Epithelial cells require attachment to extracellular matrix (ECM) to suppress an apoptotic cell death program termed anoikis. Here we describe a nonapoptotic cell death program in matrix-detached cells that is initiated by a previously unrecognized and unusual process involving the invasion of one cell into another, leading to a transient state in which a live cell is contained within a neighboring host cell. Live internalized cells are either degraded by lysosomal enzymes or released. We term this cell internalization process entosis and present evidence for entosis as a mechanism underlying the commonly observed "cell-in-cell" cytological feature in human cancers. Further we propose that entosis is driven by compaction force associated with adherens junction formation in the absence of integrin engagement and may represent an intrinsic tumor suppression mechanism for cells that are detached from ECM.

The adult, virgin mammary gland is a highly organized tree-like structure formed by ducts with hollowed lumen. Although lumen formation during pubertal development appears to involve apoptosis, the molecular mechanisms that regulate this process are not known. Here, we demonstrate that disruption of the BH3-only proapoptotic factor Bim in mice prevents induction of apoptosis in and clearing of the lumen in terminal end buds during puberty. However, cells that fill the presumptive luminal space are eventually cleared from the adjacent ducts by a caspase-independent death process. Within the filled Bim(-/-) ducts, epithelial cells are deprived of matrix attachment and undergo squamous differentiation prior to clearing. Similarly, we also detect squamous differentiation in vitro when immortalized mammary epithelial cells are detached from the matrix. These data provide important mechanistic information on the processes involved in sculpting the mammary gland and demonstrate that BIM is a critical regulator of apoptosis in vivo.

Lineage formation in the lung mesenchyme is poorly understood. Using a transgenic mouse line expressing LacZ under the control of Fgf10 regulatory sequences, we show that the pool of Fgf10-positive cells in the distal lung mesenchyme contains progenitors of the parabronchial smooth muscle cells. Fgf10 gene expression is slightly repressed in this transgenic line. This allowed us to create a hypomorphic Fgf10 phenotype by expressing the LacZtransgene in a heterozygous Fgf10 background. Hypomorphic Fgf10 mutant lungs display a decrease inβ-galactosidase-positive cells around the bronchial epithelium associated with an accumulation of β-galactosidase-expressing cells in the distal mesenchyme. This correlates with a marked reduction of α smooth muscle actin expression, thereby demonstrating that FGF10 is mostly required for the entry of mesenchymal cells into the parabronchial smooth muscle cell lineage. The failure of exogenous FGF10 to phosphorylate its known downstream targets ERK and AKT in lung mesenchymal cultures strongly suggests that FGF10 acts indirectly on the progenitor population via an epithelial intermediate. We provide support for a role of epithelial BMP4 in mediating the formation of parabronchial smooth muscle cells.

Experimental evidence is rapidly emerging that the coupling of positive regulatory signals with the induction of negative feedback modulators is a mechanism of fine regulation in development. Studies in Drosophila and chick have shown that members of the SPROUTY family are inducible negative regulators of growth factors that act through tyrosine kinase receptors. We and others have shown that Fibroblast Growth Factor 10 (FGF10) is a key positive regulator of lung branching morphogenesis. Herein, we provide direct evidence that mSprouty2 is dynamically expressed in the peripheral endoderm in embryonic lung and is downregulated in the clefts between new branches at E12.5. We found that mSprouty2 was expressed in a domain restricted in time and space, adjacent to that of Fgf10 in the peripheral mesenchyme. By E14.5, Fgf10 expression was restricted to a narrow domain of mesenchyme along the extreme edges of the individual lung lobes, whereas mSprouty2 was most highly expressed in the subjacent epithelial terminal buds. FGF10 beads upregulated the expression of mSprouty2 in adjacent epithelium in embryonic lung explant culture. Lung cultures treated with exogenous FGF10 showed greater branching and higher levels of mSpry2 mRNA. Conversely, Fgf10 antisense oligonucleotides reduced branching and decreased mSpry2 mRNA levels. However, treatment with exogenous FGF10 or antisense Fgf10 did not change Shh and FgfR2 mRNA levels in the lungs. We investigated Sprouty2 function during lung development by two different but complementary approaches. The targeted overexpression of mSprouty2 in the peripheral lung epithelium in vivo, using the Surfactant Protein C promoter, resulted in a low level of branching, lung lobe edges abnormal in appearance and the inhibition of epithelial proliferation. Transient high-level overexpression of mSpry2 throughout the pulmonary epithelium by intra-tracheal adenovirus microinjection also resulted in a low level of branching. These results indicate for the first time that mSPROUTY2 functions as a negative regulator of embryonic lung morphogenesis and growth.