HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 36, 32677-32682, September 6, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/36/32677    most recent M204010200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chavakis, T. Articles by Preissner, K. T. Search for Related Content PubMed PubMed Citation Articles by Chavakis, T. Articles by Preissner, K. T. Social Bookmarking                      What's this?

A Novel Antithrombotic Role for High Molecular Weight Kininogen as Inhibitor of Plasminogen Activator Inhibitor-1 Function^*

Triantafyllos Chavakis^§¶, Robin A. Pixley, Irma Isordia-Salas, Robert W. Colman, and Klaus T. Preissner

From the  Institute for Biochemistry, ^§ Third Department of Internal Medicine, Justus Liebig University, Giessen D-35392, Germany, and the  Sol Sherry Thrombosis Research Center, Temple University School of Medicine, Philadelphia, Pennsylvania 19140

Received for publication, April 25, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The adhesive glycoprotein vitronectin (VN) forms a function-stabilizing complex with plasminogen activator inhibitor-1 (PAI-1), the major fibrinolysis inhibitor in both plasma and vessel wall connective tissue. VN also interacts with two-chain high molecular weight kininogen (HKa), particularly its His-Gly-Lys-rich domain 5, and both HKa and PAI-1 are antiadhesive factors that have been shown to compete for binding to VN. In this study the influence of HKa and domain 5 on the antifibrinolytic function of PAI-1 was investigated. In a purified system, HKa and particularly domain 5 inhibited the binding of PAI-1 to VN and promoted PAI-1 displacement from both isolated VN as well as subendothelial extracellular matrix-associated VN. The sequence Gly⁴⁸⁶-Lys⁵⁰² of HKa domain 5 was identified as responsible for this inhibition. Although having no direct effect on PAI-1 activity itself, HKa domain 5 or the peptide Gly⁴⁸⁶-Lys⁵⁰² markedly destabilized the VN·PAI-1 complex interaction, resulting in a significant reduction of PAI-1 inhibitory function on plasminogen activators, resembling the effect of VN antibodies that prevent stabilization of PAI-1. Furthermore, high affinity fibrin binding of PAI-1 in the presence of VN as well as the VN-dependent fibrin clot stabilization by the inhibitor were abrogated in the presence of the kininogen forms mentioned. Taken together, our data indicate that the peptide Gly⁴⁸⁶-Lys⁵⁰² derived from domain 5 of HKa serves to interfere with PAI-1 function. Based on these observations potential low molecular weight PAI-1 inhibitors could be designed for the use in therapeutic interventions against thromboembolic complications.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

After vascular injury, pathological thrombosis is the critical event leading to acute myocardial infarction, unstable angina pectoris, or stroke (1). Physiological thrombus formation is balanced between the procoagulant pathway involving platelets and coagulation factors as well as the anticoagulant and fibrinolytic systems including intrinsic thrombin inhibition and clot dissolution. Intravascular fibrinolysis is initiated primarily by the protease tissue-type plasminogen activator (tPA),¹ which generates plasmin from its inactive precursor plasminogen in a fibrin-specific manner (2). Regulation of plasminogen activation in a fibrin-specific fashion is achieved by plasminogen activator inhibitor-1 (PAI-1), which forms a stable inactive complex with tPA (3). The majority of clot-responsive PAI-1 is provided during platelet degranulation and accumulates within the forming thrombus rendering it initially resistant to fibrinolysis (4-6). An elevated PAI-1 level constitutes an important thrombotic risk factor for, e.g. myocardial infarction (7, 8) or deep venous thrombosis because of an overall increased antifibrinolytic potential (9). In contrast, PAI-1 deficiency in humans is associated with abnormal bleeding, indicating an important role of PAI-1 in stabilization of the hemostatic clot (10-12).

Only in complex with vitronectin (VN), an abundant plasma and wound matrix-associated glycoprotein, PAI-1 becomes functionally stabilized by high affinity binding within the N-terminal region of the multifunctional adhesive glycoprotein (13-17). Moreover, VN plays a critical role in PAI-1 binding to fibrin (18). It is believed that PAI-1 may also induce conformational changes in VN leading to multimerization (19, 20). Finally, PAI-1 serves as an important antiadhesive factor for VN-dependent cell adhesion reactions involving either integrins or the urokinase receptor (21, 22).

Similar to PAI-1, the six domain-containing high molecular weight kininogen (HK), and especially the kinin-free two-chain form (HKa) (23), binds to VN and serves as an additional antiadhesive protein for both integrin- and urokinase receptor-dependent cellular interactions (24). The antiadhesive function of HKa is mediated by the His-Gly-Lys-rich domain 5, enabling HKa also to bind to anionic or cellular surfaces, to zinc or heparin (25-29). We have demonstrated previously that HKa and PAI-1 compete for proximal binding sites within the N-terminal "somatomedin B" domain of VN (24). Experimental evidence from in vitro and in vivo studies renders HK anti- rather than prothrombotic; a prothrombotic phenotype was reported for kininogen-deficient rats (30), and patients deficient in kininogen or other proteins of the contact phase system are at increased risk for thrombosis (31-34). In particular, HK/HKa may regulate pericellular plasmin generation by modulating plasma kallikrein-dependent formation of urokinase plasminogen activator (uPA) (35), a reaction dependent on the binding of plasma prekallikrein to HK domain 6 (36).

The competition between HKa and PAI-1 for proximal binding sites on VN prompted us to investigate whether HKa can interfere with the VN-stabilized functions of PAI-1. Our results indicate that HKa, domain 5, and especially a particular peptide Gly⁴⁸⁶-Lys⁵⁰² can inhibit the function of PAI-1 as a primary inhibitor of fibrinolysis thereby providing a novel plausible mechanism for the recently described antithrombotic properties of kininogen. Finally, based on the sequence Gly⁴⁸⁶-Lys⁵⁰² within domain 5 of HKa, potential low molecular weight PAI-1 inhibitors could be designed for therapeutic interventions in thrombotic vascular complications.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Reagents-- Single- and two-chain high molecular weight kininogen (HK and HKa) were purchased from Enzyme Research Laboratories (South Bend, IN). The purified HK and HKa (>95%) appeared as major bands of 140 and 110 kDa, respectively, on nonreduced SDS-gels. HK was digested with plasma kallikrein (HK:kallikrein, 100:1, mol/mol) for 20 min at 37 °C to obtain HKa, which appears as two bands of 62 and 46 kDa when analyzed by reduced SDS-gel electrophoresis. Glutathione S-transferase (GST) fused to domain 3, 5, or 6 of HK or to sequences derived from domain 5 were produced as described previously (37). GST was N-terminally attached to the following sequences of HK: Gly²³⁵-Met³⁵⁷ (domain 3), Lys⁴²⁰-Ser⁵¹³ (domain 5), Thr⁵⁰³-Ser⁶²⁶ (domain 6), as well as Lys⁴²⁰-Asp⁴⁷⁴ and His⁴⁷⁵-Lys⁵⁰² (N-terminal and C-terminal regions of domain 5) and purified on a glutathione column reaching more than 90% purity. Synthesis of peptides and purification by high performance liquid chromatography to more than 95% purity were performed by Dr. J. Lambris (University of Pennsylvania, Philadelphia). Refolding of cysteine-containing peptides was carried out by air oxidation for 3 days at 4 °C with continuous agitation in buffer containing 50 mmol/liter ammonium bicarbonate, pH 8.5, at a final concentration of 100 µg/ml followed by freeze drying. In addition to peptides HK406 (Gly⁴⁰⁶-Asn⁴²²), HK440 (Gly⁴⁴⁰-His⁴⁵⁵), HK475 (His⁴⁷⁵-His⁴⁸⁵), HK483 (His⁴⁸³-Gly⁴⁹⁷), and HK486 (Gly⁴⁸⁶-Lys⁵⁰²), the following scrambled peptides were used: HK483M (GHHGHKKNGKKKGNK) and HK486M (KGHKKNGKKNKGNHWGK). The properties of all peptides have been described previously (38, 39).

VN was purified from human plasma and converted to the multimeric form as described previously (40, 41). Fibrinogen was purchased from Kabivitrum (Munich, Germany). Active PAI-1 and murine monoclonal antibodies 13H1 (against human VN) and 12A4 (against PAI-1) were kindly provided by Dr. P. Declerck (University of Leuven, Belgium) (41, 42). tPA was from Roche Molecular Biochemicals, uPA was from Medac (Hamburg, Germany), and plasminogen was purified from human plasma by lysine-Sepharose adsorption. The chromogenic substrate S-2251 was from Chromogenix (Mölndal, Sweden). Thrombin, CaCl₂, and ZnCl₂ were from Sigma, and peroxidase-conjugated secondary anti-mouse and anti-rabbit immunoglobulins were from DAKO (Hamburg, Germany).

Cell Cultures-- Cultures of human umbilical vein endothelial cells (HUVEC) were established, characterized, and grown exactly as described previously (43). All cell culture media were from Invitrogen.

In Vitro Binding Assay-- Maxisorp plates (high binding capacity; Nunc) were coated with 5 µg/ml multimeric VN (dissolved in bicarbonate buffer, pH 9.6) for 16 h at 4 °C and then blocked with 3% (w/v) BSA. Binding of 2 µg/ml PAI-1 was performed in a final volume of 50 µl of Tris-buffered saline containing 0.1% Tween 20, 0.3% (w/v) BSA, for 2 h at 22 °C in the absence or presence of 10 µM ZnCl₂ and different competitors as indicated in the figure legends. After the incubation period and a washing step, bound PAI-1 was quantitated by anti-human PAI-1 monoclonal antibody 12A4 followed by the addition of 1:1,000 diluted peroxidase-conjugated goat anti-mouse IgG. The conversion of the substrate 2,2'-azino-bis(3-ethylbenzthiazoline)-6-sulfonic acid (Roche) was monitored at 405 nm in a Thermomax microtiter plate reader (Molecular Devices, Menlo Park, CA). Nonspecific binding to BSA-coated wells was used as blank and was subtracted to calculate the specific binding. To test PAI-1 displacement from VN, binding of the inhibitor was performed as described above, and after unbound PAI-1 had been washed off, different competitors were incubated for 2 h in the wells, as indicated in the figure legends, followed by quantitation of bound PAI-1 as described above.

PAI-1 Binding to Subendothelial Extracellular Matrix of HUVEC-- The preparation of the extracellular matrix was performed exactly as described previously (43, 44). Briefly, HUVEC were grown to confluence in 96-well plates and washed three times with phosphate-buffered saline containing 2% (w/v) BSA and 1 mM CaCl₂. Cells were removed with phosphate-buffered saline containing 0.5% (w/v) Triton X-100 for 15 min followed by incubation with 0.5% (w/v) Triton X-100 and 0.1 M NH₃ for another 15 min at 22 °C. PAI-1-depleted extracellular matrix was obtained by subsequent incubation with 0.1 M glycine, pH 2.3, for 2 h at 25 °C. Finally, wells were washed with phosphate-buffered saline and blocked with phosphate-buffered saline containing 3% (w/v) BSA before proceeding with binding assays. The extracellular matrix was then incubated without or together with different competitors, as indicated in the figure legends, and the remaining PAI-1 was detected as described above. Binding of PAI-1 to PAI-1-depleted extracellular matrix was performed exactly as described above.

Plasminogen Activation Assay-- 0.1-40 nM PAI-1 or buffer alone was incubated without or together with 1 µg/ml multimeric VN in the absence or presence of different competitors and for different time periods (as indicated in the figure legends) in Tris-buffered saline containing 0.1% (w/v) Tween 20 and 0.1% (w/v) BSA at 37 °C. After different time intervals, aliquots of reaction mixtures were transferred into microtiter wells, and uPA or tPA (concentrations indicated in the figure legends) was added. After an incubation period of 30 min at 22 °C, a mixture of 0.22 µM plasminogen (final concentration) and 0.8 mM substrate S-2251 was added, and the rate of plasmin generation was followed for 1 h at 405 nm.

tPA-mediated Plasmin Generation on Fibrin Matrices-- To examine the influence of different HK forms on tPA-dependent plasmin generation on fibrin matrices, experiments were performed according to a protocol published previously (18). In microtiter wells fibrin was generated by mixing aliquots of a 375 nM fibrinogen solution containing 0.5 µg/ml multimeric VN or buffer alone together with 30 nM thrombin and 1 mM CaCl₂ in a total volume of 50 µl. After incubation for 3 h at 22 °C, the fibrin-coated plates were stored at 4 °C until use. To block nonspecific binding, phosphate-buffered saline containing 3% (w/v) BSA and 0.05% (w/v) Tween 20 (blocking buffer) was added to the wells. The fibrin-coated wells were then preincubated for 1 h at 37 °C with 0.01-100 nM PAI-1 diluted in phosphate-buffered saline containing 1% (w/v) BSA, 0.1% (w/v) Tween 20 in the absence or presence of different competitors as indicated in the figure legends. After washing three times, the wells were incubated with 2.5 nM tPA, 0.54 µM plasminogen, and 1 mM S-2251 at 37 °C, and the rate of plasmin generation was followed at 405 nm.

Lysis of Purified Fibrin Clots-- Lysis of fibrin clots was studied according to a protocol described previously based on a fluorogenic assay (45, 46). Briefly, a solution of fluorescein isothiocyanate-labeled fibrinogen (100 nM) and plasminogen (50 nM), diluted in a buffer containing 0.05 M Tris-HCl, 0.15 M NaCl, and 5 mM CaCl₂, 10 µM Zn²⁺, pH 7.4, was mixed with 5 nM thrombin. Fibrin clots were also formed in the presence of 1-100 nM PAI-1 and/or 1 µg/ml VN. After clots were allowed to age for 2 h at 37 °C, they were then incubated with 2.5 nM tPA or Tris-buffered saline alone at 37 °C. The extent of tPA-induced clot lysis was quantified by following the increase in fluorescence in a fluorescent plate reader (TECAN Spectrafluor, Crailsheim, Germany) compared with base-line fluorescence of clots incubated with Tris-buffered saline alone.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Peptide Gly⁴⁸⁶-Lys⁵⁰² of HK Domain 5 Inhibits the PAI-1/VN Interaction-- We have demonstrated previously that PAI-1 competes with the Zn²⁺-dependent binding of HKa to VN, which is attributed to HK domain 5. Here, peptides derived from this domain were used to identify further the sequence epitope on HK responsible for inhibition of the PAI-1/VN interaction. Only the C-terminal portion of the domain 5, His⁴⁷⁵-Lys⁵⁰² fused to GST, but not the N-terminal portion of domain 5, Lys⁴²⁰-Asp⁴⁷⁴ fused to GST, could reproduce the function of the whole domain 5 at 10 µM Zn²⁺ (minimal required concentration). Two of five synthetic peptides derived from domain 5, namely HK486 (sequence Gly⁴⁸⁶-Lys⁵⁰²) and HK483 (sequence His⁴⁸³-Gly⁴⁹⁷) to a lower extent inhibited the binding of PAI-1 to VN, whereas peptides HK406 (Gly⁴⁰⁶-Asn⁴²²), HK440 (Gly⁴⁴⁰-His⁴⁵⁵), and HK475 (His⁴⁷⁵-His⁴⁸⁵) were not inhibitory. The sequence specificity of the peptides HK486 and HK483 containing a high proportion of Gly/His/Lys was proven by comparing them with the respective scrambled peptides, HK486M and HK483M, which exerted substantially reduced inhibitory activity (Table I).

To investigate whether preformed PAI-1·VN complexes can be dissociated by different forms of HK, PAI-1 was allowed to bind to immobilized VN followed by incubation with various HK-based competitors. Consistent with the competition studies shown before, in a concentration-dependent manner, HKa domain 5 and peptide HK486 but not HK or the scrambled peptide HK486M were able to release bound PAI-1 from VN to an appreciable degree (Fig. 1). Incubation of HUVEC subendothelial extracellular matrix with HKa domain 5 or the peptide HK486 resulted in partial PAI-1 dissociation from the matrix, whereas HK domain 3 or the scrambled peptide HK486M had no effect (data not shown). Finally, HKa domain 5 and the peptide HK486 also inhibited the binding of PAI-1 to PAI-1-depleted extracellular matrix (Fig. 2), an effect that was comparable with the inhibition provided by the antibody 13H1 against VN, described previously to block the PAI-1/VN-interaction. These data demonstrate that the sequence Gly⁴⁸⁶-Lys⁵⁰² of HK domain 5 entails an epitope that inhibits the PAI-1 binding to both isolated and endothelial extracellular matrix associated-VN and also promotes the displacement of PAI-1 from VN or the extracellular matrix.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Displacement of PAI-1 from VN by different HK forms. 2 µg/ml PAI-1 was allowed to bind onto immobilized VN for 2 h. After removing unbound PAI-1, different concentrations of HK (open circles), HKa (filled circles), GST-domain 5 (open squares), peptide HK486 (open triangles), or the scrambled control peptide HK486M (filled triangles) were allowed to react in the presence of 10 µM ZnCl2. The remaining PAI-1 was detected by the monoclonal antibody 12A4 as described under "Materials and Methods." Specific binding of PAI-1 is shown as a percent of control, which is represented by the binding of PAI-1 to VN in the absence of competitors. Data represent the mean ± S.E. (n = 3) of a typical experiment; similar results were obtained in three separate experiments.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Influence of kininogen on the binding of PAI-1 to VN in PAI-1-depleted HUVEC extracellular matrix. The binding of 2 µg/ml PAI-1 to PAI-1-depleted (2-h glycine preincubation) HUVEC extracellular matrix was tested in the absence () or presence of the monoclonal antibody 13H1 against VN (25 µg/ml), HK, HKa, GST-domain 3 (D3), GST-domain 5 (D5), the peptides HK483 and HK486, or the scrambled control peptides HK483M and HK486M (each 500 nM) in the presence of 10 µM ZnCl2. Specific binding of PAI-1 is presented as the absorbance at 405 nm, and data represent mean ± S.E. (n = 3) of a typical experiment; similar results were obtained in three separate experiments.

Inhibition of PAI-1 Function by HK Domain 5-- Only stabilized by VN, PAI-1 provides an efficient inhibition of tPA- and uPA-induced plasminogen activation. Incubation of PAI-1 alone for 3 h at 37 °C results in a complete loss of inhibitory activity, a phenomenon that can be prevented by coincubation of PAI-1 with 1 µg/ml multimeric VN (data not shown and Ref. 18). The VN-dependent inhibitory activity of PAI-1 was attenuated markedly in the presence of HKa, whereas HK had no effect at all, and the effect of HKa was reminiscent of the inhibition by the monoclonal antibody 13H1 against VN. Again, domain 5 and especially the peptide HK486 could duplicate the inhibitory activity of HKa (Table II and Fig. 3). A 10-fold molar excess of HK forms over PAI-1 was required to inhibit almost completely VN-dependent PAI-1 antifibrinolytic activity (Fig. 3). In the absence of VN, HKa domain 5 or the peptide HK486 had no effect on activity of PAI-1. The different HK forms also did not affect the activity of uPA or tPA, respectively. Thus, the inhibition of PAI-1 function by HKa is the result of dissociation of PAI-1 from VN and thereby leads to rapid inactivation of the inhibitor.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Interference of kininogen with the antifibrinolytic activity of PAI-1. A, uPA-induced plasmin formation was tested without or together with increasing concentrations of PAI-1 in the absence (filled circles) or presence of 1 µg/ml multimeric VN alone (open squares) or together with HKa (filled squares), GST-domain 5 (open circles), peptide HK486 (open triangles) (each 500 nM), or the monoclonal antibody 13H1 against VN (30 µg/ml) (filled triangles) in a buffer containing 10 µM Zn2+. PAI-1 alone or PAI-1 together with VN in the absence or presence of the different competitors was preincubated for 3 h at 37 °C. B, uPA-induced plasmin formation was tested in the presence of 40 nM PAI-1 and 1 µg/ml multimeric VN without or together with increasing concentrations of HK (open squares), HKa (filled squares), GST-domain 5 (open circles), the peptide HK486 (open triangles), or the control scrambled peptide HK486M (filled triangles) in a buffer containing 10 µM Zn2+. uPA-induced plasmin formation in the absence of any competitors (hatched circle) or in the presence of PAI-1 alone (filled circle) is also shown. PAI-1 alone or PAI-1 together with VN in the absence or presence of the different HK forms was preincubated for 3 h at 37 °C. The rate of plasmin formation is expressed as Vmax (mOD/min). Data represent the mean ± S.E. (n = 3) of a typical experiment; similar results were obtained in three separate experiments.

Because VN has also been shown to mediate the binding of PAI-1 to fibrin, we examined whether different forms of HK could inhibit PAI-1 function in this experimental setting. Fibrin clots formed in the absence or presence of VN were incubated for 1 h with PAI-1 without or together with different HK forms; after removing unbound PAI-1, tPA was added, and the rate of plasmin formation was examined. Only in the presence of VN could fibrin-bound PAI-1 inhibit subsequent plasminogen activation by tPA. The VN-dependent stabilization of PAI-1 activity was blocked by HKa domain 5 or the peptide HK486, supporting the profibrinolytic effect of these HK forms (Fig. 4).

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Influence of kininogen on the antifibrinolytic activity of PAI-1 on fibrin matrices. tPA-induced plasmin formation was tested in microtiter wells coated with fibrin formed in the absence (open bars) or presence of VN (filled bars). Prior to the addition of tPA, the wells were incubated without or together with 50 nM PAI-1 in the absence () or presence of HK, HKa, GST-domain 3 (D3), GST-domain 5 (D5), the peptide HK486, the scrambled peptide HK486M (each 500 nM), or the monoclonal antibody 13H1 against VN (30 µg/ml) in a buffer containing 10 µM Zn2+ at 37 °C. After incubation for 1 h, unbound PAI-1 was washed off, and the rate of tPA-induced plasmin formation was followed at 405 nm. The rate of plasmin formation is expressed as Vmax (mOD/min). Data represent mean ± S.E. (n = 3) of a typical experiment; similar results were obtained in three separate experiments.

To examine further the influence of kininogen on the activity of the PAI-1·VN complex in fibrin clots, tPA-mediated lysis of fluorescein-fibrinogen-labeled clots was monitored. PAI-1 incorporated into the clots had a minimal inhibitory effect on tPA-mediated clot lysis unless VN was present. When both PAI-1 and VN were incorporated into the fibrin clots, tPA-mediated clot lysis was reduced to 15-20% of control. HKa domain 5 or the peptide HK486 promoted clot lysis in two different experimental settings. (i) Formation of the clots in the presence of HKa domain 5 or the peptide HK486 markedly attenuated the antifibrinolytic effect of the PAI-1·VN complex. Clot lysis in the presence of these kininogen forms was enhanced from 15-20% to 80-85% (Fig. 5A). Peptide HK483 had a weaker effect compared with peptide HK486 (not shown), whereas the respective scrambled peptides (HK486M and HK483M) were hardly effective (HK486M is shown in Fig. 5A). HK domains 3 and 6 and the peptide HK406, HK440, or HK475 did not affect the PAI-1·VN-mediated inhibition of clot lysis. The incorporation of the different kininogen forms alone or together with PAI-1 into the fibrin clots in the absence of VN did not alter tPA-mediated clot lysis (not shown). The PAI-1·VN complex-mediated inhibition of clot lysis was also reduced if clots were formed in the presence of the monoclonal antibody 13H1 against VN. (ii) In the second setting, the influence of the different kininogen forms on clot lysis was tested when they were added together with tPA after the clots were allowed to age. In this case, HKa domain 5 or the peptide HK486 partially reversed the PAI-1·VN-mediated inhibition of clot lysis from 15-20% up to 50-55% (Fig. 5B) without affecting tPA-mediated clot lysis in the absence of PAI-1 and VN. These data are in accordance with the dissociation of a preformed PAI-1·VN complex by these kininogen forms as reported in Fig. 1. Interestingly, the anti-VN monoclonal antibody 13H1 was hardly potent in this experimental setting. Taken together, these findings strongly indicate that the peptide Gly⁴⁸⁶-Lys⁵⁰² of HK domain 5 can inhibit the VN-dependent antifibrinolytic effect of PAI-1.

View larger version (34K): [in this window] [in a new window]   Fig. 5.   Neutralization of VN-dependent PAI-1-mediated inhibition of fibrin clot lysis by kininogen. A, tPA-mediated lysis of purified fibrin clots, which were formed without (gray bar, 100% control) or together with 1 µg/ml VN (dotted bar), 30 nM PAI-1 (hatched bar), or both (filled bars) in the absence () or presence of HK, HKa, GST-domain 3 (D3), GST-domain 5 (D5), the peptides HK486 and HK486M (each 250 nM), or the monoclonal antibody 13H1 against VN (30 µg/ml), in a buffer containing 10 µM Zn2+. B, tPA-mediated lysis of purified fibrin clots that were formed without (gray bar) or together with 1 µg/ml VN (dotted bar), 30 nM PAI-1 (hatched bar), or both (filled bars). HK, HKa, GST-domain 3 (D3), GST-domain 5 (D5), the peptides HK486 and HK486M (each 250 nM), or the monoclonal antibody 13H1 against VN (30 µg/ml), respectively, was added together with tPA after the clots were allowed to age for 2 h at 37 °C. Data are expressed as a percent of clot lysis mediated by tPA alone and represent the mean ± S.E. (n = 3) of a typical experiment; similar results were obtained in three separate experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

PAI-1 is a central regulator of the fibrinolytic system, and increased levels of the inhibitor result in reduced fibrinolytic activity associated with a prothrombotic state. Thus, PAI-1 has been regarded in several studies as an important risk factor for thromboembolic complications (7-9, 47). Because VN has a major impact on stabilizing the activity of PAI-1 (13-16), any disturbance of PAI-1·VN complexes may lead to a rapid decay of the inhibitor associated with a decrease of antifibrinolytic function. In this report we define novel functions of the domain 5 of HK and especially the peptide Gly⁴⁸⁶-Lys⁵⁰², which can either inhibit the formation of or dissociate the PAI-1·VN complex and thereby contribute to a diminution of PAI-1 activity. This profibrinolytic function of HKa together with its previously described antiadhesive (24), anti-inflammatory (38), and platelet aggregation inhibitory properties (48) defines kininogen as a multifunctional endogenous antithrombotic factor involved in the regulation of the hemostatic balance.

Although having no direct effect on PAI-1 activity itself, HKa domain 5 or the peptide Gly⁴⁸⁶-Lys⁵⁰² markedly attenuated the VN-dependent inhibition of uPA- or tPA-mediated plasminogen activation by PAI-1, resembling the effect of antibodies that block complex formation between VN and PAI-1 (18). Furthermore, the high affinity PAI-1 binding to fibrin formed in the presence of VN and the VN-dependent potentiation of PAI-1-mediated inhibition of lysis of purified fibrin or plasma clots by tPA were also abrogated in the presence of these forms of kininogen. Interestingly, HKa domain 5 or the peptide HK486 inhibited the effect of the PAI-1·VN complex and promoted clot lysis not only when they were incorporated into the fibrin clot, but also when they were added together with tPA after the clot was allowed to age. A scrambled control peptide from the region Gly⁴⁸⁶-Lys⁵⁰² had markedly reduced potency in the described systems, indicating that the sequence specificity of this region and not only its Gly/His/Lys content was responsible for the ability of this defined portion of kininogen to dissociate preformed PAI-1·VN complexes. Conformational differences between HK and HKa, especially the more prominent exposure of the described region in HKa compared with HK (49), may explain the significantly higher efficiency of HKa over HK for inhibition of PAI-1 activity. Taken together, our data emphasize that the inhibition of PAI-1 function by the sequence Gly⁴⁸⁶-Lys⁵⁰² derived from domain 5 of HK provides a plausible explanation for the previously described antithrombotic role of kininogen (30, 34). Based on these observations, potential low molecular weight PAI-1 inhibitors could be designed for antithrombotic therapeutic interventions. For example, such PAI-1 inhibitors might provide a more efficient lysis of vascular thrombi by plasminogen activators, e.g. during acute therapy of myocardial infarction, stroke, or embolic occlusion of a peripheral artery.

Among the experimental strategies that were designed to counteract PAI-1 and thereby augment fibrinolysis, the following systems may become also clinically feasible: (i) PAI-1 antibodies (50, 51); (ii) a 14-amino acid peptide corresponding to the PAI-1 reactive center loop (residues 333-346) (52); (iii) two fungal metabolites (53); (iv) 11-keto-9(E),12(E)-octadecadienoic acid (fatty acid produced by Trichoderma spp.) (54); (v) the enzyme subtilisin NAT purified from Bacillus subtilis (55). However, in these approaches the concept of destabilizing the complex between VN and PAI-1 was hardly considered. Thus, the successful approach presented in this study is, to our knowledge, the first attempt to prevent the stabilization of PAI-1 by using fragments of an endogenous plasma factor, namely HKa. Based on experimental approaches in VN- or PAI-1-deficient mice, most reports indicated a prothrombotic role for both factors, and the molecular mechanisms may be attributed to their mutual interaction (56-59). Apart from stabilization of PAI-1 in its active conformation (60), VN serves as an intermolecular bridge to support high affinity binding of PAI-1 to fibrin such that both factors control tPA-mediated fibrin as well as plasma clot lysis (18). In addition, the acute phase protein ₁-acid glycoprotein has recently been shown to interact with and also stabilize PAI-1-function (61); the significance of this finding for PAI-1 function in vivo needs to be assessed in detail.

Based on our data, the inhibition of PAI-1 binding to VN by HKa (or domain 5 as well as the peptide HK486) is complete if HKa is in a 5-10-fold molar excess over PAI-1. Similarly, with respect to their antiadhesive potency in blocking integrin-mediated cell adhesion, the efficacy of PAI-1 is about 3-5-fold higher than HKa or its domain 5 (24). These differences could be attributed either to a higher affinity of PAI-1 as opposed to HKa for VN or to structural differences in the binding sites of HKa and PAI-1 on VN. For example, two binding sites for PAI-1 exist on VN showing a cooperative interaction (16, 62, 63), whereas only one binding site for HKa is presently known (24). Characterizing the exact sequences on VN which mediate the binding to PAI-1 and HKa will further clarify these issues.

Apart from its role as precursor of bradykinin, which contributes important vasodilator functions in vascular homeostasis, other characteristics of kininogen will add to the antithrombotic functions described here (24, 38), especially of the two-chain, kinin-free form (HKa). These include the following. (i) Binding to the glycoprotein Ib-IX-V complex on platelets, which is mediated by domain 3 of HK, may lead to inhibition of thrombin-dependent platelet aggregation (48). (ii) -granule-derived HK can inhibit the function of platelet calpain involved in, e.g. clot retraction, an effect mainly mediated by domain 2 of HK (64). (iii) HKa interferes with both ligand binding to _IIb3-integrin as well as _IIb3-integrin-induced signaling (65). (iv) The binding of HKa to the urokinase receptor (66) on vascular cell surfaces may approximate kallikrein and prourokinase and thereby potentiate plasmin formation. (v) HK-deficient rats are presented with a prothrombotic phenotype (30). Together, various domains/portions of HK can potentially contribute to the antithrombotic role of this plasma protein, and in particular the multifunctional domain 5 may be used as a leading substance for therapeutic interventions in vascular pathologies.

	ACKNOWLEDGEMENTS

We acknowledge gratefully the excellent technical assistance of T. Schmidt-Wöll. We appreciate the contribution of Drs. D. Johnson and Y. Lin (Philadelphia, PA) in the preparation and characterization of recombinant forms of kininogen and Dr. S. M. Kanse for helpful discussions. We also acknowledge the generous gift of reagents from Drs. P. Declerck (University of Leuven, Belgium) and W. Müller-Esterl (University of Frankfurt, Germany).

	FOOTNOTES

^* This work was supported in part by a grant from the Novartis Foundation for Therapeutical Research, Nürnberg, Germany (to K. T. P. and T. C.)The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Institut für Biochemie, Fachbereich Humanmedizin, Justus-Liebig-Universität, Friedrichstrasse 24, Giessen D-35392, Germany. Tel.: 49-641-994-7440; Fax: 49-641-994-7509; E-mail: triantafyllos.chavakis@innere.med.uni-giessen.de.

Published, JBC Papers in Press, June 24, 2002, DOI 10.1074/jbc.M204010200

	ABBREVIATIONS

The abbreviations used are: tPA, tissue-type plasminogen activator; BSA, bovine serum albumin; GST, glutathione S-transferase; HK, single chain high molecular weight kininogen; HKa, two chain (kinin-free) high molecular weight kininogen; HUVEC, human umbilical vein endothelial cell(s); PAI-1, plasminogen activator inhibitor-1; uPA, urokinase plasminogen activator; VN, vitronectin.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. McIlroy, M. O'Rourke, S. R. McKeown, D. G. Hirst, and T. Robson Pericytes influence endothelial cell growth characteristics: Role of plasminogen activator inhibitor type 1 (PAI-1) Cardiovasc Res, January 1, 2006; 69(1): 207 - 217. [Abstract] [Full Text] [PDF]

				C Roncal, J Orbe, J.A Rodriguez, M Belzunce, O Beloqui, J Diez, and J.A Paramo Influence of the 4G/5G PAI-1 genotype on angiotensin II-stimulated human endothelial cells and in patients with hypertension Cardiovasc Res, July 1, 2004; 63(1): 176 - 185. [Abstract] [Full Text] [PDF]

				M. Kawasaki, T. Maeda, K. Hanasawa, I. Ohkubo, and T. Tani Effect of His-Gly-Lys Motif Derived from Domain 5 of High Molecular Weight Kininogen on Suppression of Cancer Metastasis Both in Vitro and in Vivo J. Biol. Chem., December 5, 2003; 278(49): 49301 - 49307. [Abstract] [Full Text] [PDF]

				T. Chavakis, S. Santoso, K. J. Clemetson, U. J. H. Sachs, I. Isordia-Salas, R. A. Pixley, P. P. Nawroth, R. W. Colman, and K. T. Preissner High Molecular Weight Kininogen Regulates Platelet-Leukocyte Interactions by Bridging Mac-1 and Glycoprotein Ib J. Biol. Chem., November 14, 2003; 278(46): 45375 - 45381. [Abstract] [Full Text] [PDF]

				A. Zhao, J.-L. Lew, L. Huang, J. Yu, T. Zhang, Y. Hrywna, J. R. Thompson, N. de Pedro, R. A. Blevins, F. Pelaez, et al. Human Kininogen Gene Is Transactivated by the Farnesoid X Receptor J. Biol. Chem., August 1, 2003; 278(31): 28765 - 28770. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/36/32677    most recent M204010200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chavakis, T. Articles by Preissner, K. T. Search for Related Content PubMed PubMed Citation Articles by Chavakis, T. Articles by Preissner, K. T. Social Bookmarking                      What's this?