Thrombin Upregulates Tissue Transglutaminase in Endothelial Cells
A Potential Role for Tissue Transglutaminase in Stability of Atherosclerotic Plaque
Abstract— Atherosclerosis is characterized by thickening of the vessel wall, smooth muscle cell proliferation, macrophage infiltration, and deposition of a fibrin network. Transglutaminases are a family of enzymes catalyzing the formation of stable covalent cross-links between proteins. Here, we show that tissue transglutaminase (tTG) synthesis by human umbilical vein endothelial cells is upregulated by thrombin, the serine protease that causes fibrin formation and many cellular inflammatory effects. Thrombin upregulated tTG 2-fold at the mRNA and protein level. Cellular cross-linking activity was increased to an even greater extent; antibody to tTG neutralized the increased activity. The effect on tTG expression required active thrombin and was mediated mainly through protease-activated receptor-1, a thrombin receptor. Increased tTG antigen and activity were evident in human umbilical vein endothelial cells and extracellular matrix in situ. Thrombin treatment also led to a cellular redistribution of tTG. Normal vessel wall stained positively for tTG in the smooth muscle cells and in the subendothelium. The intensity of staining increased in vessel walls with plaque, where there was a striking increase in tTG in the smooth muscle cells immediately below the plaque. These studies indicate a role for tTG in the stabilization of atherosclerotic plaques and suggest that its local expression can be controlled by thrombin.
Transglutaminases (TGs) are a family of cross-linking enzymes that catalyze the formation of a covalent bond between glutamine and lysine residues and therefore stabilize and localize proteins. TGs are ubiquitous, with each enzyme having a distinct set of substrates, giving them distinct physiological roles. In terms of hemostasis, the most recognized member of the TG family is factor XIIIa, but there is also a potential role for tissue TG (tTG, also abbreviated TGII and TGH; EC 126.96.36.199). tTG and factor XIII differ in their location and regulation, with tTG being found in a variety of tissues and factor XIII occurring in plasma. The activity of tTG requires calcium; factor XIII also requires activation by the serine protease thrombin.1,2
tTG activity has been demonstrated to play a role in stabilization of the basement membrane and adhesion of cells,3,4 which are important processes in wound healing, angiogenesis, and bone remodeling.5–7 In particular, tTG has been shown to cross-link components of the extracellular matrix (ECM), including fibronectin, vitronectin, laminin, and collagen.8–12 tTG is expressed by a variety of cell types,1,2 including endothelial cells (ECs), smooth muscle cells (SMCs), and macrophages, which (along with fibrin) are major components of atherosclerotic lesions. During wound healing, tTG expression is upregulated in these cell types, with tTG being detected at an early stage in the lesion, suggesting a role in cell migration and ECM remodeling.6 tTG and activated factor XIII can stabilize blood clots by cross-linking fibrin strands and also by cross-linking inhibitors of fibrinolysis, α2-antiplasmin, and plasminogen activator inhibitor-2 to fibrin.13,14
ECs form a physical barrier between blood and tissues and are also the source of many molecules that influence pathological states.15 ECs are important in inflammatory lesions, including the atherosclerotic plaque, where they interact with SMCs, infiltrating macrophages, and lipid.16 The collagen and lipid content of the plaque determines its stability, a lipid-rich core offering less support for the plaque than a collagen-rich core. Therefore, we hypothesized that tTG in the plaque and surrounding tissue can cross-link supporting protein and fibrous matrices, making the plaque less prone to rupture.
tTG expression is regulated by a variety of agents in vitro. These include retinoic acid, lipopolysaccharide, interleukin-6, tumor necrosis factor-α, and transforming growth factor-β.17–21 This investigation focuses on thrombin, a serine protease that catalyzes the cleavage of fibrinogen to fibrin, which is an important component of thrombi. Thrombin has many additional biological effects that are mediated through a family of cellular receptors, protease-activated receptor (PAR)-1 through PAR-4.22 Thrombin cleaves an N-terminal peptide from the receptors, and the new N-terminus acts as the receptor ligand; a peptide that mimics the new N-terminus, SFLLRNPNDKYQPF (TRAP), is a useful tool in studies of thrombin.23 The cellular effects of thrombin include upregulation of tissue factor, plasminogen activator inhibitor-1, and monocyte chemotactic protein-1.24–26 In the present study, we show that thrombin also upregulates tTG mRNA, protein, and cross-linking activity in human umbilical vein ECs (HUVECs). We demonstrate the presence of active tTG in the vessel wall of normal and diseased aortas, including atherosclerotic lesions. The intensity of tTG staining increased as disease progressed. We hypothesize that in this way thrombin can induce stabilization of inflammatory lesions.
HUVECs were isolated from umbilical cords 0 to 3 days after delivery.27 Cells were isolated and maintained in fibronectin-coated (200 μg/mL) culture flasks in medium 199 (Sigma Chemical Co) supplemented with 20% pooled human serum, EC growth supplement (100 μg/mL), and heparin (45 μg/mL), ie, complete medium. Cells were passaged into 8-chambered slides and 25-cm2 or 75-cm2 flasks (Nunc) for experiments and were used only between passages 2 and 5. Cells were treated with complete medium with or without 1 U/mL bovine thrombin (Leo Pharmaceuticals), 10 μmol/L all-trans retinoic acid (Sigma), 20 μmol/L TRAP, or control peptide at 37°C in a 5% CO2 environment for the appropriate times. Conditioned medium was removed, cells were washed 3 times with 171 mmol/L NaCl, 3.4 mmol/L KCl, 10.1 mmol/L Na2HPO4, and 1.8 mmol/L KH2PO4 (PBS), and the cells were removed in 5 mmol/L EGTA or PBS with gentle scraping. ECM was scraped off into 0.1% SDS for SDS-PAGE or into PBS for protein and activity assays. All samples were stored at −70°C. ECs were characterized by staining for von Willebrand factor. Endotoxin-free conditions were maintained throughout by the use of pyrogen-free disposable plastic ware and filter sterilization of all reagents through a 0.2-μm filter. Cytotoxicity was monitored by measurement of release of lactate dehydrogenase by using the CytoTox96 assay (Promega).28 All experiments were repeated at least 3 times with ECs from different donors.
Measurement of tTG Antigen
Cell lysate and ECM samples were separated by SDS-PAGE, and proteins were transferred to nitrocellulose.14 tTG was detected by using a mouse monoclonal antibody to tTG (CUB7402, Neomarkers) and a rabbit anti-mouse IgG conjugated to alkaline phosphatase (Dako), with guinea pig liver tTG (Sigma) used as a standard, with the quoted 2.4 U/g tTG.
tTG Cross-Linking Activity
tTG cross-linking activity was measured by using a microtiter plate assay.29 Briefly, 96-well plates were coated with N,N-dimethylcasein. Cell samples were incubated in the plates with 5-(biotinamido)pentylamine, 10 mmol/L dithiothreitol, and 5 mmol/L CaCl2. Guinea pig liver tTG was used as a standard; the unit of activity was defined with reference to conversion of hydroxylamine plus N-α-CBZ-Gln-Gly to hydroxamate.9 Incorporation of biotinylated amine was detected colorimetrically with streptavidin-alkaline phophatase conjugate. The absorbance was read at 405 nm (reference filter 692 nm). Each sample was assayed at least twice, each time at a minimum of 3 dilutions. tTG activity was expressed relative to total cellular protein, which was measured by using the Bradford protein assay.
In Situ Antigen and Activity
tTG antigen and activity were measured in live or acetone-fixed ECs. HUVECs were passaged into 8-chambered slides (Nunc). Cells were routinely fixed in acetone for 10 minutes and incubated with 0.5 mmol/L monodansylcadaverine in PBS in the presence or absence of 2.5 mmol/L CaCl2, 100 mmol/L EDTA, or 100 mmol/L iodoacetamide at 37°C for 3 hours. For measurement of activity in live cells, the conditioned medium was removed and replaced with medium 199 plus 20% human serum containing 0.5 mmol/L monodansylcadaverine (Sigma) for 24 hours.29a The monolayer was then washed 3 times with PBS before fixation on ice in 100% acetone for 10 minutes, and tTG antigen was visualized by using 200 μg/mL CUB7402 (Neomarkers) and rabbit anti-mouse IgG conjugated to tetramethylrhodamine B isothiocyanate. Nuclei were stained with 4,6-diamidino-2-phenylindole (DAPI) for 10 minutes at room temperature before mounting with antifade mounting medium (Dako). Fluorescence was visualized by using an Olympus BX-40 microscope with a ×20 objective.
Northern Blot Analysis of tTG mRNA
mRNA was extracted from 2×106 HUVECs by the acid guanidinium thiocyanate-phenol–chloroform method and oligo(dT) beads (Dynal), electrophoresed on formaldehyde agarose gels, and vacuum-blotted. A full-length tTG cDNA (kindly supplied by Prof C.S. Greenberg, Duke University Medical Center, Durham, NC) was used to probe for tTG mRNA. The cDNA probe was labeled with [α-32P]dCTP by using the Megaprime labeling kit (Amersham Pharmacia Biotech). Membranes were prehybridized in 500 mmol/L phosphate buffer, pH 7.2, containing 10 mmol/L EDTA and 7% (wt/vol) SDS for 5 hours at 65°C. Membranes were incubated with probe overnight at 65°C, washed in 1× saline sodium citrate buffer with or without SDS at 65°C, overlaid with x-ray film, and stored at −70°C until development. In all cases, signals were standardized with reference to the GAPDH signal.30
Subcellular fractionation was carried out on 90% confluent (≈2×106) ECs.30 Cells were homogenized in 1 vol of 20 mmol/L HEPES, 2 mmol/L EDTA, and 4 mmol/L 2-mercaptoethanol, pH 7.4, containing a cocktail of protease inhibitors (homogenizing buffer, Sigma) with 20 strokes of a Dounce homogenizer (Wheaton). An equal volume of 0.66 mol/L sucrose was added to the mixture, which was centrifuged at 100g for 10 minutes to make a postnuclear supernatant. The postnuclear supernatant was layered onto 1.5 vol of 40% (wt/vol) sucrose in homogenizing buffer. Samples were centrifuged for 1 hour at 100 000g. All fractions were collected and analyzed for tTG antigen by Western blotting.
Sections of human aorta were stained for tTG (CUB7402) or for ε-γ glutamyl-Lys cross-links (153-81D1C2, abcam), with antibody to Aspergillus niger glucose oxidase as a control, by use of the alkaline phosphatase anti–alkaline phosphatase (APAAP) technique.32 Slides were washed and treated in sequence with rabbit anti-mouse IgG, APAAP complex, and substrate (Sigma fast red tablets). The slides were counterstained in hematoxylin, overlaid with a coverslip, and viewed with an Olympus BX-40 microscope.
Thrombin Upregulates tTG Antigen, mRNA, and Activity in ECs
tTG was detected in cell lysate from untreated HUVECs, and basal production was estimated at 5 μg tTG/mg total cellular protein in 90% confluent cells, as assessed by immunoblotting and comparison by densitometry with a standard curve of guinea pig liver tTG. Approximately equal quantities were detected in cells and ECM, with none detectable in conditioned medium. Treatment of cells with 1 U/mL thrombin increased tTG antigen in the cell lysate ≈2-fold (Figure 1A). These experiments were carried out 4 times, and the upregulation in tTG antigen ranged from 2- to 3-fold over untreated control, with the example given in Figure 1 being a 3-fold increase. tTG upregulation was achieved by 1 U/mL thrombin but could be seen in the range 0.5 to 5 U/mL thrombin; doses outside this range gave responses similar to the untreated control, as we have previously reported for other thrombin effects.28 Northern blotting revealed a significant increase in tTG mRNA after 6 hours of treatment with thrombin (Figure 1B). Upregulation ranged from 2- to 3-fold, with the example in Figure 1B being a 2.1-fold change by densitometry. The effect of thrombin was also visible by staining of HUVECs in situ with fluorescent antibodies to tTG (Figure 1C). The typical staining pattern shows that most tTG was present throughout the cytoplasm in untreated and treated cells, with some perinuclear staining, but little in the nucleus itself.
The cross-linking activity of tTG in cells, as shown by incorporation of biotinylated amine in the presence of calcium, was strikingly increased in response to thrombin; only trace activity was detectable in the absence of calcium. Activity was increased 6.2- to 8.2-fold after 24 hours of treatment in 4 separate experiments, which were carried out on 4 different preparations of HUVECs (Table). The increase in activity was significant at the P<0.0001 level for all 4 experiments and was neutralized by incorporation of antibody to tTG (not shown).
To visualize cross-linking activity in HUVECs, the fluorescent cross-linking substrate monodansylcadaverine was incubated with the cell monolayer, either live or acetone-fixed. Both yielded the same results, but the acetone-fixed cells are shown (Figure 2). Fluorescent staining was detected under basal conditions only after the addition of calcium. Thrombin treatment for 24 hours resulted in increased cross-linking activity. Retinoic acid treatment produced a similar result in tTG cross-linking activity (Figure 2). Activity was dependent on the presence of calcium.
Preincubation of thrombin with hirudin resulted in no increase in tTG antigen (data not shown), confirming the requirement for thrombin activity; hirudin alone had no effect. We investigated the involvement of the thrombin receptor, PAR-1, by treating cells with 20 μmol/L TRAP. This caused a 2-fold increase in tTG antigen, as assessed by densitometry, similar to the upregulation by thrombin (Figure 3).
Cellular Distribution of tTG
HUVECs were removed from ECM and fractionated on a sucrose gradient to determine the localization of tTG. In untreated cells, most cellular tTG was found in the cytosol with a small amount in the particulate and nuclear fractions (Figure 4). Thrombin induced a redistribution of tTG in the first 2 hours of treatment, with a clear decrease in the level of tTG in the particulate fraction. Therefore, thrombin has 2 distinct effects, a reorganization in the cellular location of tTG and upregulation of tTG expression. These experiments were repeated 3 times, with a reduction in particulate fraction tTG seen each time.
tTG in Vessel Wall
tTG antigen was detected in sections of the vessel wall and increased with the progression of disease. In diseased vessel with easily distinguishable plaque, the tTG staining was most obvious in the SMCs immediately below the plaque (Figure 5A) compared with SMCs of more normal regions. This intense staining was also seen in the shoulder region of the plaque. Distinct tTG staining was seen in lipid-rich areas (Figure 5C) with tTG in the surrounding macrophages, characterized by staining for CD68. There was diffuse staining for tTG in the necrotic core, and there was some positive staining for tTG in the fibrous cap, mainly localizing with fibroblasts and subendothelial matrix. Activity of tTG was detected by an antibody specific for ε-γ glutamyl-Lys cross-links and was most obvious in the fibrous cap (Figure 5B), with extensive cross-linking throughout the cap structure. Interestingly, there was less activity at the base of the plaque, which displayed the most intense staining for tTG antigen, and the necrotic core was essentially negative.
tTG is a cross-linking enzyme that covalently stabilizes proteins in the ECM and basement membrane of cells. We show that tTG is present in the wall of normal vessel and that staining increased with the onset and progression of atherosclerotic disease. We went on to investigate tTG synthesis in ECs, a major cell type in the vessel wall, to study a potential link between the increasing levels of tTG and the effect of inflammatory stimuli. Using HUVECs, we found that thrombin upregulated tTG expression at the mRNA and protein level and increased cross-linking activity. The upregulation was modest, similar to that by retinoic acid, a known regulator of tTG expression,17 but thrombin also affected the cellular location of tTG and its activity. These effects of thrombin on tTG add to the ways in which thrombin can establish a prothrombotic environment.
Only active thrombin could upregulate tTG, consistent with its action through PAR-1. PAR-1 is the predominant thrombin receptor on ECs, although PAR-2 and PAR-3 are also present.33 These other PARs may mediate part of the effect on tTG expression, but our results show that PAR-1 is the major receptor involved. Several cell types express members of the PAR family and are also known to express TGs; these include macrophages34 and smooth muscle cells.35 If thrombin also controls the expression of tTG in these cell types, then local tTG would be increased significantly in inflammatory situations. The tTG could then participate in the establishment of ECM and basement membrane for infiltrating cells, stabilizing protein assemblies and concentrating proteins at their sites of function.
Interestingly, in the present study, the increase in tTG cross-linking activity was greater than the increase in antigen. The possibility that it reflects a contribution by a different TG was ruled out by neutralization of activity by antibodies specific to tTG. The increased activity may be due to a second effect of thrombin on tTG activity, in addition to regulation of expression (for instance, by increasing intracellular calcium).36 Alternatively, it has been reported that FXIII activity is supported by Ca2+ and thrombin binding.37 It is possible that a similar phenomenon occurs with tTG such that in the presence of thrombin, more of the available pool of tTG is in an active state. The in situ cross-linking activity appeared to be largely associated with the matrix, whereas in the microtiter plate assay, there was considerable activity in the cell lysate. The apparent contrast may reflect differences in access of the cross-linking substrates. It should be noted that intracellular tTG is considered a cryptic enzyme,38 and although the substrate is freely permeable to the cells, only ECM tTG activity is seen on calcium addition.29a
Thrombin also redirected the cellular pools of tTG to different locations. We speculate that the decrease of tTG in the particulate fraction may represent secretion of tTG through a mechanism that involves tTG linking with the membrane. ECs secrete tTG basolaterally into the ECM,40 despite the fact tTG lacks a signal peptide for secretion through the classical endoplasmic reticulum-Golgi pathway.41 One possibility is that tTG localizes to the cell membrane, possibly through the recently described fibronectin-tTG-β1 integrin complex,42,43 and uses an uncharacterized mechanism to cross it. Thrombin activation could activate this secretion pathway, increasing the local tTG concentration available for cross-linking of the proteins in the ECM.
We have also demonstrated that tTG is present in the normal vessel wall, and its presence close to the lumen of the vessel supports a role for tTG in cross-linking proteins of the hemostatic system.13,14 Staining of vessel sections displaying early signs of disease showed a general increase in the tTG that was present. This increase in tTG appeared to be due in part to infiltrating macrophages, which are known to express tTG.44 There was also very distinct staining of the elastic lamina at the boundary of the intima and media. As disease progressed to the atherosclerotic plaque, there was a further increase in tTG staining, with a pattern that localized tTG to cellular areas and particularly at the base and shoulders of the plaque. This suggests that tTG may play a role in anchoring and stabilizing the plaque in the vessel wall. Conversely, staining for cross-links revealed that activity was mainly confined to the fibrous cap of the plaque, which suggests that tTG lays down a stable fibrous barrier for the plaque, again aiding its resistance to rupture.
Generation of active thrombin leads to fibrin formation and to changes in the local concentration of inflammatory and hemostatic proteins.24–26 Thrombin has a profound influence on the progression of atherosclerosis, being mitogenic for SMCs, macrophages, fibroblasts, and astrocytes45 and stimulating expression of fibrinolytic proteins in a cell-specific manner.28,46–50 As we report in the present study, thrombin also increases tTG expression and affects the location of tTG within the cell. We speculate that this tTG cross-links the supporting proteins of the plaque, such as collagen, thus anchoring it in the vessel wall. However, in lipid-rich plaques, the lower collagen content means that the plaque is intrinsically less stable and therefore more prone to rupture.51 When the plaque ruptures and thrombin levels are increased, we speculate that increased tTG expression and activity would contribute to the formation of a stable insoluble thrombus, protected from fibrinolysis through localized inhibitors.13,14,25,28 tTG is worthy of further study involving the persistence of atherosclerosis and thrombus, because the present work demonstrates a link between tTG expression, thrombin generation, and the progression of atherosclerosis. Another recent study reports localization of tTG in carotid and coronary atherosclerosis.52 It stresses the role of tTG in stabilizing the plaque and potential effects on synthesis of matrix components.
This work was supported by grants from The Carnegie Trust for the Universities of Scotland, Tenovus, and from the British Heart Foundation (FS97064). The authors thank Prof C.S. Greenberg for his kind gift of the tTG cDNA, Dr P.A.J. Brown for valuable advice on immunohistochemistry, and Prof M. Greaves and Aberdeen Maternity Hospital for material.
Received January 5, 2001; revision accepted June 22, 2001.
Greenberg CS, Birckbichler PJ, Rice RH. Transglutaminases: multifunctional cross-linking enzymes that stabilize tissues. FASEB J. 1991; 5: 3071–3077.
Jones A, Nicholas B, Mian S, Davies PJA, Griffin M. Reduced expression of tissue transglutaminase in a human endothelial cell line leads to changes in cell spreading, cell adhesion and reduced polymerisation of fibronectin. J Cell Sci. 1997; 110: 2461–2472.
Haroon ZA, Hettasch JM, Lai T-S, Dewhirst MW, Greenberg CS. Tissue transglutaminase is expressed, active, and directly involved in rat dermal wound healing and angiogenesis. FASEB J. 1999; 13: 1787–1795.
Barsigian C, Stern AM, Martinez J. Tissue (type II) transglutaminase covalently incorporates itself, fibrinogen, or fibronectin into high molecular weight complexes on the extracellular surface of isolated hepatocytes: use of 2-[(2-oxopropyl)thio] imidazolium derivatives as cellular transglutaminase inactivators. J Biol Chem. 1991; 266: 22501–22509.
Aeschlimann D, Paulsson M. Cross-linking of laminin-nidogen complexes by tissue transglutaminase: a novel mechanism for basement-membrane stabilization. J Biol Chem. 1991; 266: 15308–15317.
Sakata Y, Aoki N. Significance of cross-linking of α2-plasmin inhibitor to fibrin in inhibition of fibrinolysis and in hemostasis. J Clin Invest. 1982; 69: 536–542.
Ritchie H, Lawrie LC, Crombie PW, Mosesson MW, Booth NA. Cross-linking of plasminogen activator inhibitor 2 and α2-antiplasmin to fibrin(ogen). J Biol Chem. 2000; 275: 24915–24920.
Cines DB, Pollak ES, Buck CA, Loscalzo J, Zimmerman GA, McEver RP, Pober JS, Wick TM, Konkle BA, Schwartz BS,et al. Endothelial cells in physiology and in the pathophysiology of vascular disorders. Blood. 1998; 91: 3527–3561.
Nara K, Nakanishi K, Hagiwara H, Wakita K-I, Kojima S, Hirose S. Retinol-induced morphological changes of cultured bovine endothelial cells are accompanied by a marked increase in transglutaminase. J Biol Chem. 1989; 264: 19308–19312.
Kuncio GS, Tsyganskaya M, Zhu J, Liu SL, Nagy L, Thomazy V, Davies PJA, Zern MA. TNF-alpha modulates expression of the tissue transglutaminase gene in liver cells. Am J Physiol. 1998; 274: G240–G245.
Vollberg TM, George MD, Nervi C, Jetten AM. Regulation of type I and type II transglutaminase in normal human bronchial epithelial and lung carcinoma cells. Am J Respir Cell Mol Biol. 1992; 7: 10–18.
Bartha K, Brisson C, Archipoff G, de la Salle C, Lanza F, Cazenave JP, Beretz A. Thrombin regulates tissue factor and thrombomodulin messenger-RNA levels and activities in human saphenous-vein endothelial cells by distinct mechanisms. J Biol Chem. 1993; 268: 421–429.
Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. J Clin Invest. 1973; 52: 2745–2756.
Ritchie H, Jamieson A, Booth NA. Thrombin modulates synthesis of plasminogen activator inhibitor type 2 by human peripheral blood monocytes. Blood. 1995; 9: 3428–3435.
Slaughter TF, Achyuthan KE, Lai T-S, Greenberg CS. A microtitre plate transglutaminase assay utilizing 5-(biotinamido)pentylamine as substrate. Ann Biochem. 1992; 205: 166–171.
O’Brien PJ, Prevost N, Molino M, Hollinger MK, Woolkalis MJ, Woulfe DS, Brass LF. Thrombin responses in human endothelial cells: contributions from receptors other than PAR 1 include transactivation of PAR 2 by thrombin cleaved PAR 1. J Biol Chem. 2000; 275: 13502–13509.
Nelken NA, Soifer SJ, O’Keefe J, Vu TK, Charo IF, Coughlin SR. Thrombin receptor expression in normal and atherosclerotic human arteries. J Clin Invest. 1992; 90: 1614–1621.
Zhong C, Hayzer DJ, Corson MA, Runge MS. Molecular cloning of the rat vascular smooth muscle cell thrombin receptor. J Biol Chem. 1992; 267: 16975–16979.
Sadasivan C, Yee VC. Interaction of the factor XIII activation peptide with alpha-thrombin: crystal structure of its enzyme-substrate analog complex. J Biol Chem. 2000; 275: 36942–36948.
Smethurst PA, Griffin M. Measurement of tissue transglutaminase activity in a permeabilized cell system: its regulation by Ca2+ and nucleotides. Biochem J. 1996; 313: 803–808.
Deleted in proof
Greenberg CS, Achyyuthan KE, Borowitz MJ, Shuman MA. The transglutaminase in vascular cells and tissues could provide an alternative pathway for fibrin stabilization. Blood. 1987; 3: 702–709.
Akimov SS, Krylov D, Fleischman LF, Belkin AM. Tissue transglutaminase is an integrin-binding adhesion coreceptor for fibronectin. J Cell Biol. 2000; 148: 825–838.
Bar-Shavit R, Benezra M, Sabbah V, Bode W, Vlodavsky I. Thrombin as a multifunctional protein: induction of cell adhesion and proliferation. Am J Respir Cell Mol Biol. 1992; 6: 123–130.
Loskutoff DJ. Effect of thrombin on the fibrinolytic activity of cultured bovine endothelial cells J Clin Invest. 1979; 64: 329–332.
Gelehrter TD, Synyer-Laszuk R. Thrombin induction of plasminogen-activator inhibitor in cultured human endothelial cells. J Clin Invest. 1986; 77: 165–169.
Wojta J, Gallicchio M, Zoellner H, Hufnagl P, Last K, Filonzi EL, Binder BR, Hamilton JA, McGrath K. Thrombin stimulates expression of tissue-type plasminogen activator and plasminogen activator inhibitor type 1 in cultures human vascular smooth muscle cells. Thromb Haemost. 1993; 70: 469–474.
Levin EG, Marzec U, Anderson J, Harker LA. Thrombin stimulates tissue plasminogen activator release from cultured endothelial cells. J Clin Invest. 1984; 74: 1988–1995.
Li X-N, Varma VK, Parks JM, Benza RL, Koons JC, Grammer JR, Grenett H, Tabengwa EM, Booyse FM. Thrombin decreases the urokinase receptor and surface-localized fibrinolysis in cultured endothelial cells. Arterioscler Thromb Vasc Biol. 1995; 15: 410–419.
Davis MJ, Richardson PD, Woolf N, Katz DR, Mann J. Risk of thrombosis in human atherosclerotic plaques: role of extracellular lipid, macrophage and smooth muscle cell content. Br Heart J. 1994; 69: 377–381.