Matrix density-induced mechanoregulation of breast cell phenotype, signaling and gene expression through a FAK-ERK linkage.

Mammographically dense breast tissue is one of the greatest risk factors for developing breast carcinoma, yet the associated molecular mechanisms remain largely unknown. Importantly, regions of high breast density are associated with increased stromal collagen and epithelial cell content. We set out to determine whether increased collagen-matrix density, in the absence of stromal cells, was sufficient to promote proliferation and invasion characteristic of a malignant phenotype in non-transformed mammary epithelial cells. We demonstrate that increased collagen-matrix density increases matrix stiffness to promote an invasive phenotype. High matrix stiffness resulted in increased formation of activated three-dimensional (3D)-matrix adhesions and a chronically elevated outside-in/inside-out focal adhesion (FA) kinase (FAK)-Rho signaling loop, which was necessary to generate and maintain the invasive phenotype. Moreover, this signaling network resulted in hyperactivation of the Ras-mitogen-activated protein kinase (MAPK) pathway, which promoted growth of mammary epithelial cells in vitro and in vivo and activated a clinically relevant proliferation signature that predicts patient outcome. Hence, the current data provide compelling evidence for the importance of the mechanical features of the microenvironment, and suggest that mechanotransduction in these cells occurs through a FAK-Rho-ERK signaling network with extracellular signal-regulated kinase (ERK) as a bottleneck through which much of the response to mechanical stimuli is regulated. As such, we propose that increased matrix stiffness explains part of the mechanism behind increased epithelial proliferation and cancer risk in human patients with high breast tissue density.

Ras and Rho protein induction of motility and invasion in T47D breast adenocarcinoma cells.

There has been much interest in how Ras and Rho proteins affect cell migration and invasion. Studies of the molecular mechanisms and signaling pathways that are involved will be aided by the development and use of relevant model systems and simple in vitro assays. While a complete understanding of metastasis will ultimately need to employ in vivo studies, the migration and invasion assays presented here are good initial assessments of events that are relevant to the metastatic cascade.

Melanoma chondroitin sulphate proteoglycan regulates cell spreading through Cdc42, Ack-1 and p130cas.

Melanoma chondroitin sulphate proteoglycan (MCSP) is a cell-surface antigen that has been implicated in the growth and invasion of melanoma tumours. Although this antigen is expressed early in melanoma progression, its biological function is unknown. MCSP can stimulate the integrin-alpha4 beta1-mediated adhesion and spreading of melanoma cells. Here we show that stimulated MCSP recruits tyrosine-phosphorylated p130 cas, an adaptor protein important in tumour cell motility and invasion. MCSP stimulation also results in a pronounced activation and recruitment of the Rho-family GTPase Cdc42. MCSP-induced spreading of melanoma cells is dependent upon active Cdc42, a Cdc42-associated tyrosine kinase (Ack-1) and tyrosine phosphorylation of p130cas. Furthermore, vectors inhibiting Ack-1 or Cdc42 expression and/or function abrogate MCSP-induced tyrosine phosphorylation and recruitment of p130cas. Our findings indicate that MCSP may modify tumour growth or invasion by a unique signal-transduction pathway that links Cdc42 activation to downstream tyrosine phosphorylation and subsequent cytoskeletal reorganization.

R-Ras signals through specific integrin alpha cytoplasmic domains to promote migration and invasion of breast epithelial cells.

Specificity and modulation of integrin function have important consequences for cellular responses to the extracellular matrix, including differentiation and transformation. The Ras-related GTPase, R-Ras, modulates integrin affinity, but little is known of the signaling pathways and biological functions downstream of R-Ras. Here we show that stable expression of activated R-Ras or the closely related TC21 (R-Ras 2) induced integrin-mediated migration and invasion of breast epithelial cells through collagen and disrupted differentiation into tubule structures, whereas dominant negative R-Ras had opposite effects. These results imply novel roles for R-Ras and TC21 in promoting a transformed phenotype and in the basal migration and polarization of these cells. Importantly, R-Ras induced an increase in cellular adhesion and migration on collagen but not fibronectin, suggesting that R-Ras signals to specific integrins. This was further supported by experiments in which R-Ras enhanced the migration of cells expressing integrin chimeras containing the alpha2, but not the alpha5, cytoplasmic domain. In addition, a transdominant inhibition previously noted only between integrin beta cytoplasmic domains was observed for the alpha2 cytoplasmic domain; alpha2beta1-mediated migration was inhibited by the expression of excess alpha2 but not alpha5 cytoplasmic domain-containing chimeras, suggesting the existence of limiting factors that bind the integrin alpha subunit. Using pharmacological inhibitors, we found that R-Ras induced migration on collagen through a combination of phosphatidylinositol 3-kinase and protein kinase C, but not MAPK, which is distinct from the other Ras family members, Rac, Cdc42, and N- and K-Ras. Thus, R-Ras communicates with specific integrin alpha cytoplasmic domains through a unique combination of signaling pathways to promote cell migration and invasion.

Cdc42 and Rac1 induce integrin-mediated cell motility and invasiveness through PI(3)K.

Transformation of mammary epithelial cells into invasive carcinoma results in alterations in their integrin-mediated responses to the extracellular matrix, including a loss of normal epithelial polarization and differentiation, and a switch to a more motile, invasive phenotype. Changes in the actin cytoskeleton associated with this switch suggest that the small GTPases Cdc42 and Rac, which regulate actin organization, might modulate motility and invasion. However, the role of Cdc42 and Rac1 in epithelial cells, especially with respect to integrin-mediated events, has not been well characterized. Here we show that activation of Cdc42 and Rac1 disrupts the normal polarization of mammary epithelial cells in a collagenous matrix, and promotes motility and invasion. This motility does not require the activation of PAK, JNK, p70 S6 kinase, or Rho, but instead requires phosphatidylinositol-3-OH kinase (PI(3)K). Further, direct PI(3)K activation is sufficient to disrupt epithelial polarization and induce cell motility and invasion. PI(3)K inhibition also disrupts actin structures, suggesting that activation of PI(3)K by Cdc42 and Rac1 alters actin organization, leading to increased motility and invasiveness.

The alpha2beta1 integrin is a necessary co-receptor for collagen-induced activation of Syk and the subsequent phosphorylation of phospholipase Cgamma2 in platelets.

Although there are multiple potential collagen-binding proteins on platelets, the contribution of each to collagen-induced signaling events and platelet activation is unclear. We investigated which early platelet signaling events, if any, could be attributed specifically to the binding of collagen to one of its receptors, the alpha2beta1 integrin. Treatment of platelets with collagen induced a rapid activation of the non-receptor tyrosine kinase, Syk, as measured by an increase in phosphorylation and kinase activity. Collagen also induced the rapid phosphorylation of phospholipase Cgamma2 (PLCgamma2). The phosphorylation of both Syk and PLCgamma2, as well as platelet aggregation, was blocked by an anti-alpha2beta1 integrin monoclonal antibody (P1E6), demonstrating that collagen binding to alpha2beta1 is necessary for signaling. Cross-linking of the alpha2beta1 integrin with stimulatory monoclonal antibody against either the beta1 or alpha2 subunit stimulated the phosphorylation of both Syk and PLCgamma2. However, antibody stimulation was dependent on co-stimulation of the FcgammaII receptor (CD32) since specific F(ab')2 fragments did not induce Syk and PLCgamma2 phosphorylation. Thus, these results suggest that occupancy of alpha2beta1 by collagen is necessary, but that a co-receptor, in addition to alpha2beta1, is required for these collagen-induced signaling events. Moreover, the P1E6 antibody did not inhibit all collagen-induced tyrosine phosphorylation events, demonstrating that collagen also induces phosphorylation events that are independent of the alpha2beta1 integrin. In addition to Syk and PLCgamma2, we identified the FcgammaII receptor (FcgammaRII) as being rapidly phosphorylated in response to collagen stimulation, even in the absence of antibodies. Finally, to determine if Syk activation precedes and directly contributes to the phosphorylation of PLCgamma2, platelets were preincubated with the Syk-selective kinase inhibitor, piceatannol. A concentration of piceatannol that inhibited the phosphorylation and kinase activity of Syk, but had no effect on Src kinase activity, blocked the collagen-induced phosphorylation of PLCgamma2 and also inhibited collagen-induced platelet aggregation. Our results begin to delineate a signaling pathway whereby occupancy of the alpha2beta1 integrin is required, but not sufficient, for collagen-induced activation of Syk and the subsequent phosphorylation of PLCgamma2. These events are necessary for platelet activation and aggregation in response to collagen.

Loss of MDCK cell alpha 2 beta 1 integrin expression results in reduced cyst formation, failure of hepatocyte growth factor/scatter factor-induced branching morphogenesis, and increased apoptosis.

Cellular interactions with collagen in a model of kidney tubulogenesis were investigated using Madin-Darby canine kidney (MDCK) cells in an in vitro morphogenetic system. MDCK cells adhered to collagen types I and IV in a Mg(2+)-dependent manner, typical of the alpha 2 beta 1 integrin. Collagen-Sepharose affinity chromatography and immunoblotting demonstrated the presence and collagen binding activity of the alpha 2 beta 1 integrin on MDCK cells. To assess the function of alpha 2 beta 1 integrin, MDCK cells were transfected with a plasmid pRSV alpha 2' which allowed the expression of alpha 2-integrin subunit antisense RNA. Three G418-resistant clones showing reduced adhesion to collagen, stable genomic integration of the antisense construct, decreased alpha 2-integrin subunit mRNA and decreased alpha 2-integrin subunit protein expression were selected for analysis in morphogenetic experiments. MDCK cells and plasmid-only control transfectants, cultured in three-dimensional collagen type I gels, showed normal cyst formation, whereas the antisense RNA transfectants showed increased apoptosis and formed small rudimentary cysts. Stimulation with hepatocyte growth factor/scatter factor-containing 3T3 fibroblast-conditioned medium or recombinant hepatocyte growth factor/scatter factor resulted in extensive branching of the preformed control cysts whereas the surviving small cysts formed by antisense expressing cells increased in size but failed to elongate and branch upon stimulation. We conclude that alpha 2 beta 1 integrin collagen interactions play a crucial role in the hepatocyte growth factor/scatter factor-induced tubulogenesis and branching morphogenesis of MDCK cells in collagen gels as well as an important role in cell survival.

The spatial and temporal expression of the alpha 2 beta 1 integrin and its ligands, collagen I, collagen IV, and laminin, suggest important roles in mouse mammary morphogenesis.

To begin to determine the role of the alpha 2 beta 1 integrin and its ligands, collagen I, collagen IV, and laminin, in mammary epithelial differentiation in vivo, we determined the expression of these molecules by in situ hybridization and immunofluorescence in the developing mouse mammary gland. Expression of collagen I, collagen IV, and laminin mRNAs in the mammary gland during puberty corresponded to the period of greatest growth of the gland, 4-7 weeks postnatally. Collagen I expression preceded collagen IV expression, both of which preceded laminin expression, suggesting an important temporal sequence of extracellular matrix (ECM) production. When growth of the epithelium ceased in the adult virgin gland, expression of all three mRNAs became undetectable. Following the onset of pregnancy these molecules were re-expressed with the same chronology observed during puberty. Collagen I, collagen IV, and laminin were expressed by stromal cells immediately surrounding the developing ductal epithelium. Surprisingly, we found no expression of ECM components in the epithelial cells, suggesting the mammary epithelium does not synthesize its own basement membrane. The distribution of collagen I was consistent with a role in duct formation, since collagen I was strikingly abundant around larger mammary ducts, but was sparse around growing endbuds or alveoli. Conversely, there was abundant laminin near growing endbuds and around alveoli, and less around large ducts, suggesting its role is different than collagen I. The alpha 2 beta 1 integrin was present on the basal, lateral, and apical surfaces of the mammary epithelium throughout postnatal development and pregnancy. The alpha 2 beta 1 integrin expression was strongest at midpregnancy, suggesting a role for alpha 2 beta 1 integrin in the alveolar formation that occurs at this time. The alpha 2 beta 1 integrin expression decreased dramatically in the lactating gland. Our results suggest that alpha 2 beta 1 integrin interactions with its temporally and spatially regulated ligands, collagen I, collagen IV, and laminin, could play an important role in mammary morphogenesis in vivo.

Alteration of collagen-dependent adhesion, motility, and morphogenesis by the expression of antisense alpha 2 integrin mRNA in mammary cells.

Although integrins are known to mediate adhesive binding of cells to the extracellular matrix, their role in mediating cellular growth, morphology, and differentiation is less clear. To determine more directly the role of the alpha 2 beta 1 integrin, a collagen and laminin receptor, in mediating the collagen-dependent differentiation of mammary cells, we reduced expression of the integrin by the well differentiated human breast carcinoma cell line, T47D, by stably expressing alpha 2 integrin antisense mRNA. Flow cytometry demonstrated that the antisense-expressing clones had levels of alpha 2 beta 1 integrin on their surfaces that were decreased by 30-70%. Adhesion of antisense-expressing clones to both collagens I and IV was decreased relative to controls in a manner that correlated with the level of cell surface alpha 2 beta 1 integrin expression. Adhesion to fibronectin and laminin were not affected. Motility across collagen-coated filters in haptotaxis assays was increased for only those clones that exhibited intermediate levels of adhesion to collagen, suggesting that an intermediate density of cell-surface alpha 2 beta 1 integrin optimally supports cell motility. When cultured in three-dimensional collagen gels, T47D cells organized in a manner suggestive of a glandular epithelium. In contrast, antisense-expressing clones with decreased alpha 2 beta 1 integrin were not able to organize in three-dimensional collagen gels. The growth rate of T47D cells was reduced when the cells were cultured in three-dimensional collagen gels. Unlike adhesion, motility, and morphogenesis, growth rates were unaffected by reduction of alpha 2 beta 1 integrin expression. Our results suggest that adhesive interactions mediated by a critical level of surface alpha 2 beta 1 integrin expression are key determinants of the collagen-dependent morphogenetic capacity of mammary epithelial cells.