Thrombin Induces the Redistribution and Acute Release of Tissue Factor Pathway Inhibitor From Specific Granules Within Human Endothelial Cells in Culture
Abstract Tissue factor pathway inhibitor (TFPI) is a vascular anticoagulant that regulates the tissue factor (TF)–dependent pathway of coagulation. The majority of intravascular TFPI is thought to be noncovalently bound to the vessel wall. Our immunolocalization studies in cultures of human umbilical vein endothelial cells (HUVEC) and immortalized EA.hy926 cells show that TFPI is located in well-defined granules evenly spread over the cell surface and with apical polarization within the cytoplasm. These granules are smaller than and distinct from Weibel-Palade bodies. Upon treatment of cultured cells with low concentrations of thrombin (0.01 to 1 NIH U/mL), a marked redistribution of TFPI, occurred with patching in focal points and increased exposure of both TFPI antigen and anticoagulant activity on the surface of the stimulated endothelial cells. This redistribution was paralleled by an acute release of TFPI in the cell medium. EA.hy926 cells responded more readily to thrombin stimulation than HUVECs. The process was inhibited by both hirudin and anti-thrombin receptor antibody. Our findings demonstrate a novel mechanism by which thrombin may exert a negative feedback control on blood coagulation. Therefore, this pathway can be of physiological importance in controlling TF-mediated thrombin generation.
- Received May 12, 1995.
- Accepted August 23, 1995.
In normal conditions endothelial cells provide a natural antithrombotic and anticoagulant surface for the flowing blood (for review, see Reference 11 ), and thrombosis rarely occurs in the presence of a healthy endothelium.2 According to the revised model of blood coagulation,3 hemostasis is initiated when, at a site of blood vessel injury, FVII gains access to TF, a cell-membrane–integral protein, forming the TF-FVII/VIIa complex.
TF is constitutively expressed mainly in cells of the adventitia,4 but transient expression also occurs on monocytes/macrophages5 and on endothelial cells after perturbation by various agonists,6 7 8 9 thereby increasing the endothelial potential for thrombin generation.10 Thrombin will induce other procoagulant changes in the endothelial cell function, including release of vWf and downregulation of thrombomodulin and fibrinolytic enzymes,11 12 13 as well as enhancement of platelet and monocyte adhesion mediated by surface exposure of adhesive molecules.14 Endothelial cells also expose procoagulant surfaces that promote increased thrombin generation in diet-induced experimental hypercholesterolemia.15
The control of the highly procoagulant activity of the TF-FVIIa complex occurs through feedback inhibition by TFPI, which is considered to be the principal physiological inhibitor of the complex.3 The structure of TFPI consists of an acidic N-terminal region, followed by three tandem Kunitz-type domains and a highly basic C-terminal region. TFPI acts by initially forming a complex with FXa, which then forms a quaternary complex with TF-FVIIa.16 Thus, binding of FXa by the second Kunitz-type domain potentiates inhibition of the TF-FVIIa complex by the first Kunitz-type domain.17
In resting blood the TFPI concentration is about 3 nmol/L, of which the majority is covalently bound to HDL and LDL lipoprotein complexes,18 19 and only 10% is carrier free.20 About 2.5% of the circulating TFPI is found in platelets, from which it can be released on thrombin stimulation.21 A second form of TFPI has recently been described in placenta, with 60% homology and similar function as circulating TFPI.22 The major pool of vascular TFPI, however, is thought to be bound to the vessel wall, since there is a threefold to fourfold increase in blood levels of TFPI after intravenous infusion of heparin and related compounds.23 24 This increase is transient, and TFPI is sequestered from the blood as the level of heparin diminishes.
Studies with cultured cells have suggested that endothelial cells are the principal site of synthesis of TFPI,25 and it is considered, although not established, that the molecule is exocytosed toward the surface of the cells, where it remains anchored through its basic C-terminal domain to glycosaminoglycans in the glycocalyx.26 Since the kinetic properties of TFPI are such as to predict a relatively slow inhibition of the TF-FVIIa complex at plasma concentrations of TFPI,27 this mechanism of surface exposure of the protein would favor its proposed action as a vessel wall–bound anticoagulant.10
The present study was designed to establish the storage and secretion properties of TFPI in human endothelial cells in culture. A role of TFPI as a specialized endothelium-bound anticoagulant is supported by our findings showing that TFPI resides within endothelial cells in a secretory pool that can be transferred rapidly to the surface when the cells are stimulated with thrombin.
Cell culture media and supplements were purchased from Sigma Chemical Co. Immulon 4 microtiter plates were from Dynatech Laboratories Inc, Biotechnology Products. The polyclonal anti-TFPI antibody (CF7) was raised in rabbits against full-length TFPI (Monsanto Chemical Co) and affinity purified on Protein A–Sepharose 4B. The murine monoclonal anti-vWf was from Dakopatts, Dako Ltd. The rabbit polyclonal antibody directed against the thrombin receptor was a gift from Dr S. Stone (MRC). The conjugates goat anti-rabbit IgG coupled to FITC and horse anti-mouse IgG coupled to Texas Red, as well as Vectashield (mounting medium for fluorescence H-1000), were purchased from Vector Laboratories. Human α-thrombin was a generous gift from Dr J.-M. Freyssinet (Institut d’Hematologie et d’Immunologie; specific activity ≈3000 NIH U/mg). Recombinant hirudin was from Knoll (specific activity ≈1.7 U/μg). Human coagulation factors VIIa, Xa, and X were purchased from Enzyme Research Laboratories. Human rTFPI 1-161 was from Novo Nordisk. Chromogenic substrate Bz-Ile-Glu (piperidyl)-Gly-Arg-pNA (S-2337), specific for FXa, was obtained from Chromogenix AB. The conjugates goat anti-rabbit IgG coupled to HRP, goat anti-rabbit IgG coupled to biotin, streptavidin coupled to Cy3, and all other chemicals used were purchased from Sigma.
We used the immortalized human endothelial cell line EA.hy926 described by Edgell et al28 and HUVECs obtained from fresh umbilical cords.29 The endothelial cells split at low passage were grown on Petri dishes (35-mm diameter) precoated with gelatin, in Dulbecco’s modified Eagle’s medium containing 4 mmol/L l-glutamine, 100 U/mL penicillin, 0.1 mg/mL streptomycin, 15 mmol/L HEPES, and 1:5 diluted heat-inactivated fetal bovine serum. The cell cultures were used at morphologic confluence, on days 2 to 3 (EA.hy926 cells) or 6 to 7 (HUVECs). The normal medium was replaced with serum-free medium 24 hours before the assays.
After removing the culture medium, the monolayers were washed with warm (37°C) THB (0.14 mol/L NaCl, 2.7 mmol/L KCl, 12 mmol/L NaHCO3, 0.4 mmol/L NaH2PO4, 1 mmol/L MgCl2, 2 mmol/L CaCl2, 10 mmol/L HEPES, 5 mmol/L glucose), pH 7.35, supplemented with 3.5 g/L BSA and incubated with human α-thrombin diluted in THB at varying concentrations for the dose-response studies and for various periods of time for the time-course assays. The action of thrombin was stopped by the addition of r-hirudin at a 10-fold molar excess over the thrombin concentration.30
Matched monolayers incubated with THB only and processed in a similar manner to the thrombin-stimulated cells served as control cells.
The supernatants were removed and assayed immediately or kept at −20°C until assayed. Results obtained in these two ways were identical. The cell monolayers were processed as follows: (a) intact monolayers were used for the assay of TFPI anticoagulant potency expressed on the cell surface; (b) cell lysates, obtained by several cycles of freezing (−80°C)/thawing (37°C) of scrape-harvested cells, were used for measuring the TFPI antigen; and (c) monolayers fixed with 2% (wt/vol) paraformaldehyde solution for 1 hour at 18°C were used for immunofluorescence staining and ELISA assays.
The specificity of thrombin action on the endothelial cells was checked by incubating the cells under the same conditions as above but with thrombin inactivated with r-hirudin before the assay. To determine whether the thrombin receptor expressed on endothelial cells31 mediates the thrombin-induced TFPI release, EA.hy926 cells were mildly fixed (1% paraformaldehyde in THB, 10 minutes at 18°C), incubated with the antibody directed against the thrombin receptor (1 hour at 18°C), and then treated with thrombin (5 NIH U/mL for 30 minutes), as described above.
The cellular localization of TFPI was studied by an indirect immunofluorescence procedure before and after stimulation of the endothelial cells with thrombin (0.1 U/mL, 30 minutes at 37°C).
For detecting intracellular TFPI, cells were briefly permeabilized with 0.02% (wt/vol) saponin in THB for 10 minutes at 18°C.32
All the samples were washed in THB, blocked with 1:100 diluted normal goat or horse serum in DAKO antibody diluent, and incubated for 1 hour at 18°C with the primary antibodies—CF7 (60 μg IgG/mL) alone, or a mixture of CF7 and anti-vWf (both at 1:100 dilution) for the double immunostaining. Samples were washed with TBS (50 mmol/L Tris, 0.15 mol/L NaCl), pH 8.0, and incubated with the secondary antibodies—either goat anti-rabbit IgG/biotin (diluted 1:1000 in TBS) followed by streptavidin/Cy3 (1:140 dilution in TBS), each for 1 hour at 18°C, or a mixture of goat anti-rabbit/and horse anti-mouse/Texas Red (1:100 diluted in TBS) for the double immunolabeling. Specimens mounted in Vectashield on glass slides were examined with a Bio-Rad MRC 600 confocal laser scanning unit attached to a Nikon Diaphot inverted microscope (Bio-Rad Microscience Ltd). The light source was a Krypton/Argon laser (Ion Laser Technology) with main lines at 488, 568, and 674 nm. Samples were analyzed by serial optical sectioning in the z-axis of the cells, followed by computer-assisted reconstruction of the images.
Controls to ascertain for the specificity of the binding comprised either replacement of first antibodies with normal IgG from the same species or incubation of the cells with the secondary conjugates only.
TFPI Antigen Level Measurements
For this indirect competitive ELISA, 96-well Immulon-4 microtiter plates were coated with rTFPI (5 ng/well) in 0.1 mol/L sodium carbonate buffer, pH 9.8, and kept at 4°C until used. Endothelial cell supernatants and lysates were diluted 1:1 with PB (2.5 mmol/L NaH2PO4, 7.5 mmol/L Na2HPO4, 0.125 mol/L NaCl), pH 7.2, with 10 g/L BSA and 0.5 g/L Tween 20 added, and mixed with CF7 in PB (10 μg IgG/mL final concentration). The microtiter plates were blocked for 30 minutes at 18°C with PB supplemented with 50 g/L nonfat dry milk, 10 g/L gelatin, and 1:100 diluted normal goat serum. The sample mixtures were added to the plates (50 μL/well) and incubated for 2 hours at 37°C. After washing with PB, the plates were incubated for 1 hour at 18°C with goat anti-rabbit IgG/HRP conjugate (1:1000 dilution in PB), washed, and peroxidase was detected with 1 g/L ortho-phenylenediamine hydrochloride in citrate-phosphate buffer (35 mmol/L citric acid, 67 mmol/L Na2HPO4), pH 4.8, with 0.015% H2O2 added. The reaction was stopped with 1.5 mol/L H2SO4, and the optical density was measured at 490 nm in a Molecular Devices THERMOmax Microplate Reader (Alpha Laboratories Ltd).
The TFPI antigen concentration in the samples was extrapolated from a standard curve constructed with serial dilutions made from NHP (pool from >30 donors, stored at −80°C, heat denatured for 15 minutes at 56°C, and centrifuged for 3 minutes at 13 500g), which was assigned with an antigen concentration of 100 ng TFPI/mL.18
Assay on the Cell Monolayers
TFPI antigen associated with cell monolayers before and after thrombin stimulation was measured in a direct ELISA assay. The cells were fixed, permeabilized, quenched, and blocked for the nonspecific binding sites, then incubated with CF7 (60 μg IgG/mL) for 1 hour at 18°C, washed with PB, and developed as described above.
Considering that for validity standard curves should be processed in a manner similar to the cell samples, we coated the wells of a microtiter plate with serial dilutions of rTFPI (0.75 to 100 ng/well) and used the same protocol as above.
Functional Assay of TFPI
We used a modification of the two-stage amidolytic chromogenic assay described by Sandset et al33 in which 25 μL of samples diluted with TBS-A was incubated for 10 minutes at 37°C in the wells of a microtiter plate with 100 μL of a combined reagent containing (all final concentrations): 2.5 ng/mL FVIIa, 5 mU/mL FXa, 1:80 diluted rabbit brain thromboplastin (Sigma, Catalog No. T-0263), and 15 mmol/L CaCl2. Then, 50 μL of a mixture of 0.4 U/mL FX and 1 mmol/L synthetic substrate S-2337 in TBS-A were added to the plate, and the rate of substrate cleavage was monitored over 25 minutes at 37°C in the kinetic microplate reader, using the dual kinetic mode (L1−L2: 405 nm minus 650 nm).
Assay on Cell Monolayers
Both resting and thrombin-treated cells were washed with THB and incubated for 10 minutes at 37°C with the same combined reagent as above. Aliquots were taken off the cell culture plates and the residual activity of the FVIIa-TF complex toward FX was measured in the microtiter plate essentially as described above.
The TFPI activity was extrapolated from a standard curve constructed with serial dilutions of NHP to which a functional potency of 1 U/mL was assigned and which were processed in a way similar to the samples.
To confirm the specificity of the assay, control experiments were performed by incubating the cells with the anti-TFPI antibody (60 μg IgG/mL, 1 hour at 18°C) before the functional assay. This antibody concentration inhibited 95% of the TFPI activity in NHP.
Experiments were repeated 3 to 5 times, with 4 cell culture plates for each point and optical readings made in duplicates. For all the quantitative estimations performed, data obtained for each determination were statistically compared between each other by the paired t test and presented as mean±SD.
Immunolocalization of TFPI in Human Endothelial Cells in Culture
The localization of TFPI in HUVECs and EA.hy926 cells was studied by means of the indirect immunofluorescence labeling approach, and results are illustrated in Figs 1⇓ and 2⇓. Fluorescence localization was achieved through serial optical sectioning of the cells and computer-assisted reconstruction of the image. Staining of nonpermeabilized cells showed that TFPI was uniformly distributed over the entire surface of the cells (Fig 1a⇓, 1 through 5, and 1b, vertical section), in discrete granular structures (Fig 1a⇓, inset, arrows). Intracellular TFPI was found within well-defined granules in resting cells permeabilized with saponin (Fig 1c⇓, arrows). The serial optical sectioning technique revealed that TFPI distribution is polarized, with the bulk confined to the apical part of the cells (Fig 1d⇓, 4 through 6). Results of double immunolabeling for TFPI (green) and vWf (red) are illustrated in Fig 1e⇓, where almost no colocalization of the two antigens (yellow) can be observed.
The normal immunostaining pattern was dramatically changed for both surface-exposed and intracellular TFPI after thrombin stimulation of the cells. As shown in Fig 2a⇑, a strong patching of the granules occurred over the cell surface, leading to big areas of intense fluorescence (magenta color, arrows), which generally concentrated to the periphery of the cells. This pattern is also illustrated by images obtained after optical vertical sectioning through the z-axis of the cells: The TFPI granules accumulated in strong fluorescent patches, with a certain tendency of polarization toward the junction areas (Fig 2b⇑ and 2c⇑, arrows). Whereas resting endothelial cells express TFPI only on the apical surface, the thrombin-stimulated cells expose TFPI on both apical and basolateral areas.
The normal distribution of TFPI granules was altered within the interior of the cells also: The fluorescence patched in high-intensity points (Fig 2d⇑, magenta color, arrows), and areas of the cells became completely devoid of fluorescence as a consequence of the polarization of pooled TFPI toward the lateral parts (compare Fig 2e⇑, 4 through 6, with matched sections in Fig 1d⇑).
Control experiments gave negative results for all the conjugates tested.
Thrombin-Induced Acute Release of TFPI From EA.hy926 Cells
Exposure of endothelial cells to α-thrombin resulted in an acute release of TFPI. The level of TFPI antigen increased significantly (P<.05 for all the points) in the cell medium, in a time- and dose-dependent manner (Fig 3a⇓ and 3b⇓). We found that 30 seconds of stimulation and even low concentrations of thrombin (0.05 NIH U/mL) were sufficient to cause release of TFPI, mirrored by a corresponding decrease of TFPI antigen in cell lysates (Fig 3b⇓).
The ELISA assay performed on cell monolayers permitted us to discriminate between the surface-only-expressed and the intracellularly located TFPI antigen. After thrombin stimulation, the decrease of intracellular TFPI was paralleled by an increased exposure of the antigen on the cell surface (Fig 3c⇑).
TFPI Activity on Thrombin-Stimulated EA.hy926 Cell Monolayers
The functional potency of TFPI associated with the surface of thrombin-stimulated endothelial cells was significantly enhanced (P<.01 for all points) compared with resting cells (Fig 3d⇑ through 3f). The increase was manifest within 30 seconds of stimulation and at a concentration of 0.05 U/mL of thrombin (Fig 3d⇑ and 3e⇑). After incubating the cells with the anti-TFPI antibody, TFPI activity was reduced by >50% (Fig 3f⇑).
As illustrated in Fig 4a⇓ and 4b⇓, thrombin inactivated with hirudin before the assay failed to induce the release of TFPI from the endothelial cells, even with high concentrations of thrombin (5 U/mL for 30 minutes).
Fig 4c⇑ and 4d⇑ reveals that both the release of TFPI in the medium and its enhanced expression on the endothelial cell surface were almost completely prevented by the anti-thrombin receptor antibody. Similar results were obtained when nonfixed EA.hy926 cells were kept at 4°C during the incubation with the anti-thrombin receptor antibody (data not shown).
Thrombin stimulation of HUVECs had the same effect in terms of TFPI release, although it was less than that observed with EA.hy926 cells (data not shown).
The present study demonstrates that TFPI is intracellularly stored in human endothelial cells in culture and actively released on acute stimulation, a process similar to that of other proteins synthesized in endothelial cells, such as vWf,34 P-selectin,35 and tPA.36
In resting endothelial cells, a part of TFPI is spread in granular structures uniformly over the cell surface, and another part is located intracellularly, with apical polarization. No colocalization between the TFPI-containing granules and the WPb could be observed. TFPI granules were seen as circular-shaped structures smaller than WPb, with estimated dimensions, from the immunofluorescence studies, of 250 to 500 nm in diameter. The ultrastructural characteristics of the TFPI storage granules are currently under investigation.
From our findings, we propose that the bulk of TFPI resides within a storage pool in the endothelial cell, from which it can be released in a manner appropriate to its putative role as a vessel wall–bound anticoagulant protein. In quiescent cells, there is a significant and persistent exposure of TFPI on the surface of the cells, perhaps reflecting the tendency of the pool to be continuously exocytosed.
When endothelial cells in culture were challenged with thrombin, this secretory property was more marked, with focal patching and accumulation of TFPI-containing granules at the cell margins and junction areas. This was associated with an over-exposure of TFPI on the cell surface and enhanced release into the medium, both of which were dose and time dependent. The process occurred rapidly, probably within seconds after stimulation, and at low concentrations of thrombin (0.05 NIH U/mL, ≈1 nmol/L).
If these properties are compared with those of other proteins secreted by the endothelium (eg, vWf and tPA) then two pathways may also exist for TFPI secretion, namely constitutive and/or stimulated. This possibility is at present under investigation. Patching of TFPI granules in our cells resembles in some aspects the observations made by Richardson et al,37 who found that the mechanism by which vWf is discharged from thrombin-stimulated endothelial cells involves fusion of WPb with each other to form larger vacuoles that encroach on, and subsequently fuse with, the luminal membrane of the endothelium.
Endothelial cells express the thrombin receptor on their surface, and most of the activities induced by thrombin are mediated by the receptor.31 This seems to be equally true for the thrombin-induced release of TFPI from the EA.hy926 cells, since the release was blocked by a specific anti-thrombin receptor antibody.
In considering the physiological role of TFPI, it has been shown that it acts as a major FXa binding protein on the surface of cultured hepatoma cells,26 a property that confers to these cells the ability to effectively inhibit the TF-FVIIa complex. Besides TFPI, another major FXa binding protein on the HepG2 cell surface is protease nexin-1.26 Our results showing ≈50% inhibition of TFPI anticoagulant activity by CF7 antibody are consistent with the presence of another anti-Xa activity on the surface of endothelial cells also.
We have observed that, after thrombin treatment, both the TFPI anticoagulant activity and antigen levels were increased by ≈25% on the endothelial cell surface. These data support the hypothesis that part of the secreted TFPI remains associated with structures on the cell surface, possibly glycosaminoglycans in the cell glycocalyx. Indeed, the exposure of enhanced TFPI activity on the endothelial cell surface was completely prevented (data not shown) when cells were preincubated with polybrene (hexadimethrine bromide), a synthetic polycation that neutralizes the effect of heparin24 and strongly binds to heparan sulfate and other glycosaminoglycans, blocking their negative charge.38
Although the properties of TFPI as an inhibitor of coagulation are well established in vitro, the role it plays in cardiovascular physiopathology is still largely unknown. Besides acting as an antihemostatic agent, whereby dysfunctionality would result in thrombosis, TFPI can modulate the TF system, which plays a major role in the development of thrombotic complications associated with atherosclerosis.4 Increased plasma TFPI levels seen in patients with familial hypercholesterolemia may act as a compensatory mechanism to prevent the activation of blood coagulation.18 39 Our own observations showing human atherosclerotic plaques to exhibit increased levels of both TFPI antigen and activity40 tend to support this proposal.
Clarification of the mechanisms that regulate TFPI and TF expression and function on the cell membrane surface will be of great importance. A number of mechanisms have already been identified, including the inhibition of TF activity on phosphatidylserine-rich surfaces,41 the particular topographic location of TF on the abluminal surface of endothelial cells,32 which prevents activation of the TF pathway under physiological conditions, and the control of the proteolytic activity of the TF-FVIIa complex by antithrombin III and TFPI.3 10 It is noteworthy that TFPI is the only known endothelial anticoagulant protein the expression of which is not downregulated but even slightly upregulated during an inflammatory response.25
In conclusion, we have shown that TFPI is stored intracellularly within endothelial cells in culture and, on acute stimulation with thrombin, redistributes to the surface of the cells, where it would be available for the inhibition of the TF-FVIIa complex. In addition, substantial amounts of TFPI are released in the cell medium. The thrombin generation mechanism in vivo is replete with many positive and negative feedback loops.42 We believe that the process described herein could be physiologically relevant for the function of TFPI as a regulator of the activation of coagulation by arresting FX activation by the TF-FVIIa complex at low TF levels during inflammatory reactions or host defense responses. Thus, further thrombin formation through this pathway may be prevented. This new negative feedback mechanism can be compared with another one previously described,43 which involves thrombin-dependent inactivation of FVa and FVIIIa by activated protein C generated by the thrombomodulin-thrombin complex.
Selected Abbreviations and Acronyms
|F||=||human coagulation factor|
|HUVEC||=||human umbilical vein endothelial cell|
|NHP||=||normal human plasma|
|rTFPI||=||recombinant tissue factor pathway inhibitor|
|tPA||=||tissue-type plasminogen activator|
|TBS-A||=||TBS supplemented with 1 g/L BSA|
|TFPI||=||tissue factor pathway inhibitor|
|vWf||=||von Willebrand factor|
This work was supported by stipends from the Thrombosis Research Trust and British Heart Foundation grant PG/95027. We thank Professor Fedor Bachmann for the critical reading of the manuscript and invaluable help in editing.
Stern DM, Handley D, Nawroth PP. Endothelium and the regulation of coagulation. In: Simionescu N, Simionescu M, eds. Endothelial Cell Biology in Health and Disease. New York, NY: Plenum Press; 1988:275-306.
Pearson JD. Vessel wall interactions regulating thrombosis. Br Med Bull. 1994;50:776-788.
Wilcox JN, Smith KM, Schwartz SM, Gordon D. Localization of tissue factor in the normal vessel and in the atherosclerotic plaque. Proc Natl Acad Sci U S A. 1989;86:2839-2843.
McGee MP, Li L. Functional differences between intrinsic and extrinsic coagulation pathways-kinetics of factor X activation on human monocytes and alveolar macrophages. J Biol Chem. 1991;266:8079-8085.
Bevilacqua MP, Pober JS, Majeau GR, Cotran RS, Gimbrone MA. Interleukin 1 (IL-1) induces biosynthesis and cell surface expression of procoagulant activity in human vascular endothelial cells. J Exp Med. 1984;160:618-623.
Colucci M, Balconi G, Lorenzet R, Pietra A, Locati D, Donati MB, Semeraro N. Cultured human endothelial cells generate tissue factor in response to endotoxin. J Clin Invest. 1983;71:1893-1896.
Nawroth PP, Stern DM. Modulation of endothelial cell hemostatic properties by tumor necrosis factor. J Exp Med. 1986;163:740-745.
Levine JD, Harlan JM, Harker LA, Joseph ML, Counts RB. Thrombin-mediated release of factor VIII antigen from human vein endothelial cells in culture. Blood. 1982;60:531-534.
Rabiet MJ, Plantier JL, Dejana E. Thrombin-induced endothelial cell dysfunction. Br Med Bull. 1994;50:936-945.
Nawroth PP, Handley DA, Esmon CT, Stern DM. Interleukin-1 induces endothelial cell procoagulant while suppressing cell surface anticoagulant activity. Proc Natl Acad Sci U S A. 1986;83:3460-3464.
Van Obberghen-Schilling E, Pouyssegur J. Signalling pathways of the thrombin receptor. Thromb Haemost. 1993;66:163-168.
Broze GJ, Warren LA, Novotny WF, Higuchi DA, Girard JJ, Miletich JP. The lipoprotein-associated coagulation inhibitor that inhibits the factor VII–tissue factor complex also inhibits factor Xa: insight into its possible mechanism of action. Blood. 1988;71:335-343.
Hansen J-B, Huseby N-E, Sandset PM, Svensson B, Lyngmo V, Nordoy A. Tissue-factor pathway inhibitor and lipoproteins: evidence for association with and regulation by LDL in human plasma. Arterioscler Thromb. 1994;14:223-229.
Novotny WF, Girard TJ, Miletich JP, Broze GJ. Purification and characterization of the lipoprotein-associated coagulation inhibitor from human plasma. J Biol Chem. 1989;264:18832-18837.
Novotny WF, Girard TJ, Miletich JP, Broze GJ. Platelets secrete a coagulation inhibitor functionally and antigenically similar to the lipoprotein associated coagulation inhibitor. Blood. 1988;72:2020-2025.
Sprecher CA, Kisiel W, Mathewes S, Foster DC. Molecular cloning, expression, and partial characterization of a second human tissue-factor-pathway inhibitor. Proc Natl Acad Sci U S A. 1994;91:3353-3357.
Abildgaard U. Heparin/low molecular weight heparin and tissue factor pathway inhibitor. Haemostasis. 1993;23(suppl 1):103-106.
Ameri A, Kuppuswamy N, Basu S, Bajaj SP. Expression of tissue factor pathway inhibitor by cultured endothelial cells in response to inflammatory mediators. Blood. 1992;79:3219-3226.
Kazama Y, Komiyama Y, Kisiel W. Tissue factor pathway inhibitor and protease nexin-1 are major factor Xa binding proteins on the HepG2 cell surface. Blood. 1993;81:676-682.
McGee MP, Foster S, Wang X. Simultaneous expression of tissue factor and tissue factor pathway inhibitor by human monocytes: a potential mechanism for localized control of blood coagulation. J Exp Med. 1994;179:1847-1854.
Edgell CJS, McDonald CG, Graham JB. Permanent cell line expressing human factor VIII-related antigen established by hybridization. Proc Natl Acad Sci U S A. 1983;80:3734-3737.
Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J Clin Invest. 1973;52:2745-2756.
Garcia JGN, Patterson C, Bahler C, Aschner J, Hart CM, English D. Thrombin receptor activating peptides induce Ca mobilization barrier dysfunction, prostaglandin synthesis, and platelet-derived growth factor mRNA expression in cultured endothelium. J Cell Physiol. 1993;156:541-549.
Ryan J, Brett J, Tijburg P, Bach RR, Kisiel W, Stern DM. Tumor necrosis factor-induced endothelial tissue factor is associated with subendothelial matrix vesicles but is not expressed on the apical surface. Blood. 1992;80:966-974.
Wagner DD, Olmsted JB, Marder VJ. Immunolocalization of von Willebrand protein in Weibel-Palade bodies of human endothelial cells. J Cell Biol. 1982;95:355-360.
McEver RP, Beckstead JH, Moore KL, Marshall-Carlson L, Bainton DF. GMP-140, a platelet alpha-granule membrane protein, is also synthesized by vascular endothelial cells and is localized in Weibel-Palade bodies. J Clin Invest. 1989;84:92-99.
Emeis JJ. Regulation of the acute release of tissue-type plasminogen activator from the endothelium by coagulation activation products. In: Brakman P, Kluft C, eds. Plasminogen Activation in Fibrinolysis, in Tissue Remodeling and in Development. Ann N Y Acad Sci. 1992:249-258.
Richardson M, Tinlin S, De Reske M, Webster S, Senis Y, Giles AR. Morphological alterations in endothelial cells associated with the release of von Willebrand factor after thrombin generation in vivo. Arterioscler Thromb. 1994;14:990-999.
Sandset PM, Lund H, Norseth J, Abildgaard U, Ose L. Treatment with hydroxymethylglutaryl-coenzyme A reductase inhibitors in hypercholesterolemia induces changes in the components of the extrinsic coagulation system. Arterioscler Thromb. 1991;11:138-145.
Lupu F, Lupu C, Scully MF, Das S, Kakkar VV. Localisation of tissue factor pathway inhibitor in normal and atherosclerotic vessels. Thromb Haemost. 1995;73:1184. Abstract.
Esmon CT. The roles of protein C and thrombomodulin in the regulation of blood coagulation. J Biol Chem. 1989;264:4743-4746.