Deposited footprints let cells switch between confined, oscillatory, and exploratory migration

Abstract

For eukaryotic cells to heal wounds, respond to immune signals, or metastasize, they must migrate, often by adhering to extracellular matrix (ECM). Cells may also deposit ECM components, leaving behind a footprint that influences their crawling. Recent experiments showed that some epithelial cell lines on micropatterned adhesive stripes move persistently in regions they have previously crawled over, where footprints have been formed, but barely advance into unexplored regions, creating an oscillatory migration of increasing amplitude. Here, we explore through mathematical modeling how footprint deposition and cell responses to footprint combine to allow cells to develop oscillation and other complex migratory motions. We simulate cell crawling with a phase field model coupled to a biochemical model of cell polarity, assuming local contact with the deposited footprint activates Rac1, a protein that establishes the cell’s front. Depending on footprint deposition rate and response to the footprint, cells on micropatterned lines can display many types of motility, including confined, oscillatory, and persistent motion. On two-dimensional (2D) substrates, we predict a transition between cells undergoing circular motion and cells developing an exploratory phenotype. Small quantitative changes in a cell’s interaction with its footprint can completely alter exploration, allowing cells to tightly regulate their motion, leading to different motility phenotypes (confined vs. exploratory) in different cells when deposition or sensing is variable from cell to cell. Consistent with our computational predictions, we find in earlier experimental data evidence of cells undergoing both circular and exploratory motion.

Confinement, jamming, and adhesion in cancer cells dissociating from a collectively invading strand

“…In the absence of chemotaxis (k θ = 0), the angle θ i will simply diffuse with angular diffusion coefficient D r . This representation of cell polarity is a very simplified one, and could be extended by models of Rho GTPase dynamics [17,29] or stochastic protrusion dynamics [30].…”

“…Interestingly, the timescale of rupture appears to be independent of channel width in the experiments. We construct a model of this dissociation process by describing the cells as deformable crawling objects using a phase field model [7,17]. This model allows us to describe the shape of each individual cell, which is key to unjamming [7][8][9][10], while allowing us to model cells detaching from one another.…”