Increased Plasmin and Serine Proteinase Activity During Flow-Induced Intimal Atrophy in Baboon PTFE Grafts
High blood flow causes intimal atrophy and loss of extracellular matrix in PTFE aortoiliac grafts. We have investigated whether matrix-degrading proteinases are altered in this baboon model of atrophy using zymography, western analysis, and a versican degradation assay. After four days of high flow, urokinase was increased and plasminogen activator inhibitor-1 was decreased in the intima. Plasminogen was increased after seven days. Pro-matrix metalloproteinase (MMP)-2, activated MMP-2, and proMMP-9 levels were modestly increased by high flow at 7 days, whereas MMP-3 and tissue inhibitor of metalloproteinases-1 were not altered. Extracts of 4-day high-flow intimas degraded more 35S-methionine–labeled versican than low-flow intimal extracts, and this activity was inhibited by AEBSF, a serine proteinase inhibitor, and a plasmin antibody. In contrast, this activity was not inhibited by the MMP inhibitor, BB-94 (Batimastat). These data suggest that serine proteinases, including plasmin, may be largely responsible for extracellular matrix degradation in this primate model of flow-induced intimal atrophy.
In response to changes in blood flow and shear stress, arteries adjust lumen size acutely by changes in vasomotor tone and chronically by remodeling the structure of the intima and media.1,2⇓ In rigid prosthetic grafts or stented arteries, which cannot dilate or contract, blood flow alters the size of the lumen by inducing growth or regression of the neointima.3,4⇓ PTFE grafts used to replace segments of iliac artery in baboons under high-flow conditions (created by placement of a femoral arteriovenous fistula) develop less neointima than those under normal flow.5 When normal flow is returned to a graft that has been allowed to heal under high-flow conditions, message levels of platelet-derived growth factor (PDGF)-A increase, intimal smooth muscle cells (SMCs) begin to proliferate, and the neointima thickens.6 The SMCs also synthesize extracellular matrix so that the volume of matrix per cell is similar between high- and normal-flow conditions (approximately 65% of intimal volume).7 Conversely, increasing the blood flow to a graft that has healed under normal flow induces atrophy of the neointimal layer within 2 months.8 Because atrophy involves the loss of extracellular matrix as well as cell death, we have investigated the regulation of matrix-degrading proteinases. We demonstrate a large increase of serine proteases, specifically urokinase and plasmin.
Sixteen male baboons (Papio cynocephalus anubis) received bilateral PTFE aortoiliac bypass grafts as described.8 After eight weeks, a unilateral arteriovenous fistula was created downstream between the common femoral artery and vein. Midstream graft blood velocities, measured with duplex ultrasonography as previously described,7 were increased acutely 2.6-fold. Animals were euthanized 4, 7, or 14 days later. The grafts of one 4-day animal were occluded with thrombosis and not used further. Graft intimas were either fixed in buffered 10% formalin or frozen for extraction. Final numbers of intimal extracts used for this study were 5, 5, and 4 for days 4, 7, and 14, respectively.
Zymography and Western Analysis
Frozen portions of the neointima were pulverized, extracted (0.05 mol/L Tris, pH 7.5, 0.01 mol/L CaCl2, 2.0 mol/L guanidine HCl, and 0.2% Triton X-100), dialyzed (0.1 mol/L Tris, 5 mmol/L EDTA, 0.2% Triton X-100, pH 7.4), and 10 μg of protein (bicinchoninic acid method with a bovine albumin standard; Pierce) were loaded per lane for SDS-PAGE (10%) and Western blotting. Gelatin, casein, and casein/plasminogen zymography was performed as described.9,10⇓ Mouse monoclonal antibodies used were: plasminogen (1 μg/mL; NeoMarkers), α2-antiplasmin (2 μg/mL; American Diagnostica), and plasminogen activator inhibitor type 1 (PAI-1; 1 μg/mL; American Diagnostica). Rabbit antibodies to tissue inhibitor of metalloproteinases (TIMP)-1 and TIMP-2 were kindly provided by Dr. Howard Welgus, Parke Davis, Ann Arbor, Mich. Blots were developed using chemiluminescent reagents (Pharmacia). Western blots and zymograms were quantified as described.9
Human SMCs were labeled with 40 μCi 35S-methionine/mL for 24 hours, and versican was isolated from the conditioned medium by diethylaminoethyl cellulose and Sepharose CL-2B chromatography as described.11–13⇓⇓ Ten micrograms of intimal extract protein (in 25 μL of 100 mmol/L Tris, 10 mmol/L EDTA, 0.2% Triton X-100, pH 7.4) was added to 30,000 dpm of 35S-versican with or without various proteinase inhibitors (in 25 μL of 50 mmol/L Tris, 25 mmol/L CaCl2, pH 7.4). The anti-plasminogen antibody that blocks activity (2D1.2.2; a gift from Lindsey Miles, La Jolla, Calif) and the control immunoglobulin G1 were used at 450 μg/mL. The samples were incubated at 37°C for 20 to 24 hours and quick frozen. Chondroitin sulfate (5 μg) was added to samples with 4 volumes of 1.3% potassium acetate in 95% ethanol and left at −70°C for ≥3 hours. Samples were centrifuged, and the precipitate was treated with chondroitin ABC lyase (ICN Biomedicals; 0.03 U/mL in 50 mmol/L Tris, 3 mmol/L Na acetate, 0.1 mg/mL bovine serum albumin, pH 8, for 3 hours at 37°C) and run on a 4% to 12% gradient SDS-polyacryamide gel. Dried gels were exposed to photographic film or PhosphorImager and quantitation done using ImageQuant (Molecular Dynamics).
Data are expressed as the mean±SEM of the indicated number of intimal extracts from individual animals. Statistical differences between high- and normal-flow groups were tested with the Wilcoxon signed rank test (SPSS 8.0 for Windows).
Matrix Degrading Proteinases
To determine if matrix proteases in intimal extracts were affected by flow, we investigated the expression of the plasminogen activators and matrix metalloproteinases (MMP), which have been shown to play major roles in tissue involution and remodeling.14–16⇓⇓ Urokinase (48-kd band on casein zymography; inset Figure 1A) was increased by high flow at day 4 (Figure 1A) and at 14 days, although the latter was not statistically significant. Plasminogen was observed as 85-kd and 76-kd bands (inset Figure 1B), which are consistent with Glu- and Lys-plasminogen.17 There was no significant effect of flow on the proportion of these two species (data not shown), and the total of Glu- plus Lys-plasminogen was increased at day 7 by high flow (Figure 1B).
Gelatin zymography demonstrated the presence of MMP-2 (68-kd proform and 60-kd activated form18) and MMP-9 (100-kd proform and 89-kd activated form18) in the graft intima (inset Figure 1C). High flow increased proMMP-2 and proMMP9 at day 7 by 23% and 44%, respectively (Figure 1C). Although activated MMP-2 was also slightly increased at day 7 (1.21±0.08 high-to-normal flow ratio, P=0.043, n=5), active MMP-9 was not (0.66±0.22, 1.31±0.22, and 1.00±0.22 high-to-normal flow ratio for days 4, 7, and 14, respectively; P>0.20, n=4 to 5). MMP-3, which was observed as a 50-kd band in casein zymograms (Figure 1D) and non-reducing Western blots (data not shown), was not significantly changed by flow. The caseinolytic bands between 70 and 100 kd are plasmin, based on inhibition by AEBSF (data not shown), molecular weight, and response to high flow.
We also investigated some inhibitors of the plasminogen activators and the MMPs. PAI-1 was observed as doublets at ≈44 kd and ≈57 kd and variably observed as a doublet at ≈102 kd (inset Figure 2A; the ≈68-kd band is nonspecific). These represent intact PAI-1 (≈57 kd), PAI-1 clipped by plasminogen activators (≈44 kd),19,20⇓ and a plasminogen activator–PAI-1 complex (≈100 kd).21 The ≈44-kd and 57-kd doublets were decreased by high flow by ≈50% at 4 and 14 days (Figure 2A). This trend was also apparent for the ≈102 kd doublet (0.47±0.17, 2.81±0.98, and 0.36±0.02 for the high-to-normal flow ratio for days 4, 7, and 14, respectively; P=0.138, 0.138, and 0.068, respectively; n=4 to 5). The inhibitor of MMPs, TIMP-1, was mostly present as 4 to 5 bands between ≈120 kd and ≈230 kd with much less of the 28 kd free form (inset, Figure 2B). Neither the complexes nor the free TIMP-1 (Figure 2B) were significantly altered by high flow. No specific staining for TIMP-2 was observed (data not shown). Finally, the plasmin inhibitor α2-antiplasmin was observed as three bands between ≈50 kd and 68 kd (inset, Figure 2C); the 68-kd form represents intact α2-antiplasmin, and the others are partially cleaved forms.22 There was no statistically significant difference seen between normal- and high-flow samples in the total amount of all forms of α2-antiplasmin (Figure 2C) or in the amount of cleaved forms (data not presented).
35S-Versican Degradation In Vitro
Neointimal extracts were examined for proteolytic activity by incubation with 35S-versican, because versican is a major vascular proteoglycan present in abundance in the graft intima. Versican core protein prepared from human arterial SMC cultures runs on SDS-polyacrylamide electrophoresis as two major bands, which we have identified as the V0 and V1 isoforms.23 Extracts of 4-day high-flow intimas consistently degraded the V0 and V1 core proteins to a greater degree (Figure 3A and inset) than did the normal-flow intimal extracts. Extracts of 7-day high-flow intimas demonstrated variable activity compared with normal-flow extracts (two of four extracts had increased activity). Extracts of 14-day high-flow intimas consistently demonstrated higher activity than normal-flow extracts, but the results were not statistically significant. The versican-degrading activity in the day 4 high-flow extracts was completely inhibited by the serine proteinase inhibitor AEBSF, while EDTA had some inhibitory effect (Figure 3B). However, BB-94 at concentrations known to inhibit all MMPs and RG101625 (a MMP inhibitor specific for MMP-7) did not prevent versican degradation (Figure 3B). In addition, TIMP-1 and TIMP-2 at 750 nmol/L had little or no effect (data not presented). The aspartate and cysteine proteinase inhibitors, pepstatin A and E-64, also had no effect (Figure 3B). Because the serine proteinase zymogen plasminogen was detected by Western analysis in extracts, a plasmin-blocking antibody was tested and found to significantly inhibit the versican-degrading activity (Figure 3C).
We have presented evidence consistent with the idea that increased proteolytic activity caused by increased blood flow contributes to PTFE graft intimal atrophy. Our results suggest that serine proteinases are a major factor in matrix degradation in this model of intimal atrophy. Plasmin and urokinase plasminogen activator (uPA) are increased more than 3-fold over normal-flow levels, whereas MMP-2 and MMP-9 are increased less than 50%. The ability of a plasmin-blocking antibody to inhibit versican degradation by intimal extracts further supports the role of this protease in graft intimal atrophy. Despite the fact that MMP-1,24 MMP-2,24,25⇓ MMP-3,24 MMP-713 (Figure 3), and MMP-925 can degrade versican and MMPs are involved in arterial remodeling in response to changes in blood flow,26,27⇓ BB-94 had no effect on versican degradation by the intimal extracts. Although this result suggests that active MMPs are not available in the intima, this may be an artifact of the extraction process during which active enzyme may be brought together with inhibitors. It is also possible that MMPs not inhibited by BB-94 or that MMPs with specificity for substrates other than versican may be involved in matrix removal.
Although non-hepatic tissue, such as kidney, testis, intestine, and brain,28–30⇓⇓ can produce plasminogen mRNA, there is no evidence for the synthesis of plasminogen by vascular SMCs. Therefore, our data suggest that high blood flow increases the transport of plasminogen across the endothelial barrier and/or increases retention of plasminogen in the intima. Numerous studies have demonstrated increased endothelial cell permeability resulting from increased shear stress supporting the former possibility. Data supporting the latter possibility are that apoptosis of SMCs is increased 2- to 3-fold by high blood flow in this model (S.A. Berceli, M.G. Davies, R.D. Kenagy, A.W. Clowes, unpublished data, 2001) and that dead and dying cells possess increased proteolytic capacity. Endothelial cells committed to apoptosis after growth factor removal have increased MMP-1, activated MMP-2, and uPA compared with starving, but viable, endothelial cells (B. Levkau, R.D. Kenagy, unpublished data, 1999). Apoptotic U937 and epithelial cells express increased uPA and plasminogen binding31,32⇓ mediated by the uPA receptor and plasminogen receptors.33 Therefore, dying SMCs might bind plasminogen from the plasma and mediate the removal of extracellular matrix.
An unusual aspect of this study is the biphasic time course of uPA, PAI-1, and versican degradation. Although the changes are internally consistent, increased uPA and decreased PAI-1 with increased degradation of versican, this result is not easily explained. Because macrophages can express uPA at high levels,34 one possibility is that increased flow may induce an early influx of monocytes. A loss of macrophages might then be followed later by increased expression of SMC urokinase. However, the number of monocyte/macrophages in the intima was <0.1% of intimal cells and was not changed by high flow (S.A. Berceli et al, unpublished data, 2001). Clearly more study is required to understand the mechanisms regulating protease induction in this model.
Our data suggest that by increasing plasmin activity the restenotic process that leads to failure in many vascular grafts may be reversed. Consistent with this idea is the observation that plasmin is required for the positive arterial remodeling observed after arterial injury.15,35⇓ In addition, arterial lesion development mediated by hyperlipidemia is accelerated by the absence of plasminogen,36 indicating a protective role in atherogenesis. However, in the normolipemic context, plasmin is required for the intimal hyperplasia observed after arterial injury37 and it has been demonstrated to play a role in aneurysmal expansion.37,38⇓ It is clear that plasmin is a multifunctional protease. Whether the regulation of this enzyme can be of value therapeutically will depend on the context in which its activity is modified.
Supported by grants from the National Institutes of Health, US Public Health Service (HL30946, RR00166, HL07828, and HL18645); PTFE grafts were provided by W.L. Gore & Associates; Dr Davies was a recipient of an National Institutes of Health Cardiovascular Training Grant Fellowship (HL07828). The authors thank Holly Lea and Selina Vergel for technical assistance, Sue Perigo-Potter for preparing the 35S-methionine–labeled SMC versican, and Lindsey Miles for the blocking antibody to plasmin.
Received November 21, 2001; revision accepted January 4, 2002.
- ↵Langille BL. Blood flow-induced remodeling of the artery wall.In: Bevan JA, Kaley G, Rubanyi GM, eds. Flow-Dependent Regulation of Vascular Function. New York, NY: Oxford University Press; 1995: 277–299.
- ↵Wentzel JJ, Krams R, Schuurbiers JCH, Oomen JA, Kloet J, Van der Giessen WJ, Serruys PW, Slager CJ. Relationship between neointimal thickness and shear stress after wall stent implantation in human coronary arteries. Circulation. 2001; 103: 1740–1745.
- ↵Kohler TR, Kirkman TR, Kraiss LW, Zierler BK, Clowes AW. Increased blood flow inhibits neointimal hyperplasia in endothelialized vascular grafts. Circ Res. 1991; 69: 1557–1565.
- ↵Kraiss LW, Geary RL, Mattsson EJR, Vergel S, Au YPT, Clowes AW. Acute reductions in blood flow and shear stress induce platelet-derived growth factor-A expression in baboon prosthetic grafts. Circ Res. 1996; 79: 45–53.
- ↵Geary RL, Kohler TR, Vergel S, Kirkman TR, Clowes AW. Time course of flow-induced smooth muscle cell proliferation and intimal thickening in endothelialized baboon vascular grafts. Circ Res. 1994; 74: 14–23.
- ↵Mattsson EJR, Kohler TR, Vergel SM, Clowes AW. Increased blood flow induces regression of intimal hyperplasia. Arterioscler Thromb Vasc Biol. 1997; 17: 2245–2249.
- ↵Kenagy RD, Hart CE, Stetler-Stevenson WG, Clowes AW. Primate smooth muscle cell migration from aortic explants is mediated by endogenous platelet-derived growth factor and basic fibroblast growth factor acting through matrix metalloproteinases 2 and 9. Circulation. 1997; 96: 3555–3560.
- ↵Clowes AW, Clowes MM, Au YPT, 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: 61–67.
- ↵Schonherr E, Jarvelainen HT, Sandell LJ, Wight TN. Effects of platelet-derived growth factor and transforming growth factor-beta1 on the synthesis of a large versican-like chondroitin sulfate proteoglycan by arterial smooth muscle cells. J Biol Chem. 1991; 266: 17640–17647.
- ↵Olin KL, Potter-Perigo S, Barrett PHR, Wight TN, Chait A. Lipoprotein lipase enhances the binding of native and oxidized low density lipoproteins to versican and biglycan synthesized by cultured arterial smooth muscle cells. J Biol Chem. 2000; 274: 34629–34636.
- ↵Halpert I, Sires UI, Roby JD, Potter-Perigo S, Wight TN, Shapiro SD, Welgus HG, Wickline SA, Parks WC. Matrilysin is expressed by lipid-laden macrophages at sites of potential rupture in atherosclerotic lesions and localizes to areas of versican deposition, a proteoglycan substrate for the enzyme. Proc Natl Acad Sci U S A. 1996; 93: 9748–9753.
- ↵Talhouk RS, Bissell MJ, Werb Z. Coordinated expression of extracellular matrix-degrading proteinases and their inhibitors regulates mammary epithelial function during involution. J Cell Biol. 1992; 118: 1271–1282.
- ↵Drew AF, Tucker HL, Kombrinck KW, Simon DI, Bugge TH, Degen JL. Plasminogen is a critical determinant of vascular remodeling in mice. Circ Res. 2000; 87: 133–139.
- ↵Mignatti P. Extracellular matrix remodeling by metalloproteinases and plasminogen activators. Kidney Int. 1995; 47 (suppl 49): S12–S14.
- ↵Twining SS, Wilson PM, Ngamkitidechakul C. Extrahepatic synthesis of plasminogen in the human cornea is up-regulated by interleukins-1α and -1β. Biochem J. 1999; 339: 705–712.
- ↵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: 1373–1382.
- ↵Knudsen BS, Nachman RL. Matrix plasminogen activator inhibitor: modulation of the extracellular proteolytic environment. J Biol Chem. 1988; 263: 9476–9481.
- ↵Bartha K, Declerck PJ, Moreau H, Nelles L, Collen D. Synthesis and secretion of plasminogen activator inhibitor 1 by human endothelial cells in vitro: effect of active site mutagenized tissue-type plasminogen activator. J Biol Chem. 1991; 266: 792–797.
- ↵Sandy JD, Westling J, Kenagy RD, Iruela-Arispe ML, Verscharen C, Rodriguez-Mazaneque JC, Zimmermann DR, Lemire JM, Fischer JW, Wight TN, et al. Versican V1 proteolysis in human aorta in vivo occurs at the Glu441-Ala442 bond, a site that is cleaved by recombinant ADAMTS-1 and ADAMTS-4. J Biol Chem. 2001; 276: 13372–13378.
- ↵Perides G, Asher RA, Lark MW, Lane WS, Robinson RA, Bignami A. Glial hyaluronate-binding protein: a product of metalloproteinase digestion of versican?. Biochem J. 1995; 312: 377–384.
- ↵De Smet BJGL, De Kleijn D, Hanemaaijer R, Verheijen JH, Robertus L, Van der Helm YJM, Borst C, Post MJ. Metalloproteinase inhibition reduces constrictive arterial remodeling after balloon angioplasty: a study in the atherosclerotic Yucatan micropig. Circulation. 2000; 101: 2962–2967.
- ↵Zhang L, Seiffert D, Fowler BJ, Jenkins JS, Thinnes TC, Loskutoff DJ, Parmer RJ, Miles LA. Plasminogen has a broad extrahepatic distribution. Thromb Haemost. In Press.
- ↵Ramharack R, Spahr MA, Kreick JS, Sekerke CS. Expression of apolipoprotein[a] and plasminogen mRNAs in cynomolgus monkey liver and extrahepatic tissues. J Lipid Res. 1996; 37: 2029–2040.
- ↵Xiao Q, Danton MJS, Witte DP, Kowala MC, Valentine MT, Bugge TH, Degen JL. Plasminogen deficiency accelerates vessel wall disease in mice predisposed to atherosclerosis. Proc Natl Acad Sci U S A. 1997; 94: 10335–10340.