Compressive mechanical stress triggers MAPK-dependent loss of cell polarity

Abstract

Compressive mechanical force is an important physiological signal that underlies cellular processes in development and disease. Cells become compressed as they migrate through narrow constrictions or if they grow in confined spaces with steric hinderance from either neighboring cells or the extracellular matrix during embryogenesis and tumor growth. Not only is this seen in multicellular organisms, but also in simpler eukaryotes such as yeast when they grow as colonies or biofilms in limited spaces. However, the underlying mechanisms that mediate their responses to compressive signals remain elusive. This is primarily due to technological limitations in applying specific stimuli to trigger mechano-sensitive pathways. To circumvent this limitation, we have designed a novel microfluidic device coupled to live-cell microscopy, enabling us to both apply compressive stress to budding yeast cells and to perform quantitative single-cell analyses of their responses. Using this platform, we discovered that yeast cells sense compressive stress through the Ca2+-channel Mid1/Cch1 and the cell surface protein Mid2. These proteins activate calcium signaling and the Pkc1/Mpk1 pathway respectively, cooperatively enabling cell survival under compressive mechanical conditions. The application of mechanical stress also disrupts polarized actin cables in a Mid2-dependent manner, thereby antagonizing polarized growth. Failure to depolymerize the actin cytoskeleton leads to cell lysis when subjected to compressive mechanical stress. We further found that compressive stress-induced activation of Pkc1 also prevents pheromone-induced cell polarization in budding yeast. Mechanistically, Pkc1 phosphorylates the MAP kinase scaffold Ste5 at a conserved Serine 185 residue within its RING-H2 domain. This phosphorylation prevents Ste5 from binding to Gβγ heterodimers, leading to the dispersal of Ste5 from shmoo tips. In contrast, a non-phosphorylatable Ste5 mutant protein remains at the shmoo tips even under mechano-stress, leading to cell lysis. During mating, the residence time of non-phosphorylatable Ste5 at the of site cell-cell fusion is longer compared to that of the wild type, thereby contributing to an increased frequency of cell-cell fusion defects. After showing that compressive stress prevents cell polarization in budding yeast in two physiological contexts, we studied the conservation of this mechanism in a mammalian system. Interestingly, mammalian cells reorganize their actin cytoskeleton and lose cell-cell contacts and cell polarity during epithelial-mesenchymal transition (EMT). As EMT is a critical step during the progression of tumors of epithelial origin, we hypothesized that growth-induced compressive forces in tumors could trigger EMT. To test this hypothesis, we established a transwell-based compression platform and applied mechanical forces to epithelial monolayers formed by NMuMG E9 and PyMT-1099 cells. Indeed, our preliminary data indicate that, like in yeast, sustained compression triggers the reorganization of the actin cytoskeleton, which leads to the acquisition of a spindle-shaped morphology, a loss of cell-cell contacts, and dissolution of cobblestone-like morphology. Under these conditions, cells also showed reorganization of E-cadherin, nuclear relocation of ZEB1 and upregulation of vimentin, suggesting a compressive stress-induced EMT phenotype. Furthermore, YAP1, a transcription factor regulated by the Hippo signaling pathway, relocated into the nucleus when subjected to compressive stress. Interestingly, YAP1 relocation was dependent on the activity of ERK5, the homologue of yeast Mpk1. Taken together, our results suggest that mammalian and yeast cells use conserved signaling mechanisms to trigger the loss of cell polarity under compressive mechanical stress conditions, and our data further emphasize the importance of sensing mechanical stimuli in cellular physiology.