Aye Aye Thant

Fibronectin activates matrix metalloproteinase-9 secretion via the MEK1-MAPK and the PI3K-Akt pathways in ovarian cancer cells.

Cell adhesion to the extracellular matrix appears to trigger a cascade of intracellular signalings. We have previously shown that treatment of ovarian cancer cells, NOM1, with fibronectin (FN) stimulated matrix metalloproteinase (MMP)-9 secretion and thereby activated the invasiveness of cells via the FAK/Ras signaling pathway. By use of chemical inhibitors, we investigated the downstream effectors critical for FN-dependent secretion of MMP-9. Treatment of cells with MEK1 inhibitors, U0126 and PD98059, dramatically suppressed the secretion of MMP-9 activated by FN. Similarly, PI-3 kinase inhibitors, Wortmannin and LY294002, strongly suppressed the FN-dependent secretion of MMP-9 together with the inhibition of Akt activation. In contrast, a specific PKC inhibitor (GF109203X) showed no inhibitory effect on the FN-dependent MMP-9 secretion. Moreover, we found that both the MEK1 inhibitor and the PI3-K inhibitor, but not the PKC inhibitor, strongly suppressed the invasiveness of NOM1 cells. Taken together, our results suggest that activation of dual signaling pathways, MEK1-MAPK and PI3K-Akt, is required for the FN-dependent activation of MMP-9 secretion. Our results suggest the importance of these signaling molecules as a chemotherapeutic target for cancer.

Secretion of matrix metalloproteinase‐9 by the proinflammatory cytokine, IL‐1β: a role for the dual signalling pathways, Akt and Erk

Background: Matrix metalloproteinases including MMP‐9 mediate matrix destruction during chronic inflammatory diseases such as arthritis and atherosclerosis. MMP‐9 up‐regulation by inflammatory cytokines involve interactions between several transcription factors including activator protein‐1 and NFκB. The upstream regulatory pathways are less well understood.

A role for FAK in the Concanavalin A-dependent secretion of matrix metalloproteinase-2 and -9.

To study the signaling pathway critical for the secretion of matrix metalloproteinases (MMPs), we examined the role of focal adhesion kinase (FAK) in Concanavalin A (Con A)-stimulated cells. We established a cell line in which FAK gene was conditionally inducible by use of FAK-null fibroblasts and the tetracycline repression system. In this cell line, FAK expression was undetectable in the presence of tetracycline but induced within 1 day by the removal of the drug. We found that FAK expression augmented the Con A-dependent secretion of MMP-9 and MMP-2. In contrast, proteolytic activation of MMP-2 by Con A-treatment did not require FAK expression. In addition, activation of MMP-secretion and tyrosine phosphorylation of FAK by Con A, but not the proteolytic activation of MMP-2, required attachment of the cells to the extracellular matrix. Taken together, our results suggest that the FAK signaling pathway play a pivotal role in the secretion of MMPs.

A Role for Focal Adhesion Kinase Signaling in Tumor Necrosis Factor-α–Dependent Matrix Metalloproteinase-9 Production in a Cholangiocarcinoma Cell Line, CCKS1

We have previously reported that tumor necrosis factor-alpha (TNF-alpha) stimulation of CCKS1, a cell line established from cholangiocarcinoma with i.p. dissemination, dramatically increased matrix metalloproteinase-9 (MMP-9) production and tumor invasion. We investigated the role of focal adhesion kinase (FAK) in TNF-alpha-dependent production of MMP-9 in CCKS1 and FAK-null mouse fibroblast cells. TNF-alpha stimulation of CCKS1 or wild-type fibroblasts substantially activated FAK phosphorylation and increased MMP-9 production. In contrast, FAK-null fibroblasts could not respond well to TNF-alpha stimulation. Conditional expression of wild-type FAK in FAK-null cells restored the TNF-alpha-dependent production of MMP-9. TNF-alpha treatment activated the kinase activity of FAK and its phosphorylation especially at Y397 and Y925. Phosphorylated FAK accumulated at focal adhesions and formed a complex with growth factor receptor binding protein 2 and SOS. In contrast, Y397F FAK and Y925F FAK, whose Y397 and Y925 were replaced with phenylalanine, respectively, as well as KD FAK, whose kinase was inactivated, could not restore the MMP-9 production. In addition, small interfering RNA against FAK drastically suppressed the TNF-alpha-dependent production of MMP-9 and inhibited the TNF-alpha-dependent invasion of CCKS1. Taken together, our results suggest the pivotal role of FAK in TNF-alpha-dependent production of MMP-9 and subsequent activation of tumor invasion.

C-RAS IS REQUIRED FOR THE ACTIVATION OF THE MATRIX METALLOPROTEINASES BY CONCANAVALIN A IN 3Y1 CELLS

Concanavalin A (Con A) is known to trigger augmented secretion and proteolytic activation of the matrix metalloproteinases (MMPs) in fibroblasts. To study the signaling pathway critical for the activation of MMPs in fibroblasts, we examined the effects of dominant negative ras (S17N ras) expression under the control of conditionally inducible promoter in Con A‐activated 3Y1 cells. We found that augmented secretion and proteolytic activation of MMP‐2 and MMP‐9 together with expression of MT1‐MMP in Con A‐activated 3Y1 were dramatically suppressed by S17N ras expression. These results strongly suggest that c‐Ras plays a critical role in the augmented expression and proteolytic activation of MMPs in fibroblasts.

Ras pathway is required for the activation of MMP-2 secretion and for the invasion of src-transformed 3Y1

To search for the signaling pathway critical for tumor invasion, we examined the effects of dominant negative ras (S17N ras) expression on the activation of matrix metalloproteinase-2 (MMP-2) in src-transformed 3Y1, SR3Y1, under the control of conditionally inducible promoter. In SR3Y1 clones transfected with S17N ras, augmented secretion and proteolytic activation of MMP-2 were dramatically suppressed by S17N Ras expression, while tyrosine phosphorylation of cellular proteins was not suppressed. We found that invasiveness of SR3Y1 cells assayed by the modified Boyden Chamber method was strongly suppressed by S17N Ras expression. In contrast, cell morphology reverted partially and glucose uptake remained unchanged by S17N Ras expression. In addition, treatment of SR3Y1 with manumycin A, a potent inhibitor of Ras farnesyltransferase, strongly suppressed both augmented secretion and proteolytic activation of MMP-2. Contrary, treatment of SR3Y1 with wortmannin or TPA showed no clear effect on MMP-2 activation. Thus, these results strongly suggest that Ras-signaling, but neither P13 kinase- nor protein kinase C-signalings, plays a critical role in activation of MMP-2 and, subsequently, in the invasiveness of src-transformed cells.

v-Src suppresses SHPS-1 expression via the Ras-MAP kinase pathway to promote the oncogenic growth of cells.

We investigated the effect of cell transformation by v-src on the expression and tyrosine phosphorylation of SHPS-1, a putative docking protein for SHP-1 and SHP-2. We found that transformation by v-src virtually inhibited the SHPS-1 expression at mRNA level. While nontransforming Src kinases including c-Src, nonmyristoylated forms of v-Src had no inhibitory effect on SHPS-1 expression, transforming Src kinases including wild-type v-Src and chimeric mutant of c-Src bearing v-Src SH3 substantially suppressed the SHPS-1 expression. In cells expressing temperature sensitive mutant of v-Src, suppression of the SHPS-1 expression was temperature-dependent. In contrast, tyrosine phosphorylation of SHPS-1 was rather activated in cells expressing c-Src or nonmyristoylated forms of v-Src. SHPS-1 expression in SR3Y1 was restored by treatment with herbimycin A, a potent inhibitor of tyrosine kinase, or by the expression of dominant negative form of Ras. Contrary, active form of Mek1 markedly suppressed SHPS-1 expression. Finally, overexpression of SHPS-1 in SR3Y1 led to the drastic reduction of anchorage independent growth of the cells. Taken together, our results suggest that the suppression of SHPS-1 expression is a pivotal event for cell transformation by v-src, and the Ras-MAP kinase cascade plays a critical role in the suppression.

A role for SHPS-1/SIRPα1 in IL-1β- and TNFα-dependent signaling

We investigated the role of SHPS-1/SIRPα1 in IL-1β- and TNFα-dependent signaling that leads to the activation of Erk 1/2 and Akt. Treatment of Balb3T3 cells with IL-1β or TNFα activated tyrosine phosphorylation of SHPS-1, its association with SHP-2 and the phosphorylation of Erk 1/2 and Akt. PP1, a specific inhibitor for the Src family protein tyrosine kinases, strongly inhibited tyrosine phosphorylation of SHPS-1 and complex formation of SHPS-1 with SHP-2 by IL-1β. In addition, PP1 substantially inhibited the IL-2β- and TNFα-dependent activation of Erk 1/2 and Akt. Exogenous expression of either SHPS-1 mutants that lack SHP-2 binding function or a dominant negative mutant of SHP-2 markedly inhibited the activation of Erk 1/2 and Akt by IL-1β, whereas wild type SHPS-1 did not. Moreover, IL-1β-stimulation induced association of SHPS-1 with IL-1RAcP, a second subunit of IL-1 receptor, whereas expression of SHPS-1 mutant that lack SHP-2 binding function clearly blocked the association and tyrosine phosphorylation of endogenous SHPS-1. Taken together, our results strongly suggest that activation of Erk 1/2 and Akt by proinflammatory cytokines requires tyrosine phosphorylation of SHPS-1 and subsequent association of SHPS-1 with SHP-2.

Cysteine residues in the C-terminal lobe of Src: their role in the suppression of the Src kinase

To evaluate the function of cysteine residues of the Src kinase, we constructed a series of Src mutants in which some of cyteines were replaced to alanines. With these mutants, we studied the effect of SH-alkylating agents, N-[p-(2-benzimidazolyl)phenyl] maleimide (BIPM) and N-(9-acridinyl) maleimide (NAM), on their kinase activity. Of 10 cysteine residues scattered over v-Src, either a single mutation at Cys520 or multiple mutations at the four clustered cyteines, Cys483, Cys487, Cys496 and Cys498, yielded clear resistance to the treatment with 10 μM BIPM or 1 μM NAM. In contrast, other cysteines including those in the SH2 domain and those in the catalytic cleft of the kinase domain were dispensable for the inactivation by BIPM and NAM. Similarly, deletion of SH2 and SH3 did not confer the resistance to v-Src, suggesting the inactivation by the SH-alkylating agents is SH2/SH3-independent. Although Cys520-mutated v-Src was resistant to 1 μM NAM, it was inactivated by 5 μM NAM. However, combined mutation including all of Cys483, Cys487, Cys496, Cys498 and Cys520 yielded clear resistance to 5 μM NAM. Among these mutants, those with double mutations in the four clustered cysteines yielded a temperature sensitive phenotype in the transfected cells, whereas Cys520 did not, suggesting that Cys520 has, at least in part, a discrete function. In contrast to v-Src, c-Src, which lacks cysteine at position 520, was resistant to 1 μM NAM but sensitive to 5 μM NAM. While replacement of Phe520 of c-Src to cysteine made it sensitive to 1 μM NAM, double mutation in clustered cysteines again yielded resistance to 5 μM NAM. Taken together, our results strongly suggest that the multiple cysteine residues clustered at the end of the C-terminal lobe are critical for the inhibition by the SH-alkylating agents and, thereby, have an allosteric repressor effect on the catalytic activity of Src in a SH2-phosphoTyr527 independent manner.

Suppression of cell spreading by v-Crk requires Ras-MEK-MAP kinase signaling

We investigated the attachment and spreading of v-Crk-transformed cells, v-Crk3Y1, on fibronectin. Transformation by v-Crk virtually suppressed the spreading, but not the attachment, of cells on fibronectin. This suppression of cell spreading was not correlated with the suppression of integrin α5 and β1 expression. However, the spreading of v-Crk3Y1 on fibronectin was dramatically restored by either expression of dominant-negative Ras or treatment with manumycin A, a Ras farnesyltransferase inhibitor. Moreover, both expression of dominant-negative MEK1 and treatment of cells with U0126, a MEK1 inhibitor, restored the cell spreading of v-Crk3Y1. In contrast, neither treatment with LY294002, a PI3K inhibitor, nor expression of dominant-negative C3G showed no effect on cell spreading on fibronectin. Taken together, our results suggest that, among multiple signaling pathways activated by v-Crk, the Ras-MEK1-MAP kinase cascade plays a pivotal role in the suppression of cell spreading on fibronectin, but C3G and the PI3 kinase do not.