Plasminogen Activator Inhibitor Type 1 and Tissue Inhibitor of Metalloproteinases-2 Increase After Arterial Injury in Rats
Abstract Vascular injury induced by angioplasty causes smooth muscle cells to migrate, proliferate, and form a neointima. The neointima is further enlarged by the accumulation of matrix molecules synthesized by smooth muscle cells. Smooth muscle cell migration and matrix accumulation are associated with an increase in the expression of matrix-degrading enzymes and might be regulated by the balance of protease and anti-protease activity. We have studied the inhibitors of two major classes of matrix-degrading enzymes, the plasminogen activators and the matrix metalloproteinases (MMPs) to understand better the regulation of proteolytic activity following balloon catheter injury in the rat carotid artery. At various times after injury, protease inhibitor expression was analyzed by Northern blotting, reverse zymography, immunohistochemistry, and Western blotting. During the first month after injury, we found that the expression of two proteinase inhibitors (plasminogen activator inhibitor type 1 [PAI-1] and tissue inhibitor of metalloproteinases-2 [TIMP-2]) was modulated. PAI-1 mRNA expression reached a maximum 6 hours after injury before tapering off to baseline levels by 3 days. PAI-1 activity, as measured by reverse zymography, followed the same temporal profile. PAI-1, localized by immunohistochemistry, was expressed at low levels in the media of control arteries and was increased after injury primarily in the medial smooth muscle cells. TIMP-2 mRNA levels began to increase 24 hours after injury and reached a maximum at day 7. TIMP-2 activity, measured by reverse zymography, peaked at day 3 after injury. TIMP-2 protein was increased in the intima compared with the media and adventitia at day 7 after injury. The increase of PAI-1 and TIMP-2 after injury supports the hypothesis that changes in the proteolytic balance play an important role in smooth muscle cell migration after arterial injury.
- smooth muscle cell
- plasminogen activator inhibitor type 1
- tissue inhibitor of metalloproteinases-2
- arterial injury
Cell migration is a central event in many biological processes. Tumor invasion, development, and tissue remodeling all require carefully controlled cell migration. Migration of vascular SMCs plays a major role in the formation of arterial intimal lesions in human atherosclerosis and the formation of the neointima associated with restenosis following balloon angioplasty.1 SMCs in the arterial wall are surrounded by ECM composed of collagen, fibronectin, proteoglycans, and other glycoproteins. SMCs, in order to migrate, must detach themselves from this surrounding cage of ECM by proteolytic degradation of their pericellular matrix.
Vascular SMCs are known to produce both serine proteases and MMPs. These proteases are capable of degrading all the components of the ECM. Plasmin, a serine protease, is formed from plasminogen by PAs (tissue PA and urokinase PA). Studies from this as well as other laboratories have shown that PA activity increases after balloon injury and that by blocking plasmin generation, neointima formation is reduced.2 3 Plasmin is a serine protease of broad specificity that is also able to activate other MMPs as well as degrade matrix. Plasmin generation is primarily controlled by the balance between the PAs and their physiological inhibitors, the PAIs. In addition to PAs, MMPs are produced by vascular SMCs in response to balloon injury.4 5 Inhibition of MMP activity reduces neointima formation5 and limits invasiveness in tumor invasion models.6 7 8 These previous results show that inhibiting protease activity can limit neointima formation and suggest that protease inhibitors may serve to regulate SMC migration.
We now have investigated whether protease inhibitors are modulated in the injured rat carotid artery and whether their expression may serve to alter the proteolytic balance at key time points during the formation of the neointima. In the present study, we examined two classes of inhibitors of PAs and MMPs and demonstrate that in these two classes of inhibitors only two specific protease inhibitors, PAI-1 and TIMP-2, are increased in injured arteries and that their location and time of expression in the vessel wall may help define their role in the response to injury.
Materials and Methods
Animals, Surgery, and Tissue Preparation
Three-month-old male Sprague-Dawley rats were obtained from Bantin & Kingman Inc, Seattle, Wash. Rats were anesthetized with 1.0 mL/kg of an intramuscular solution containing 1 mg/mL acepromazine (Fermenta Animal Health Corp), 50 mg/mL ketamine (Aveco Inc), and 5 mg/mL Rompun (Mobay Corp) in a saline solution. Balloon injury to the common carotid artery was performed by the passage of a 2F embolectomy catheter as previously described.9 Animals were killed with an overdose of pentobarbital. The arteries were flushed clear of blood with PBS, pH 7.4, removed, and stripped of the surrounding connective tissue and fatty material, leaving the adventitia. Samples intended for Northern analysis were immediately frozen in liquid nitrogen and stored at −70°C. Arteries that were to be examined immunohistochemically were perfusion-fixed in 10% neutral buffered formalin at physiological pressure for 5 minutes after being flushed with lactated Ringer’s solution (Baxter). The carotid arteries were stored in formalin overnight before being embedded in paraffin. Rats were cared for according to the Principles of Laboratory Animal Care (formulated by the National Society for Medical Research) and the Guide for the Care and Use of Laboratory Animals (NIH publication No. 86-23, revised 1985.)
Histological sections (4 μm) were cut from paraffin-embedded rat carotid arteries, prepared, and placed on poly-l-lysine–coated microscope slides. The sections were deparaffinized by three changes in xylene for 5 minutes each. Slides were rehydrated in a graded alcohol series as follows: three changes of 100% ethanol for 5 minutes total, two changes of 95% ethanol for 4 minutes total, and one change of 75% ethanol for 1 minute. Slides were prepared for immunohistochemistry using the Vectastain Elite ABC kit (Vector Laboratories). Endogenous peroxidase activity was blocked by a 5-minute incubation in 3% hydrogen peroxide. Slides were washed in two changes of PBS, pH 7.2. Nonspecific binding of the secondary antibody was blocked by incubation in 10% normal goat serum for 10 minutes, followed by two changes of PBS. The rabbit anti-rat PAI-1 antibody (American Diagnostica) or control rabbit IgG (Vector Laboratories) was added at a concentration of 25 μg/mL in 1% BSA. PAI-1 antibody and control rabbit IgG were incubated overnight at 4°C. The primary antibody was washed in three changes of PBS for 1 minute each. Biotinylated anti-rabbit secondary antibody was added for 30 minutes. Slides were washed in three changes of PBS for 1 minute each, followed by avidin-biotin-alkaline phosphatase for 30 minutes. The slides were then washed in one change of PBS, followed by one wash in 0.05 mol/L Tris, pH 7.6, at 37°C for 5 minutes each. Slides were incubated at 37°C for 30 minutes in 175 mL 0.05 Tris with 4 mL DAB (132 mL of 0.05 mol/L Tris, pH 7.6, and 5 g DAB; aliquot stored at −70°C) and alkaline phosphatase substrate (Vector Laboratories), followed by one wash in distilled water for 1 minute. Slides were counterstained in hematoxylin, cleared in Histoclear (National Diagnostics), and mounted in Histomount (National Diagnostics). Immunohistochemical staining was visualized under bright-field illumination on a Leitz Dialux 20 EB and photographed using Ectachrome Elite 100 ASA.
Isolation of RNA and Northern Blot Analysis
Total cellular RNA was isolated from 12 rat common carotid arteries in pooled groups of four per time point by guanidinium isothiocyanate–phenol–chloroform extraction as previously described,10 except that tissue was first finely crushed under liquid nitrogen in a mortar and pestle and then homogenized for 30 seconds in a Brinkman PT10-35 homogenizer. Isolated RNA was resuspended in 0.5% SDS and quantified spectrophotometrically. RNA samples were separated in a 1% agarose/formaldehyde gel.11 RNA was transferred to Zeta-probe nylon membrane (Bio-Rad) as described by the manufacturer and cross-linked to the membrane using UV light (Stratagene). Filter hybridizations were carried out as described by Church and Gilbert.12 In general, hybridizations were performed in a solution of 1% BSA, 7% SDS, 0.5 mol/L sodium phosphate, pH 7.0, 1 mmol/L EDTA, and 20% formamide at 65°C. Filters were washed to a final stringency of 0.1× SSPE at 65°C (0.33 mol/L NaCl and 0.2 mol/L NaH2PO4, pH 7.4) and exposed to Kodak XAR x-ray film at −70°C using an intensifying screen and quantified on the PhosphorImager at the Markey Molecular Medicine Facility, Seattle, Wash.
Isolation of Arterial Protein and Western Analysis
Seven days after injury, carotid arteries were removed, and the intima, media, and adventitia were separated. The intima was separated by cutting open the artery longitudinally and then peeling out the intima with forceps. The media was separated from the adventitia by holding onto the media and pulling the adventitia; this process leaves some tightly connected adventitia remaining with the media. Tissue was then snap-frozen and processed as described for reverse zymography. Protein (10 μg per lane) was loaded and run on a 10% SDS-polyacrylamide gel and transferred to nitrocellulose. Rabbit anti–TIMP-2 antibody was added at 1:1000 for 30 minutes and visualized with 1:7500 dilution of anti-rabbit IgG (Fc)-alkaline phosphatase conjugate and subsequent color development (Promega).
The rat PAI-1 cDNA probe used in the present study was generously provided by Dr T. Gelehrter.13 The TIMP-2 cDNA and TIMP-2 antibody were provided by Dr Keith Langley (Amgen, Thousand Oaks, Calif). For Northern hybridizations, plasmids were radiolabeled by the nick-translation method14 using the large fragment of Klenow DNA polymerase and [32P]dCTP (3000 Ci/mmol).
Carotid arteries were removed, stored, and homogenized in 0.05 mol/L Tris, 0.01 mol/L CaCl2, 2.0 mol/L guanidine HCl, and 0.2% Triton X-100, pH 7.5, as described by Dean et al.15 Homogenate was centrifuged at 10 000g for 1 minute, and the supernatant was dialyzed against 3 vol of 0.05 mol/L Tris-HCl, pH 7.5, and 0.2% Triton X-100 over 2 days (SpectraPore, Spectrum). Aliquots were frozen at −70°C. Protein concentration was determined by the BCA assay (Pierce) as described by the manufacturer. For PAI analysis, 20-μg samples were run on a 10% SDS-PAGE as described by Laemmli16 and then washed for 10 minutes in 2.5% Triton X-100 (Sigma Chemical Co) and 10 minutes in PBS. The gel was overlaid onto a 1.25% agar (Difco) containing 40 μg/mL plasminogen (American Diagnostica), 250 mU/mL human recombinant HMW urokinase (Calbiochem), and 2% nonfat dry milk powder (wt/vol) (Carnation) in 0.1 mol/L Tris, pH 8.1. The reverse zymogram was developed at 4°C overnight and then at 37°C for 3 to 6 hours. Visualization was by dark-field illumination. For TIMP reverse zymography, 100 μg of protein was run on a 10% SDS-polyacrylamide gel containing 1 mg/mL gelatin and 1 mL of MMP solution (University Technologies Intl Inc). The gel was washed in 2.5% Triton X-100 and developed for 22 hours at 37°C before staining with Coomassie blue.
PhosphorImager data from Northern blots were normalized to the signal from 28S rRNA labeling, as a measure of loading, and expressed as fold increase compared with uninjured carotid arteries. The mean increase is reported with two-tailed 95% confidence intervals by one-way ANOVA compared with the 0-hour injured control vessel. Error bars are standard deviations of the mean.
Effects of Arterial Injury on PAI Expression
To establish whether inhibitors of PA activity were present in the injured vessel wall, extracts of injured rat carotid arteries were analyzed by reverse zymography. Reverse zymography provides a qualitative measure of inhibitors of PA activity, including inactive PAI-1, which is activated by denaturants in the polyacrylamide gel. After injury, the only detectable inhibitor of PA activity was PAI-1 (50 kD). PAI-2 was not detectable. Six hours after injury, PAI-1 reached maximal levels (Fig 1⇓). This activity decreased by 24 hours, after which it was undetectable in this assay.
Northern Analysis of PAI-1 mRNA
To determine the kinetics of PAI-1 induction in the rat carotid, total mRNA from the carotids of balloon-injured rats was analyzed on Northern blots using 32P-labeled cDNA probes specific for rat PAI-1. Rat PAI-1 mRNA has only one transcript, whereas PAI-1 in many other species has 3.2- and 2.3-kb transcripts.13 17 PAI-1 mRNA was induced by injury in the rat common carotid as shown in Fig 2⇓. This induction was transient and returned to baseline levels by 3 days after injury. The induction of message was maximal at 6 hours (18-fold increase; n=3; SD, 10.6; two-tailed 95% confidence interval, −7 to 45). Although the mRNA levels were increased on three independent blots, the increase was not statistically different from that for the control 0-hour injured carotid artery. This induction was dramatically decreased by 24 hours and returned to baseline levels by day 3 after injury.
Immunohistochemical Analysis of PAI-1 Antigen in Injured Rat Carotid Arteries
To determine the cell type responsible for the increase in PAI-1 expression, cross sections of injured carotid arteries were stained for PAI-1. PAI-1 expression was increased in a subpopulation of cells in the media at 6 hours after injury (Fig 3B⇓). PAI-1 staining was not seen by 3 days (Fig 3C⇓). PAI-1 was absent in the 15-minute injured vessel and in the normal uninjured vessel. Adventitial staining was present in the IgG controls (Fig 3D⇓, 3E⇓, and 3F⇓). Adventitial staining was also present in sections stained with anti–PAI-1 preincubated with human PAI-1 antigen (data not shown). Taken together, these data from Northern blotting, reverse zymography, and immunohistochemistry clearly demonstrate that medial SMCs increase transiently the expression of PAI-1 in response to injury.
Effect of Arterial Injury on Inhibitors of MMPs
The carotid artery extracts were also analyzed for inhibitors of MMPs. TIMP-2 (18 kD) was the only MMP inhibitor detected by reverse zymography and appeared to be modulated by injury (Fig 4⇓). TIMP-1 (Mr, 28 kD) activity was not detectable in the carotid extracts (Fig 4⇓). Immediately after injury through 24 hours, TIMP-2 was present at low levels. Three days after injury, TIMP-2 inhibitory activity increased and reached maximal activity around day 5. By 14 days after injury, TIMP-2 was decreased.
Northern Analysis of TIMP-2 mRNA After Arterial Injury
TIMP-2 levels began to increase 24 hours after balloon injury and reached a maximal 3-fold induction at day 7 (n=3; SD, 0.65; two-tailed 95% confidence interval, 1.4 to 4.7) (Fig 5⇓). TIMP-2 mRNA was not significantly elevated at 2 weeks. Rat TIMP-2 has two transcripts, 1.0 and 3.5 kb. In response to injury, only the 1.0-kb transcript was significantly induced, whereas the 3.5-kb transcript was not modulated and remained constitutively expressed.
Western Analysis of TIMP-2 Antigen in the Injured Carotid Artery
The location of TIMP-2 after injury was determined by physically separating the intima, media, and adventitia. Protein from each region was then extracted and analyzed by Western blotting for TIMP-2 antigen (Fig 6⇓). Seven days after injury, TIMP-2 antigen was confined to the intima and was not detected in the media or the adventitia. The data from Northern blotting, reverse zymography, and Western blotting clearly demonstrate that TIMP-2 is induced by balloon injury in the rat carotid artery and that TIMP-2 expression is limited to the intima at day 7 after injury.
Previous investigators have suggested that changes in protease activity are required for cell migration following arterial injury.2 3 4 5 The question addressed by the present study is whether protease inhibitors are regulated by balloon injury in the rat carotid artery. The main findings of the present study are that members of the serine protease inhibitor (serpin) and MMP inhibitor families are increased after balloon catheter injury in the rat carotid artery. We show that two members of these protease inhibitor families, PAI-1 and TIMP-2, are increased and that the expression of these inhibitors follows different spatial and temporal profiles. Other PAIs and TIMPs were not detected.
PAI-1 mRNA, antigen, and activity were induced rapidly after injury, reached a maximum by 6 hours, and then quickly returned to baseline after 24 hours. TIMP-2 was induced at 1 day, reached maximal levels in the intima around day 7, and remained elevated through 2 weeks. TIMP-2 might be complexed at late time points with proteases (eg, MMP-2). This possibility might explain the apparent decline in TIMP-2 activity despite the presence of TIMP-2 mRNA at 14 days. The increase in protease inhibitors, in conjunction with previous observations of regulated protease activity, suggests that changes in proteolytic and fibrinolytic activity may be central events in the response to injury.
There is strong experimental evidence both in vivo and in vitro that regulation of matrix proteolysis is important for tumor invasion, tissue remodeling, angiogenesis, and development. Changes in the proteolytic balance between protease and anti-protease expression have been associated with diseases such as atherosclerosis,18 19 metastasis,20 and thromboembolism.21 Proteinases and their inhibitors also play an important role in tissue remodeling,22 23 wound repair,24 and many other normal processes related to reproduction, such as lactation,25 uterine implantation,26 and gestation.27 All of these events require controlled expression of PAs or MMPs or both in conjunction with their respective inhibitors. Vascular repair in response to injury might also require a highly controlled pattern of protease and anti-protease expression by vascular SMCs. Vascular SMCs degrade ECM by plasminogen-dependent and plasminogen-independent pathways.28 Plasmin and the MMPs form an interactive proteolytic cascade, since they affect the activation of one another. This proteolytic cascade is able to degrade most of the ECM components, including collagen types I, III, and IV, laminin, fibronectin, elastin, and proteoglycans. A shift in the balance between the proteases and their respective inhibitors is probably necessary for vascular remodeling to occur.
PAI-1 Expression After Arterial Injury
PAI-1 is a single-chain polypeptide with an Mr of 50 kD that inhibits urokinase PA as well as single- and two-chain forms of tissue PA by rapidly forming a 1:1 stoichiometric complex. PAI-1 is an early growth response gene in fibroblasts and hepatocytes and is induced during the transition from the quiescent state (G0) to the G1 phase of the cell cycle.29 In the mouse, the highest concentrations of PAI-1 are found in the aorta, consistent with observations that PAI-1 is a major product of smooth muscle cells.30 The 18-fold increase in expression of PAI-1 at 6 hours after injury is consistent with the time course of PAI-1 induction seen by other investigators using cytokines31 or with other forms of injury.29 PAI-1 is found in both active and latent forms. Northern analysis, reverse zymography, and immunohistochemistry do not distinguish between the two forms. This limits our ability to determine whether PAI-1 is having a physiological effect after injury. However, PAI-1 as measured in the present study might be biologically active, because it can be stabilized in the vessel wall through association with serum-derived vitronectin.32
Plasma PAI-1 increases after trauma and inflammation as one of several APR proteins.33 PAI-1 is synthesized by a wide variety of tissues in vivo, unlike most APR proteins, which are synthesized in the liver.30 In the rat, PAI-1 is a tissue-associated inhibitor of fibrinolysis at sites of vascular disruption and inflammation. Balloon catheter injury inflicts extensive damage on the vessel wall. The immediate priority of the vessel after this type of extensive injury is to prevent hemorrhage. Control of vascular disruption and hemorrhage is achieved through the formation of thrombi. At early times after arterial injury, there is platelet adherence and limited fibrin deposition on the luminal surface, but the luminal surface clears after 1 day. An early increase in tissue-associated PAI-1 may serve to limit fibrinolysis during this early repair period. Additionally, PAs are expressed in the injured rat carotid artery and are necessary but not sufficient for SMC migration. Tranexamic acid, a synthetic PAI, blocks SMC migration.3 PAI-1 may have a similar effect by limiting proteolytic activity at early time points after injury and as a result block migration.
TIMP-2 and Arterial Injury
MMPs, as well as PAs, are induced in rat injured carotid arteries.4 5 A 92-kD collagenolytic activity thought to be MMP-9 is increased 24 hours after injury and then decreases to baseline levels. Gelatinase (72 kD) is expressed constitutively, and its activated form is increased 5 days after injury and decreases after 14 days.4 5 A low-molecular-weight MMP (24.5 kD) with broad proteolytic activity and high elastinolytic activity is found in the adventitia and is increased 5 days after injury. This 24.5-kD MMP may be involved in elastin turnover in the adventitia. This increased activity of MMPs coincides with increased TIMP-2 expression. Increased protease inhibitors could block cell migration from the media. This theory is supported by experiments in which a synthetic MMP inhibitor, GM 6001, is able to block 97% of the SMCs migrating into the intima by blocking MMP activity.5
TIMP-2 is a nonglycosylated 21-kD protein that selectively forms a 1:1 complex with the latent and activated form of 72-kD type IV collagenase. TIMP-2 preferentially binds to the 72-kD proenzyme, whereas TIMP-1 preferentially binds to the 92-kD proenzyme form of type IV collagenase. Therefore, increased TIMP-2 expression may specifically limit the activity of 72-kD type IV collagenase. Although TIMP-1 and TIMP-2 have preferential binding patterns, they are able to inhibit all members of the MMP subclasses that have been tested.34 The preference of inhibitors for specific types and forms of MMPs confers an added level of control for proteolytic degradation.
There are two TIMP-2 transcripts expressed by SMCs. The 3.5-kb transcript of TIMP-2 is constitutively expressed after injury, whereas the 1.0-kb transcript is increased 3-fold. The two transcripts are thought to arise from alternative splicing of 5′-untranslated regions.35 Both transcripts have been shown to encode identical proteins.36 37 The opposite pattern of differential regulation of the TIMP-2 message has been described in vitro by Testa,38 who reported that the 3.5-kb transcript was modulated and the 1.0-kb transcript was minimally induced. Khokha et al39 have shown that even a 29% increase in TIMP-1 mRNA expression has a significant effect on the invasive phenotype. These results support the conclusion that the increase in TIMP-2 mRNA and activity seen in the present study is likely to produce a strong biological effect in the vessel wall and may serve to limit migration.
We have also examined TIMP-1 and TIMP-3 mRNA expression in the rat carotid artery after balloon injury and found no significant modulation of TIMP-1 and no detectable TIMP-3 expression (data not shown). These findings indicate that TIMP-1 and TIMP-2 are not regulated in parallel, a conclusion reached by other investigators.35 TIMP-1 is present at lower levels in rat arteries than in baboon arteries.40 This observation may suggest that different species preferentially use different proteases and their respective inhibitors to perform similar functions.
In summary, we have demonstrated that two endogenous inhibitors of protease activity, PAI-1 and TIMP-2, are increased in response to arterial injury in the rat carotid artery. Taken together with our previous observations on the induction of proteases during SMC migration, these results support our conclusion that changes in proteolytic activity are necessary for SMC migration and may be a critical part of the response to injury in the carotid artery.
Selected Abbreviations and Acronyms
|APR protein||=||acute phase–reactant protein|
|SMC||=||smooth muscle cell|
|TIMP||=||tissue inhibitor of MMPs|
This study was supported by National Institutes of Health grants HL-18645 and HL-52459 and training grant T32 HL-07312 (Dr Hasenstab). We are grateful to Dr Keith E. Langley, Amgen Inc, Thousand Oaks, Calif, for the TIMP-2 cDNA and antibody, Dr Thomas D. Gelehrter for the PAI-1 cDNA, and Holly Lea, University of Washington, for excellent surgical work.
Reprint requests to David Hasenstab, University of Washington, Department of Surgery, 356410, Seattle, WA 98195.
- Received December 13, 1995.
- Accepted December 19, 1996.
- © 1997 American Heart Association, Inc.
Clowes AW, Clowes MM, Kirkman TR, Jackson CL, Au YPT, Kenagy R. Heparin inhibits the expression of tissue-type plasminogen activator by smooth muscle cells in injured rat carotid artery. Circ Res. 1992;70:1128-1136.
Bendeck MP, Zempo N, Clowes AW, Galardy RE, Reidy MA. Smooth muscle cell migration and matrix metalloproteinase expression after arterial injury in the rat. Circ Res. 1994;75:539-545.
Pauly RR, Passaniti A, Bilato C, Monticone R, Cheng L, Papadopoulos N, Gluzband YA, Smith L, Weinstein C, Lakatta EG, Crow MT. Migration of cultured vascular smooth muscle cells through a basement membrane barrier requires type IV collagenase activity and is inhibited by cellular differentiation. Circ Res. 1994;75:41-54.
Stettler-Stevenson WG, Liotta LA, Kleiner DE. Extracellular matrix 6: role of matrix metalloproteinases in tumor invasion and metastasis. FASEB J. 1993;7:1434-1441.
Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989:7.43.
Church GM, Gilbert W. Genomic sequencing. Proc Natl Acad Sci U S A. 1984;81:1991-1995.
Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 1989:10.6.
Ginsburg D, Zeheb R, Yang AY, Rafferty UM, Andreasen PA, Nielsen L, Dano K, Lebo RV, Gelehrter TD. cDNA cloning of human plasminogen activator-inhibitor from endothelial cells. J Clin Invest. 1986;78:1673-1680.
Schneiderman JA, 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 EKO. Localization and production of plasminogen activator inhibitor-1 in human healthy and atherosclerotic arteries. Arterioscler Thromb. 1993;13;1090-1100.
Lang IM, Marsh JJ, Olman MA, Moser KM, Loskutoff DJ, Schleef RR. Expression of type 1 plasminogen activator inhibitor in chronic pulmonary thromboemboli. Circulation. 1994;89:2715-2721.
Woessner JF. Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J. 1991;5:2145-2154.
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.
Brenner CA, Adler RR, Rappolee DA, Pederson RA, Werb Z. Genes for extracellular matrix-degrading metalloproteinases and their inhibitor, TIMP, are expressed during early mammalian development. Genes Dev. 1989;3:848-859.
Sperti G, van Leeuwen RTJ, Quax PHA, Maseri A, Kluft C. Cultured rat aortic vascular smooth muscle cells digest naturally produced extracellular matrix: involvement of plasminogen-dependent and plasminogen-independent pathways. Circ Res. 1992;71:385-392.
Thornton AJ, Bruzdzinski CJ, Raper SE, Gelehrter TD. Plasminogen activator inhibitor-1 is an immediate early response gene in regenerating rat liver. Cancer Res. 1994;54:1337-1343.
Sawdey MS, Loskutoff DJ. Regulation of murine type 1 plasminogen activator inhibitor gene expression in vivo. J Clin Invest. 1991;88:1346-1353.
Reilly CF, McFall RC. Platelet derived growth factor and transforming growth factor-beta regulate plasminogen activator inhibitor-1 synthesis in vascular smooth muscle cells. J Biol Chem. 1991;266:9419-9427.
Stettler-Stevenson WG, Brown PD, Onisto M, Levy AT, Liotta LA. Tissue inhibitor of metalloproteinases-2 (TIMP-2) mRNA expression in tumor cell lines and human tumor tissues. J Biol Chem. 1990;265:13933-13938.
Testa JE. Loss of the metastatic phenotype by a human epidermoid carcinoma cell line, HEp-3, is accompanied by increased expression of tissue inhibitor of metalloproteinase 2. Cancer Res. 1992;52:5597-5603.
Khokha R, Waterhouse P, Yagel S, Lala PK, Overall CM, Norton G, Denhardt DT. Antisense RNA-induced reduction in murine TIMP levels confers oncogenicity on Swiss 3T3 cells. Science. 1989;243:947-950.