Urokinase-type plasminogen activator induces tyrosine phosphorylation of a 78-kDa protein in H-157 cells

G. Jayarama Bhat, Jagadambika J. Gunaje, Steven Idell

Abstract

Studies from our laboratory have shown that exposure of human lung epithelial cells to urokinase plasminogen activator (uPA) induces their proliferation. This effect of uPA is likely to occur via activation of signal transduction pathways. To elucidate uPA-induced signal transduction mechanisms, we exposed H-157 cells to uPA and determined the induced tyrosine phosphorylation profile of proteins. We demonstrate that, in these cells, uPA prominently induced tyrosine phosphorylation of a 78-kDa protein. This effect was observed as early as 30 min and was sustained for at least 24 h. Treatment of cells with agents that abrogate uPA receptor (uPAR) function, including neutralizing anti-uPAR antibody, phosphatidylinositol-specific phospholipase C, or a selective antagonist that blocks the association of uPA with uPAR (Å5 compound), all failed to prevent uPA-induced tyrosine phosphorylation. B-428, an active site inhibitor of uPA activity, prevented the uPA effect. Treatment of cells with hepatocyte growth factor, vascular endothelial growth factor, or transforming growth factor-β, all of which are known to be activated by a uPA-dependent pathway, did not stimulate tyrosine phosphorylation of the 78-kDa protein. uPA induced an increase in [3H]thymidine incorporation into DNA, and cell numbers were unaffected in the presence of Å5. These results demonstrate that, in H-157 cells, uPA induces tyrosine phosphorylation of a 78-kDa protein via a proteolysis-dependent but uPAR-independent mechanism. This novel signaling pathway represents a putative mechanism by which uPA could influence epithelial cell proliferation.

  • cell signaling
  • urokinase plasminogen activator receptor
  • tyrosine phosphorylation

the urokinase plasminogen activator (uPA) is a serine protease that plays an important role in physiological and pathological processes in which cell adhesion, migration, or tissue remodeling is required (4, 8). It converts plasminogen to plasmin and degrades fibrin and other protein components of the extracellular matrix (e.g., fibronectin). Through proteolysis, uPA also activates pro-hepatocyte growth factor (HGF; see Ref. 14) and vascular endothelial growth factor (VEGF; see Ref. 15), and, via generation of plasmin, it activates transforming growth factor-β (TGF-β; see Ref. 18).

Beyond its role as a facilitator of extracellular proteolysis and activator of growth factors, uPA is reported to generate intracellular signals via both a high-affinity receptor (uPA receptor or uPAR)-dependent (1, 2, 5-7, 13, 16, 22) or -independent (12) mechanism. The high-affinity uPAR has been cloned (17); however, recent evidence suggests that uPA may also interact with other membrane proteins to initiate signaling (9, 12). uPAR is a heavily glycosylated protein linked to the plasma membrane by a glycosylphosphatidylinositol (GPI) moiety. uPAR lacks both cytoplasmic and transmembrane domains, but binding of uPA to uPAR is known to activate several signaling intermediates. These include tyrosine phosphorylation of a 38-kDa protein in U-937 cells (6), activation of protein kinase C-ε in WISH cells (2), diacylglycerol formation in epidermal cells (5), activation of Janus kinase-signal transducers and activators of transcription (JAK-STAT) signaling in the epithelial cell line TCL-598 (13) and vascular smooth muscle cells (7), Src kinases in THP-1 and monocytic cells (1, 16), and p125 focal adhesion kinase (FAK) and p42/44 mitogen-activated protein (MAP) kinase in endothelial cells (22). These signaling intermediates, either individually or collectively, may contribute to uPA-mediated cellular responses, including cellular proliferation.

We have previously demonstrated that H-157 (a human lung epithelial cell line) and M9K (a human mesothelioma cell line) cells express high levels of uPAR and that exposure of these cells to uPA induces their proliferation (19, 20). The potent growth effects of uPA prompted us to examine the signaling pathway initiated by uPA in these cells. In this study, we demonstrate that uPA prominently induces tyrosine phosphorylation of a 78-kDa protein via a uPAR-independent signaling pathway. This requires uPA proteolytic activity and is not mediated by secondary activation of growth factors such as VEGF, HGF, and TGF-β. This novel pathway may contribute to uPA-induced cellular responses in epithelial and mesothelioma cells, including cell migration, cellular proliferation, or adhesion during tissue remodeling or metastasis.

MATERIALS AND METHODS

Materials. Tissue culture flasks, FBS, and cell culture media were purchased from Life Technologies (Gaithersburg, MD). Nitrocellulose membranes were purchased from Amersham (Arlington Heights, IL); [3H]thymidine was purchased from NEN (Boston, MA). Anti-phosphotyrosine antibody was purchased from Upstate Biotechnology (Lake Placid, NY), and goat anti-mouse IgG coupled to horseradish peroxidase was purchased from Bio-Rad (Melville, NY). The two-chain high-molecular-mass uPA was a generous gift from Dr. Jack Henkin, Abbott Laboratories (Abbott Park, IL). The low-molecular-mass uPA, amino-terminal fragment (ATF), and neutralizing anti-uPAR antibody were obtained from American Diagnostics (Greenwich, CT). HGF, VEGF, and TGF-β were purchased from R&D Systems (Minneapolis, MN); Å5 compound was a gift from Dr. A. Mazar (Ångstrom Pharmaceuticals, San Diego, CA); all other chemicals were purchased from Sigma (St. Louis, MO).

Cell culture. H-157, a cell line derived from human squamous cell lung carcinoma, cells were obtained from American Type Culture Collection (Manassas, VA); M9K mesothelioma cells were provided by Dr. Brenda Gerwin (National Cancer Institute, Bethesda, MD). Both cell types were grown in RPMI medium containing 10% FBS for 12–24 h and were serum starved for 12 h in RPMI medium before the addition of uPA or other agents.

Total cell extraction and Western blots. Cells were treated with various agents for indicated times and were washed in PBS. Cells were scraped and lysed in buffer (10 mM Tris-Cl, pH 7.4, 100 mM NaCl, 1% Triton X-100, 15% glycerol, 1 mM sodium orthovanadate, and 10 μg/ml aprotinin and leupeptin). Twenty-five micrograms of protein were run on an 8% polyacrylamide gel, transferred to nitrocellulose, and probed with anti-phosphotyrosine antibodies. Alternatively, 200 μg of the samples were immunoprecipitated with respective antibodies, and complexes were run on an 8% polyacrylamide gel and immunoblotted with anti-phosphotyrosine antibodies. Tyrosine-phosphorylated proteins were detected by enhanced chemiluminescence.

Subcellular fractionation. Cells were scraped and washed with PBS two times and swelled in hypotonic buffer (10 mM Tris-Cl, pH 7.5, 1 mM MgCl2, 10 mM KCl, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, and 3 μg/ml aprotinin). The cells were then homogenized with a Dounce homogenizer and centrifuged at 600 gfor 10 min at 4°C. The resultant supernatant was centrifuged at 100,000 g for 1 h. The pellets were referred to as membrane fraction, and the supernatant was referred to as the cytosolic fraction. The plasma membrane was isolated as previously described (11). For this, after thorough washing with PBS, the cells were homogenized in 0.3 M sucrose at 4°C in a Dounce homogenizer. After sedimentation at 1,500g for 15 min, the homogenized membranes were resuspended in 1.56 M sucrose, overlaid with 0.3 M sucrose, and centrifuged at 105,000 gfor 2 h. The plasma membrane fraction was collected from the interface and was stored at −80°C until used.

[3H]thymidine incorporation assay.

H-157 cells were plated on 24-well plates and were grown to 80% confluence (19). The cells were then serum starved for 24 h and treated with various agents and uPA for 36 h. Cells were then pulse labeled with 1 μCi/ml [3H]thymidine. The cells were then fixed with 1 ml of 15% TCA. This was followed by 10-min washes with 10% TCA, first at 4°C and then at room temperature. Cells were then dissolved in 1 N NaOH and placed in Ready-safe scintillation fluid, and radioactivity was counted using a scintillation counter. To determine cell numbers, the cells were detached and counted using a hemacytometer.

RESULTS

uPA induces tyrosine phosphorylation of a 78-kDa protein in H-157 and M9K cells. Signal transduction mechanisms via membrane receptors commonly involve protein tyrosine phosphorylation. To elucidate uPA-induced signal transduction mechanisms in H-157 cells, we treated cells with uPA (the high-molecular-mass, two-chain form) for various lengths of time (from 15 min to 24 h), and tyrosine-phosphorylated proteins were visualized in Western blots using anti-phosphotyrosine antibody. Figure1 Ademonstrates that, in these cells, uPA prominently induces tyrosine phosphorylation of a 78-kDa protein. This was detected as early as 15 min and was sustained for at least 24 h. In a few blots (2 out of 7), we observed weak tyrosine phosphorylation of two additional proteins of molecular masses 75 and 58 kDa (Fig.1 A). The reason for the weaker induction of these proteins by uPA is not clear at this stage; however, it may be related to their expression levels at the time of cellular isolation. It is also possible that differences in the levels of secretion of endogenous uPA by these cells (19) at the time of isolation may cause a variation in the ability of exogenously added uPA to induce signaling. Because tyrosine phosphorylation of the 78-kDa protein was prominently detected in all experiments, we sought to characterize the activation of this protein by uPA in the present study. We also observed that uPA induced tyrosine phosphorylation of a 78-kDa protein in the mesothelial cell line M9K (Fig.1 B). Using H-157 cells for more detailed analyses, we next characterized uPA-induced tyrosine phosphorylation of the 78-kDa protein. Figure1 C demonstrates that the effect of uPA is concentration dependent and is observed at a concentration as low as 0.1 nM. We also sought to determine if uPA-mediated tyrosine phosphorylation could be blocked by herbimycin A, a tyrosine kinase inhibitor. Figure 1 D demonstrates that herbimycin A inhibits uPA-mediated tyrosine phosphorylation. Similar results were obtained in M9K cells (data not shown). These results demonstrate that, in H-157 and M9K cells, uPA, via activation of a signal transduction pathway, induces tyrosine phosphorylation of a 78-kDa protein.

Fig. 1.

Urokinase plasminogen activator (uPA)-induced tyrosine phosphorylation of proteins in H-157 (A) and M9K (B) cells. Serum-starved cells were left untreated or were treated with uPA (10 nM) for different periods of time, and total cell extracts were prepared and immunoblotted with anti-phosphotyrosine antibody. The position of the 78-kDa protein tyrosine phosphorylated by uPA and positions of tyrosine-phosphorylated 75- and 58-kDa proteins are shown by arrows.C: concentration-dependent effect of uPA on tyrosine phosphorylation of the 78-kDa protein. Serum-starved cells were treated with different concentrations of uPA for 3 h. Total cell lysates were then prepared and immunoblotted with anti-phosphotyrosine antibody. D: effect of herbimycin A on uPA-induced tyrosine phosphorylation. Serum-starved cells were left untreated (lane 1) or were treated with uPA (10 nM;lane 2), herbimycin A (1 μM;lane 3), or first with herbimycin A (1 μM) for 30 min and then with uPA (10 nM; lane 4). After 3 h of treatment with uPA, total cell lysates were prepared and immunoblotted with anti-phosphotyrosine antibody. Blots are representative of the findings of 3 independent experiments.

The tyrosine-phosphorylated 78-kDa protein associates with intracellular membrane fractions in uPA-treated cells. To localize the 78-kDa tyrosine-phosphorylated protein in uPA-stimulated H-157 cells, we isolated particulate (membrane), cytoplasmic, and nuclear fractions. Proteins from these fractions were analyzed in Western blots by probing with anti-phosphotyrosine antibody. Figure2 A demonstrates that the 78-kDa protein is present in the particulate fraction in uPA-stimulated cells (lane 2). It is not detectable in cytoplasmic (Fig. 2 A, lane 4) or nuclear (Fig. 2 A, lane 6) fractions. Further analysis showed that the uPA-induced protein is not present in plasma membrane fractions (Fig. 2 B).

Fig. 2.

Localization of the 78-kDa protein in the particulate fraction.A: serum-starved H-157 cells were left untreated or were treated with uPA (10 nM) for 4 h. After treatment with uPA, cytoplasmic, particulate, nuclear, and plasma membrane fractions were prepared and solubilized, and proteins were immunoblotted with anti-phosphotyrosine antibody.B: as a positive control for plasma membrane preparation, we used cells treated with epidermal growth factor (EGF; 10 ng/ml) for 10 min. Experiments are representative of 4 separate experiments. EGFR, EGF receptor.

Enzymatic activity is required for uPA-mediated tyrosine phosphorylation. Depending upon the cell type analyzed, uPA-mediated signaling effects are reported to occur by both proteolysis-dependent (10) or -independent (16, 22) mechanisms. For example, it was shown in THP-1 cells that the addition of the ATF of uPA was sufficient to activate p56/p59hck kinases and promote their association with uPAR (16). Interestingly, the 33-kDa lower-molecular-mass form of uPA was ineffective. The uPA ATF does not contain enzyme activity; however, it binds to uPAR with high affinity (4). The low-molecular-mass form of uPA does not bind uPAR; however, it possesses enzyme activity. To establish if enzyme activity is required for uPA-induced tyrosine phosphorylation in H-157 cells, we tested the ability of ATF and low-molecular-mass uPA to mimic the effects of high-molecular-mass two-chain uPA. Figure 3demonstrates that ATF failed to induce tyrosine phosphorylation of the 78-kDa protein (lane 2), whereas low-molecular-mass uPA induced this phosphorylation reaction (lane 5). Preexposure of cells to ATF (which binds uPAR) did not prevent the ability of uPA to induce tyrosine phosphorylation (Fig. 3, lane 4). Figure 4 demonstrates that B-428, an active site inhibitor of uPA activity (23), completely abolished the uPA effect. These results demonstrate that uPA enzymatic activity is required for tyrosine phosphorylation of the 78-kDa protein in H-157 cells.

Fig. 3.

Effect of amino-terminal fragment (ATF) and small-molecular-mass uPA on induction of tyrosine phosphorylation. Serum-starved H-157 cells were left untreated (lane 1) or were treated with ATF (10 nM; lane 2), uPA (10 nM; lane 3), ATF (10 nM) + uPA (10 nM; lane 4), or low-molecular-mass uPA (10 nM; lane 5). ATF was added 30 min before the addition of uPA for 4 h. Cell lysates were prepared, and proteins were immunoblotted with anti-phosphotyrosine antibody. Blots are representative of 3 separate experiments.

Fig. 4.

Effect of B-428 on uPA-induced tyrosine phosphorylation. Serum-starved H-157 cells were left untreated (lane 1) or were treated with uPA (10 nM;lane 2), B-428 (10 μM;lane 3), or B-428 (10 μM) + uPA (10 nM; lane 4). After 4 h of treatment, cell lysates were prepared, and proteins were immunoblotted with anti-phosphotyrosine antibody. Blots are representative of the findings of 3 separate experiments.

uPA-induced tyrosine phosphorylation occurs via a uPAR-independent mechanism. The ability of uPA to stimulate tyrosine phosphorylation in cells preexposed to ATF (Fig. 3) suggests that the response may occur via a uPAR-independent mechanism. We used three independent experimental approaches to establish the involvement of uPAR in uPA-mediated tyrosine phosphorylation in H-157 cells. First, we exposed the cells to neutralizing anti-uPAR antibody and then sought to determine if uPA would induce tyrosine phosphorylation of the 78-kDa protein under these conditions. Figure5 Ademonstrates that the neutralizing antibody did not block the uPA effect (lanes 2 and4), suggesting that a uPAR-independent signaling pathway may be involved in the induction of tyrosine phosphorylation of the 78-kDa protein. Densitometric analysis of Fig. 5 A showed that treatment with anti-uPAR antibody slightly increased (15%) uPA-mediated tyrosine phosphorylation of the 78-kDa protein. The reason for this increased uPA-induced response in the presence of anti-uPAR antibody is not clear. However, one possibility is that binding of uPAR antibody to uPAR may increase the total availability of uPA for uPAR-independent signaling. In the second approach, we used the recently developedÅ5 compound (a uPAR antagonist) to determine if it would block uPA-induced tyrosine phosphorylation at different time points (30 min to 3 h). Å5 is an 11-mer cyclic peptide that inhibits the binding of human uPA to uPAR with an IC50 of 11 nM (Dr. A. Mazar, personal communication). Exposure of cells to Å5 failed to block the uPA effect at all of the time points examined (Fig.5 B). This also suggests that localization of uPA activity by uPAR is not required for generating the signals. In a third approach, we stripped uPAR from cells by treating with phosphatidylinositol-phospholipase C (PI-PLC) and then tested to see if uPA would stimulate tyrosine phosphorylation of the 78-kDa protein. PI-PLC cleaves GPI-linked proteins from the cell surface, including uPAR. We observed that, in PI-PLC- treated cells, uPA still induced tyrosine phosphorylation (Fig.5 C, lane 4). In these cells, PI-PLC completely removed uPAR from the cell surface (Fig. 5 D, positive control), providing further evidence that the effect is not mediated by the association of uPA with uPAR at the cell surface.

Fig. 5.

Effect of anti-uPA receptor (uPAR) antibody (A), Å5 compound (B), and phosphatidylinositol-phospholipase C (PI-PLC;C) on uPA-induced tyrosine phosphorylation. In A, H-157 cells were left untreated (lane 1) or were treated with uPA (lane 2), anti-uPAR antibody (lane 3), or first with anti-uPAR antibody (1 μg/ml) and then with uPA (lane 4). Anti-uPAR antibody was added 30 min before the addition of uPA for 4 h. Total cell lysates were prepared, and proteins were immunoblotted with anti-phosphotyrosine (anti-P-tyr) antibody.B: H-157 cells were left untreated (lane 1) or were treated with uPA for different time points (lanes 2–5) or first with Å5 for 30 min and then with uPA over the same time intervals (lanes 7–10). Lane 6represents protein from cells treated with Å5 for 3 h and 30 min (control). Total cell lysates were prepared and immunoblotted with anti-phosphotyrosine antibody. C: cells were left untreated (lane 1) or were treated with PI-PLC (PLC; 10 U/ml; lane 2), uPA (10 nM; lane 3), or first with PLC for 1 h and then with uPA (lane 4). After 4 h of treatment with uPA, proteins from the cell lysates were immunoblotted with anti-phosphotyrosine antibody. D: samples from C were immunoblotted with anti-uPAR (human) antibody. uPAR band appears heterogeneous due to heavy glycosylation (4). Blots are representative of the findings of 3 separate experiments.

uPA-induced signaling is not mediated by HGF, VEGF, or TGF-β. uPA has been reported to convert the inactive pro-HGF form to the mature, active form of HGF in MRC-5 fibroblasts (14). In CHO-K1 cells, uPA was demonstrated to convert the 50-kDa nonmitogenic form of VEGF to the mature 38-kDa mitogenic form by proteolytic action (15). In another study, it was shown that plasmin generated by uPA activity in the extracellular matrix can convert latent TGF-β to the active species in cocultures of endothelial and smooth muscle cells (18). Based upon these observations, we next considered the possibility that uPA may induce phosphorylation of the 78-kDa protein via a secondary activation of HGF, VEGF, or TGF-β in the extracellular matrix. To determine if this was the case, we tested the ability of these growth factors to stimulate tyrosine phosphorylation of the 78-kDa protein in H-157 cells. Figure6 demonstrates that only uPA, but not HGF, VEGF, or TGF-β, induces tyrosine phosphorylation in these cells. We also investigated the possibility that uPA mediates its effect through the generation of plasmin. To examine this possibility, we performed experiments to see if uPA would induce tyrosine phosphorylation of the 78-kDa protein in the presence of the protease inhibitor aprotinin. Figure 7 Ademonstrates that aprotinin did not block the ability of uPA to induce tyrosine phosphorylation. Figure 7 Bdemonstrates that exposure of cells to plasmin did not induce tyrosine phosphorylation of the 78-kDa protein (lane 3). Also, treatment of cells with another protease, namely α-thrombin, failed to mimic the uPA effect (lane 4). These observations demonstrate that tyrosine phosphorylation of the 78-kDa protein is specifically mediated by uPA and that plasmin is not involved in uPA-mediated signaling effects in H-157 cells.

Fig. 6.

Effect of transforming growth factor (TGF)-β, vascular endothelial growth factor (VEGF), and hepatocyte growth factor (HGF) on protein tyrosine phosphorylation. In A, cells were left untreated (lane 1) or were treated with uPA (lane 2) for 4 h or TGF-β (10 ng/ml; lane 3), VEGF (20 ng/ml) (lane 4), or HGF (20 ng/ml;lane 5) for 20 min. Cell lysates were prepared and immunoblotted with anti-phosphotyrosine antibody. Blots are representative of the findings of 3 separate experiments.

Fig. 7.

uPA-mediated induction of tyrosine phosphorylation is not mediated via generation of plasmin. A: H-157 cells were left untreated (lane 1) or were treated with uPA (10 nM) for 1 h (lane 2), aprotinin (20 μg/ml) for 1 h (lane 3), or aprotinin and uPA (10 ng/ml) for 1 h (lane 4). Cell lysates were prepared and immunoblotted with anti-phosphotyrosine antibody. B: serum-starved cells were left untreated or were treated with uPA (lane 2), plasmin (lane 3), or α-thrombin (lane 4). Total cell lysates were prepared and immunoblotted with anti-phosphotyrosine antibody. Blots are representative of the findings of 3 separate experiments.

Å5 compound does not block the uPA-mediated increase in [3H]thymidine incorporation and cell number.

To determine if a uPAR-independent signaling pathway in H-157 cells potentially contributes to the uPA-induced proliferative effects, we measured [3H]thymidine incorporation in DNA in the presence of uPA or Å5 plus uPA. Table 1 demonstrates that uPA induces an ∼1.6-fold stimulation of [3H]thymidine incorporation in H-157 cells. Å5 did not block the uPA-induced increase in [3H]thymidine incorporation. In addition, Å5 did not block the ability of uPA to increase the cell number (∼2-fold), suggesting that uPA-induced cellular proliferation does not require binding of uPA to uPAR.

View this table:
Table 1.

Effect of Å5 on uPA-induced uptake of [3H]thymidine and cell proliferation

A possible role for the 78-kDa protein in uPA-induced increase in cell proliferation. To establish a correlation between uPA-induced tyrosine phosphorylation of the 78-kDa protein and cell proliferation, we investigated if inhibition of tyrosine phosphorylation with herbimycin A would prevent the uPA-mediated increase in [3H]thymidine incorporation and cell number. Table 2demonstrates that herbimycin A inhibits the uPA-induced increase in DNA synthesis and cell number. Pretreatment of the cells with vanadate (a tyrosine phosphatase inhibitor) had no effect on the uPA-induced response. Pretreatment of cells with B-428 (inhibitor of uPA) completely blocked uPA-induced DNA synthesis and cell number. This suggests that uPA enzyme activity is required for uPA-induced cellular responses. Also, treatment with plasmin did not induce DNA synthesis and cell number. These results in aggregate suggest that the 78-kDa protein may be involved in uPA-mediated effects on DNA synthesis and proliferation in H-157 cells.

View this table:
Table 2.

Effect of B-428, herbimycin A, and vanadate on uPA-induced uptake of [3H]thymidine and cell proliferation

DISCUSSION

In this study, we have identified a proteolysis-dependent but uPAR-independent signaling pathway mediated by uPA in H-157 and M9K cells. Activation of this pathway involves tyrosine kinases and results in tyrosine phosphorylation of a 78-kDa protein. This protein associates with the intracellular membrane fraction in uPA-stimulated cells. Its absence in the nucleus suggests that it is not a transcription factor. Although the identity of this protein is not known at this stage, it may represent a kinase or a structural protein phosphorylated by a kinase during uPA-induced cellular signaling. This uPAR-independent signaling pathway represents a putative mechanism by which uPA could influence cellular responses in epithelial and mesothelioma cell types.

Studies performed in other cell types have demonstrated the involvement of uPAR in uPA-mediated signaling. For example, activation of FAK and MAP kinases in cultured endothelial cells is clearly a uPAR-dependent process (22). Similarly, activation of a 38-kDa protein in U-937 cells occurs via uPAR (6). In both of these cases, signaling by uPA was abolished when cells were treated with PI-PLC. Other studies also implicate uPAR in signal transduction processes. For example, one report demonstrated the association of uPAR with protein kinase C-ε and cytokeratin 8 and 18 in uPA-treated WISH cells (2). uPAR also was shown to associate with β2-integrins and members of the Src kinase family in uPA-treated human monocytes / THP-1 cells (1, 16). In aortic vascular smooth muscle cells, uPAR was shown to associate with JAK1, Tyk2, and Src tyrosine kinases (7). In the kidney epithelial tumor cell line TCL-598, an association of uPAR with the signal transducer protein gp130 and JAK1 has been demonstrated (13). In these cells, the assembly of uPAR with tyrosine/serine kinases and membrane-spanning β2-integrins in one receptor complex indicates functional cooperation between these molecules. However, it is to be noted that the uPA-induced signaling effects that have been reported are not observed in all cell lines examined and, in fact, are quite diverse (4). The reason for such differential activation of signaling intermediates in different cell types is not clear. However, this phenomenon may be related to the expression of different levels of signaling proteins in each cell type. Our current demonstration of a novel pathway of uPAR-independent signaling in H-157 and M9K cells adds further diversity to the previously described mechanisms by which uPA-mediated signaling can be accomplished.

The mechanism by which uPA induces tyrosine phosphorylation of the 78-kDa protein in H-157 and M9K cells is not known at this time. Clearly, it is not due to the secondary activation of HGF, VEGF, and TGF-β. The response is probably not mediated via generation of plasmin because exposure of cells to aprotinin did not block tyrosine phosphorylation of the 78-kDa protein. Also, treatment of cells with plasmin failed to induce phosphorylation of the 78-kDa protein. One possibility is that uPA, through proteolysis, may activate a transmembrane protein to initiate this signaling pathway in H-157 cells. In this context, it is important to note that uPA was shown to interact with a novel, high-affinity binding protein in platelet membranes (9). Another recent study indicates that the mitogenic effects of uPA on IF6 and M14 melanoma cells are independent of high-affinity binding of uPA to the uPAR (12). In that study, it was shown that blocking the proteolytic activity decreased the mitogenic effect by 30%; however, blocking the interactions of uPA with uPAR, using a specific monoclonal antibody, did not alter the mitogenic effect induced by uPA. Thus, in melanoma cells, uPA appears to bind to an unidentified membrane-associated protein to mediate signal transduction. Although a similar scenario may be operative in H-157 cells, the delineation of the exact mechanisms by which uPA initiates signal transduction in these cells requires further study.

In immunoprecipitation experiments, antibodies against protein kinase C-α, protein kinase C-δ, protein tyrosine phosphatase (PTP)-1C, PTP-1D, pp60src substrate, and phosphatidylinositol 3-kinase (p85) all failed to recognize the 78-kDa protein (unpublished results). Additional studies are required to identify this protein and to determine how it affects uPA-induced cellular responses in H-157 and M9K cells. However, we have developed evidence to suggest that uPAR-independent signaling has the potential to contribute to uPA-induced cellular responses. In support of this idea, we observed that uPA addition to H-157 cells induced an ∼1.6-fold increase in [3H]thymidine incorporation and a comparable 2-fold increase in cell number. Preexposure of cells to the uPAR antagonist Å5 did not prevent this uPA-induced response, suggesting that uPA-induced cellular proliferation does not require binding of uPA to uPAR. Inhibition of tyrosine phosphorylation of the 78-kDa protein with herbimycin A decreased the uPA-induced DNA synthesis and cell number. This observation establishes a link between activation of the 78-kDa protein and cell proliferation in uPA-treated cells. Direct confirmation of the role of this 78-kDa protein in uPA-induced cellular proliferation will require its future identification.

Acknowledgments

We thank Dr. Andrew Mazar, Ångstrom Pharmaceuticals, San Diego, CA, for providing Å5 compound; Dr. Jack Henkin, Abbott Laboratories, Abbott Park, IL, for providing the two-chain high-molecular-mass urokinase plasminogen activator; and Dr. Brenda Gerwin, National Cancer Institute, Bethesda, MD, for providing M9K cells.

Footnotes

  • Address for reprint requests and other correspondence: G. J. Bhat, Dept. of Specialty Care Services, The Univ. of Texas Health Center at Tyler, 11937 US Hwy. 271, Tyler, TX 75708 (E-mail:jbgunaje{at}uthct.edu).

  • This work was supported by National Heart, Lung, and Blood Institute Grant HL-45018 (to S. Idell) and by American Heart Association (National Center) Grant no. 96006630 (to G. J. Bhat).

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

REFERENCES

View Abstract