The Jak/Stat Pathway and Urokinase Receptor Signaling in Human Aortic Vascular Smooth Muscle Cells*

The binding of urokinase plasminogen activator (uPA) to its specific receptor (uPAR) facilitates migration of vascular smooth muscle cells (VSMC). However, the signaling cascade utilized by the urokinase receptor is only incompletely understood. We investigated intracellular uPA/uPAR signaling in human aortic VSMC from the cell membrane to the nucleus. uPA binding to VSMC induced a rapid and pronounced increase in tyrosine phosphorylation of several proteins with molecular masses of 53–60, 85–90, and 130–140 kDa. By using co-immunoprecipitation techniques and in vitro kinase assays, the uPAR-associated proteins were identified as Janus (Jak) and Src non-receptor protein-tyrosine kinases (PTK) Jak1, Tyk2, and p59 fyn , p53/56 lyn , p53/59 hck , and p55 fgr . Furthermore, uPA induced a time-dependent reversible translocation of the Stat1 (signal transducer and activator of transcription) protein to the VSMC nuclei, as shown by confocal microscopy studies. Using an electrophoretic mobility shift assay, we then demonstrated that Stat1 is rapidly activated in response to stimulation with uPA and specifically binds to the DNA regulatory elements GAS (interferon-γ activation site) and ISRE (interferon-stimulated response element). Mobility supershift experiments confirmed DNA-protein complexes containing Stat1 protein. Migration experiments with double immunofluorescence staining revealed polarization of uPAR, and colocalization with Jak1 and Tyk2 to the leading edge of the migrating cells. Under the same conditions, Jak2, Jak3, and the Src-PTKs remained randomly distributed over the entire body of the cells. Our studies therefore suggest that, in VSMC, the uPAR-signaling complex utilizes at least two different mechanisms, a direct signaling pathway utilizing the Jak/Stat cascade and a second signal transduction mechanism via Src-like protein-tyrosine kinases. uPA-induced signaling via Jak/Stat is most likely involved in the regulation of cell migration, while the functional purpose of the uPA-associated Src-PTK activation remains to be elucidated.

Vascular smooth muscle cells (VSMC) 1 are the predominant cell type in the mammalian arterial wall. In response to vascular injury such as balloon angioplasty, or in chronic pathological processes such as atherosclerosis, the differentiated VSMC dedifferentiate. The associated functional changes allow medial VSMC to migrate from the media into the intima where they proliferate and contribute to neointima formation and restenosis. To date, a number of factors which contribute to the regulation of VSMC migration and/or proliferation have been identified, including numerous growth factors and cytokines (1,2). In addition to these mechanisms, a substantial body of evidence implicates regulatory proteases, such as urokinasetype plasminogen activator (uPA), to play a decisive role in cell migration and tissue remodeling (3,4).
uPA binding to the cell surface via uPAR is a critical step in uPA function, allowing uPA to act as a membrane-associated ectoenzyme, focusing proteolysis to the immediate pericellular environment (5,6). In non-migrating cells, uPAR is distributed over the entire cell surface, while uPAR is polarized toward the leading edge of migrating monocytes (6,7), endothelial cells (8), and VSMC (9,10). Furthermore, beyond merely facilitating extracellular proteolysis, the binding of uPA to its receptor also induces an intracellular signal (7,11). Moreover, intracellular signals were also induced by stimulation with the NH 2 -terminal fragment of uPA, which retains receptor binding properties, but is devoid of enzymatic activity (12). However, because of the receptor's special topology in the cell membrane implying the extracellular localization of uPAR and its membrane attachment via a glycosyl phosphatidylinositol (GPI) anchor, the mechanisms responsible for extra-/intracellular signaling via uPAR still remain unclear (13,14). The finding that cytoplasmic protein-tyrosine kinases (PTK) are noncovalently associated with GPI-anchored proteins (15) including uPAR (12,16) has encouraged a search for probable signal transduction mechanisms by which this class of molecules induces intracellular signals.
Since uPA-induced serine phosphorylation of two cytokeratins has been observed in migrating human epithelial cells (17), Src-PTKs are not the only kinases that may be utilized in uPAR-mediated signaling. Furthermore, uPAR-related changes in diacylglycerol formation (18) and the activation of the c--fos gene (19) via an unknown signaling pathway have also been demonstrated. The important question as to whether the different functions of uPAR are mediated through different signaling mechanisms and/or by kinases in addition to Src, has not yet been addressed. One potential candidate for alternative uPAR-related signaling is the Jak/Stat pathway, a system that has recently been demonstrated to be involved in the signaling induced by a variety of growth factors, cytokines, and interferons (20 -22). An important functional role for certain Stat proteins has been demonstrated in VSMC (23) after injury.
We here report data which suggest that, in human aortic VSMC, the uPAR-signaling complex in addition to its close association with Src-PTK directly activates the Jak/Stat signaling cascade. uPA induces Jak1 and Tyk2 activation and causes a rapid translocation of Stat1 to the nucleus, where Stat1 selectively binds to the interferon-␥ activation site/interferon-stimulated response element (GAS/ISRE) DNA transcription elements. In addition, we demonstrate that in migrating VSMC uPAR is localized to the leading edge of the cells where it is colocalized with the Jak, but not with the Src kinases. These observations suggest the involvement of Jak1 and Tyk2 in the regulation of cell migration.
Cell Culture-Human aortic VSMC were obtained from Clonetics (San Diego, CA). The cells were grown in SmGM2 medium (Clonetics) supplemented with 5% fetal bovine serum and were used between passages 3 and 6. For uPA stimulation experiments, the cells were cultured for 48 h in serum-free SmGM2 medium and were then treated with uPA as described below.
Wound Assay-The wound assay was performed as described elsewhere (9). Briefly, VSMC were plated confluently on 14-mm diameter glass coverslips (Menzel-Glä ser, Braunschweig, Germany) and were placed in 35-mm tissue culture dishes (Nunc, Roskilde, Denmark). A wound width of approximately 3 mm was made using a razor blade. VSMC were allowed to recover for 4 -24 h and processed as noted below.
Confocal Microscopy-The techniques used for confocal microscopy are described in detail elsewhere (24,25). The images were acquired with NORAN Instrument Odyssey XL supported with Intervision 1.4.1 software, and with a Bio-Rad MRC 600 confocal imaging system (Bio-Rad Laboratories, Freiburg, Germany) with an argon-krypton laser. At least 20 -40 cells from at least seven separate experiments were examined under each experimental condition. The observers were unaware of the experimental design and antibodies used.
Immunoprecipitation and Immune Complex Kinase Assay-For the immunoprecipitation, cell lysates containing 400 -800 g of protein were precleared for 90 min at 4°C with Gamma-Bind Sepharose (Pharmacia Biotech) and were then immunoprecipitated overnight at 4°C by using 10 g of anti-uPAR mAb or pAb coupled to protein A-agarose (Santa Cruz Biotechnology, Inc.). Precipitates were washed twice in lysis buffer without EDTA and were used for PAGE and Western blotting as indicated above. Kinase assay was performed as described elsewhere (27), in 50 l of kinase buffer containing 5 mM p-nitrophenylphosphate and 5-10 Ci of [␥-32 P]ATP (3,000 Ci/mmol) for 5-10 min at room temperature. Phosphorylated proteins were analyzed by SDS-PAGE and autoradiography. The sequential immunoprecipitations was performed as described elsewhere (28). Nuclear Extract Preparation and Electrophoretic Mobility Shift Assay (EMSA)-Nuclear extracts were prepared from VSMC that were either left untreated or treated with 1 nM uPA. Cells were put on ice, washed with ice-cold HEPES-buffered saline containing the protease inhibitors, and harvested by scraping. After centrifugation, the cell pellet was resuspended in buffer A (10 mM HEPES, pH 7, 9, 1.5 mM MgCl 2 , 10 mM KCl, 0.5 mM DTT), containing the protease inhibitors and incubated on ice for 15 min. After homogenization in a Wheaton 0.1-ml homogenizer, the nuclei were collected by centrifugation. The pellet was resuspended in buffer B (20 mM HEPES, pH 7.9, 25% glycerol, 1.5 mM MgCl 2 , 420 mM NaCl, 0.2 mM EDTA, 0.5 mM DTT), containing protease inhibitors, and incubated on ice for 30 min, followed by centrifugation at 13,000 ϫ g (5 min, 4°C). The supernatant was dialyzed against buffer C (20 mM HEPES, 20% glycerol, 100 mM KCl, 0.2 mM EDTA, 0.5 mM DTT), containing protease inhibitors for 2 h at 4°C, followed by centrifugation by 13,000 ϫ g for 5 min at 4°C. The supernatant proteins were used immediately or aliquoted and stored at Ϫ80°C.
EMSA was performed for 30 min on ice in a volume of 20 l, containing 4 g of nuclear protein extracts, 40 ng of poly(dI-dC), 4 l of 5 ϫ binding buffer (1 ϫ binding buffer: 20 mM HEPES, pH 7.9, 50 mM KCl, 5 mM MgCl 2 , 1 mM EDTA, 1 mM DTT, 10% glycerol) with or without 20 -50 fold excess of cold competitor or of unrelated competitor, and a radiolabeled probe (3 ϫ 10 4 cpm). In supershift EMSA, protein extracts were incubated with 2 g of experimental or isotypic control antibody, prior to the addition of 32 P-labeled probe. DNA-protein complexes were separated on 5% polyacrylamide gel in Tris/glycine buffer.

RESULTS
uPA-induced Tyrosine Phosphorylation in VSMC-We first examined tyrosine phosphorylation in VSMC after stimulation with uPA indicating uPA-induced transmembrane events. Cells were preincubated with uPA for various periods of time (from 1 to 30 min), washed, scraped, lysed, and subjected to PAGE followed by Western blotting using anti-phosphotyrosine monoclonal antibodies. The results of these experiments are shown in Fig. 1a. After 5 min of uPA application, a pronounced increase in tyrosine phosphorylation of several proteins with molecular masses of ϳ38 -42, 53-60, 85-90, and 130 -140 kDa was observed (Fig. 1a, lanes 3-6). The increase in phosphorylation was rapid, peaked between 5 and 15 min, and decreased after 30 min. The control VSMC medium did not increase tyrosine phosphorylation. To exclude a potential contribution of proteolytic activity to the mechanism of tyrosine phosphorylation, diisopropyl fluorophosphate-inactivated uPA was used for cell stimulation. This material provided the same effect (data not shown), confirming the involvement of proteolytically inactive uPA and amino-terminal fragments of uPA devoid of proteolytic activity in signaling events. VSMC pretreatment for 15 min with the tyrosine kinase inhibitor genistein abolished the uPA-induced phosphorylation completely (Fig. 1a, lane 2). The 38 -42-kDa phosphorylated doublet in the VSMC lysates (Fig.  1a) may contain members of the mitogen-activated protein kinase family and the presence of 53-60-kDa phosphorylated proteins after uPA treatment suggested identity with the pre-viously described Src-PTK. Of interest were the non-identified bands of phosphorylated proteins with molecular masses 85-90 and 130 -140 kDa.
To determine whether the phosphorylated 85-90-kDa and/or 130 -140-kDa proteins were associated with uPAR, cells were lysed under mild conditions (see "Experimental Procedures"), and immunoprecipitation was performed using anti-uPAR polyclonal antibodies. SDS-PAGE-separated immunoprecipitates were immunoblotted with anti-phosphotyrosine monoclonal antibodies. A similar pattern of phosphorylated 130 -140-kDa proteins and a weak band of 85-90-kDa proteins were obtained (Fig. 1b, lane 1). The control IgG did not precipitate any of the tyrosine-phosphorylated products in this range of molecular masses (Fig. 1b, lane 2).
Identification of the Kinases Associated with uPAR in VSMC-To identify the kinases providing tyrosine phosphorylation after uPA treatment, VSMC lysates were subjected to immunoprecipitation using monoclonal anti-uPAR antibodies. The immunoprecipitate products were analyzed by SDS-PAGE and Western blotting using polyclonal antibodies against individual Jak proteins and Src-PTK. The presence of two Jak proteins, namely Jak1 and Tyk2, was clearly demonstrated in the immunoprecipitates (Fig. 2, a and b). Immunoprecipitation with these antibodies for 2 h did not show any qualitative difference in the protein pattern compared with overnight immunoprecipitation. The bands were rather weak, indicating only the presence of quantitative differences. The antibodies against two other members of Jak proteins, Jak2 and Jak3, did not show any positive reaction (not shown). The control reversed co-immunoprecipitation experiments were also performed, using the anti-Jak antibodies, and probing the blots with anti-uPAR mAb. The uPAR protein was found only in the immunoprecipitates performed with anti-Jak1 and anti-Tyk2 antibodies. The application of anti-Src-PTK antibodies revealed four bands corresponding to p59 fyn , p53/56 lyn , p55 fgr , and p55/59 hck (Fig. 2c). These findings were further confirmed by re-precipitation of in vitro phosphorylated material from uPAR immunoprecipitates by means of specific antibodies (Fig.  2d). This sequential immunoprecipitation confirmed that 130 -140-kDa proteins were Jak1 and Tyk2 (Fig. 2c, lanes 1 and 2), which were tyrosine-phosphorylated after uPA treatment and co-isolated with uPAR. The 53-60-kDa phosphoproteins coprecipitated with uPAR were re-precipitated with antibodies to p59 fyn , p53/56 lyn , p55 fgr , and p55/59 hck (Fig. 2d, lanes 3-6).
uPA-induced Nuclear Translocation of Stat1 in VSMC-To determine whether or not the activation of Jak proteins upon uPA treatment induced the activation of transcription factors, Stat1 nuclear translocation experiments were performed. Stat1 was chosen because of its recently demonstrated role in VSMC signaling and transcriptional regulation (23). For the induction and activation of Stat1 (p91/84), VSMC were treated with 1 nM uPA. The VSMC were washed, fixed, and stained before and after the uPA stimulation. Immunochemical staining with anti-Stat1 mAb revealed that Stat1 protein was localized diffusely within the cytoplasm and to a lesser extent in the nucleus of untreated VSMC (Fig. 3a). After 5 min of treatment with 1 nM uPA, anti-Stat1 staining was intensively increased in the perinuclear space. Furthermore, some Stat1 protein was also found in the nucleus (Fig. 3b). Ten and 15 min of VSMC activation resulted in predominantly nuclear staining (Fig. 3, c  and d). After 20 min of uPA treatment, VSMC immunoreactivity decreased (Fig. 3e). These experiments demonstrate that uPA activation promotes Stat1 translocation in the nucleus.
uPA Induces a Specific GAS/ISRE Binding Activity in VSMC-An adequate test for Stat activation is a DNA binding assay (EMSA) analyzing the mobility of the oligonucleotides to which activated Stat proteins bind. In our EMSA, 32 P-labeled GAS/ISRE oligonucleotide containing a consensus binding site for Stat1 was used as a probe to analyze the nuclear extracts from VSMC activated with uPA. As seen in Fig. 4, stimulation of cells with uPA led to the induction of DNA-binding protein complex within 15 min. The observed band was specific since it could be inhibited by unlabeled oligonucleotide but not by an unrelated oligonucleotide (Fig. 4, lanes 5-8). The presence of  2. Identification of uPAR-associated PTK. uPAR-signaling complex was immunoprecipitated from the cell lysates using monoclonal anti-uPAR antibody and analyzed by immunoblotting with the indicated individual anti-Jak (a and b) and anti-Src (c) polyclonal antibodies (0.5 g/ml), followed by chemiluminescent staining. d, lysates from the same cells were immunoprecipitated (IP) as above with anti-uPAR antibody, subjected to the in vitro kinase assay, and subjected to the second round of immunoprecipitation followed by 7.5% SDS-PAGE and autoradiography. The material after the first immunoprecipitation was reimmunoprecipitated (re-IP) with anti-Jak1 (lane 1), anti-Tyk2 (lane 2), anti-p59 fyn , anti-p53/56 lyn , anti-p55 fgr , and anti-p55/59 hck (lanes 3-6).
Stat1 in the observed complexes was tested by using anti-Stat1 antibody in gel retardation assays. Addition of antibody abolished the corresponding band (Fig. 4, lane 10); the effect was consistent with the characteristic of anti-Stat1 antibody used in these experiments (Santa Cruz Biotechnology commercial recommendations). The kinetics of DNA binding activity (Fig.  4, lanes 2-4) correlated very well with Stat1 nucleus translocation experiments done by confocal microscopy (Fig. 3). In both cases the effects were rather transient; the maximum of uPA-induced activation was observed within 10 -15 min and declined significantly after 20 -30 min.
Jak1 and Tyk2 Are Polarized with uPAR to the Leading Edge of Migrating VSMC-One of the most established uPAR functions is cell migration. The association between uPAR and cell migration is presumably independent of extracellular matrix proteolysis; however, the process may also involve signal transduction (10 -12). To explore the possible role of uPAR signaling on cell migration, we used a migration wound assay. VSMC migration in a wounded monolayer is shown on Fig. 5. VSMC migration in one general direction following wounding was observed by 4 h after injury and was pronounced at 24 h (Fig.   FIG. 3. uPA-  5c). In confluent VSMC, uPAR was distributed evenly over the body of the cell in a punctate pattern, with some increased staining at the periphery. However, the staining was not restricted to a specific site on the cell membrane (Fig. 6a). In response to injury, about 25% of migrating cells demonstrated a highly asymmetric distribution of uPAR along the leading edge of the cell membrane in a polarized fashion (Fig. 6b).
To obtain evidence that Jak proteins and Src-PTK exist in the same physical complex as uPAR upon cell migration, double immunolabeling experiments were performed using anti-uPAR mAb and pAb to individual kinases. Pronounced polarization and colocalization with uPAR at the leading edge of the migrating cells was found for Jak1 (Fig. 7, a-c) and Tyk2 (Fig.  7, aЈ-cЈ). Antibodies to Jak2 and Jak3 demonstrated a homogeneous punctate staining pattern, with increased staining at the sites of cell-cell contacts. However, the staining was not restricted to the leading edge of the cells either in wounded and nonwounded cultures (Fig. 8c, shown for Jak3). Strongly pronounced polarization of Jak1 and Tyk2 to the leading edge of migrating cells and colocalization with uPAR could be seen already 6 h after wounding. Control antibodies showed no staining above background autofluorescence (Fig. 7, d and e).
Finally, we performed double staining for uPAR and p59 fyn , p53/56 lyn , p55 fgr , and p55/59 hck , four Src family PTK found to be coimmunoprecipitated in our experiments with uPAR. This staining demonstrated a uniform patch-like pattern over the body of the cell, with some mildly increased staining at the periphery for all four kinases in both subconfluent and wounded confluent cultures (Fig. 8, a and aЈ, shown for p55 fgr and p55/59 hck ). uPAR displayed high polarization to the leading edge of migrating cells (Fig. 8, b and bЈ). DISCUSSION We here demonstrate that uPAR can directly activate the Jak/Stat signaling pathway in human aortic VSMC. Coimmunoprecipitation experiments showed the association of uPAR with two Janus kinases: Jak1 and Tyk2, and uPA-dependent activation of these two Jak kinases was demonstrated by in vitro kinase assay. In addition, we demonstrate that uPA induces rapid translocation of Stat1 to the nucleus, where it selectively binds to the GAS/ISRE DNA transcriptional elements. To our knowledge, these findings provide the first evidence that uPA activates a definite chain of signal transduction from the cell membrane to the nuclear transcriptional machinery.
Since in human monocytes (16) and monocytic THP-1 cells (12), the PTK of the Src family coprecipitate with uPAR, recent research on uPA/uPAR signal transduction was focused on these PTK. These earlier findings are confirmed by our study in VSMC. Using co-immunoprecipitation, a standard technique for the characterization of molecular interactions with GPI proteins, and an in vitro kinase assay, we were able to demonstrate close association of uPAR and the four members of the Src-PTK family p59 fyn , p53/56 lyn , p55 fgr , and p55/59 hck in VSMC. However, information concerning the distal components of this signaling cascade is still lacking. uPA-induced c-fos gene expression has been demonstrated in previous studies (19). This suggests c-fos as a downstream component of Src-PTK phosphorylation in response to uPA stimulation. Nonetheless, the possibility that the stimulation of c-fos expression was a result of uPA-induced activation of mitogenactivated protein kinases cannot be excluded, and the question of whether or not the 38 -42 kDa proteins, which we also found to be tyrosine-phosphorylated in response to uPA actually are mitogen-activated protein kinases, needs further study.
The recently identified Jak/Stat signal transduction pathway is utilized by cytokines and growth factors in the regulation of accelerated gene expression (20 -22). With this background, we hypothesized uPA/uPAR-associated signaling via the Jak/Stat cascade. In VSMC, uPA treatment induced a specific time-dependent increase in the tyrosine phosphorylation of 85-90-and 130 -140-kDa proteins. These proteins are of similar molecular mass as the Jak family of protein-tyrosine kinases and the Stat family of transcription factors. The latter are known to be substrates for Jak kinases in a number of cytokine receptor systems (21). Actually, immunoprecipitation experiments, combined with in vitro kinase assays, allowed us to identify two members of Janus kinases, namely Jak1 and Tyk2, to be associated with uPAR in one signal transducing complex.
The activation of Jak by cytokines or growth factors leads to the rapid phosphorylation of Stat proteins on tyrosine residues. The phosphorylated Stat proteins translocate immediately to the nucleus and bind to specific DNA sequences such as GAS, ISRE, or Sis-inducible element, and hence activate specific gene expression (20 -23). To confirm that uPA via uPAR does induce Jak/Stat signaling, we further examined uPA-induced translocation of Stat1 to the nuclei of VSMC and its binding to the DNA sequences GAS/ISRE. Stat1 was selected because of its previously shown functional importance for VSMC (23). These studies allowed us to demonstrate that uPA activation actually leads to the rapid reversible translocation of Stat1 to the nucleus. Gel shifts of VSMC nuclear fractions were performed using the GAS/ISRE consensus binding sequence for interferon-␥ activation factor. They showed the specific formation of protein-DNA complexes in response to stimulation with uPA. In addition, the kinetics of uPA-induced DNA binding activity in the cell nuclear extracts was rapid and time-dependent and had a rather transient character correlating with the kinetics of uPA-dependent Stat1 translocation into the nucleus. In electrophoretic mobility supershift experiments, anti-Stat1 antibody abolished the protein-DNA complex, thus demonstrating that these complexes did indeed contain Stat1 protein.
These results suggest a novel mechanism for uPAR-dependent signal transduction and provide further evidence that links uPAR to a dynamic intracellular event. These findings are in accordance with other studies showing that the signaling machinery induced by cytokines and growth factors may involve two or more interacting pathways, the activation of Jak/Stat system and the tyrosine phosphorylation of Src-PTK (29 -31).
To further explore the probable functions of both these signaling pathways, VSMC migration experiments were performed using a wound assay. In this assay, a highly asymmetric distribution of uPAR was found in response to wounding, localizing uPAR to the leading edge of the cells. This finding is in agreement with the observation that uPAR polarization occurs upon injury in human umbilical vein VSMC (9, 10) and in endothelial cells (8). In addition, similar polarization of uPAR has been observed in human monocytes in response to chemotactic gradients (6,7). Our experiments also demonstrated pronounced redistribution of both uPAR-associated Janus kinases, Jak1 and Tyk2, to the leading edge of migrating cells, where they colocalized with uPAR. Under the same experimental conditions, Jak2, Jak3, and the members of the Src family remained randomly distributed over the entire body of the cell.
Redistribution of the molecules participating in the regulation of cell migration to the leading edge, as compared with a random distribution, is of advantage for the migrating cell. Focal concentration of the molecules potentially increases the information content of the signal and thus accelerates its transduction. Induced proximity therefore may be regarded as a very important event in intracellular signaling (32). Redistribution of the molecules responsible for cell migration may thus be a highly conserved mechanism for maximizing signal transduction. Our data on the redistribution and colocalization of Jak1 and Tyk2 with uPAR to the leading edge of migrating cells indicate that the Jak/Stat pathway does play in important role in the uPAR-related regulation of cell migration. This notion is well in agreement with recent findings on the requirement of Jak1 kinase for cell migration and anterior specification in zebra fish embryos (33). This work also implies that, in addition to its role in signal transduction of cytokines in adult tissues, Jak1 has a role in early embryonic development and tissue remodeling. The involvement of the Jak/Stat signaling pathway in uPAR signaling thus strongly suggests a mechanism that optimizes gene induction during cell migration. Further research that targets to the identification of Stat1 transcriptionally regulated genes should therefore be able to define their potential roles in VSMC functional responses to uPA. The functional role of the Src kinases in uPAR-related signaling as well as the nature of the mechanism providing extra-/intracellular conduction of the uPAR-related signal also remains to be established. The later may include adapter molecules, co-receptor functions, and other more indirect interactions (34).