Cell-to-cell variability of dynamic CXCL12-CXCR4 signaling and morphological processes in chemotaxis

Chemotaxis drives critical processes in cancer metastasis. While commonly studied at the population scale, metastasis arises from small numbers of cells that successfully disseminate, underscoring the need to analyze chemotaxis at single-cell resolution. Here we focus on chemotaxis driven by the CXCL12-CXCR4 pathway, a signaling network that promotes metastasis in more than 20 different human cancers. CXCL12-CXCR4 activates ERK and Akt, kinases known to promote chemotaxis, but how cells couple signaling to chemotaxis remain poorly defined. To address this challenge, we implemented single-cell analysis of MDA-MB-231 breast cancer cells migrating in a chemotaxis device towards chemokine CXCL12. We integrated live, single-cell imaging with advanced computational analysis methods to discover processes defining subsets of cells that move efficiently toward a CXCL12 gradient. We identified dynamic oscillations in ERK and Akt signaling and associated morphological transitions as key determinants of successful chemotaxis. Cells with effective chemotaxis toward CXCL12 exhibit faster and more persistent movement than non-migrating cells, but both cell populations show similar random motion. Migrating cells exhibit higher amplitude fluctuations in ERK and Akt signaling and greater frequencies of generating lateral cell membrane protrusions. Interestingly, computational analysis reveals less correlated network coupling of signaling and morphological changes in migrating cells. These data reveal processing events that enable cells to convert a signaling input to chemotaxis and highlight how cells in a uniform environment produce heterogeneous responses. ### Competing Interest Statement G.D.L. has a research contract from InterAx AG administered through the University of Michigan. All other authors declare no competing interests.