Recently a mechanism was described whereby human cells internalize into neighboring cells, called entosis. Entosis underlies the formation of 'cell-in-cell' structures, where viable cells are engulfed inside of others. These unusual cell structures have been reported in human tumors for decades, but their physiological role remains unknown. Entosis is induced by detachment of cells from extracellular matrix in vitro, and is prevalent in anchorage-independent growth assays in soft agar. In breast tumors, cell-in- cell structures are found in early-stage (DCIS) tumors, and also in late stage invasive tumors, in matrix-deprived regions, suggesting that this process could affect the formation or metastatic spread of cancers. Although cells internalized by entosis are initially viable, most eventually undergo cell death, suggesting that entosis could be a mechanism of tumor suppression. Cell death occurs by a nonapoptotic mechanism that can eliminate cells which are resistant to apoptosis. Entosis may therefore act as a backup or cooperative tumor suppressive mechanism to apoptosis to prevent transformed growth. The identification of this cellular program, whose evidence in vivo far predates the in vitro mechanism, was made possible only by real-time imaging of a classical assay of tumorigenicity, where the basic cellular programs that control the ability to grow are not defined. This proposal describes plans to examine tumorigenic transformation by real-time imaging, to elucidate the molecular mechanisms of entosis, and to examine the role of this process in human cancers.