Plasminogen Activator Inhibitor 1 and Vitronectin Protect Against Stenosis in a Murine Carotid Artery Ligation Model
Objective— We previously reported that plasminogen activator inhibitor 1 (PAI-1), in the presence of vitronectin (VN), inhibits thrombin activity in vitro. Furthermore, we demonstrated in human atherosclerotic plaques the colocalization of thrombin, PAI-1, and VN, as well as activity of thrombin and PAI-1. Here, we show that PAI-1 is a local thrombin inhibitor in vivo.
Methods and Results— We used the murine carotid artery ligation model to assess the role of PAI-1 and VN in stenosis by using PAI-1–deficient (PAI-1−/−) and VN−/− mice. Ligation resulted in a smooth muscle cell (SMC)-rich intima without infiltrating cells. We show that PAI-1−/− and VN−/− mice generate a larger intima than wild-type mice as the result of more extensive SMC proliferation, as evidenced by cell counting and staining for proliferating cell-nuclear antigen.
Conclusions— In PAI-1−/− mice, excessive intima formation is prevented by the thrombin-specific inhibitor hirudin. Finally, immunohistochemical analysis revealed PAI-1, VN, and (pro)thrombin antigen in intimal lesions. Our observations are compatible with inhibition of thrombin-mediated SMC proliferation by PAI-1/VN complexes.
Plasminogen activator inhibitor 1 (PAI-1) is a multifunctional protein that is best known for its regulation of plasminogen activation.1 It rapidly forms inactive complexes with the serine proteases tissue-type and urokinase-type (uPA) plasminogen activator, thereby preventing the conversion of plasminogen into plasmin. Furthermore, PAI-1 binds to vitronectin (VN), an adhesive, extracellular matrix protein. It has been shown that the binding of PAI-1 to VN affects the migration of smooth muscle cells (SMCs) by preventing the binding of VN to the αvβ3 integrin.2
See page 1943
Delineating the role of PAI-1 in atherogenesis and restenosis is particularly cumbersome, partially as a result of the different design of experimental models that often accentuate a single feature of the human pathology. For instance, comparison of mechanical and electrical injury-induced intima formation in PAI-1–deficient (−/−) and wild-type mice revealed that PAI-1 blocks intimal thickening by inhibiting the migration of SMCs.3 Moreover, PAI-1–overexpressing SMCs, seeded on a denuded carotid artery, inhibited intima formation as the result of decreased SMC migration.4 In contrast, by using a copper- or ferrichloride-induced oxidative vascular injury model, PAI-1 deficiency attenuated lesion formation.5–7⇓⇓ Contradictory observations have been noted in high-fat, macrophage-rich/SMC-poor models (apoE−/− and/or LDLR−/− mice), revealing that PAI-1 deficiency may reduce, enhance, or not influence lesion formation, possibly depending on strain differences.8–10⇓⇓ In addition, in a balloon-injured rat carotid artery model, PAI-1 overexpression resulted in increased SMC proliferation that was associated with fibrin accumulation.11 In the murine carotid artery ligation model, formation of a fibrin matrix apparently accounted for increased intima formation in uPA−/− mice, whereas PAI-1−/− mice responded like wild-type mice.12 However, in similar study that used the carotid artery ligation model, decreased intima formation was observed in VN−/− and PAI-1−/− mice.7 A satisfactory explanation for these discrepancies is presently not available, but it is possible that the different models and/or methodologies accentuate the various, distinct functions of PAI-1.
Here, we studied the role of PAI-1 as a local thrombin inhibitor in SMC proliferation by using the carotid artery ligation model, which mimics restenosis after angioplasty. In this model, the endothelium remains intact, and lesions virtually lack infiltrating cells.13 It is assumed that reduced shear stress, hypoxia, and arterial wall pressure are determinants for intima formation as a result of SMC proliferation.14 A possible function for PAI-1 in controlling SMC proliferation is suggested by the following observations and explored in this study. First, PAI-1 inhibits thrombin in vitro, provided either VN or heparin is present as a cofactor.15–17⇓⇓ Second, thrombin is a potent mitogen for cultured SMCs.18,19⇓ Accordingly, the thrombin-specific inhibitor hirudin significantly reduces intima formation in various animal models.20–22⇓⇓ Third, the presence of active thrombin, as well as active PAI-1, was reported in human atherosclerotic lesions.23,24⇓ Moreover, the latter studies revealed colocalization of thrombin, PAI-1, and VN, supporting the potential significance of PAI-1/VN complexes in regulating the mitogenic activity of thrombin.23 Importantly, the (atherosclerotic) vessel wall is virtually devoid of the traditional plasma thrombin inhibitors,25 whereas PAI-1 is synthesized by vascular cells and abundantly present in lesions.26,27⇓ Similarly, large amounts of prothrombin and VN have been encountered in the human vessel wall.25,28⇓ In addition, components of the intrinsic and extrinsic coagulation pathway are encountered in diseased vessels, suggesting that prothrombin can be activated.25 Hence, the requirements are met for controlling thrombin activity by PAI-1/VN complexes in the vessel wall.
Here, we used the carotid artery ligation model in PAI-1−/−, VN−/−, and wild-type mice to explore the effect of PAI-1 and VN on intima formation.13 We report that increased intima formation is observed in PAI-1−/− and VN−/− mice as compared with wild-type mice. Excessive intima formation in PAI-1−/− mice is inhibited by the thrombin-specific inhibitor hirudin. These results provide support for a function of PAI-1/VN complexes in thrombin-mediated SMC proliferation.
Mice and Experimental Protocol
Breeding of PAI-1−/− mice and backcrossing with wild-type C57Bl6/6J has been described.29 VN−/− mice were kindly provided by Randal Westrick and Dr. David Ginsburg (University of Michigan, Ann Arbor,).30 VN−/− mice had been backcrossed for at least 8 generations with C57Bl/6J (Charles River Laboratories Inc., Iffa-Credo, l’Arbresle, France). All mice were maintained on a standard chow diet. The mice used were 2 to 4 months old and weighed 20 to 30 g. Ligation of the left carotid artery, near the distal bifurcation, was performed as described previously.13 The mice were sacrificed either at 2.5 or 4 weeks after ligation of the carotid artery. Four to five mice were used for each time point. In a subsequent carotid artery ligation experiment, only wild-type and PAI-1−/− mice were used, which were then subjected to daily subcutaneous injections with placebo (50 mg/mL D-mannitol; Merck) or polyethylene glycol (PEG)-hirudin (1 mg/kg body weight; kindly provided by Dr. Klaus Rübsamen and Dr. Wilfried Hornberger, Knoll AG, Ludwigshafen, Germany) in 50 mg/mL D-mannitol. Seven to eight mice were used in this experiment. All mice were operated and sacrificed under ketamine/xylazine anesthesia and subsequently fixed by perfusion with 4% (v/v) paraformaldehyde in PBS. The carotid arteries were excised, immersion fixed in 70% (v/v) ethanol, and embedded in paraffin. The nonligated, right carotid artery served as internal control and was embedded together with the ligated vessel. Serial sections of 5 μm were cut for morphometric analysis. After sectioning and mounting on Superfrost Plus slides (Menzel-Gläser), the specimens were subjected to hematoxylin-eosin staining. Animal care and experimental procedures were approved by the local Ethical Committee on Animal Experiments.
The ligated carotid artery was sectioned from the ligature toward the aortic arch. A standardized reference point was set at which location the ligature did not distort the vessel and where elastic laminae were intact. The reference point is situated between 0.05 mm and 0.13 mm from the ligation. Cross sections at 0.2, 0.3, and 0.4 mm from the reference point were morphometrically analyzed by using the QWin software (Leica Microsystems) on digital images of the vessel, obtained with a Sony DXC-950 3CCD color videocamera. The circumferences of the lumen, internal elastic lamina (IEL), and external elastic lamina (EEL) were measured by tracing along the luminal surface, IEL, and EEL. The medial area (M) was calculated by subtracting the area within the IEL from the area defined by the EEL. Similarly, the intimal area (I) was calculated by subtracting the luminal area from the IEL area. The circumference of the control nonligated carotid artery is independent of the ligation time and identical between deficient and wild-type mice. To establish vascular remodeling, the total vessel area of the ligated carotid arteries (EEL of the left carotid) is expressed as percentage of the total vessel area of the nonligated artery (EEL of the right carotid).
Paraffin sections (5 μm) were deparaffinized, rehydrated, and incubated with 0.3% hydrogen peroxide and blocked with 10% (v/v) preimmune goat serum (DAKO). Subsequently, sections were incubated overnight at 4°C with primary antibodies, followed by incubation with biotinylated secondary rat antimurine light-chain antiserum or goat antirabbit antiserum, which were detected with streptavidin-horseradish peroxidase conjugates (DAKO) and, subsequently, with amino-ethylcarbazole and hydrogen peroxide. After counterstaining with hematoxylin, sections were embedded in glycergel (Sigma). Proliferating cell-nuclear antigen (PCNA) staining was performed by using an anti-PCNA monoclonal antibody (DAKO) with 0.1% (v/v) Triton X-100. Antigen retrieval was performed by boiling three times for 5 minutes at 700 Watt in a microwave in citrate buffer (pH 6). A polyclonal anti-PAI-1 antiserum was kindly provided by Dr. D.J. Loskutoff (The Scripps Research Institute, La Jolla, Calif) and used as in the PCNA-staining protocol. A polyclonal antimurine VN antiserum was obtained from Molecular Innovations Inc. A polyclonal antiserum against murine thrombin was kindly provided by Dr. J. Timmerman (this institute) and affinity-purified on a thrombin-containing column.
Thrombin Activity and Hirudin Assays
Thrombin-mediated coagulation was determined with the Pathromtin SL kit to measure the activated partial thromboplastin time (APTT) of citrated plasma samples from wild-type and PAI-1−/− mice by using the BCS Coagulation System (Dade Behring). Plasma was incubated with phospholipids and calcium to trigger coagulation, and time to clot formation was recorded. The activity of PEG-hirudin was measured by using a hirudin-activity assay in the BCS Coagulation System. Hirudin inactivates a fixed amount of added thrombin: Residual thrombin activity is determined by continuously measuring the increase in absorbance (405 nm) caused by conversion of the chromogenic substrate tosylglycyl-l-prolyl-l-arginyl-5-amino-2-nitrobenzoic acid isopropylamide. On the final experimental day, the mice received a last placebo or PEG-hirudin injection. Blood collection and subsequent sacrifice of the mice started at least 1 hour after the above-mentioned injections.
All statistical analyses were performed with the SPSS version 10.0.5 software. Experimental values are expressed as mean±standard deviation. The significance of differences in intima formation between the PAI-1−/−, VN−/−, and the corresponding wild-type groups was determined by using a nonparametric test (Mann–Whitney U 2-tailed test with a significance of P<0.05).
Vascular Remodeling in Ligated Carotid Arteries of VN−/−, PAI-1−/−, and Wild-Type Mice
The total vessel area was measured in ligated carotids at set distances from the reference point (Figure 1A). By comparing the vessel area of nonligated with the ligated carotid arteries of wild-type mice (Figure 1B and 1C), we confirmed that ligation results in narrowing of arteries (constrictive remodeling).13,14⇓ After 1 week of ligation, equal constrictive remodeling to approximately 50% of its original size was observed in the ligated carotids of wild-type, VN−/−, and PAI-1−/− mice (data not shown). Yet, after 2.5 and 4 weeks of ligation, less constrictive remodeling was observed in both VN−/− (Figure 1D) and PAI-1−/− mice (Figure 1F) as compared with the wild-type mice (Figure 1C and 1E), which is significant as shown (Figure 1G and 1H). This observation is presumably the result of the increased number of cells present in the intima of the deficient mice that counteracts the remodeling process at these latter time points. It reveals that luminal narrowing in ligated carotids is not exclusively because of constrictive remodeling. A significant part of the lumen is narrowed by a stenosis, largely composed of SMCs (Figure 1C through 1F). Consequently, the VN−/− and PAI-1−/− mice display a substantially larger intima than wild-type mice.
Intima of Ligated Carotid Arteries of VN−/−, PAI-1−/−, and Wild-Type Mice
We measured the length of the lesion by counting the number of sections containing intimal SMCs, and determined the intima/media (I/M) ratio for the deficient and wild-type mice. Clearly, the lesions of both VN−/− (Figure 2A) and PAI-1−/− mice (Figure 2B) were substantially longer than those of wild-type mice. At 2.5 and 4 weeks after ligation, the intimal lesion in VN−/− mice was approximately 1.5-fold longer than that of the corresponding wild-type mice. PAI-1−/− mice displayed a 2-fold longer lesion than their wild-type counterparts.
The I/M ratios were calculated on morphometric analysis of sections of ligated carotid arteries that were taken at a fixed distance (0.3 mm) from the reference point. For VN−/− mice, the I/M ratio after 2.5 weeks of ligation was 5-fold higher than for wild-type mice (Figure 2C). After 4 weeks of ligation, a 2.5-fold increase was observed. The wild-type mice had a similar I/M ratio after 2.5 and 4 weeks of ligation. After 2.5 weeks of ligation, the I/M ratio of PAI-1−/− mice was 5-fold higher than that of the corresponding wild-type mice (Figure 2D). After 4 weeks of ligation, a further enlargement of the intimal area caused a 7-fold increase of I/M ratio. Taken together, we conclude that after carotid artery ligation, a longer lesion and a higher I/M ratio is generated in PAI-1−/− and VN−/− mice as compared with the corresponding wild-type mice.
To determine whether larger intimal lesions are caused by SMC proliferation, we counted the nuclei present in sections that were also used for morphometric measurements. For VN−/− mice, the cell number after 2.5 weeks of ligation was 4-fold higher than for the wild-type mice (Figure 3A). After 4 weeks of ligation, a 2.5-fold increase was seen. Wild-type mice had a comparable number of cells present in the intima after 2.5 and 4 weeks of ligation. In PAI-1−/− mice, an increased cell number was seen after both 2.5 and 4 weeks of ligation (Figure 3B). After 2.5 weeks, the cell number was 9-fold higher than that of the corresponding wild-type mice. Although the lesion had increased after 4 weeks of ligation in the PAI-1−/− mice, the increase in cell number was only 6-fold compared with the wild-type mice. This is explained by the 2-fold increase in cell number of the wild-type mice. To further substantiate the increase in cell number observed and to establish whether these cells are actively engaged in proliferation, PCNA staining was performed. After 2.5 weeks of ligation, both VN−/− (Figure 3C) and PAI-1−/− mice (Figure 3D) had an increased number of PCNA-stained cells, whereas wild-type mice only showed an occasional PCNA-stained cell in their minute lesions. After 4 weeks of ligation, the PAI-1−/− mice still had an increased number of PCNA-stained cells. However, at this time point, the number of PCNA-staining cells was diminished in the VN−/− mice, being the same as for the wild-type counterparts. In conclusion, the data on cell numbers measured by two different methods agree with those recorded for I/M ratios.
Hirudin Affects Intima Formation
We measured thrombin-mediated coagulation activity in plasma as a surrogate value for prothrombin levels in the vessel wall.25 The clotting values for VN−/−, PAI-1−/−, and wild-type genotypes are not significantly different (see online Figure I, which can be accessed at http://atvb.ahajournals.org). Therefore, prothrombin availability is considered similar in all genotypes studied.
To provide evidence for a role of thrombin in intima formation in PAI-1−/− and wild-type mice, we assessed the effect of the thrombin-specific inhibitor hirudin. To that end, we injected PEG-hirudin and measured its bioactivity in plasma of placebo- and PEG-hirudin–treated mice. In PEG-hirudin–treated mice, 3% to 5% (0.3 to 0.5 μg/mL plasma) of the injected PEG-hirudin (10 μg/mL plasma) was recovered (Figure 4A). No significant difference was observed between wild-type and PAI-1−/− mice. Yet, the amount of PEG-hirudin encountered is responsible for a 2-fold longer APTT (Figure 4B).
Carotid artery ligations were performed and I/M ratios were calculated. The I/M ratio of placebo-treated PAI-1−/− mice was 2-fold higher than that of placebo-treated wild-type mice (Figure 4C). In wild-type mice, PEG-hirudin did not cause an alteration of the I/M ratio as compared with placebo-treated wild-type mice. However, in PAI-1−/− mice, PEG-hirudin significantly reduced the I/M ratio, reaching the level of placebo- or hirudin-treated wild-type mice.
Presence of VN, PAI-1, and Prothrombin
To verify the presence of the components studied, immunohistochemical detection of VN, PAI-1, and prothrombin was performed in ligated carotids. VN antigen was predominantly present in the adventitia and in the intima of wild-type and PAI-1−/− mice (Figure 5A). PAI-1 antigen was found almost exclusively in the intima of wild-type and VN−/− mice after 2.5 and 4 weeks of ligation (Figure 5B). Furthermore, PAI-1 was encountered both cell associated and in the extracellular matrix. Prothrombin, a plasma-derived protein, was unevenly distributed in the intima and media (Figure 5C). Consequently, all three proteins were present in the intima of ligated arteries, thus potentially positioned to interact.
We used the murine carotid artery ligation model to study the role of PAI-1 and VN in (re)stenosis.13 In this model, the size of the intima relies greatly on SMC proliferation because the contribution of other vascular cell types is negligible. The main conclusion on the data obtained in this study is that PAI-1 and VN prevent intima formation, exemplified by substantially longer lesions and I/M ratios in PAI-1−/− and VN−/− mice than in the corresponding wild-type mice. Although PAI-1 has been implicated in various processes relevant to atherogenesis and (re)stenosis, ranging from (indirect) control of plasmin-mediated transforming growth factor-β activation and subsequent induction of platelet-derived growth factor,31 metalloproteinase activation,32 inhibition of uPA-mediated cell migration,33 to cellular adhesion, and detachment by virtue of its overlapping VN-binding site with αv integrins and the urokinase receptor,34 our data point to a role for PAI-1, in conjunction with VN, in SMC proliferation. Significantly, excessive intima formation in PAI-1−/− mice is largely dependent on mitogenic activity of thrombin, because hirudin reduces intima formation to that of corresponding wild-type mice.
Seemingly contradictory data on the role of PAI-1 have been reported by using adenovirus-mediated overexpression of PAI-1 in a rat vascular injury model.10 Excess of PAI-1 resulted in SMC proliferation that correlated with persistent fibrin(ogen) accumulation, presumably caused by the PAI-1–mediated inhibition of fibrinolysis.11 In contrast to the rat vascular injury model, carotid artery ligation is not based on injury because the endothelial-cell layer remains intact.13 Consequently, after 1 week of ligation, a thrombus is rarely observed in any of our studied genotypes, confirming previous observations.14,35⇓ At this time point, virtually no intima has been generated. Thus, in our hands, there are no indications that thrombus formation plays a prominent role in the carotid artery ligation model. By contrast, Peng and collaborators7 recently used the carotid artery ligation model and observed decreased intima formation in PAI-1−/− mice as compared with wild-type mice. These authors assumed that thrombi had been generated and attributed a cell migration-promoting role to fibrin, which would be diminished in the absence of PAI-1, because of increased fibrinolysis. Apart from fibrin formation, an essential methodological discrepancy is that their measurements of intima size were performed at a considerably larger distance from the ligature (2 to 7 mm) than in our experiments (0.05 to 1.5 mm), apparently spanning an area that does not harbor proliferating SMC.
Hirudin-mediated reduction in intima formation is observed in PAI-1−/− mice; however, no effect of hirudin was seen on the small lesions of wild-type mice. It is conceivable that, simultaneous with thrombin, other determinants function to ultimately induce SMC proliferation (eg, stasis, hypoxia, arterial wall pressure). Such factors are insensitive to hirudin. Yet, the discrepancy might also be explained by previous observations with cultured SMCs, showing that thrombin induces PAI-1 synthesis in these cells.36 In addition, we and others15–17⇓⇓ demonstrated previously that thrombin is a target serine proteinase for PAI-1/VN complexes in vitro. Hence, in this environment, thrombin provides for control of its own activity. Obviously, this feedback loop is not existent in PAI-1−/− mice, resulting in abundant thrombin activity and, consequently, excessive SMC proliferation. Collectively, apart from other functions of PAI-1 and VN, the interaction between thrombin and PAI-1/VN complexes is conceivably relevant for the generation of SMC-rich intima.
The interaction between thrombin and PAI-1/VN complexes is characterized by a so-called suicide-substrate mechanism.17 This mechanism consists of two separate pathways for the fate of PAI-1, namely the generation of covalent thrombin/PAI-1 complexes (ie, suicide reaction) and the cleavage of PAI-1 and recovery of active thrombin (ie, substrate reaction). In the latter case, thrombin removes active PAI-1 from the vascular compartment and thereby eliminates not only its own inhibitor but also that of other target proteases, notably uPA. Because uPA is primarily responsible for the migration of SMCs,33 the ultimate effect of the substrate reaction is to remove the uPA inhibitor and, consequently, an indirect promotion of migration by thrombin. Alternatively, if the suicide reaction prevails, then PAI-1 predominantly forms covalent complexes with thrombin, resulting in inhibition of thrombin-mediated SMC proliferation. We believe that the data reported here favor the latter explanation because we observed an increased cell number and PCNA-expressing cells in the intima of PAI-1−/− and VN−/− mice.
Finally, it should be noted that the kinetics of intima formation in PAI-1−/−and VN−/− mice are not identical. This variation is not caused by differences in apoptosis because nuclear morphology and the number of active caspase 3-positive SMCs did not differ between the various genotypes (data not shown). In summary, observations with the murine carotid artery ligation model lead us to conclude that PAI-1 and VN protect against excessive SMC proliferation. Our data provide support for a mechanism by which PAI-1, presumably in conjunction with the essential cofactor VN, controls the mitogenic activity of thrombin in the vessel wall.
This study was supported by a grant (M93.007) from the Netherlands Heart Foundation. We are grateful for the excellent technical assistance of Willy M. Morriën (Department of Vascular Medicine, Academic Medical Center, Amsterdam, The Netherlands).
Received August 9, 2002; revision accepted September 24, 2002.
- ↵Van Meijer M, Pannekoek H. Structure of plasminogen activator inhibitor 1 (PAI-1) and its function in fibrinolysis: an update. Fibrinol Proteolysis. 1995; 9: 263–276.
- ↵Carmeliet P, Moons L, Lijnen R, Janssens S, Lupu F, Collen D, Gerard RD. Inhibitory role of plasminogen activator inhibitor-1 in arterial wound healing and neointima formation: a gene targeting and gene transfer study in mice. Circulation. 1997; 96: 3180–3191.
- ↵Hasenstab D, Lea H, Clowes AW. Local plasminogen activator inhibitor type 1 overexpression in rat carotid artery enhances thrombosis and endothelial regeneration while inhibiting intimal thickening. Arterioscler Thromb Vasc Biol. 2000; 20: 853–859.
- ↵Zhu Y, Farrehi PM, Fay WP. Plasminogen activator inhibitor type 1 enhances neointima formation after oxidative vascular injury in atherosclerosis-prone mice. Circulation. 2001; 103: 3105–3110.
- ↵Peng L, Bhatia N, Parker AC, Zhu Y, Fay WP. Endogenous vitronectin and plasminogen activator inhibitor-1 promote neointima formation in murine carotid arteries. Arterioscl Thromb Vasc Biol. 2002; 22: 934–939.
- ↵Eitzman DT, Westrick RJ, Xu Z, Tyson J, Ginsburg D. Plasminogen activator inhibitor-1 deficiency protects against atherosclerosis progression in the mouse carotid artery. Blood. 2000; 96: 4212–4215.
- ↵Luttun A, Lupu F, Storkebaum E, Hoylaerts MF, Moons L, Crawley J, Bono F, Poole AR, Tipping P, Herbert J-M, Collen D, Carmeliet P. Lack of plasminogen activator inhibitor-1 promotes growth and abnormal matrix remodeling of advanced atherosclerotic plaques in apolipoprotein E-deficient mice. Arterioscl Thromb Vasc Biol. 2002; 22: 499–505.
- ↵Sjoland H, Eitzman DT, Gordon D, Westrick R, Nabel EG, Ginsburg D. Atherosclerosis progression in LDL receptor-deficient and apolipoprotein E-deficient mice is independent of genetic alterations in plasminogen activator inhibitor-1. Arterioscl Thromb Vasc Biol. 2000; 20: 846–852.
- ↵DeYoung MB, Tom C, Dichek DA. Plasminogen activator inhibitor type 1 increases neointima formation in balloon-injured rat carotid arteries. Circulation. 2001; 104: 1972–1977.
- ↵Kawasaki T, Dewerchin M, Lijnen HR, Vreys I, Vermylen J, Hoylaerts MF. Mouse carotid artery ligation induces platelet-leukocyte-dependent luminal fibrin, required for neointima development. Circ Res. 2001; 88: 159–166.
- ↵Kumar A, Lindner V. Remodeling with neointima formation in the mouse carotid artery after cessation of blood flow. Arterioscler Thromb Vasc Biol. 1997; 17: 2238–2244.
- ↵Kumar A, Hoover JL, Simmons CA, Lindner V, Shebuski RJ. Remodeling and neointima formation in the carotid artery of normal and P-selectin-deficient mice. Circulation. 1997; 96: 4333–4342.
- ↵Ehrlich HJ, Gebbink RK, Keijer J, Linders M, Preissner KT, Pannekoek H. Alteration of serpin specificity by a protein cofactor: vitronectin endows plasminogen activator inhibitor 1 with thrombin inhibitory properties. J Biol Chem. 1990; 265: 13029–13035.
- ↵Naski MC, Lawrence DA, Mosher DF, Podor TJ, Ginsburg D. Kinetics of inactivation of alpha-thrombin by plasminogen activator inhibitor-1: comparison of the effects of native and urea-treated forms of vitronectin. J Biol Chem. 1993; 268: 12367–12372.
- ↵Van Meijer M, Smilde A, Tans G, Nesheim ME, Pannekoek H, Horrevoets AJ. The suicide substrate reaction between plasminogen activator inhibitor 1 and thrombin is regulated by the cofactors vitronectin and heparin. Blood. 1997; 90: 1874–1882.
- ↵Patterson C, Stouffer GA, Madamanchi N, Runge MS. New tricks for old dogs: nonthrombotic effects of thrombin in vessel wall biology. Circ Res. 2001; 88: 987–997.
- ↵Sarembock IJ, Gertz SD, Gimple LW, Owen RM, Powers ER, Roberts WC. Effectiveness of recombinant desulphatohirudin in reducing restenosis after balloon angioplasty of atherosclerotic femoral arteries in rabbits. Circulation. 1991; 84: 232–243.
- ↵Gerdes C, Faber-Steinfeld V, Yalkinoglu O, Wohlfeil S. Comparison of the effects of the thrombin inhibitor r-hirudin in four animal models of neointima formation after arterial injury. Arterioscler Thromb Vasc Biol. 1996; 16: 1306–1311.
- ↵Stoop AA, Lupu F, Pannekoek H. Colocalization of thrombin, PAI-1 and vitronectin in the atherosclerotic vessel wall: a potential regulatory mechanism of thrombin activity by PAI-1/vitronectin complexes. Arterioscler Thromb Vasc Biol. 2000; 20: 1143–1149.
- ↵Westmuckett AD, Lupu C, Goulding DA, Das S, Kakkar VV, Lupu F. In situ analysis of tissue factor dependent thrombin generation in human atherosclerotic vessels. Thromb Haemost. 2000; 84: 901–911.
- ↵Schneiderman J, Sawdey MS, Keeton MR, Bordin GM, Bernstein EF, Dilley RB, Loskutoff DJ. Increased type 1 plasminogen activator inhibitor gene expression in atherosclerotic human arteries. Proc Natl Acad Sci U S A. 1992; 89: 6998–7002.
- ↵Lupu F, Bergonzelli GE, Heim DA, Cousin E, Genton CY, Bachmann F, Kruithof EK. Localization and production of plasminogen activator inhibitor-1 in human healthy and atherosclerotic arteries. Arterioscler Thromb. 1993; 13: 1090–10100.
- ↵Zheng X, Saunders TL, Camper SA, Samuelson LC, Ginsburg D. Vitronectin is not essential for normal mammalian development and fertility. Proc Natl Acad Sci U S A. 1995; 92: 12426–12430.
- ↵He C, Wilhelm SM, Pentland AP, Marmer BL, Grant GA, Eisen AZ, Goldberg GI. Tissue cooperation in a proteolytic cascade activating human interstitial collagenase. Proc Natl Acad Sci U S A. 1989; 86: 2632–2636.
- ↵Carmeliet P, Moons L, Herbert JM, Crawley J, Lupu F, Lijnen R, Collen D. Urokinase but not tissue plasminogen activator mediates arterial neointima formation in mice. Circ Res. 1997; 81: 829–39.
- ↵Weiler H, Lindner V, Kerlin B, Isermann BH, Hendrickson SB, Cooley BC, Meh DA, Mosesson MW, Shworak NW, Post MJ, Conway EM, Ulfman LH, von Andrian UH, Weitz JI. Characterization of a mouse model for thrombomodulin deficiency. Arterioscl Thromb Vasc Biol. 2001; 21: 1531–1537.
- ↵Noda-Heiny H, Fujii S, Sobel BE. Induction of vascular smooth muscle cell expression of plasminogen activator inhibitor-1 by thrombin. Circ Res. 1993; 72: 36–43.