Immunoelectron microscopic localization of type 1 plasminogen activator inhibitor on the surface of activated endothelial cells.

Immunogold EM was employed to compare the distribution of type 1 plasminogen activator inhibitor (PAI-1) on the surface of agonist-activated human umbilical vein endothelial cells (HUVECs) with that of control, unactivated cells. As previously observed, (Schleef, R.R., T.J. Podor, E. Dunne, J. Mimuro, and D.J. Loskutoff. J. Cell Biol. 110:155-163), analysis of cross-sections of nonpermeabilized control HUVEC monolayers stained first with affinity-purified rabbit antibodies to PAI-1 and then with gold-conjugated goat anti-rabbit IgG, revealed the presence of relatively few gold particles (less than 1-2% of the total) on the apical cell surface. The majority of gold particles were detected primarily in the extracellular matrix between the culture substratum and the cell membrane. In contrast, treatment of HUVECs with tumor necrosis factor alpha (TNF alpha; 200 U/ml, 24 h) or with lipopolysaccharide (LPS; 10 micrograms/ml, 24 h) resulted in an increased staining of PAI-1 not only in the extracellular matrix, but also on the apical cell surface (10-fold increase). Immunoabsorption of the rabbit anti-PAI-1 with purified PAI-1, or treatment of HUVECs with tissue-type plasminogen activator (2.5 micrograms/ml, 2 h, 4 degrees C) reduced the amount of staining both on the apical surface and in the extracellular matrix of agonist-activated HUVECs by 80-95%. The topographical location of PAI-1 on the cell surface was examined further by coupling immunogold staining with high resolution surface replication. Transmission EM of surface replicas from TNF alpha- or LPS-activated HUVECs revealed a general increase in PAI-1 staining both on planar regions and within indentations of the apical cell surface. Nonactivated HUVECs revealed PAI-1-specific immunogold particles only in areas of exposed extracellular matrix between the cells and occasionally at regions of cell-cell contacts. Analysis of activated bovine aortic endothelial cells by immuno-electron microscopy, immunologic assays, and flow cytometry revealed similar increases in surface PAI-1. These increases in surface PAI-1 could be detected by 3 h and continued over a 24-h period. The expression of PAI-1 on the luminal surface of endothelial cells during immune or inflammatory reactions could reduce endothelial fibrinolytic activity, thus, promoting the localized, pathologic formation of intravascular thrombi.

The majority of type 1 plasminogen activator inhibitor associated with cultured human endothelial cells is located under the cells and is accessible to solution-phase tissue-type plasminogen activator.

The interactions between exogenously added tissue-type plasminogen activator (t-PA) and the active form of type 1 plasminogen activator inhibitor (PAI-1) produced by and present in cultured human umbilical vein endothelial cells (HUVECs) were investigated. Immunoblotting analysis of the conditioned media obtained from monolayers of HUVECs treated with increasing concentrations of t-PA (less than or equal to 10 micrograms/ml) revealed a dose-dependent formation of both t-PA/PAI-1 complexes, and of a 42,000-Mr cleaved or modified form of the inhibitor. Immunoradiometric assays indicated that t-PA treatment resulted in a fourfold increase in PAI-1 antigen present in the conditioned media. This increase did not result from the release of PAI-1 from intracellular stores, but rather reflected a t-PA-dependent decrease in the PAI-1 content of the Triton X-100 insoluble extracellular matrix (ECM). Although the rate of t-PA-mediated release of PAI-1 was increased by the removal of the monolayer, similar quantities of PAI-1 were removed in the presence or absence of the cells. These results suggest that the cells only represent a semipermeable barrier between ECM-associated PAI-1 and exogenous t-PA. Treatment of HUVECs with t-PA (1 microgram/ml, 2 h) to deplete the ECM of PAI-1 did not affect the subsequent rate of PAI-1 production and deposition into the ECM. Immunogold electron microscopy of HUVECs not only confirmed the location of PAI-1 primarily in the region between the culture substratum and ventral cell surface but failed to demonstrate significant (less than 1%) PAI-1 on the cell surface. Thus, the majority of PAI-1 associated with cultured HUVEC monolayers is present under the cells in the ECM and is accessible to solution-phase t-PA.

Regulation of the fibrinolytic system of cultured human vascular endothelium by interleukin 1.

We examined the effects of human interleukin 1 (IL-1) on the production of fibrinolytic components by cultured human vascular endothelium. Conditioned media collected from IL-1-treated (5 U/ml, 24 h) monolayers exhibited decreased tissue-type plasminogen activator (tPA) activity and increased plasminogen activator inhibitor (PAI) activity, as assessed by fibrin and reverse fibrin-autography. Quantitative immunological assays revealed a 35% decrease in tPA antigen and a 360% increase in active PAI antigen, after incubation for 24 h with 0.6 U/ml IL-1. Maximal effects (approximately 50% decrease in tPA antigen; 400-800% increase in active PAI antigen) were observed with 2.5-5 U/ml IL-1. Changes in tPA and PAI reached a maximum at approximately 24 h and persisted for greater than 48 h. IL-1 induction of endothelial procoagulant activity was more rapid and transient, peaking by 6 h and subsiding by 24 h. Natural monocyte-derived IL-1 and two species of recombinant IL-1 had comparable effects. Heat and polymyxin-B treatments differentiated IL-1 actions from those of endotoxin, which promoted similar endothelial alterations. IL-1 effects on endothelial procoagulant and fibrinolytic activities may contribute to the generation and maintenance of fibrin in pathophysiological settings in vivo.

Bleeding diathesis due to decreased functional activity of type 1 plasminogen activator inhibitor.

We evaluated an elderly patient with a lifelong history of severe bleeding after surgery or trauma and with evidence of persistent hyperfibrinolysis. Routine coagulation studies were normal. Serum plasminogen (40%, normal 72-128%) and alpha 2-antiplasmin (55%, normal 70-145%) activities were decreased. Euglobulin clot lysis was abnormally shortened (50 min) and normalized in vitro with epsilon-aminocaproic acid (EACA). The patient was treated with EACA with prompt cessation of bleeding. Patient tissue-plasminogen activator (t-PA) levels in serum were normal (4.7 ng/ml, control 3.5-7.2) as detected by a two-site immunoradiometric assay (IRMA). Patient fibrinolytic inhibitor activities were assessed by incubating 125I-labeled t-PA with either whole blood or serum followed by SDS-PAGE and autoradiography to identify the resultant protease/protease inhibitor complexes. In comparison to blood samples obtained from normal donors, patient plasma and serum demonstrated reduced binding of a fast-acting plasminogen activator inhibitor to 125I-labeled t-PA. Immunoprecipitation experiments indicated diminished complex formation between type 1 plasminogen activator inhibitor (PAI-1) in patient serum and 125I-labeled t-PA. Low patient PAI-1 activity was confirmed in serum (0.36 U/ml, control 0.87-1.81; n = 3) and in platelet lysates using a functional IRMA to quantitate PAI-1 binding to immobilized t-PA. However, patient serum PAI-1 antigen was within the normal range when analyzed by IRMA (31.8 ng/ml, control 19.6-42.2); this result was confirmed in both serum and platelets by Western blot (n = 3). Mixing experiments using purified PAI-1 as well as patient and control sera did not show evidence for an inhibitor against PAI-1. We conclude that this patient's bleeding diathesis was due to hyperfibrinolysis and defective PAI-1. This patient provides the first demonstration of a link between decreased in vivo PAI-1 activity and disordered hemostasis, and supports a role for PAI-1 in control of vivo fibrinolysis.

Modulating the fibrinolytic system of peripheral blood mononuclear cells with adenovirus.

Gene therapy utilizing leukocytes is an unexplored therapeutic strategy for targeting tissue-type plasminogen activator (t-PA) to fibrin and sites of inflammation. In this study, five cationic lipids were observed to enhance the adenovirus (Ad)-mediated expression of t-PA in human peripheral blood mononuclear cells (PBMCs) in a dose-dependent manner between 1000 and 15,000 lipid molecules per Ad particle (efficiency:LipofectAMINE > GenePORTER > Effectene > SuperFect > DMRIE-C). PBMCs treated with Ad/t-PA * LipofectAMINE complexes displayed elevated t-PA expression over a 4-day period and the t-PA-expressing cells facilitated the lysis of plasma clots in vitro. Functional and immunologic assays revealed that the Ad * LipofectAMINE infection protocol did not affect monocyte adhesion in vitro or elevate the expression of procoagulant activity, interleukin 8, or tumor necrosis factor alpha. The potential of this system was documented with an in vivo rat model system that involved the injection of lipopolysaccharide into the peritoneal cavity to induce an inflammatory response. Infusion of Ad/t-PA-infected rat PBMCs into the vasculature of lipopolysaccharide-treated animals was found to increase local fibrinolytic activity by 4-fold. These data provide a framework for utilizing adenovirus to transfer genes into PBMCs.

Recombinant rabbit Fab with binding activity to type-1 plasminogen activator inhibitor derived from a phage-display library against human alpha-granules.

The display of panels of antibody (Ab) fragments on the surface of filamentous bacteriophage offers a way of making Ab with defined binding specificities. Because rabbit Ab are routinely utilized as immunologic probes in a variety of biological techniques, the aim of this study was to design and utilize primers for the amplification of mRNAs encoding rabbit kappa light and gamma heavy chains for the construction of an Ab library from this species. Using the polymerase chain reaction, a diverse Ab library with a repertoire of 2 x 10(7) clones was derived from the spleen and bone marrow of a rabbit that had been immunized with purified human platelet alpha-granules. From this library, specific clones were isolated after three rounds of affinity selection with binding activity to type-1 plasminogen activator inhibitor, a trace protein contained in platelet alpha-granules. These data indicate that recombinant phage-displayed Ab libraries obtained after immunization with complex biological antigens can be employed for the isolation of rabbit monoclonal Fab against specific antigens contained in the biological sample.

Functional display of human plasminogen-activator inhibitor 1 (PAI-1) on phages: novel perspectives for structure-function analysis by error-prone DNA synthesis.

The synthesis of the human plasminogen-activator inhibitor 1 (PAI-1) protein in the cytoplasm of transformed Escherichia coli cells results in inactive protein preparations that can be activated by denaturation and renaturation. We have used the phagemid pComb3, designed for combinatorial immunoglobulin repertoire cloning, for routing of PAI-1 to the periplasm and subsequent exposure on the surface of filamentous phages. Phage-displayed PAI-1 specifically binds to immobilized polyclonal and monoclonal anti-human PAI-1 antibodies. In addition, PAI-1 retains its capacity to form equimolar complexes with its target serine protease tissue-type plasminogen activator (t-PA), as well as its ability to inhibit t-PA activity. Finally, we have explored and manipulated the error-prone property of TaqI DNA polymerase during PCR amplification of the full-length PAI-1 cDNA to generate a large library of predominantly single, random PAI-1 mutants. In addition, a computer simulation program has been devised that converts the number of mutations per codogenic region (in this case PAI-1) into actual mutant proteins. The PAI-1-phage mutant library is composed of 46% single and 34% double mutants and 20% wild-type PAI-1 and can be employed to isolate mutants defective in interactions of PAI-1 with other components. The method described here is applicable to other studies on the structure-function analysis of eukaryotic proteins.

Human glioma U-251 cells contain type 1 plasminogen activator inhibitor in a rapidly releasable form.

Because recent information suggests that the localized deposition of protease inhibitors is one mechanism by which cells regulate pericellular proteolysis during tissue invasion, the distribution of type 1 plasminogen activator inhibitor (PA1-1) associated with the invasive human glioma cell line U-251 was investigated. Direct and reverse fibrin zymography indicated the presence of urokinase-like plasminogen activator (u-PA) and PAI-1 in U-251 conditioned media and cell lysates. PA1-1 antigen was detected immunologically in cytoplasmic granules present within cellular processes of U-251 cells and these organelles could be isolated on Percoll density gradients in a high density band. In contrast, u-PA activity and another secreted protein, amyloid beta-protein precursor, were only present in the low density region of the gradients. Functional analysis of PAI-1 in the granules contained within the high density fractions revealed the presence of active PAI-1. Incubation of U-251 cells with the secretagogue, 8-bromoadenosine 3':5'-cyclic monophosphate, resulted in a 3-fold increase in the release of PAI-1 in the media conditioned by these cells. These data suggest that the human glioma cell line U-251 contains PAI-1 in a rapidly releasable form, which may provide another mechanism by which these tumors could regulate proteolytic activity in a localized manner.

Adenovirus-mediated expression and packaging of tissue-type plasminogen activator in megakaryocytic cells.

Platelets release large quantities of plasminogen activator inhibitor 1 (PAI-1) that plays an important role in maintaining the integrity of fibrin-rich thrombi. In addition, tissue-type plasminogen activator (t-PA), a key physiological regulator of fibrinolysis, has been detected in platelet alpha-granules at low abundance. This information raises the possibility of enhancing t-PA expression in megakaryocytes as a means to enhance the fibrinolytic properties of platelet alpha-granules and target PAs directly to fibrin clots. This study was initiated to investigate adenovirus (Ad)-mediated expression and packaging of t-PA into alpha-granules-like structures in the megakaryocytic cell line MEG-01. Ad/t-PA infection of phorbol myristate acetate (PMA)-differentiated MEG-01 cells increased cellular t-PA levels by 120 fold (1580 +/- 130 ng/10(6) cells at 5 MOI) in comparison to non-or Ad/beta-gal-infected cells. Fluorescence-activated cell sorter (FACS) analysis indicates that Ad/t-PA-infected cells yielded a homogenous shift in the t-PA staining profile with a 4-fold shift in mean fluorescence in comparison to non- or Ad/beta-gal-infected cells. For the isolation of alpha-granule-like structures, MEG-01 cell homogenates were fractionated by differential centrifugation and two consecutive Percoll density gradients. Fibrin autography of storage granules revealed a prominent lytic zone at Mr 66 kD comigrating with free t-PA. Quantitative analyses indicate that a 16-fold elevation in t-PA antigen within storage granules in comparison to non- or Ad/beta-gal-infected cells. To document the ability of t-PA to be stored in a rapidly-releasable form in these cells, we isolated platelet-like particles from the supernatant of differentiated cells and determined that particles from Ad/t-PA-infected cells display a 4-8 fold enhanced secretion of t-PA following treatment with the clasical secretagogue calcium ionphore 23187, ADP, or thrombin. Confocal immunofluoresence microscopy analysis indicates that Ad/t-PA mediated productive expression of t-PA in murine megakaryocytes. These data provide support for the concept of increasing the expression of t-PA in megakaryocytes as a means to alter the hemostatic properties of alpha-granules.

Effect of sodium chloride on limulus amebocyte lysate. Inhibition of endotoxin activation of procoagulase.

Concentrations of sodium chloride up to 3 M increase the time necessary for the clot formation from Limulus amebocyte lysate (LAL) induced with endotoxin. Sodium chloride at a concentration of 4 M prevents clot formation by either precipitation or denaturation of procoagulase. The time necessary for the activation of procoagulase by endotoxin is increased by a change in the sodium chloride concentration from 0.15 M to 0.588 M. No effect on the proteolytic phase or the polymerization phase of the clotting reaction is detected by the increase in sodium chloride concentration from 0.15 M to 0.588 M. The authors conclude that increased sodium chloride concentrations may aid the isolation of procoagulase.

Characterization of two monoclonal antibodies against human tissue-type plasminogen activator.

A series of hybridoma clones, each producing monoclonal antibodies to human tissue-type plasminogen activator (t-PA), were prepared from mice by standard procedures. Two of these clones were selected for further study. One HI72A1, produced antibodies that bound to t-PA and strongly inhibited its activity, whereas another, LI72D1, produced antibodies that bound to t-PA but did not affect its activity. The specificity of these antibodies was assessed in immunoabsorption experiments. Both immunoprecipitated 125I-labeled t-PA, and both were specific since only t-PA was recognized in conditioned media collected from Bowes melanoma cells cultured in the presence of 3H-leucine. Neither antibody recognized urokinase. t-PA was desorbed from antibody HI72A1-Sepharose columns with 0.5 M NaCl, consistent with its relatively low association constant (Ka = 9.37 X 10(7) M-1). In contrast, a strong chaotropic agent (i.e., 2 M KI) was required to elute t-PA from antibody LI72D1 columns (Ka = 2.08 X 10(9) M-1). This latter high affinity antibody was employed to develop an immunoradiometric assay for t-PA having a sensitivity of 0.5 ng/ml.

A monoclonal antibody that does not recognize tissue-type plasminogen activator bound to its naturally occurring inhibitor.

Hybridoma clones specific for tissue-type plasminogen activator (tPA) were screened in solid-phase radioimmunoassays for their reactivity toward free tPA and tPA complexed to the endothelial cell-derived, beta-migrating plasminogen activator inhibitor (beta-PAI). Two monoclonal antibodies (MABs) were identified with quite distinct properties. The first, MAB LI72D1, bound to free tPA but did not recognize tPA complexed to the beta-PAI, whereas the second, MAB HI72C1, bound both to free tPA and to tPA in complex with beta-PAI. These properties were maintained when the MABs were immobilized to plastic microtiter wells, thus permitting the development of immunoradiometric assays (IRMAs) to quantitate free tPA and total tPA antigen in various samples. The IRMAs were employed to analyze the tPA in media conditioned by several human cell types. The results indicate that in some cases, tPA may be present entirely as a free and active enzyme, while in others it apparently exists entirely in complex with beta-PAI. Interestingly, some cells appear to contain both forms of tPA.

The dexamethasone-induced inhibitor of plasminogen activator in hepatoma cells is antigenically-related to an inhibitor produced by bovine aortic endothelial cells.

Glucocorticoids decrease plasminogen activator (PA) activity in HTC rat hepatoma cells by inducing a specific inhibitor of PA activity (PAI). This inhibitor is similar in several biochemical properties to the PAI purified from bovine aortic endothelial cells (BAEs). We have used reverse fibrin autography and antiserum against BAE PAI to establish more fully the biochemical and immunological relationship of these inhibitors. Both inhibitors migrated with an apparent Mr of approximately 50,000, and the activity of both PAIs was stimulated by treatment with SDS suggesting that each of these molecules exists in both an active and a latent form. Antiserum to the BAE PAI immunoprecipitated all of the HTC PAI demonstrable by reverse fibrin autography. Finally, using this antiserum in a functional immunoassay, we have demonstrated that dexamethasone increases both active and latent PAI made by HTC cells. These results indicate that HTC PAI and BAE PAI are antigenically as well as biochemically related molecules.

Prothrombotic activation of pulmonary arterial endothelial cells in a patient with tuberculosis.

Activation of endothelial cells occurs in response to numerous physiological stimuli and results in the concerted expression of endothelial cell proteins that change the nonthrombogenic intimal surface of a vessel into a thrombogenic surface, with the subsequent development of local thrombosis. For example, both type 1 plasminogen activator inhibitor and tissue factor expression are mediated by endothelial cell stimulation in vitro; however, in contrast to type 1 plasminogen activator inhibitor, it has been difficult to detect tissue factor associated with endothelial cells in vivo. This case study describes the presence of both type 1 plasminogen activator inhibitor and tissue factor antigen associated with pulmonary arterial endothelial cells of a patient exhibiting a mycobacterial infection. The disease was associated with chronic hemoptysis and characterized by extensive tissue destruction and local thrombosis within the pulmonary artery. The data show that conditions occur in vivo in which local thrombosis is associated with increased levels of type 1 plasminogen activator inhibitor and tissue factor.

The effect of fibrin on endothelial cell migration in vitro.

Endothelial cells are known to migrate and come into contact with fibrin during numerous physiological processes, such as in wound healing and in tumor growth. The present study was initiated to investigate the effect of fibrin on endothelial cell migration in vitro. Endothelial cell migration was assayed by wounding confluent monolayers of bovine aortic endothelial cells with a razor blade and counting the number of cells crossing the wound per unit time. Wound-induced proliferation of endothelial cells was inhibited by mitomycin C-treatment without affecting endothelial cell migration, indicating that in this assay migration could be measured independent of proliferation. Migration of endothelial cells in vitro was inhibited by fibrin in a concentration dependent manner. Endothelial cell migration under fibrin was further reduced by plasminogen depletion of the serum, and fibrin still inhibited the migration of mitomycin C-treated endothelial cells. Kadish et al. (Tissue and Cell, 11, 99, 1979) previously reported that fibrin did not affect EC migration in vitro. The inability to inhibit EC migration with fibrin appears to be due to their assay system which employed agarose, since pre-treating the wounded monolayer with agarose eliminated the inhibition of EC migration by fibrin. The present results indicate that EC migration in vitro can be used as a model system for studying the interaction of fibrin with EC.

Fibrinolytic system of vascular endothelial cells. Role of plasminogen activator inhibitors.

The regulation of the fibrinolytic system is of critical importance during hemostasis, wound repair, neoplasia, inflammation, and a variety of other biologic processes. This control is achieved in a large part through the action of specific plasminogen activator inhibitors (PAIs). Cultured endothelial cells (ECs) produce type 1 PAI (PAI-1), the physiologic inhibitor of tissue-type plasminogen activator. PAI-1 is one of the most highly regulated of the fibrinolytic components produced by ECs. Its synthesis is modulated by a variety of compounds including endotoxin, thrombin, transforming growth factor beta interleukin 1, and tumor necrosis factor alpha. Recent studies suggest that PAI-1 is synthesized by ECs as an active molecule, but that it spontaneously decays into a latent form in solution. Specific components present in the extracellular matrix of ECs and in plasma bind to PAI-1 and prevent this inactivation. The unexpected finding that cultured ECs also produce type 2 PAI (PAI-2) introduces a previously unsuspected level of complexity to our understanding of this system and raises the possibility that the altered fibrinolytic activity of ECs following various treatments, or of blood in certain individuals, may reflect changes in either one of these inhibitors.

Recombinant Semliki Forest virus enhanced plasminogen activator inhibitor 1 expression and storage in the megakaryocytic cell line MEG-01.

Platelet plasminogen activator inhibitor I (PAI-1), a trace alpha-granule protein, is a key physiological regulator of fibrinolysis. Because information on the packaging of PAI-1 into alpha-granules during megakaryocytopoiesis may reveal novel approaches for controlling hemostasis, this study investigated basal, plasmid-mediated, and alphavirus-mediated PAI-1 packaging into alpha-granules-like structures in the megakaryocytic cell line MEG-01. Differentiation of MEG-01 cells with phorbol myristate acetate (PMA) was observed to result in a four-fold increase in both secreted and cell-associated PAI-1 antigen over a four day period. Subcellular fractionation of PMA-treated MEG-01 cells on 45% self-forming Percoll gradients was employed to separate low density membrane and Golgi-rich fractions from a high density granule-containing region. A subsequent 30-60% pre-formed Percoll gradient was employed to remove contaminating lysosomes from the PAI-1/glycoprotein IIbIIIa-containing granules. Electron microscopy showed that these MEG-01 granules share a similar size distribution (350-600 nm) and morphology to platelet alpha-granules. PAI-1 (40 ng/mg protein) in isolated MEG-01 storage granules was approximately 10% of the levels present in isolated platelet alpha-granules. To elevate PAI-1 production/storage, two expression systems were investigated. Experiments with plasmids encoding PAI-1 and beta-galactosidase resulted in low transfection efficiency (0.001%). In contrast, Semliki Forest virus (SFV)-mediated gene transfer increased cellular PAI-1 by 31-fold (1,200 ng/10(6) cells at 10 MOI) in comparison to mock-infected cells. Pulse-chase experiments demonstrated that SFV/PAI-1 mediated gene expression could enhance PAI-1 storage 6-9-fold, reaching levels present within platelets. To document the ability of PAI-1 to be stored in a rapidly releasable form in MEG-01 cells, we isolated platelet-like particles from the media conditioned by the cells and examined secretagogue-induced release of PAI-1. Particles from SFV/PAI-1 infected cells display a 5-fold enhanced secretion of PAI-1 following treatment with ADP in comparison to particles incubated in the absence of secretagogue. These results suggest that SFV mediated gene expression in MEG-01 cells provides a useful framework for analyzing the production and storage of alpha-granule proteins.

Two functionally distinct pools of vitronectin (Vn) in the blood circulation: identification of a heparin-binding competent population of Vn within platelet alpha-granules.

The biological functions of vitronectin (Vn) are dependent on its conformation. Whereas plasma Vn is present in a conformation that does not bind to heparin, platelet Vn has been recognized to be in a multimeric, conformationally altered form. To further understand the characteristics of platelet Vn, the molecules present in plasma and total and size-fractionated platelet releasates were compared (1) immunologically using three conformationally sensitive epitope-defined monoclonal antibodies, (2) functionally for their ability to interact with heparin, and (3) structurally using denaturing and nondenaturing polyacrylamide gel electrophoresis (PAGE). Our data indicate that Vn is present in platelet releasates in two molecular weight (M(r) forms. The high M(r) fractions contain conformationally and structurally altered Vn capable of interacting with heparin, and this form is distinct from plasma Vn and purified denatured Vn. In contrast, the lower M(r) forms of Vn are similar to plasma Vn. To determine if the presence of multimeric Vn requires platelet activation, platelets were disintegrated by sonication and fractionated by density gradients. Combined sodium dodecyl sulfate-PAGE (SDS-PAGE) and immunoblotting analysis showed a codistribution of multimeric Vn and type 1 plasminogen activator inhibitor in alpha-granule-rich fractions. Thus, platelet Vn is stored in a structurally and functionally distinct form from the molecule in plasma, raising the possibility that platelet-derived heparin-binding competent Vn will accumulate in areas of vascular injury.