Ras induces NBT-II epithelial cell scattering through the coordinate activities of Rac and MAPK pathways

Cell dissociation and cell migration are the two main components of epithelium-mesenchyme transitions (EMT). We previously demonstrated that Ras is required for the accomplishment of both of these processes during the EGF-induced EMT of the NBT-II rat carcinoma cell line in vitro. In this study, we examined the downstream targets of Ras that are responsible for the dissociation and motility of NBT-II cells. Overexpression of activated forms of c-Raf and MEK1 (a component of the mitogen-activated protein kinase pathway, MAPK) led to cell dissociation, as inferred by the loss of desmosomes from the cell periphery. By contrast, active PI3K, RalA and RalB did not induce desmosome breakdown. The MEK1 inhibitor PD098059 inhibited EGF- and Ras-induced cell dispersion, whereas the PI3K inhibitor LY294002 had no effect. Accordingly, among the partial loss-of-function mutants of Ras (RasV12) that were used to distinguish between downstream targets of Ras, we found that the Raf-specific Ras mutants RasV12S35 and RasV12E38 induced cell dissociation. The PI3K- and RalGDS-activating Ras mutants had, in contrast, no effect on cell dispersion. However, MEK1 was unable to promote cell motility, whereas RasV12S35 and RasV12E38 induced cell migration, suggesting that another Ras effector was responsible for cell motility. We found that the small GTPase Rac is necessary for EGF-mediated cell dispersion since overexpression of a dominant-negative mutant of Rac1 (Rac1N17) inhibited EGF-induced NBT-II cell migration. All stimuli that promoted cell migration also induced Rac activation. Finally, coexpression of active Rac1 and active MEK1 induced the motility of NBT-II cells, suggesting that Ras mediates NBT-II cell scattering through the coordinate activation of Rac and the Raf/MAPK pathway.