynamics and intend to assess its ability to reproduce diverse spheroid phenotypes.