No abstract

Despite recent efforts to understand homeostasis in epithelial tissues, there are many unknowns surrounding this steady state. It is considered to be regulated by mechanoresponse, but unlike for single cells, this remains heavily debated for tissues. Here, we show that changes in matrix stiffness induce a non-equilibrium transition from tubular to squamous Madin-Darby Canine Kidney II tissues. Nonetheless, despite different cell morphologies and densities, all homeostatic tissues display equivalent topologies, which, hence, must be actively targeted and regulated. On the contrary, the mechanoresponse induces dramatic changes in the large-scale organization of the colonies. On stiff gels, this yields an unreported cooperative state of motile cells displaying higher densities than in the arrested homeostatic state. This suggests a more complex relation between cell density and motility than previously anticipated. Our results unequivocally relate the mechanosensitive properties of individual cells to the evolving macroscopic structures, an effect that could be important for understanding the emergent pathologies of living tissues.

Despite the primary role of cell proliferation in tissue development and homeostatic maintenance, the interplay between cell density, cell mechanoresponse, and cell growth and division is poorly described in theoretical models of tissues. As a consequence, the predictive power of such models is largely undermined, and the lack of systematic statistical and experimental analysis in controlled experiments does not allow to calibrate and validate already existing tools. In this article, we first report an experimental investigation of cell proliferation on all time- and length-scales and quantify the role of cell-cell and cell-matrix interactions. Based on these results, we build a minimal density-driven mean-field model of cell proliferation within 2D epithelia which can account for mechanoresponse and which is based on a separation of cell population into growing and dividing cells. We show that the model can capture the in-vitro experimental results and is further validated by in-silico experiments, and we highlight the importance of the separation of time scale and populations in the description of the cell life cycle. Additionally, we show that the mechanoresponse observed in the proliferation patterns is responsible of large-scale density structures across the tissues that we can capture by nonlinear delayed Fisher-Kolmogorov-like equations. This work sets a first building stone in a theoretical description of tissues on long time- and length-scale while accounting for proliferation.