Focal adhesion kinase (FAK) is a major mediator of integrin signaling pathways. The mechanisms of regulation of FAK activity and its associated cellular functions are not very well understood. Here, we present data suggesting that a novel protein FIP200 functions as an inhibitor for FAK. We show the association of endogenous FIP200 with FAK, which is decreased upon integrin-mediated cell adhesion concomitant with FAK activation. In vitro- and in vivo-binding studies indicate that FIP200 interacts with FAK through multiple domains directly. FIP200 bound to the kinase domain of FAK inhibited its kinase activity in vitro and its autophosphorylation in vivo. Overexpression of FIP200 or its segments inhibited cell spreading, cell migration, and cell cycle progression, which correlated with their inhibition of FAK activity in vivo. The inhibition of these cellular functions by FIP200 could be rescued by coexpression of FAK. Last, we show that disruption of the functional interaction between endogenous FIP200 with FAK leads to increased FAK phosphorylation and partial restoration of cell cycle progression in cells plated on poly-L-lysine, providing further support for FIP200 as a negative regulator of FAK. Together, these results identify FIP200 as a novel protein inhibitor for FAK.

The current dogma of G 1 cell-cycle progression relies on growth factor-induced increase of cyclin D:Cdk4/6 complex activity to partially inactivate pRb by phosphorylation and to sequester p27 Kip1 triggering activation of cyclin E:Cdk2 complexes that further inactivate pRb. pRb oscillates between an active, hypophosphorylated form associated with E2F transcription factors in early G 1 phase and an inactive, hyperphosphorylated form in late G 1 , S and G 2 /M phases. However, under constant growth factor stimulation, cells show constitutively active cyclin D:Cdk4/6 throughout the cell cycle and thereby exclude cyclin D:Cdk4/6 inactivation of pRb. To address this paradox, we developed a mathematical model of G 1 progression using physiological expression and activity profiles from synchronized cells exposed to constant growth factors and included a metabolically responsive, activating modifier of cyclin E:Cdk2. Our mathematical model accurately simulates G 1 progression, recapitulates observations from targeted gene deletion studies and serves as a foundation for development of therapeutics targeting G 1 cell-cycle progression.

Using the Diagrammatic Cell Language trade mark, Gene Network Sciences (GNS) has created a network model of interconnected signal transduction pathways and gene expression networks that control human cell proliferation and apoptosis. It includes receptor activation and mitogenic signaling, initiation of cell cycle, and passage of checkpoints and apoptosis. Time-course experiments measuring mRNA abundance and protein activity are conducted on Caco-2 and HCT 116 colon cell lines. These data were used to constrain unknown regulatory interactions and kinetic parameters via sensitivity analysis and parameter optimization methods contained in the DigitalCell computer simulation platform. FACS, RNA knockdown, cell growth, and apoptosis data are also used to constrain the model and to identify unknown pathways, and cross talk between known pathways will also be discussed. Using the cell simulation, GNS tested the efficacy of various drug targets and performed validation experiments to test computer simulation predictions. The simulation is a powerful tool that can in principle incorporate patient-specific data on the DNA, RNA, and protein levels for assessing efficacy of therapeutics in specific patient populations and can greatly impact success of a given therapeutic strategy.

No abstract

Polymerization of soluble fibronectin into extracellular matrix fibers occurs through the interaction between the amino terminus of fibronectin contained within a 70 kDa fragment and ‘matrix assembly sites’ on the cell surface. The present studies were performed to localize the ‘matrix assembly sites’ (defined by 70 kDa binding sites) on newly adherent cells and on cells containing preformed fibronectin matrix. Matrix nucleation sites on newly spread cells were visualized using Texas Red conjugated 70 kDa fragment and were found to colocalize with vinculin and substrate fibronectin fibrils. Cells plated onto vitronectin coated coverslips did not exhibit any 70 kDa binding sites although these cells were well-spread with fully developed focal adhesions. Time course studies indicated that 70 kDa binding sites could be detected on newly adherent cells within 30–40 minutes following cell plating onto fibronectin coated coverslips, prior to the reorganization of substrate fibronectin into fibrils. Similarly, exogenous fibronectin conjugated with Texas Red was also colocalized with vinculin when added to newly adherent cells. The disruption of actin filaments with cytochalasin D both prevented the expression of 70 kDa binding sites and also resulted in the loss of established 70 kDa binding sites on newly spread cells. After 3 days in culture, cells organized an extensive fibronectin matrix and 70 kDa was colocalized with two distinct types of matrix fibronectin fibers: fine linear cell-associated fibers which co-stained with the beta1 integrin and coarse extracellular fibers which did not stain for the beta1 integrin. There was also a third type of fibronectin fiber which was organized into a meshwork structure. There was no localization of either beta1 or 70 kDa to these structures. Treatment of 3-day cells with cytochalasin D resulted in the disruption of cell-matrix fibers and cell-associated 70 kDa binding sites. In contrast, the coarse extracellular matrix fibers as well as the meshwork fibers were unaffected by cytochalasin. In the presence of cytochalasin D, 70 kDa bound to sites which colocalized with the coarse extracellular matrix fibers. These data suggest that de novo assembly of fibronectin matrix occurs at sites of focal adhesion and as fibronectin polymerization proceeds, matrix nucleation sites colocalize along cell associated fibronectin fibers. At later times 70 kDa is localized to a subset of more mature fibronectin-containing fibers. These results suggest that there are at least three morphologically distinct 70 kDa binding sites on adherent cells: one which colocalizes with beta1 to focal adhesions, a second which colocalizes with beta1 and fibronectin in matrix contacts, and a third which localizes to extracellular matrix fibers.