VEGF(165) requires extracellular matrix components to induce mitogenic effects and migratory response in breast cancer cells.

The expression of VEGF and the relapse-free survival rate of breast cancer patients are inversely related. While VEGF induces the proliferation and migration of vascular endothelial cells, its function in breast cancer cells is not well studied. We reported previously that fibronectin increased VEGF-dependent migration in breast cancer cells. Since VEGF has an extracellular matrix (ECM)-binding domain and possesses binding affinity for heparin, we sought to determine the effects of VEGF in breast cancer cells and the role of heparin and/or fibronectin in VEGF-induced signaling. Cells grown on plastic were compared to those grown on fibronectin or to those grown on plastic in the presence of heparin, and analysed for intracellular signaling, proliferation and migration in response to VEGF(165). Both heparin and fibronectin enhanced the binding of VEGF to T47D cells. After treatment with VEGF, [(3)H]thymidine incorporation, c-fos induction, and the number of migrating cells were significantly higher ( approximately twofold) in cells grown on fibronectin or in cells grown on plastic in the presence of heparin when compared to those grown on plastic only. Likewise, tyrosine phosphorylation of VEGF receptors, MAPK activity and PI3-kinase activity were all several-fold higher in cells seeded on fibronectin or in the presence of heparin as compared to cells exposed to VEGF alone. VEGF-dependent c-fos induction was found to be regulated through a MAPK-dependent, but PI3-kinase-independent pathway. In contrast, the migration of T47D cells in response to VEGF, in the presence of ECM, was regulated through PI3-kinase. Therefore, VEGF requires ECM components to induce a mitogenic response and cell migration in T47D breast cancer cells.

Role of vascular endothelial growth factor in the stimulation of cellular invasion and signaling of breast cancer cells.

The expression of vascular endothelial growth factor (VEGF) by breast tumors has been previously correlated with a poor prognosis in the pathogenesis of breast cancer. Furthermore, VEGF secretion is a prerequisite for tumor development. Although most of the effects of VEGF have been shown to be attributable to the stimulation of endothelial cells, we present evidence here that breast tumor cells are capable of responding to VEGF. We show that VEGF stimulation of T-47D breast cancer cells leads to changes in cellular signaling and invasion. VEGF increases the cellular invasion of T-47D breast cancer cells on Matrigel/ fibronectin-coated transwell membranes by a factor of two. Northern analysis for the expression of the known VEGF receptors shows the presence of moderate levels of Flt-1 and low levels of Flk-1/KDR mRNAs in a variety of breast cancer cell lines. T-47D breast cancer cells bind 125I-labeled VEGF with a Kd of 13 x 10(-9) M. VEGF induces the activation of the extracellular regulated kinases 1,2 as well as activation of phosphatidylinositol 3'-kinase, Akt, and Forkhead receptor L1. These findings in T-47D breast cancer cells strongly suggest an autocrine role for VEGF contributing to the tumorigenic phenotype.

Heparan sulfate chains with antimitogenic properties arise from mesangial cell-surface proteoglycans.

Heparan sulfate (HS) chains accumulate in both the medium and the cell layer of mesangial cell cultures. When given in fresh medium to quiescent cultures at naturally occurring concentrations, they suppress entry into the cell cycle and progression to DNA synthesis. We have attempted to identify the proteoglycan (PG) source of the antimitogenic HS chains from mesangial cell layers (HS(c)) and medium (HS(c)). When cells were labeled for 16 hours with [35S]sulfate, 25% of the label was found in intracellular HS chains and 5% in extracellular HSPGs. Cell-surface HSPGs accounted for the remaining 70% of the label associated with cell-layer HS and were released by either trypsin or 2% Triton X-100. About 20% of this cell-surface fraction was released by treatment with phosphatidylinositol-specific phospholipase C (PI-PLC), and probably represents glypican-like PG; glypican mRNA was present in the cells. The remainder of this fraction could be incorporated into liposomes, indicating the presence of hydrophobic transmembrane regions suggestive of syndecans. Upon purification and deglycosylation, an antiserum to rat liver HSPGs that reacts primarily with syndecan-2 showed a strong signal corresponding to this protein and three weaker bands that may represent additional syndecans. mRNAs for syndecan-1, -2, and -4 were present in the cultures. Syndecan-1 and -2 mRNAs were increased 30 minutes after stimulation of quiescent rat mesangial cells (RMCs) with serum. Heparin, HS(c), and HS(m) all prevented this increase. Syndecan-4 mRNA was not affected by serum, heparin, or HS. In pulse-chase experiments, the amount of 35S appearing in the cellular protein-free HS fraction was accounted for almost entirely by cell-surface PGs, as matrix-associated label was a minor contribution at the end of the pulse-labeling. The appearance of [35S]HS in cell extracts was unaffected by phospholipase C treatment, indicating that turnover of the newly labeled syndecan fraction is the source of the antimitogenic HS chains.

Inactivation of kinase cascades in mesangial cells grown on collagen type I.

Growth on collagen type I gels is known to suppress the mitogenic responsiveness of mesangial cells. Because these cells proliferate in some renal diseases and themselves synthesize collagen type I, we examined the influence of growth on collagen upon several kinase signaling cascades involved in mesangial cell proliferation. Quiescent mesangial cells grown on collagen type I and then stimulated with serum showed a markedly diminished induction of the protooncogene c-fos, compared with their counterparts on plastic or fibronectin. This effect was accompanied by decreased activation of mitogen-activated (Erk family) and Ca2+/calmodulin-dependent protein kinases. Cells on collagen showed lower basal protein kinase C (PKC) activity and diminished levels of PKC-alpha and -zeta isoforms. Global phosphorylation of tyrosine residues was diminished on collagen, and tyrosine phosphorylation of Erk and focal adhesion kinase in response to serum was not detected, in contrast to cells on plastic. We conclude that attachment of mesangial cells to collagen type I results in a broad suppression of protein phosphorylation that is reflected in diminished induction of the c-fos gene and probably underlies the conversion of cultured mesangial cells to a nonproliferative phenotype.

Cadmium and calcium-dependent c-fos expression in mesangial cells.

Cadmium is a carcinogenic metal known to increase the expression of several protooncogenes in a variety of cells. although the underlying mechanisms are unknown. Renal mesangial cells are smooth muscle cells in which Ca2+ signaling pathways regulate the induction of c-fos through both cAMP-dependent and mitogen-activated protein kinase- (MAPK-) dependent pathways. We report that c-fos is induced in these cells by both protein kinase C- (PKC-) dependent (phorbol ester, platelet-derived growth factor), and independent (serum, ionomycin) mechanisms. In all cases, prevention of an increase in cytosolic [Ca2+] with the chelator BAPTA prevented this induction. CdCl2 (10 microM) caused an accumulation of c-fos mRNA over 30 min that was sustained for at least 8 h. Cycloheximide inhibits turnover of c-fos mRNA and shows a synergistic effect with Cd2+ on transcript levels. Together with a similar half life of the transcript whether accumulated in response Cd2+ or induced by phorbol ester, this suggests induction of c-fos by Cd2+ rather than an effect of Cd2+ on transcript stability. Cadmium increased MAPK activity by 5 min; this was sustained for at least 8 h, consistent with the time course of c-fos mRNA accumulation. The MAPK kinase inhibitor PD98059 caused a marked decrease in the induction of c-fos by Cd2+, but did not eliminate the phenomenon completely. Although Cd2+ has been reported to activate PKC in vitro, no effect was found on PKC activity in Cd2+ -treated cells, indicating the activation of MAPK by Cd2+ is through an unidentified PKC-independent pathway. We conclude that Cd2+ can cause a sustained induction of c-fos in part through sustained activation of MAPK, that contrasts with the transient activation of these species in response to physiological mitogenic stimuli.

Heparin inhibits Ca2+/calmodulin-dependent kinase II activation and c-fos induction in mesangial cells.

Like vascular smooth-muscle cells, rat mesangial cells (RMCs) display an anti-mitogenic response to heparin. In particular, heparin partially suppresses the ability of quiescent RMCs to enter the cell cycle and induce c-fos expression. When the mitogenic stimulus is serum, phorbol ester or platelet-derived growth factor, this response appears to result from the ability of heparin to suppress activation of the extracellular-signal-regulated kinase family of mitogen-activated protein kinases. However, we have also shown that heparin suppresses c-fos expression in response to ionophores such as ionomycin, an event independent of mitogen-activated protein kinase [Miralem, Wang, Whiteside and Templeton (1996) J. Biol. Chem. 271, 17100-17106]. Here we identify this second heparin-sensitive pathway as involving Ca2+/calmodulin-dependent kinase (CaMK) II. Ionomycin (100 nM) caused a transient rise in intracellular Ca2+ concentration ([Ca2+]i) in quiescent RMCs to 386+/-55 nM, with an increase in CaMK II activity that peaked 30 s later. The accumulation of c-fos mRNA that ensued 30 min later was prevented when the increase in [Ca2+]i was prevented with the intracellular Ca2+ chelator, 1,2-bis-(2-aminophenyoxy)ethane-N,N,N',N'-tetra-acetic acid. The broad-specificity CaMK inhibitor, KT 5926, inhibited ionomycin-dependent c-fos induction at a concentration at which it was without effect on induction by serum or phorbol ester. The CaMK II-specific inhibitor, KN-93, likewise inhibited c-fos induction by ionomycin, but not by serum or phorbol ester. ML-7, an inhibitor of the CaMK-related myosin light-chain kinase (MLCK), was without effect. Heparin (1 microg/ml) suppressed ionomycin-dependent c-fos induction. It was without effect on [Ca2+]i, but inhibited the development of autonomous CaMK II activity. However, when heparin was added to the CaMK II assay solution in vitro, it was without effect on autonomous activity. Furthermore, heparin did not prevent full activation of CaMK II by the Ca2+-calmodulin complex in vitro. Heparin did not affect myosin light-chain phosphorylation or RMC contraction, processes mediated by MLCK. We conclude that ionomycin induces c-fos in RMCs through the CaMK II pathway, and that heparin prevents CaMK II activation by an indirect process mediated by other cell components. Heparin does not affect activation of the closely related CaMK, MLCK.

Chelation of intracellular calcium prevents mesangial cell proliferative responsiveness.

Mesangial cell transformation into a proliferative phenotype, observed in many glomerular diseases, occurs in response to growth factors and cytokines. This study tests the hypothesis that intracellular calcium is necessary for stimulation of mesangial cell proliferative responsiveness to a variety of growth factors. Furthermore, these experiments tested whether nonspecific calcium entry via a calcium ionophore was sufficient to elicit the same response. Rat primary mesangial cells (passages 5 to 10) were growth-arrested for 48 h in 0.5% fetal bovine serum (FBS), then stimulated with 0.1 microM endothelin-1, 1.9 microM platelet-derived growth factor (PDGF)-BB, 0.5% FBS, or 0.1 microM ionomycin, with or without the intracellular calcium chelator 1,2-bis-(2-aminophenoxy)-ethane-N,N,N',N'-tetra-acetic acid (BAPTA). Calcium signaling was measured in Fura-2-loaded cells on coverslips by dual-wavelength spectrofluorometry and in Fluo-3-loaded cells by confocal fluorescence laser microscopy. [3H]-Thymidine incorporation was measured after 12 to 24 h of stimulation with each test agent. Expression of c-fos mRNA was analyzed by Northern blot. All test agents stimulated a significant increase in cytosolic and nuclear calcium, which were both effectively inhibited with BAPTA. All agents stimulated a significant increase in [3H]-thymidine incorporation and enhanced c-fos mRNA expression (no detectable c-fos mRNA was observed in quiescent cells). BAPTA prevented the enhanced [3H]-thymidine incorporation stimulated by endothelin-1 and PDGF, and partial inhibition of FBS-stimulated incorporation with BAPTA was observed. BAPTA inhibited c-fos expression observed in response to these agents. Phorbol ester induction of c-fos mRNA in the absence of raised cytosolic or nuclear calcium was also suppressed by BAPTA. Cell viability as measured by thiazolyl blue and trypan blue was not altered by BAPTA. It is concluded that normal regulation of intracellular calcium is necessary for mesangial cell proliferative responsiveness.

Heparin inhibits mitogen-activated protein kinase-dependent and -independent c-fos induction in mesangial cells.

Heparin suppresses mitogenic responses in renal mesangial cells, and when quiescent mesangial cells are stimulated with serum, heparin blocks the induction of c-fos seen at 15 min. Because heparin is taken up by cells over a much longer time course, we addressed mechanisms whereby extracellular heparin might suppress c-fos induction at such early times. Quiescent cells were treated with serum, 12-O-tetradecanoylphorbol-13-acetate, or low concentrations of Ca2+ ionophores that produced increases in intracellular Ca2+ concentration ([Ca2+]i) in the physiological range. Each treatment caused an increase in c-fos mRNA, but they did so by different mechanisms. Serum activated mitogen-activated protein kinase (MAPK) and increased [Ca2+]i without affecting protein kinase C. Activation of protein kinase C with phorbol ester activated MAPK without much effect on [Ca2+]i. Ionophores increased [Ca2+]i without affecting basal levels of protein kinase C or MAPK. Heparin (1 microg/ml) suppressed the induction of c-fos initiated by all three treatments. It did not affect the activity of protein kinase C, but inhibited activation of MAPK by either serum or phorbol ester, suggesting a common site of action at or below the probable convergence of the induced signals at Ras/Raf-1 activation. Heparin also inhibited the serum-stimulated entry of extracellular Ca2+ to the same extent as verapamil, consistent with the ability of verapamil to block L-type Ca2+ channels and the known presence of these channels in mesangial cells. However, this effect does not appear to be related to heparin's ability to inhibit induction of c-fos. First, verapamil had no effect on induction of c-fos by serum. Second, heparin had no effect on changes in [Ca2+]i achieved by ionophores. We conclude that heparin suppresses induction of c-fos in mesangial cells by blocking at least two different points in signal transduction cascades, one upstream of MAPK and the other independent of MAPK, but dependent on intracellular Ca2+.

Collagen type I enhances endothelin-mediated contraction and induces nonproliferating phenotype in mesangial cells.

Accumulation of glomerular extracellular matrix is a characteristic accompaniment of mesangial cell proliferation in progressive renal disease. We examined how growth on several matrices affected the proliferative phenotype of cultured rat mesangial cells. Compared with growth on plastic, Matrigel, or mesangial matrix, collagen type I caused a decreased cell number at 72 h, decreased total DNA per culture, and a decrease in the incorporation of [3H]thymidine during S phase in cells released from quiescence. These antiproliferative and antimitogenic effects of collagen type I required growth on a collagen gel; soluble collagen or collagen fragments were without effect. Because a number of agents elicit both proliferative and contractile responses in mesangial cells, we examined the effect of growth on collagen on contractility. Compared with plastic, cells grown on collagen type I were more contractile, showed a higher Ca2+ signal in response to endothelin, and responded to endothelin with a more rapid myosin light-chain kinase-dependent phosphorylation of myosin light chain. We conclude that growth on a collagen type I gel uncouples contractility from a proliferative response in mesangial cells, suppressing proliferation while enhancing contraction and Ca2+ signaling in response to endothelin.

Effect of aluminum chloride, -citrate, and -maltol on the calcium-mediated degradation of neurofilament proteins.

Aluminum (Al) has been observed to cause neurofilament protein accumulation in both experimental animals and cultured cells. Impairment of axonal transport is thought to be a mechanism of toxicity. Inhibition of the degradation of neurofilament proteins, however, resulting in accumulation of these proteins may be an alternative mechanism for Al toxicity. In the present study, the effect of calcium (Ca) on the proteolysis of the neurofilament triplet proteins by calcium-activated neutral proteases (CANP) was studied in the isolated sciatic nerve explants. The extent of the degradation was found to be dependent on the Ca concentration. The effect of Al chloride, -citrate and -maltol on the calcium-induced degradation was studied. No effect of any of the Al compounds was observed, suggesting that the metal may exert its neurotoxic effect via a mechanism other than impairment of neurofilament proteolysis. Maltol itself was found to enhance the effect of Ca on the degradation of neurofilament proteins, probably by facilitating the movement of Ca across the neuronal membrane.