Emergence of multiple collective motility modes in a physical model of cell chains

Abstract

Collective cell migration underpins key (patho)physiological processes, ranging from embryonic development to wound healing and cancer metastasis. While notable progress has been made in elucidating mechanisms that drive collective cell motility, the classification remains incomplete. In this study, we focus on the migration patterns of small cell chains, specifically cohesive pairs of cells migrating after each other on flat surfaces. Experiments with Dictyostelium discoideum (Dd) cells, which typically display amoeboid motility, revealed two distinct motility modes in cell pairs: the individual contributor (IC) mode, where each cell generates its own traction force dipole, and the supracellular (S) mode, characterized by a single dipole. Intriguingly, the IC mode dominates in Dd pairs, but the S mode prevails in Madin-Darby canine kidney (MDCK) cell doublets, which typically undergo mesenchymal motility. This observation highlights an apparent discrepancy in emergent motility modes between cell types. To uncover the physical mechanisms driving these diverse motility modes, we developed a two-dimensional biophysical model incorporating mechanochemical details such as cell-cell adhesion, combined with membrane-cortex contractility, and cell-matrix adhesion. Our model could recapitulate many experimental observations; the IC mode emerged naturally in amoeboid doublets when both cells exerted similar traction stresses, while the S mode dominated with "stronger" leaders that essentially pull on trailers. In contrast, in our simulations, mesenchymal MDCK-like pairs largely migrated in supracellular arrangement (S mode), with traction stress patterns representative of a rear-drive system with a "pushy" trailer, rather than a front-drive system. Our findings also showed that increasing cell-matrix adhesion predisposes amoeboid cell chains to act autonomously (IC mode), but the chains motility mode was largely insensitive to changes in cell-cell adhesion parameters. Contrary to amoebas, MDCK-like cell chains showed a bias towards S mode when increasing cell-matrix adhesion and a preference on IC mode when increasing cell-cell adhesion. Extending the model to longer cell chains, we showcase the models applicability across scales, providing a foundation for exploring collective migratory behavior in other contexts.