Atherosclerosis 159 (2001) 297– 306 www.elsevier.com/locate/atherosclerosis Urokinase plasminogen activator augments cell proliferation and neointima formation in injured arteries via proteolytic mechanisms Olga Plekhanova a, Yelena Parfyonova a,*, Robert Bibilashvily a, Sergei Domogatskii a, Victoria Stepanova a, Dietrich C. Gulba b, Alex Agrotis c, Alex Bobik c, Vsevolod Tkachuk a a b Molecular Endocrinology Laboratory, Institute of Experimental Cardiology, Cardiology Research Center, Moscow, 121552, Russia Franz-Volhard Clinic at Charite, Clinic of Humboldt Uni6ersity, Max-Delbruech Centrum for Molecular Medicine, Berlin, Germany c Cell Biology Laboratory, Baker Medical Research Institute, Alfred Hospital, Melbourne, 3181, Australia Received 18 November 2000; accepted 15 March 2001 Abstract Urokinase plasminogen activator (uPA) has been implicated in the healing responses of injured arteries, but the importance of its various properties that influence smooth muscle cell (SMC) proliferation and migration in vivo is unclear. We used three recombinant (r-) forms of uPA, which differ markedly in their proteolytic activities and abilities to bind to the uPA receptor (uPAR), to determine, which property most influences the healing responses of balloon catheter injured rat carotid arteries. After injury, uPA and uPAR expression increased markedly throughout the period when medial SMCs were rapidly proliferating and migrating to form the neointima. Perivascular application of uPA neutralizing antibodies immediately after injury attenuated the healing response, significantly reducing neointima size and neointimal SMC numbers. Perivascular application of r-uPAwt (wild type uPA) or r-uPA/GDF (r-uPA with multiple mutations in its growth factor-like domain) doubled the size of the neointima. Four days after injury these two uPAs nearly doubled neointimal and medial SMC numbers in the vessels, and induced greater reductions in lumen size than injury alone. Proteolytically inactive r-uPA/H/Q (containing glutamine rather than histidine-204 in its catalytic site) did not affect neointima or lumen size. Also, in contrast to the actions of proteolytically active uPAs, tissue plasminogen activator (tPA) did not affect the rate of neointima development. We conclude that uPA is an important factor regulating the healing responses of balloon catheter injured arteries, and its proteolytic property, which cannot be mimicked by tPA, greatly influences SMC proliferation and early neointima formation. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Vessel injury; Plasminogen activators; Proteolysis; Neointima; Proliferation 1. Introduction Percutaneous coronary artery angioplasty and intraluminal stenting are common procedures used to increase blood flow to an ischaemic myocardium. However, restenosis due to severe lumen narrowing greatly limits their usefulness, necessitating repeat interventions in up to 30% of cases [1,2]. The mechanisms responsible for this narrowing can involve vessel remodeling and/or the development of a large neointima [3]. The cellular responses to vessel injury include exces* Corresponding author. Tel.: + 7-095-4146712; fax: + 7-0954146712. E-mail address: yeparfyon@cardio.ru (Y. Parfyonova). sive smooth muscle cell (SMC) proliferation and migration, resulting in an occlusive neointima [4,5]. Both metalloproteinases [6] and the plasminogen activator system (PAS) have been implicated in these processes [7]. The PAS is composed of two activators, a tissue-type PA (tPA) and a urokinase-type PA (uPA), specific uPA cell surface receptors (uPAR) and multiple plasminogen activator inhibitors (PAIs) [8]. In uninjured arteries, tPA appears important for maintaining vessel patency, whilst a haemostatic barrier is provided by SMC derived PAI-1 [9]. After balloon catheter injury to a vessel, tPA and uPA expression by the medial SMCs is rapidly increased, with high uPA expression during 0021-9150/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 0 2 1 - 9 1 5 0 ( 0 1 ) 0 0 5 1 1 - 1 298 O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 SMC mitogenesis and tPA expression greatest during migration [10]. Despite these associations, studies with uPA and tPA knockout mice indicate that uPA rather than tPA participates in SMC migration and neointima formation after vessel injury [11]. The role of uPAR in responses to injury is less defined, despite its ability to influence the migration of cultured SMCs [12]. For example, the expected mitogenic and chemotactic effects of fibroblast growth factor-2 (FGF-2), a growth factor implicated in the early proliferative and migratory responses of SMCs to injury [13] are absent in uPAR-deficient cells, or when cultured SMCs are exposed to uPAR antibodies [14]. Similarly, the presence of the growth factor-like domain of uPA, a region that specifically interacts with uPAR is essential for its chemotactic action on cultured SMCs [15]. Other studies indicate that the ability of uPA to stimulate mitogenesis and cell migration is not dependent on plasmin generation [16], a key factor in the activation of tissue proteolysis. In this study, we determined the extent to which such uPA-dependent proteolytic and non-proteolytic mechanisms affect neointima formation after balloon catheter injury to the rat carotid artery. By examining the effects of recombinant forms of native (i.e. wild type) and mutated uPAs on SMC proliferation and the size of the developing neointima, we find that the predominant structural region of uPA responsible for regulating SMC proliferation, neointima formation and probably SMC migration is the proteolytic domain. 2. Materials and methods 2.1. Animals and experimental design Male Wistar-Kyoto rats (4– 5 months old) were obtained from a colony maintained at the Cardiology Research Center, Moscow, Russia. Their left common carotid artery was subjected to balloon catheter injury using surgical procedures approved by the Cardiology Research Center’s Animal Experimentation Committee. Initially we determined the time course of changes in uPA/uPAR expression in relation to SMC proliferation and neointima formation during the ensuing 4 days after balloon catheter injury. These were examined 6, 24, 48 and 96 h after vessel injury and compared with uninjured vessels. Assessment of the significance of uPA’s proteolytic domain and the domain, which interacts with uPAR, for SMC proliferation, migration and vessel structure after injury was investigated by applying to the adventitial side of seven to eight arteries 20 nmol/kg of either, (1) recombinant wild type uPA (r-uPAwt); (2) recombinant uPA with mutations in the growth factor-like domain (r-uPA/GFD), which prevents uPAR-dependent cultured SMC migration [15]; (3) proteolytically inactive recombinant uPA, in which glutamine replaced histidine in position-204 (r-uPA/H/Q); (4) recombinant tPA (r-tPA). We also examined the effects of applying a uPA neutralizing monoclonal IgG antibody (500 mg per vessel) or control non-specific mouse IgG (500 mg per vessel). The peptides were dissolved in saline containing 40% gel F-127 (Pluronic, BASF) [17,18]. Four days later the animals were culled and isolated arteries subjected to immunohistochemical and morphometric analyses. 2.2. Surgery Rat carotid arteries were injured with an inflated balloon catheter [17,18]. Briefly, after anesthetizing the rats with ketamine (100 mg/kg body wt., i.p., Pfaffen – Schwabenheim), a midline incision was made in the neck to expose the left external carotid artery. A 2 F Fogarty arterial embolectomy catheter (Baxter Healthcare) was introduced into this artery through an arteriotomy and passed into the common carotid artery to the aortic arch. The balloon was inflated and then slowly rotated while pulling the catheter back towards the external carotid artery. This was repeated three times and then the external carotid artery was ligated. The contralateral right carotid artery as well as uninjured left carotid arteries from sham-operated rats served as controls. To peri-adventitially administer various agents, the arteries were gently dissected free of their surrounding connective tissue and then 0.5 ml of the desired Pluronic solution was placed around the vessel as described earlier [17,18]. The incisions were closed and the animals were allowed to recover. 2.3. Tissue collection and processing The animals were deeply anesthetized with sodium pentobarbital (100 mg/kg body wt.; Sanofi, Sante Animale), and Evans blue (60 mg/kg body wt.; IV) administered, so that removal of endothelium in the damaged vessels could be confirmed. Then the animals were perfused (120 mmHg) with saline solution, followed by 4% formaldehyde solution for 10 min [18,19]. Left and right common carotid arteries were removed, cleaned of extraneous material and cut into three equal segments before embedding in paraffin. Cross-sections (5 mm for immunohistochemistry and 10 mm for morphometry) were cut from each block, at 100– 200 mm intervals. The vessels for mRNA analyses were obtained from animals perfused only with saline. These were placed in ice cold saline and stripped of adventitia before storing at − 72°C. O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 2.4. Morphometry Morphometry was carried out as described earlier [18,19]. Briefly, cross-sectional areas of the lumen, neointima, media and area, encompassed by the external elastic laminae (EEL), of formaldehyde fixed toluidine blue stained sections were measured by a blinded histologist, using a Zeiss microscope coupled to a ProgRess-3008 camera (Kontron Elektronik) and computer with an ‘Optimas’ morphometric program (Optimas Corporation). Morphometry was performed on multiple sections of the distal and proximal segments of the arteries, and the data for each artery averaged. 2.5. Immunohistochemistry Sections were deparaffinized with xylene, rehydrated and treated with 3% hydrogen peroxide to quench endogenous peroxidase. Serial sections were used to localize uPA and uPAR immunoreactive peptides [17,19]. The sections were incubated in 10% serum (ICN) from the same species as the secondary biotinylated antibodies, and then with either the anti-uPA monoclonal antibody (10 mg/ml), the anti-uPAR polyclonal antibody (10 mg/ml), the anti-PCNA monoclonal antibody (4.8 mg/ml), the anti-alpha smooth muscle actin monoclonal antibody (3.6 mg/ml) or the appropriate control non-immune IgGs (either mouse or rabbit) in concentrations, coinciding with those of each immune IgG, for 1 h in a humidified chamber. After multiple washings in physiological buffered saline (PBS, pH 7.4), the sections were incubated with a biotinylated anti-rabbit or anti-mouse antibody (15 mg/ml). Antigens were detected using the ABC method (Vector laboratories Inc.) and the chromogen 3,3%-diaminobenzidine tetrahydrochloride, before lightly staining the sections with haematoxylin. 2.6. RT-PCR analysis uPA and uPAR mRNA in vessels were determined using a semi-quantitative reverse-transcriptase polymerase chain reaction (RT-PCR) procedure as described earlier [17,19,20]. Oligonucleotide primer pairs for detection by PCR of uPA and uPAR cDNA fragments were designed to the following criteria, GC content 45–55%, melting point (m.p.) 76– 83.5°C, filtering hairpins and 3%-homologies. The sequences for the uPA and uPAR primers were based on GenBank accession nos. X65651 and X71899, respectively. The ‘uPA primers’ were sense: 5%-TCGTGAATCAGCCAAAGAAGGAAGAGTACG-3% (bp 680–709 in the cDNA) and antisense: 5%-CAACTGACATTTTCAGGTTC-3% (bp 993–1012 in the cDNA), to theoretically amplify a 333 bp cDNA fragment. The ‘uPAR primers’ were sense: 5%-CAGAACACTGTATTGAAGTG- 299 GTGACGCTCC-3% (bp 431–460 in the cDNA), antisense: 5%-TCCAAGCACTGATTCATTGGTCCCCG-3% (bp 712– 737 in the cDNA) to amplify a 307 bp cDNA fragment. Oligonucleotide primers for L7 were those earlier used by us [17]. Amplification parameters for semi-quantitative PCR were 94°C for 30 s, 60°C for 1 min and 72°C for 2 min [17,19]. The optimum number of cycles for uPA was 28 and 26 for uPAR. These numbers were chosen to ensure that the product amplification-RNA relationship was always in the log-linear range [17]. Confirmation of identical amounts of RNA for each RT-PCR was carried out by amplifying in 23 PCR cycles the L7 cDNA fragment; the L7 cDNA encodes a non-inducible cell cycle-independent ribosomal protein [17]. PCR products were electrophoresed in 3% agarose gels containing ethidium bromide, photographed under ultraviolet light using positive/negative film (Polaroid 665), and then the intensities of the bands were quantitated using laser densitometry (LKB 2202 Ultrascan Bromma, LKB). Those for uPA and uPAR were normalized to the L7 band, and the results expressed as fold increases above values from uninjured vessels. 2.7. Recombinant mutated uPAs, antibodies and reagents A recombinant wild type human uPA (r-uPAwt) and a recombinant uPA form with a mutated growth factorlike domain (r-uPA/GFD) were produced as described earlier [15,20– 22]. Recombinant-uPA/GFD differs from r-uPAwt in the first 24 amino acids, SNELHQVPSNCDCLNGGTCVSNKY, of the N-terminus, which were replaced with ITPSLHACRSTLD [15]. Recombinant-uPA/GFD does not stimulate cultured rat aortic SMCs to migrate, contrasting with the stimulatory activity of r-uPAwt [15]. After activation with plasmin the proteolytic activities of r-uPAwt and r-uPA/GFD ranged from 1 to 1.2× 105 U/mg protein [15]. A proteolytically inactive uPA, r-uPA/H/Q was prepared by mutating His-204 within the catalytic center to Gln [21,22]. Recombinant-uPA/H/Q was mapped with antibodies for different uPA epitopes. It did not possess proteolytic activity, measured using S2444 (Chromogenix). The ability of r-uPA/H/Q to bind to human uPAR is identical to r-uPAwt and it was as effective as r-uPAwt in displacing 125I-r-uPAwt bound to cultured rat aortic medial SMC (IC50’s20 nM) [21]. All the recombinant forms of uPA appeared as single proteins on SDS-electrophoresis, with apparent molecular weights ranging from 40 to 43 kDa. Their purity was greater than 95%. An anti-uPA monoclonal IgG1 antibody was prepared by immunizing mice with human urokinase, purified from urine. The antibody has a high affinity for human and rat uPA and recognized all forms of uPA. 300 O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 It is capable of neutralizing human recombinant uPA activity (1 mg/ml neutralizes 0.8 nM uPA). In rat tissues, it detects uPA with an apparent molecular weight of 48 kDa. Non-specific total mouse IgG from pooled serum (Inpharm Inc, Russia) was used as control for peri-adventitial anti-uPA antibody. An anti-uPA monoclonal mouse IgG (Inpharm Inc, Russia) raised against human uPA was used for immunohistochemistry. It also detects uPA in rat tissues. A specific uPAR rabbit polyclonal IgG antibody that detects glycosylated forms of human uPAR (39– 66 kDa) [23] and rat uPAR (35– 58 kDa, Stepanova V., unpublished), was purchased from American Diagnostica Inc. A specific PCNA mouse monoclonal IgG antibody (PC 10) was purchased from Santa Cruz Biotechnology. The anti-a-smooth muscle actin (a-SM actin) monoclonal antibody was from Dako. Control nonimmune rabbit and mouse IgGs, a biotinylated horse anti-mouse rat adsorbed IgG antibody and a biotinylated goat anti-rabbit IgG were purchased from Vector Laboratories. Recombinant tissue-type plasminogen activator (tPA) was from Boehringer Ingelheim Pharma KG. 2.8. Assessment of efficacy of peri6ascular administration The efficacy of perivascular delivery of the uPA neutralizing antibody was assessed using a rat-adsorbed biotinylated horse anti-mouse antibody, a polyclonal horse anti-goat antibody as negative control and an avidin –biotin immunoperoxidase kit. The delivery of control non-specific total mouse IgG was also assessed. The efficacy of perivascular delivery of the recombinant forms of uPA was assessed 2 and 4 days after their application to the arteries, using r-uPA forms conjugated to biotin. Recombinant-uPAs applied to the vessels were later detected using (a) Western blots and NeutrAvidin (Pierce) and (b) immunohistochemistry with the Vector Laboratories’ streptavidin-peroxidase kit. All substances were present in the media and adventitia of the arteries 2 days after their application, but by 4 days they were no longer detectable (not shown). 2.9. Data analyses The intensities of immunohistochemical stains detecting uPA and uPAR were graded in a semi-quantitative blinded manner by two histologists, using a scale of 0 –4, with ‘0’ indicating no staining (background); ‘+ 1’, low and variable staining in a specific region; ‘ +2’, weak staining; ‘ +3’, consistent positive staining; ‘+ 4’, intense staining. The neointima (region between the endothelium and the internal elastic laminae) and media (region between the internal and external elastic laminae) were assessed in this manner (three sections per rat) [24,25]. PCNA positive cells and total cell numbers (i.e. hematoxylin stained nuclei in the media and neointima) were determined by counting (three sections per rat). A PCNA labeling index was calculated using the equation, PCNA labeling index= [PCNA −positive cells per three cross sections/total cells per three cross sections]× 100. The percentage of a-SM actin positive cells in the neointima was determined in the same manner. All results are means9 standard error of mean (S.E.M.). Comparisons between multiple groups were performed using the one-way analysis of variance (ANOVA) and the Student–Newman – Keuls test for multiple comparisons. Single comparisons were made using Student’s t-test. A value of PB 0.05 was considered statistically significant. All statistical analyses were performed using ‘Jandel SigmaStat’. 3. Results 3.1. uPA and uPAR expression after experimental angioplasty Urokinase PA immunoreactive peptides were readily detected in the media of uninjured carotid arteries and localized predominantly to SMCs (Fig. 1). About 48 h after injury, uPA was elevated nearly two-fold (Fig. 2, top panel) and distributed uniformly throughout the media (Fig. 1). This elevation became apparent as early as 6 h after injury and persisted for at least 96 h (Figs. 1 and 2, top panel), i.e. throughout the time when SMCs proliferate rapidly in the media and migrate to form the neointima [10]. At 96 h uPA was also highly expressed in the developing neointima (Figs. 1 and 2, top panel). There were also associated increases in uPA mRNA, indicated by the greater abundance of RTPCR transcripts for uPA (333 bp fragment), compared with those obtained from uninjured vessels, 2 and 4 days after the injury (Fig. 3). Increases in uPAR immunoreactive peptides were also apparent after injury (Fig. 1). They were more gradual than for uPA (Fig. 2, bottom panel), with peak levels being reached at 48 h nearly three times higher than those in the uninjured vessels (Fig. 2, bottom panel). Immunohistochemistry on serial sections indicated a similar distribution pattern for uPAR and uPA in the media and in the neointima (Figs. 1 and 2, bottom panel). The injured vessels also contained higher levels of uPAR mRNA than uninjured vessels, detected by RT-PCR as a 307 bp cDNA fragment, 2 and 4 days after injury (Fig. 3). At 4 days 979 1% of the cells in the media and 9792% in the developing neointima expressed a-SM actin. O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 3.2. Urokinase PAs and early neointima de6elopment Since uPA and its receptor are expressed early after injury, when the neointima commences to develop and the media is being repaired, we next investigated, (1) how elevations in uPA affect neointima formation and vessel structure; (2) the importance of uPA’s receptor binding property and (3) the importance of its proteolytic property in modulating vessel structure. Periadventitial application of r-uPAwt doubled the size of the neointima in the carotid arteries 4 days after injury (PB 0.05 from control; Fig. 4, top panel). The size of the lumen was reduced (P B0.05 from control; Fig. 4, middle panel). There was also a tendency for the size of the media to increase (P\ 0.05; Fig. 4, lower panel). Periadventitial application of r-uPA/GFD also increased neointima size (P B0.05), to the same level as achieved with r-uPAwt (Fig. 4, top panel). Recombi- 301 nant-uPA/GFD reduced lumen size after injury (P B 0.05 from injured control vessels; Fig. 4, middle panels). Here, there was also a tendency for medial cross sectional area to be increased, although not statistically significant (P \0.05; Fig. 4, lower panel). In contrast, peri-adventitial application of r-uPA/H/Q, which lacks proteolytic activity, did not increase neointima size after balloon catheter injury (P for difference \0.05 from vehicle treated vessels; Fig. 4, top panel). Recombinant-uPA/H/Q also had no effect on lumen or media size (P\ 0.05; Fig. 4, middle and lower panels). Periadventitial application of the uPA neutralizing antibody attenuated neointima size 4 day after vessel injury, by about 35% (P B0.05; Fig. 4, top panel). Although there was a tendency for lumen size to be reduced, as well as for medial cross-sectional area, these effects were not statistically significant (P \0.05; Fig. 4, middle and lower panels). Peri-adventitial application Fig. 1. Top panels: expression of uPA and uPAR immunoreactive peptides in serial sections of uninjured carotid arteries; controls shown represent non-immune rabbit IgG controls for the uPAR polyclonal antibody; identical results (not shown) were obtained with the control mouse IgG for the uPA monoclonal antibody. Center panels: expression 48 h after balloon catheter injury. Bottom panels: expression 96 h after injury. Immunoreactive peptides are represented by the brown coloration. L represents the lumen. Elastic laminae are stained blue, cell nuclei are stained with haematoxylin. Scale bar 50 mm. Magnifications × 450. 302 O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 sis [14,15], we compared the effects of the recombinant forms of uPA on SMC accumulation in the neointima. As reported earlier [26], the number of SMCs in the neointima increased significantly 4 days after injury (PB 0.05). Perivascular r-uPAwt doubled the number of SMCs in the developing neointima (P B0.05; Fig. 5, top panel). The number of SMCs in the healing media also increased by nearly 50% (P B0.05; Fig. 5, bottom panel). The increases in neointima SMC numbers were also observed after r-uPA/GFD (P B0.05; Fig. 5, top panel). However, the small increase ( 25%) in SMC numbers in the media was not statistically significant (P \0.05; Fig. 5; bottom panel). Since r-uPA/GFD Fig. 2. Time course of changes in the expression of uPA and uPAR during the 96 h after balloon catheter injury to the rat carotid artery. Control represents values obtained in uninjured arteries. The intensity of staining was graded from 0 to +4 as described in Section 2 and the results for each animal averaged. Results are the means 9 S.E.M. of four animals in each group. * PB 0.05 from control. of control non-specific mouse IgG did not induce changes in vessel structure compared with vehicletreated injured arteries (not shown). To determine whether the proteolytic dependent effects of uPA on vessel structure were specific to uPA, we examined the effects of tPA. Four days after its application to the balloon catheter injured arteries, the size of the neointima was identical to that observed in vehicle-treated injured arteries (tPA: 0.01390.001 mm2; vehicle: 0.01490.001 mm2; P\ 0.05). 3.3. Urokinase PAs and neointimal SMC numbers Since uPA stimulates cultured SMCs to proliferate and migrate by mechanisms not dependent on proteoly- Fig. 3. Panel A, typical agarose gels demonstrating the effects of injury to the carotid artery on mRNAs encoding uPA, uPAR and L7. Top bands represent uPA RT-PCR transcripts obtained from the RNA of injured (I) left carotid and uninjured (U) right carotid arteries of the same animals 48 and 96 h after the balloon catheter injury. Middle and bottom bands, RT-PCR transcripts of uPAR and L7, obtained from the same RNA. Panel B, magnitude of the changes in uPA and uPAR mRNAs observed 48 and 96 h after injury. Results are the average from four experiments,* PB 0.05. O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 303 mouse IgG did not affect neointimal nor medial SMC numbers compared with vehicle-treated arteries (data not shown). 3.4. uPAs and neointima SMC cycle status Both proliferating and non proliferating SMCs are known to migrate from the injured media to form the neointima [27]. To determine whether uPA increased the number of neointimal and medial cells, entering the cell cycle, we compared the proportion of PCNA positive SMC in the neointima and media of uPA and vehicle treated arteries 4 days after injury. Urokinase PA increased the frequency of PCNA expressing SMCs in the neointima from 26.791.4 to 36.99 2.3% (P B 0.05), while PCNA labeling index in the media was not affected compared with vehicle-treated injured arteries (P \0.05). 4. Discussion Fig. 4. Bar graphs summarizing the effects of perivascularly applied r-uPAwt, r-uPA/GFD, r-uPA/H/Q and uPA neutralizing antibody on neointima cross sectional area (upper panel), lumen area (middle panel) and media area (bottom panel) 4 days after injury to carotid arteries. Control represents vessels, which treated only with Pluronic gel/saline. All injured vessels were compared with their corresponding right uninjured carotid artery in the same animals and the differences calculated. Results are the mean 9 S.E.M. of six to seven animals in each group. * P B 0.05 from control. does not stimulate SMC migration in vitro [15], its ability to increase neointima SMC numbers appears mostly due to increases in SMC proliferation. In contrast, perivascular r-uPA/H/Q did not increase neointimal SMC numbers; rather there tended to be a small but not statistically significant decrease (P\ 0.05; Fig. 5, top panel). It also did not affect SMC numbers in the media (P \0.05 from control; Fig. 5, lower panel). Perivascular application of the uPA neutralizing antibody reduced neointimal SMC numbers by approximately 45% (P B0.05; Fig. 5, top panel), without affecting SMC numbers in the media (P\ 0.05; Fig. 5, bottom panel). Peri-adventitial control non-specific It is well accepted that the uPA system can regulate vascular SMC migration as well as participate in neointima growth in injured vessels and the development of various human vascular disorders [10,11,15,16,18,28– 30]. Although cell culture studies indicate that uPA stimulates SMC DNA biosynthesis and migration by mechanisms apparently not dependent on proteolysis or uPAR occupancy [21,31– 33], in balloon catheter injured arteries the proteolytic property of uPA, independent of the interaction with uPAR, is the prime mechanism by which uPA enhances neointima growth. Enhanced neointimal SMCs proliferation appears mostly responsible for the increase in neointima size; effects due to an increase in SMC migratory activity appear to be less evident. This proteolytic-dependent effect of uPA on the size of the neointima of the injured arteries could not be induced by tPA. Two lines of evidence indicate that the proteolytic property of uPA alone, despite the early expression of uPAR in the balloon catheter injured arteries, appears to be the major determinant influencing SMC proliferation and possibly migration. First, proteolytically inactive r-uPA/H/Q did not increase early neointima formation, contrasting markedly with the effects of proteolytically active recombinant uPAs. Second, proteolytically active r-uPA/GFD, which does not stimulate SMC migration in vitro [15], is similar to r-uPAwt in its ability to augment neointima size and neointimal SMC numbers. Differences in environmental conditions of cultured SMCs and SMCs in the injured vessels could account for the different responses of SMCs in arteries and cell culture to the recombinant forms of uPA [31,32]. However, the probable explanation is that the proteolytic property of uPA predominantly regulates 304 O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 SMC proliferation in vivo. Our conclusion on the significance of interactions of uPA with uPAR for SMC proliferation, early neointima formation and possibly migration are consistent with recent observations on the healing responses of injured arteries in uPAR knockout mice [12]. Whilst uPAR can be involved in SMC migration, in its absence other non-uPAR dependent mechanism appears to take over [34]. Whether this latter mechanism is the consequence of uPA interacting with yet to be characterized receptors in the vascula- ture, such as those reported on platelets [35] and rat parenchymal liver cells [36] is not known. Correlations have been reported between the early expression of uPA and SMC proliferation and tPA and SMC migration in balloon catheter injured arteries [10]. Also, other studies in knockout mice [11] and primate arterial tissue [37] suggest that uPA participates in neointima development. Our study indicates that uPA markedly enhances neointimal SMCs proliferation. Exogenous r-uPAwt considerably enhanced cell prolifera- Fig. 5. Bar graphs summarizing the effects of r-uPAwt, r-uPA/GFD, r-uPA/H/Q and a uPA neutralizing antibody on the number of smooth muscle cells in the neointima (top panel) and the media (bottom panel) of carotid arteries 4 days after injury. Control represents injured vessels to which only Pluronic solution was applied. Results are the mean 9 S.E.M. of six to seven animals in each group. * PB 0.05 from control. O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 tion and neointima formation, whilst uPA neutralizing antibody attenuated neointima formation in the injured arteries. In uPA-knockout mice, the reduction in neointimal SMC numbers after vessel injury was similar to those reported here (45%) [11]. Our finding that the elevation in uPA is a significant contributor to the early SMC proliferation and neointima formation might be related with the demonstrated role for FGF-2 in the healing of mechanically injured vessels. FGF-2 is rapidly released from SMCs in arteries following balloon catheter injury [13,38]. Also, in cultured SMCs FGF-2 elevates uPA levels four to five-fold [14]. Urokinase PA also appears necessary for FGF-2 to exert its maximal mitogenic and chemotactic effects on vascular smooth muscle [14]. We have already shown that the ability of PDGF-BB and FGF-2 to stimulate cultured SMC to migrate is dependent on uPA [15]. Proteolysis initiated via uPA in injured arteries has the potential to stimulate SMC proliferation and migration by a number of highly coordinated mechanisms. Rapid up-regulation of metalloproteinases (MMPs) occurs after balloon catheter injury to the vessel wall, coinciding with high rates of medial SMC proliferation, and rapid SMC migration into the intima [39]. The conversion of pro-MMPs to active MMPs is dependent on urokinase-generated plasmin activities, which in turn degrade extracellular matrix (ECM) proteins including collagens [29]. Matrix bound growth factors, such as FGF-2 and latent transforming growth factor-beta (TGF-b) are released during ECM degradation [40,41], and these have the potential to further augment SMC proliferation and cell migration. FGF-2 is a potent mitogen in injured vessels [13,14], and TGF-b1 also augments neointima formation after balloon catheter injury to carotid arteries [42]. Whilst these effects might be considered largely due to actions on SMCs, adventitial fibroblasts can also migrate through arterial layers after balloon catheter injury and contribute to neointima growth [43]. It is possible that similar actions on adventitial fibroblasts might also contribute to uPA’s ability to augment neointima growth. To summarize, we have shown that the proteolytic activity of uPA is the major contributor to its ability to augment neointima formation early after balloon catheter injury to the rat carotid artery. Our experiments suggest that this property of uPA, which can not be mimicked by tPA, contributes to neointima growth, SMC accumulation and proliferation in the neointima of healing arteries after balloon catheter injury. Urokinase PA may be a specific functional target for attenuating lesion growth. Acknowledgements Dr Ye. Parfyonova and Dr O. Plekhanova have 305 participated in the Russian/Australian Exchange Program in Medicine and Public Health. These studies have in part been funded by the National Health and Medical Research Council (NHMRC) of Australia and Russian Basic Research Foundation, grant no. 99-0448772. The authors are grateful for assistance with research of Svetlana Mukhina, Alexey Poliakov, Marina Solomatina and Sergei Ragozin. References [1] Libby P, Schwartz D, Brogi E, Tanaka H, Clinton SK. A cascade model for restenosis. A special case of atherosclerotic lesion progression. Circulation 1992;86(Suppl. III):47 –52. [2] Schwartz RS. The vessel wall reaction in restenosis. Semin Intervent Cardiol 1997;2:83 – 8. [3] Schwartz RS. Pathophysiology of restenosis: interaction of thrombosis, hyperplasia, and/or remodeling. Am J Cardiol 1998;81(7A):14E– 17. [4] Schwartz SM. Smooth muscle migration in atherosclerosis and restenosis. J Clin Invest 1997;99:2814 – 7. [5] Clowes AW, Reidy MA. Prevention of stenosis after vascular reconstruction: pharmacological control of intima hyperplasia — a review. J Vasc Surg 1991;13:885 – 9. [6] Zempo N, Koyama N, Kenagy RD, Lea HJ, Clowes AW. Regulation of vascular smooth muscle cell migration and proliferation in vitro and in injured rat arteries by a synthetic matrix metalloproteinase inhibitor. Arterioscler Thromb Vasc Biol 1996;16:28 – 33. [7] Reidy MA, Irvin C, Lindner V. Migration of arterial wall cells. Expression of plasminogen activators and inhibitors in injured rat arteries. Circ Res 1996;87:405 – 14. [8] Tkachuk V, Stepanova V, Little PJ, Bobik A. Regulator and role of urokinase plasminogen activator in vascular remodelling. Clin Exp Pharmacol Physiol 1996;23:759 – 65. [9] Loskutoff DJ, van Aken BE, Seiffert D. Abnormalities in the fibrinolytic system of the vascular wall associated with atherosclerosis. Ann NY Acad Sci 1995;748:177 – 83. [10] Clowes AW, Clowes MM, Au YP, Reidy MA, Belin D. Smooth muscle cells express urokinase during mitogenesis and tissue-type plasminogen activator during migration in injured rat carotid artery. Circ Res 1990;67:61 –7. [11] Carmeliet P, Moons L, Herbert J-M, Crawley J, Lupu F, Lijnen R, Collen D. Urokinase but not tissue plasminogen activator mediates arterial neointima formation in mice. Circ Res 1997;81(5):829 –39. [12] Carmeliet P, Moons L, Dewerchin M, Mackman N, Luther T, Breier G, Ploplis V, Muller M, Nagy A, Plow E, Gerard R, Edgington T, Risau W, Collen D. Insights in vessel development and vascular disorders using targeted inactivation and transfer of vascular endothelial growth factor, the tissue receptor, and the plasminogen system. Ann NY Acad Sci 1997;811:191 – 206. [13] Lindner V, Reidy MA. Expression of basic fibroblast growth factor and its receptor by smooth muscle cells and endothelium in injured rat arteries. Circ Res 1993;73:589 –95. [14] Herbert JM, Lamarche I, Carmeliet P. Urokinase and tissue-type plasminogen activator are required for the mitogenic and chemotactic effects of bovine fibroblast growth factor and plateletderived growth factor-BB for vascular smooth muscle cells. J Biol Chem 1997;272:23585 – 91. [15] Stepanova V, Bobik A, Bibilashvily R, Belogurov A, Rybalkin I, Domagatsky S, Little PJ, Goncharova E, Tkachuk V. Urokinase 306 [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] O. Plekhano6a et al. / Atherosclerosis 159 (2001) 297–306 plasminogen activator induces smooth muscle cell migration: key role of growth factor-like domain. FEBS Lett 1997;414:471 – 4. Okada SS, Grobmyer SR, Barnathan ES. Contrasting effects of plasminogen activators, urokinase receptor, and LDL receptorrelated protein on smooth muscle cell migration and invasion. Arterioscler Thromb Vasc Biol 1996;16:1269 – 76. Ward MR, Agrotis A, Kanellakis P, Dilley R, Jennings G, Bobik A. Inhibition of protein tyrosine kinases attenuates increases in expression of transforming growth factor b isoforms and their receptors following arterial injury. Arterioscler Thromb Vasc Biol 1997;17:2461 – 70. Plekhanova O, Parfyonova Ye, Bibilashvily R, Stepanova V, Bobik A, Tkachuk V. Urokinase plasminogen activator enhances intima and media growth and reduces lumen size in carotid arteries. J Hypert 2000;18(8):1065 – 9. Wong J, Rauhoft C, Dilley RJ, Agrotis A, Jennings GL, Bobik A. Angiotensin-converting enzyme inhibition abolishes medial smooth muscle PDGF-AB biosynthesis and attenuates proliferation in injured carotid arteries: relationships to neointima formation. Circulation 1997;96:1631 – 40. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate – phenol– chloroform extraction. Anal Biochem 1987;162:152 – 9. Poliakov AA, Mukhina SA, Traktouev DO, Bibilashvily RS, Gursky YG, Minashkin MM, Stepanova VV, Tkachuk VA. Chemotactic effect of urokinase plasminogen activator: a major role for mechanisms independent of its proteolytic or growth factor domains. J Recept Signal Transduct Res 1999;19(6):939 – 51. Mukhina S, Stepanova V, Traktouev D, Poliakov A, Bibilashvilly R, Gursky Y, Minashkin M, Shevelev A, Tkachuk V. The chemotactic action of urokinase on smooth muscle cells is dependent on its kringle domain. Characterization of interactions and contribution to chemotaxis. J Biol Chem 2000;275:16450– 8. Stahl A, Mueller BM. Binding of urokinase to its receptor promotes migration and invasion of human melanoma cells in vitro. Cancer Res 1994;54(11):3066 –71. Giddings JC, Banning AP, Ralis H. Redistriburion of von Willebrand factor in porcine carotid arteries after balloon angioplasty. Arterioscler Thromb Vasc Biol 1997;17:1872 –8. Galis ZS, Sukhova GK. Increased expression of matrix metalloproteinases and matrix degrading activity in vulnerable regions of human atherosclerotic plaques. J Clin Invest 1994;94:2493 – 503. Clowes AW, Reidy MA, Clowes MM. Kinetics of cellular proliferation after arterial injury. I: Smooth muscle growth in the absence of endothelium. Lab Invest 1983;49:327 – 33. Clowes AW, Schwartz SM. Significance of quiescent smooth muscle migration in the injured rat carotid artery. Circ Res 1985;56:139 – 45. Padro T, Emeis JJ, Steins M, Schmid KW, Kienast J. Quantification of plasminogen activators and their inhibitors in the aortic vessel wall in relation to the presence and severity of atherosclerotic disease. Arterioscler Thromb Vasc Biol 1995;15:893 – 902. Carmeliet P, Moons L, Lijnen R, Baes M, Lemaitre V, Tipping P, Drew A, Eeckhout Y, Shapiro S, Lupu F, Collen D. Urokinase-generated plasmin activates matrix metalloproteinases during aneurysm formation. Nat Genet 1997;17:439 – 44. [30] Estreicher A, Mullhauser JL, Carpentier L, Orci L, Vassalli JD. The receptor for urokinase type plasminogen activator polarizes expression of the protease to the leading edge of migrating monocytes and promotes degradation of enzyme inhibitor complexes. J Cell Biol 1990;111:783 – 92. [31] Tang H, Kerins DM, Hao Q, Inagami T, Vaughan DE. The urokinase-type plasminogen activator receptor mediates tyrosine phosphorylation of focal adhesion proteins and activation of mitogen-activated protein kinase in cultured endothelial cells. J Biol Chem 1998;273:18268 – 72. [32] Kanse SM, Benzakour O, Kanthou C, Kost C, Lijnen HR, Preissner KT. Induction of vascular SMC proliferation by urokinase indicates a novel mechanism of action in vasoproliferative disorders. Arterioscler Thromb Vasc Biol 1997;17:2848 – 54. [33] Stepanova V, Mukhina S, Kohler E, Resink TJ, Erne P, Tkachuk VA. Urokinase plasminogen activator induces human smooth muscle cell migration and proliferation via distinct receptor-dependent and proteolysis-dependent mechanisms. Mol Cell Biochem 1999;195(1 – 2):199– 206. [34] Yu W, Kim J, Ossowski L. Reduction in surface urokinase receptor forces malignant cells into a protracted state of dormancy. J Cell Biol 1997;137:767 – 77. [35] Jiang Y, Pannell R, Liu J-N, Gurewich V. Evidence for a novel binding protein to urokinase-type plasminogen activator in platelet membranes. Blood 1996;87:2775 – 81. [36] van der Kaaden ME, Rijken DC, Groeneveld E, van Berkel TJC. Native and non-glycosylated recombinant single-chain urokinase-type plasminogen activator are recognized by different receptor systems on rat parenchymal liver cells. Thromb Haemost 1995;74:722 – 9. [37] Kenagy RD, Vergel S, Mattsson E, Bendeck M, Reidy MA, Clowes AW. The role of plasminogen, plasminogen activators, and matrix metalloproteinases in primate arterial smooth muscle cell migration. Arterioscler Thromb Vasc Biol 1996;16(11):1373 – 82. [38] Jackson CL, Reidy MA. Basic fibroblast growth factor: its role in the control of smooth muscle cell migration. Am J Pathol 1993;143:1024 –31. [39] Jenkins GM, Crow MT, Bilato C, Gluzband Y, Ryu W-S, Li Z, Stetler-Stevenson W, Nater C, Froehlich JP, Lakatta EG, Cheng L. Increased expression of membrane-type matrix metalloproteinase and preferential localization of matrix metalloproteinase-2 to the neointima of balloon-injured rat carotid arteries. Circulation 1998;97:82 –90. [40] Wells JM, Strikland S. Regulated localization confers multiple functions on the protease plasminogen activator. J Cell Physiol 1997;171:217– 25. [41] Agrotis A, Bobik A. Vascular remodelling and molecular biology: new concepts and therapeutic possibilities. Clin Exp Pharmacol Physiol 1996;23(5):363 – 8. [42] Smith JD, Bryant SR, Couper LL, Vary CPH, Gotwals PJ, Koteliansky VE, Lindner V. Soluble transforming growth factorb type II receptor inhibits negative remodeling, fibroblast transdifferentiation and intimal lesion formation but not endothelial growth. Circ Res 1999;84:1212 – 22. [43] Li G, Chen SJ, Oparil S, Chen YF, Thompson JA. Direct in vivo evidence demonstrating neointimal migration of adventitial fibroblasts after balloon injury of rat carotid arteries. Circulation 2000;101(12):1362– 5.