WO1999066939A1

WO1999066939A1 - Phospholipid vesicle-tissue factor complex preparations and methods of making and using same

Info

Publication number: WO1999066939A1
Application number: PCT/US1999/014183
Authority: WO
Inventors: Victor J. Marder; Katalin Varadi; Juergen Siekmann; Peter L. Turecek; Hans-Peter Schwarz
Original assignee: Marder Victor J; Katalin Varadi; Juergen Siekmann; Turecek Peter L; Schwarz Hans Peter
Priority date: 1998-06-23
Filing date: 1999-06-23
Publication date: 1999-12-29
Also published as: AU4709399A

Abstract

The inventive subject matter described herein provides complexes comprised of tissue factor bound to a phospholipid vesicle. The phospholipid vesicle can comprise 1,2-dioleyl-sn-glycero-3-phosphocholine (DOPC) and 1-palmitoyl-2-oleyl-sn-glycero-3-phosphoserine (POPS), and have a diameter of about 200 nm to about 300 nm, for example. The vesicles can be unilamellar, and the complex can comprise a multiple number of molecules of tissue factor (preferably 10 or more) bound to the phospholipid vesicle. Optionally, the vesicles also can have thrombomodulin bound thereto. Methods of using the complexes and preparations comprising these complexes are also provided herein.

Description

PHOSPHOLIPID VESICLE-TISSUE FACTOR COMPLEX PREPARATIONS AND METHODS OF MAKING AND USING SAME

This application claims priority to U.S. Provisional Patent Application No. 60/090,417, filed June 23, 1998, the entirety of which is hereby incorporated by reference.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates to lipid vesicle-tissue factor complexes, preferably phospholipid-tissue factor (PL-TF) complexes (which optionally include thrombomodulin) and preparations containing such complexes. The complexes according to the invention can be used as a new assay system for simultaneous measurement of thrombin and APC generation in plasma or blood, among other things.

2. Description of the Field

The physiologic response to vascular trauma is a finely ordered cascade of cellular and molecular events culminating in hemostatic plug formation. Natural inhibitory mechanisms, most notably the endothelial-associated thrombomodulin/protein C/protein S pathway, provide built-in checks on hemostasis, especially to limit its location to the injury site. Disruption of the inhibitory mechanisms allows the process to occur at inappropriate sites, which might lead to venous thrombosis following otherwise innocuous venous stasis or to widespread microvascular occlusion and purpura fulminans, which may occur after inflammatory events.

Physiologic hemostasis and often pathologic thrombosis may be initiated by exposure of blood to the trans-membrane protein tissue factor (TF). Depending upon the cell type and the stimulus for TF expression and exposure, a slow, gradual process or a dramatic initiation of coagulation may follow. Generated thrombin is bound to thrombomodulin (TM) by a process which not only eliminates its fibrinogen clotting ability but also activates the protein C pathway, which induces feedback inhibition of further clotting by inactivating factors Va and Villa.

Although there have been major advances in the field of blood chemistry, much of the workings of the blood coagulation system remain unknown. The blood coagulation is quite complex, which in large part is due to the multiple roles played by the proteins involved in the blood coagulation system. It therefore is often difficult to ascertain the true nature of a disease state or other abnormal condition. Accordingly, there is a need to advance the state of the art in terms of approaches for investigating and evaluation the blood coagulation system.

This need is satisfied by the invention disclosed herein, which advantageously employs a non-enzymatic, surface density based inhibition of APC by tissue factor. Tissue factor, when in high concentrations and surface density on a phospholipid vesicle exerts complementary effects of promoting the regular procoagulant cascade while inhibiting the effects of the protein C pathway. Thrombomodulin in conjunction with tissue factor also can be employed according to the invention.

SUMMARY OF THE INVENTION

It therefore an object of the present invention to provide improved compositions and methodologies for investigating and evaluating the blood coagulation system.

It is another object of the present invention to provide complexes which facilitate the activation of thrombin and/or protein C.

It is yet another object of the present interference to provide methods, preparations, complexes, and assay kits that are useful for monitoring the effects of antithrombotic therapies, ascertaining the levels and balance of blood proteins in a patient's blood or plasma, determining the activity level of blood coagulation preparations, and the like. To achieve these and other objects, there are provided, in accordance with one aspect of the invention, a complex comprised of tissue factor bound to a lipid vesicle, preferably a phospholipid vesicle. Thrombomodulin also can be bound to the phospholipid vesicle to form a complex comprised of a phospholipid vesicle, tissue factor and thrombomodulin.

The phospholipid vesicle can comprise l,2-dioleyl-sn-glycero-3-phosphocholine (DOPC) and l-palmitoyl-2-oleyl-sn-glycero-3-phosphoserine (POPS), and have a diameter of about 200 nm to about 300 nm, for example (vesicles outside of this range can be employed and can be readily attained by the skilled person in view of the teachings contained herein). The vesicles can be unilamellar, and the complex can comprise a multiple number of molecules of tissue factor (preferably 10 or more) bound to the phospholipid vesicle. In certain embodiments, the complex also can comprise thrombomodulin bound to the phospholipid vesicle.

In accordance with another aspect of the present invention, there are provided assay kits for evaluating blood or plasma samples, wherein the kit comprises a complex comprised of tissue factor bound to a phospholipid vesicle. Again, the phospholipid vesicle can comprise l,2-dioleyl-sn-glycero-3-phosphocholine (DOPC) and l-palmitoyl-2-oleyl-sn- glycero-3-phosphoserine (POPS), and have a diameter of about 200 nm to about 300 nm, for example (vesicles outside of this range can be employed and can be readily attained by the skilled person in view of the teachings contained herein). The vesicles can be unilamellar, and the complex can comprise a multiple number of molecules of tissue factor (preferably 10 or more) bound to the phospholipid vesicle. In certain embodiments, the complex also can comprise thrombomodulin bound to the phospholipid vesicle.

In accordance with still another aspect of the present invention, there are provided methods of assaying active blood proteins in a blood or plasma sample from a test subject, comprising contacting the blood or plasma sample with a complex comprising tissue factor bound to a phospholipid vesicle, and determining the amount of active blood proteins in the sample from a test subject (such as a person who possesses a lupus inhibitor and/or as an approach to diagnose or characterize pathological thrombosis), wherein the active blood proteins are at least one selected from the group consisting of activated protein C and thrombin. Depending upon the practitioner's goals, the contacting step can include contact with thrombomodulin, which along with thrombin is involved in the activation of protein C, and can be used to determine the presence of activated protein C. If thrombomodulin is not part of the contacting step or the complex, typically only the presence of thrombin is ascertained or detected. In certain embodiments, the complex also can comprise thrombomodulin bound to the phospholipid vesicle.

In accordance with yet another aspect of the present invention, there are provided methods of monitoring the effects of antithrombotic therapy (such as the use of coumadin) in a patient, comprising contacting a blood or plasma sample from the patient with a complex comprising tissue factor bound to a phospholipid vesicle, and determining the amount of active blood proteins in the sample, wherein the active blood proteins are at least one selected from the group consisting of activated protein C and thrombin. Similar to that explained above, thrombomodulin can be employed in order to determine the activation of protein C. In certain embodiments, the complex also can comprise thrombomodulin bound to the phospholipid vesicle. Without the use or inclusion of thrombomodulin, typically only the level of thrombin is determined.

In accordance with a further aspect of the present invention, there are provided methods of assaying the level of activity in a preparation containing active blood proteins, comprising contacting the preparation with a complex comprising tissue factor bound to a phospholipid vesicle, and determining the amount of active blood proteins in the sample, wherein the active blood proteins are at least one selected from the group consisting of activated protein C and thrombin. The active blood proteins can be obtained from purified sources and comprise at least one protein selected from the group consisting of thrombin and activated protein C. These proteins can be obtained from natural sources, be recombinantly produced, and be made by peptide synthesis, in whole or in part. The complexes employed can be like those described above, and can be optionally used in conjunction with thrombomodulin, as described above. In certain embodiments, the complex also can comprise thrombomodulin bound to the phospholipid vesicle.

These and other aspects of the present invention will become apparent to the skilled person in view of the disclosure, data, tables, and figures contained herein. BRIEF DESCRIPTION OF THE DRAWINGS

FIGURE 1, panels A-H, depict thrombin activity of the PL-TF complexes identified in TABLE 1. Thrombin activity is measured in the presence of 27 nm TM (closed squares ■) or in the absence of TM (open squares D). FIGURE 2, panel A depicts the initial rate of thrombin generation relative to TF concentration, and panel B depicts the TF surface density. This data is based upon the data of FIGURE 1.

FIGURE 3 depicts the effect of the initial rate of thrombin generation on maximum concentration of thrombin and APC, and on the feedback inhibition of thrombin generation. In panel A, thrombin activity is measured in the absence of TM. In panel B, the maximum APC activity in the mean APC activity observed after 10 and 15 minutes of coagulation in the presence of TM. Panel C depicts the percent of inhibition of thrombin generation , which was calculated from the peak thrombin activities detected in the presence and absence of TM. FIGURE 4 depicts the rate of inactivation of purified human factor Va by APC, and shows that the inactivation rate depends on phospholipid vesicle concentration. Panel A shows the time course of phospholipid vesicle concentration of 5 μm (closed circle, •), 20 μm (closed triangle, ^), 60 μm (closed square, ■) and 120 μm (closed, upside down triangle, T). Panel B show the calculated pseudo first order rate constants as a function of the phospholipid content of the vesicles.

FIGURE 5 depicts the equivalent effect of TF surface density on the rate of APC- mediated factor Va inactivation and the feed inhibition of thrombin generation in plasma. Pseudo first order rate constants if AP-mediated factor Va inactivation were measured using purified human proteins (panel A) and percent of thrombin generation inhibition in normal human plasma (panel B) in the presence of PL-TF complexes with low TF surface density (open squares D) or high TF surface density (closed squares, ■).

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

The present invention provides complexes of phospholipid vesicles and tissue factor (PL-TF complex). The PL-TF complex can be used in the presence or absence of thrombomodulin (TM) in an assay system to measure prothrombotic and antithrombotic tendencies in a test sample, such as plasma from a patient. Thrombomodulin also can be bound to the phospholipid vesicle to form a complex comprised of a phospholipid vesicle, tissue factor and thrombomodulin. The PL-TF complex (optionally with thrombomodulin accompanying the complex or as part of the complex itself by binding thrombomodulin to the phospholipid vesicle) thus serves as a reagent in this context, for example. Although theories and hypothetical explanations are set forth herein about how tissue factor and phospholipid vesicles interact in the blood coagulation pathways, the inventive subject matter is not limited to or defined by any type of theory or explanation. It is to be understood that the present inventions function by virtue of biological phenomenon and processes, and a complete understanding of the full nature of such biological phenomenon and processes are not necessary to successfully practice the inventive subject matter taught herein.

The PL-TF complex comprises phospholipid vesicles with tissue factor bound thereto (optionally including thrombomodulin bound to the vesicle). The word "bound" in its various grammatical forms refers to an association between the tissue factor and the phospholipid vesicle that is of sufficient strength and integrity to permit the complex to be considered an entity that is not unduly disrupted or separated under physiologic conditions. For example, a single phospholipid vesicle may have 10 or more tissue factor molecules bound (i.e. , connected/attached) to the vesicle.

The phospholipid vesicles can be formed by a number of ways, such as: the extrusion method according to Olson et al, Biochem. Biophys. Ada 557: 9-23 (1979), reversed phase evaporation technique (F. Szoka and D. Paphadjopoulos, Proc. Nat'l Acad. Sci. USA 75: 4194-98 (1978)), sonication of phospholipid dispersions (Barenholz et al, Biochemistry 16: 2806-10 (1977); Gregoriades et al, FEBSLett. 14: 95-99 (1971)), the ethanol injection method (S. Batzri and E.D. Korn, Biochem. Biophys. Acta 298: 1015-19 (1973)), and the removal of detergents from mixed micelles by dialysis methods (O. Zumbuehl and H.G. Weder, Biochem. Biophys. Acta 640: 252-62 (1981)). Preferred vesicle production and tissue factor attachment protocols are discussed below. The PL-TF complex also can be used in conjunction with thrombomodulin (which optionally also can be bound to the phospholipid vesicle) for the generation of thrombin and activated protein C (APC), while dampening the effect of APC. Thus, the PL-TF complex can be used to interfere with the APC-mediated inactivation of Factor Va and Factor Villa, thereby permitting the generation of further thrombin.

The PL-TF complex can be part of a preparation and an assay kit. The PL-TF complex can be used in methods for evaluating the effects of antithrombotic compounds (for example, coumadin). Additionally, the complexes can be used to ascertain the levels/activities of blood proteins in the blood or plasma of a test subject or in a preparation containing blood or plasma.

The invention is further explained by referring to the following examples, which do not limit the invention in any manner or way.

EXAMPLE 1: Preparation of the PL-TF complex and other compositions

Bovine serum albumin (BSA) was purchased from Calbiochem (La Jolla, CA). Viper venom Agkistrodon rhodostoma (Ancrod) and purified bovine thrombin were obtained from Sigma (St. Louis, MO). Chromogenic substrate S-2366 and 1-2581 thrombin inhibitor were obtained from Chromogenix (Mδlndal, Sweden). Recombinant human tissue factor (TF) and recombinant soluble thrombomodulin (TM) were purchased from American Diagnostica, Inc. (Greenwich, CT). l,2-dioleyl-sn-glycero-3-phosphocholine (DOPC) and l-palmitoyl-2-oleyl-sn-glycero-3-phosphoserine (POPS) were obtained from Avanti Polar Lipids (Alabaster, AL). Virus-inactivated human factor Va (Tans et al, Blood 11: 2641 (1991)) was kindly provided by J. Rosing (Maastricht, The Netherlands). Prothrombin was purified from a human prothrombin complex fraction prepared according to Brummelhuis, METHODS OF PLASMA PROTEIN FRACTIONATION (Curling ed., Academic Press 1980) and subsequent purification to homogeneity as described in Turecek et al, Ann. Haemotol. 74 (Supp. II): All (1997). Human factor Xa was prepared from highly purified factor X followed by activation with Russell's viper venom and affinity purification as described above. Turecek, loc. cit.

Defibrinated plasma samples were prepared as follows: Pooled normal plasma was a commercial product of George King Bio-Medical, Inc. (Overland Park, KS). The frozen samples were thawed at 37°C, defibrinated with Ancrod (Sultan et al, J. Lab. Clin. Med. 121: 444 (1993)), aliquoted and frozen at -80°C, then thawed at 37°C just before use. Preparation of phospholipid vesicle - tissue factor (PL-TF) complex was done in the following way: Phospholipid vesicles composed of 80% (w/w) of DOPC and 20% (w/w) of POPS were prepared by the extrusion method (Hope et al., Biochim. Biophys. Acta. 812: 55 (1985)) as follows: A phospholipid thin film was hydrated with 20 mM Tris, 150 mM NaCl, pH 7.4 (TBS) freeze-dried and reconstituted with the appropriate volume of distilled water. The phospholipid suspension was repeatedly extruded through two stacked 1000 nm polycarbonate filters, then through two 400 nm polycarbonate filters using an extrusion device of Lipex Biomembranes, Inc. (Vancouver, Canada). The vesicle preparation was diluted with TBS to a concentration of 1 .2 mM and freeze-dried after addition of 5 % sucrose (w/v). After reconstitution, the vesicles had a mean diameter of about 260 nm (polydispersity 0.23), as determined by dynamic light scattering (Zetasizer 4, Mai vern Instruments, Worcestershire, UK) (Ruf et al., Meth. Enzymol. 172: 364-90 (1989). Electron microscopic studies with negative staining or thin section preparations showed that about ninety percent of the vesicles in the preparation are unilamellar. Recombinant TF (non-lipidated, American Diagnostica, Inc., Greenwich, CT) was added to the reconstituted vesicle preparation, frozen at -20 °C overnight, thawed for 30 minutes at 25 °C and diluted 6.7-fold with TBS to yield final concentrations of 0.28 to 105 nM TF and 120 μM PL. After an additional 30 minutes of incubation at 25°C, the mixtures were centrifuged at 67,000 x g for 10 minutes, the supernatant fluid was aspirated and the complex reconstituted in the original volume of TBS and stored at 4°C. Where desired, thrombomodulin also can be included in this and other production methodologies to yield complexes comprising phospholipid vesicles with tissue factor and thrombomodulin bound thereto.

EXAMPLE 2: Characterization of PL-TF complexes

Freshly dissolved pellets contained vesicle-aggregates of 500 to 2000 nm diameter, as determined by dynamic light scattering. Electron microscopic studies (negative staining and thin sections) showed these vesicle-aggregates in freshly dissolved pellets and the vesicles were unilamellar. Like reconstituted phospholipid vesicle preparations without TF, the vesicles bound with TF (that is, a PL-TF complex), whether derived from pellets or supernatant, also were composed of unilamellar vesicles with a few collapsed ones in the electron microscopic studies. Vesicle size decreased upon storage at 4°C for several days to about 200 to about 300 nm, which is similar to the size of vesicles present in the supernatants, suggesting that the aggregates dissociate to monomeric forms. Other ranges of vesicles, such as 100 nm to 500 nm, also can be employed.

The actual TF concentration of pellets and supernatants was measured with the IMUBIND ELISA system (American Diagnostica, Inc., Greenwich, CT) and the phospholipid concentration was calculated on the basis of the phosphatidylcholine content using an enzymatic assay (Test Combination Phospholipids, Boehringer Mannheim, Germany) for all preparations. No difference was observed in the TF content of the vesicles measured in the absence or presence of detergents, suggesting that all TF bound onto the surface of the PL vesicles, and not inside. With increasing amounts of TF, phospholipids were distributed increasingly in the supernatants, suggesting that PL-TF complexes with high TF concentration are less likely to aggregate to form a pellet. Based on these data, TF surface density was calculated as TF molecules per vesicle, assuming that all phospholipids are arranged as unilamellar spheres of an average diameter of about 260 nm and that TF is fully bound to phospholipid vesicles (Pitlick et al., Biochemistry 9: 5105- 13 (1970)) and that TF is evenly distributed on a lipid bilayer composed of 5 x 10⁶ molecules/μm². See Alberts et al, MOLECULAR BIOLOGY OF THE CELL (Garland Publishing, Inc. 1989).

The TF surface density was comparable in the corresponding pellets and supernatants derived from the same preparations. For subsequent testing and analysis the PL-TF complexes are divided into low and high TF surface density groups, as shown in Table 1 below.

TABLE 1. Composition of PL-TF complexes

TF surface density^*

Low high

A B C D E F G H

TF(nM) 0.03 0.26 0.28 0.36 1.3 1.6 13 32

PL(μM) 14.5 69.2 107 49.2 17.5 5.8 100 110

TF/PL (molecules/vesicle) 2.2 3.9 2.8 7.7 78 289 137 306

'Expressed as ratio of TF to PL (molecules/vesicle) EXAMPLE 3: Thrombin and APC Assay Systems Thrombin and APC generation in plasma was measured simultaneously as follows:

A volume of 100 μl of defibrinated plasma was preincubated with 50 μl of PL-TF complex "mix" or "pellet" in the presence of absence of 50 μl TM (270 nM) in 25 mM HEPES, 175 mM NaCl, 1 mg/ml BSA, pH 7.35 (HNaBSA) buffer (final volume: 450 μl) at 37°C for 5 minutes. The coagulation pathway was initiated with 50 μl of 25 mM CaCl₂ and subsamples were withdrawn after pre-specified incubation intervals at 37°C. Thrombin activity was measured in 10 μl subsamples that were added to 300 μl TH-1 chromogenic substrate containing 3 mM EDTA to stop further thrombin generation.

After five minutes, chromogenic substrate conversion was terminated with 100 μl of acetic acid (75% v/v). APC generation was measured in 30 μl subsamples, added to 300 μl S-2366 chromogenic substrate containing 3 mM EDTA and 0.1 mM 1-2581 thrombin inhibitor to block thrombin activity. After 10 minutes, 100 μl of acetic acid (75% v/v) was added to terminate the chromogenic substrate conversion. After each reaction was stopped, 300 μl reaction mixture was transferred to an ELISA plate, and the absorbance was measured at 405 nm in an ELISA-reader. The molar concentrations of the generated thrombin and APC were calculated from reference curves constructed by measuring the activity of purified human thrombin (0 to 400 nM) and APC (0 to 20 nM) with the appropriate substrates. EXAMPLE 4: APC-mediated Inactivation of Factor Va

Purified factor Va (0.375 nM) was incubated with 0.05 mM APC in the presence of phospholipid vesicles alone or PL-TF complex and 5 mM CaCl₂ in HNaBSA buffer at 37°C. Factor Va activity was measured in a prothrombinase assay as follows: Subsamples (10 μl) were removed at pre-specified time intervals, added to a 140 μl mixture of 0.05 nM factor Xa, 75 nM of prothrombin, 10 μM PL (PC:PS vesicles, 80:20) and 5 mM CaCl₂ in HNaBSA buffer. After a 5 minute incubation at 37°C, thrombin activity was measured in 10 μl subsamples as noted above using TH-1 chromogenic substrate. Without the addition of APC, factor Va was stable for at least 10 minutes at 37°C, and the activity measured was taken at 100% .

The reaction conditions of the prothrombinase assay were chosen such (low factor Xa concentration) that APC-mediated cleavage at Arg₅₀₆ already results in a complete loss of factor Va activity and that slow cleavage at Arg₃₀₆ negligibly contributes to factor Va inactivation. Nicolaes et al., J. Biol. Chem. 270: 21158 (1995). Accordingly, time courses of factor Va inactivation were single exponential curves, and the pseudo first order inactivation rate constant (k'jinin ¹) was calculated using the equation:

[Va]_(t=x) = [Va]_(t=0) . exp^("kt), where [Va]_(t=x) is the residual factor Va activity measured after incubation with APC, and [Va]_(t=0) is the factor Va activity measured without APC.

EXAMPLE 5: Thrombin and APC generation

As shown in the following FIGURE 1 with the complexes of TABLE 1 , peak thrombin activity in plasma is attained after 2 to 5 minutes, followed by a rapid decline that is probably the result of thrombin neutralization. See Kessel et al, Comput. Biol. Med. 24: 277 (1994); Willems et al, Haemostas. 21: 197 (1991). Plasma APC is generated somewhat more slowly, presumably as the result of thrombin complex formation with thrombomodulin (TM) and reaction with protein C, reaching stable plateau levels after about 10 minutes. As soluble TM is used in this assay system, such TM-mediated APC formation occurs in solution. In contrast to separate assay systems for thrombin (Beguin et al, Thromb. Haemostasis 68: 136 (1992); Duchemin et al, Thromb. Haemostasis 71: 331 (1994); and Hemker et al, Thromb. Haemostasis 56: 9 (1986)) and APC generation (Bauer et al, J. Clin. Invest. 74: 2033 (1994); Francis et al, Am. J. Clin. Pathol. 87: 619 (1987); and Ohishi et al, Thromb. Haemostasis 70: 423 (1993)), simultaneous measurements in the same plasma reaction mixture allow the assessment of the possible impact of APC-mediated attenuation of thrombin generation under physiologic conditions, since APC is formed after activation of protein C by in situ generated thrombin.

EXAMPLE 6: High TF prevents APC-mediated thrombin inhibition

Prior reports suggest that decreased thrombin generation in the presence of TM is due to "feedback" inhibition by the protein C pathway. Ohishi noted that factor Va inactivation in plasma was more rapid in the presence of TM, and Duchemin observed quantitatively less feedback inhibition in plasma that was deficient in protein C or protein S. As shown in FIGURE 1, PL-TF complexes (TABLE 1) with low TF concentrations ( < 1 nM) and with low surface density (< 10 TF molecules/PL vesicle) induced a low rate of thrombin generation, and resulted in substantially less maximal thrombin generation in the presence of TM, then in the absence of it. In contrast, when thrombin generation was induced with PL-TF complexes that had high TF concentrations (1.3-32 nM) and high TF surface densities (78-306 TF molecules/phospholipid vesicle), the rate of thrombin generation and peak thrombin activities were higher. Moreover, no difference was found when TM was present.

The initial rate of thrombin generation was identical in the absence or presence of TM and correlated strongly with TF concentration (FIGURE 2, Panel A) but much less so with TF surface density (FIGURE 2, Panel B). The phospholipid concentration itself had no substantial influence on the rate of thrombin formation (data not shown).

The maximum thrombin generated in the absence of TM was a function of the initial rate of thrombin generation (FIGURE 3, Panel A). In the presence of TM, APC generation started simultaneously with thrombin activation, reaching plateau levels within 10 to 15 minutes (data not shown). No APC generation was detected in the absence of TM. As would be expected, both the rate of activation (not shown) and the maximum of APC generated was in direct correlation with the initial rate of thrombin generation (FIGURE 3, Panel B). Inhibition of thrombin generation, calculated from the maximum thrombin concentration obtained with and without TM, would be expected to be highest in the presence of the highest APC concentration. While this effect was noted for lower rates of initial thrombin generation induced with PL-TF complexes with low TF surface density, there was a lack of inhibition at higher rates of above 100 nM/min (FIGURE 3, Panel C), despite the rapid generation of the maximum amount of APC (FIGURE 3, Panel B). The lack of feedback inhibition suggests that the APC pathway is ineffective in the presence of high TF surface density vesicles.

In summary, where thrombin generation is induced in normal plasma in the absence and presence of soluble TM, inhibition of thrombin generation was only observed when PL-TF complex with low TF content and surface density on the vesicles were applied. In contrast, PL-TF complex with higher TF content and surface density on the vesicles prevented the APC-mediated inhibition of thrombin generation (FIGURE 1). As the initial rate of thrombin generation directly correlated with the TF concentration (FIGURE 2) and peak thrombin activity was higher at higher initial rates of thrombin generation (FIGURE 3, Panel A), a logical scenario for the absence of feedback inhibition would be that thrombin formation is completed before significant APC is generated. However, with the increasing thrombin generation rate, the generated APC also is increased (FIGURE 3, Panel B), which is in contrast to the threshold-like lack of feedback inhibition observed above an initial thrombin generation rate of 100 nM/min (FIGURE 3, Panel C). The seeming discrepancy of an absent feedback inhibition in the presence of high concentrations of APC suggest that even low APC concentrations are sufficient for feedback inhibition when the initial rate of thrombin generation is slow, and that in the presence of PL-TF complex with high TF content and surface density, APC activity is attenuated.

Thus, TF pathway-induced thrombin generation is mainly regulated by the rate of factor Va activation via a thrombin feedback effect and by the rate of factor Va inactivation by APC. Such inactivation requires the presence of TM, which complexes with generated thrombin and activates protein C in plasma. To better understand the effect of the TF content on the negative feedback regulation of thrombin generation, the PL-TF complexes were assessed for factor Va inactivation by APC using purified proteins in a non-plasma milieu. EXAMPLE 7: Mechanism of APC-mediated thrombin inhibition

The mechanism by which TF may influence feedback inhibition of thrombin generation by APC was further investigated by directly measuring the effect of TF on APC- mediated factor Va inactivation in model systems containing purified human proteins. In agreement with earlier reports (Bakker et al, Eur. J. Biochem. 208: 171 (1992)) phospholipid concentration is a critical determinant for factor Va inactivation (FIGURE 4) indicating that the phospholipid vesicle surface serves as a binding platform for interaction of the reaction components.

The time course of APC-mediated factor Va inactivation was measured in the presence of constant amounts of purified human factor Va and APC and increasing concentrations of phospholipid vesicles without TF (FIGURE 4, Panel A). The initial rate of inactivation was directly related to phospholipid concentration and the calculated pseudo first order rate constants (k') of factor Va inactivation appear to be a function of the concentration of vesicles (FIGURE 4, Panel B). Factor Va inactivation measured in the presence of phospholipid preparations with low TF surface density showed comparable inactivation rates as calculated for the corresponding phospholipid vesicles without TF. However, the factor Va inactivation rate constants measured in the presence of high TF surface density phospholipid vesicles were markedly lower than obtained for phospholipid vesicles without TF (TABLE 2).

Table 2. Pseudo first order rate constants of APC-mediated factor Va inactivation

^*Rate constants were estimated from the reference curve shown in Fig. 4, Panel B. The rate of purified factor Va by APC in the non-plasma system showed the same relationship to TF surface density (FIGURE 5, Panel A) as did the degree of thrombin generation inhibition in the plasma system (FIGURE 5, Panel B). This comparison of the same PL-TF complexes in the two reaction systems indicates that TF surface density controls feedback inhibition of thrombin generation by limiting APC-mediated factor Va inactivation.

Thus, when PL vesicles were replaced with PL-TF complexes, the rate of factor Va inactivation was strongly influenced by the TF surface density on the vesicles. On PL-TF complexes with a low TF surface density, the apparent first order rate constants of APC- mediated factor Va inactivation were comparable with those determined on phospholipid vesicles without TF. In contrast, when TF surface density exceeded a threshold of more than 10 TF molecules per vesicle, the rate of factor Va inactivation was greatly reduced and became independent of the phospholipid concentration.

The PL-TF complexes that exhibited retardation of APC-mediated inactivation of factor Va in the model system also prevented the feedback inhibition of thrombin generation in plasma (FIGURE 5). Again, although not to be bound by any theory, this observation suggests that the lack of feedback inhibition of thrombin generation in plasma is the consequence of the impairment of APC activity on PL surfaces with a high TF surface density. This negative influence of TF on APC-mediated factor Va inactivation had not been previously considered to be a principal function of TF. See, for example, Morrissey et al., Thromb. Haemostasis 78: 112 (1997). The observations reported herein can be explained by surface inhibition in which TF occupies space that is required for the binding of APC and its substrate factor Va, a situation which is consistent with Nemerson's model of a PL membrane that is "crowded with substrate and product molecules." Nemerson et al.,

Haemostasis 26 (Supp. 4): 98 (1996). Another explanation could be that TF might have a direct inhibitory effect on the APC pathway. Regardless of the modality of action, the PL- TF complex has demonstrated the ability to interfere with factor Va inactivation by APC. EXAMPLE 8: Summary of Data and Applications of the PL-TF Complex in Hemostasis

The data show the following:

1. Maximal thrombin concentration in the absence or presence of TM is observed at 2-5 minutes, followed by a decrease to about 20% . APC was generated only in the presence of TM, reaching plateau levels at about 10-15 minutes. The initial rate of thrombin generation is directly related to the concentration of tissue factor (nM).

2. At relatively low tissue factor density on the surface of the vesicle, less thrombin is generated in the presence than in the absence of thrombomodulin (a type of "feedback inhibition").

3. The rate of factor Va inactivation by APC is inversely related to the tissue factor surface density (molecules/vesicle) of the complexes, providing an explanation for the "feedback inhibition" effect.

The data herein demonstrate that two functions of TF act in concert to promote clot formation (hemostasis) or, in pathologic states, to enhance thrombus formation. On phospholipid vesicle surfaces with high TF content, TF induces maximal early thrombin generation. Although APC also is generated maximally, the high TF surface density would limit available vesicle surface for APC-mediated factor Va inactivation. Both processes are conducive to effective physiologic hemostatic plug formation. Under conditions of low TF presence, initial thrombin generation rates are low and APC-mediated factor Va inactivation has ample phospholipid vesicle surface for rapid action, effects which would limit thrombin generation and fibrin formation. Abnormalities of the protein C pathway would not induce an imbalance of hemostasis under conditions of high TF concentration and surface density and maximal thrombin generation. However, in situations of slow thrombin generation, such as may exist in veins with static flow, for which APC generation should inhibit the clotting cascade, the imbalance caused by any defects in the protein C pathway allows otherwise harmless thrombin generation to continue relatively unabated, leading to venous thrombosis Dahlback, Thromb. Res. 11: 1 (1995). As TF is normally found on cell surfaces outside the vasculature (Martin et al, Thromb. Res. 90: 1-25 (1998)) and would be exposed to blood at the site of vessel wall injury, the abnormal expression of TF on blood contacting cell surfaces induced by inflammatory cytokines, bacterial endotoxins or in association with malignant diseases (Ruf et al, FASEB J. 8: 386-90 (1994); Carmeliet et al, Int'l J.

Biochem. Cell Biol. 30: 661-67 (1998)) with increased TF surface density, might overhelm the protein C pathway, resulting in multiple microvascular thrombi formation leading to DIC. Weiss et al, J. Vet. Intern. Med. 12: 317-24 (1998).

Thus, TF modulates both thrombin formation and factor Va inactivation, upregulating the former while downregulating the latter for maximal hemostasis or setting imits on the former while allowing the latter in situations of natural inhibition of clot formation by the protein C pathway.

EXAMPLE 9: Reagents and assays 1. The PL-TF complex should be employed with a defined TF concentration and surface density. Preferably, the vesicle is unilamellar and has a diameter of 100 to 500 nm, preferably 200 nm to 300 nm, although vesicles outside these ranges also can be readily employed by the skilled person in view of the teachings contained herein. , and thus are within the scope of the invention. The phospholipid vesicles can be made with a variety of phospholipids, such as l,2-dioleyl-sn-glycero-3-phosphocholine (DOPC) and l-palmitoyl-2- oleyl-sn-glycero-3-phosphoserine (POPS). One preferred embodiment employs vesicles comprised of 80% (w/w) DOPC and 20% (w/w) POPS.

Assuming phospholipid vesicles having a diameter of 200 to 300 nm, the presence of bound tissue factor should be sufficiently large (at least about 10 TF molecules per vesicle) so as to permit the capabilities of the invention to be best realized.

The PL-TF complex can be part of an assay kit, which can include, among other things, thrombomodulin. Such assays kits are useful for the simultaneous thrombin and activated protein C (APC) generation. Using the PL-TF complexes in the presence or absence of soluble thrombomodulin (TM), thrombin is generated by calcium initiation of the coagulation pathways, following which the generated thrombin initiates PC activation and APC accumulation. The complexes also can comprise bound thrombomodulin such that the vesicle has tissue factor and thrombomodulin bound thereto.

The assay kits, including the PL-TF complex, also are useful for factor Va inactivation. APC-mediated factor Va inactivation also can be measured in purified systems using PL-TF complexes for the surface-oriented reaction.

EXAMPLE 10: Exemplary Uses of PL-TF complexes

PL-TF complexes can be used in a variety of contexts. For example, these complexes can be used to simultaneously measure and evaluate prothrombotic and antithrombotic tendencies. The following are illustrative (and non-limiting) of the uses for PL-TF complexes:

1. Patient studies a. Evaluation of patient blood samples for possible imbalance between thrombin and APC generation, as an explanation for pathologic thrombosis. These pathologies include hereditary defects in amount or integrity of a protein, including abnormalities such as protein C, protein S, factor V Leiden, and other abnormalities involving prothrombin and thrombin. In addition, acquired disorders that affect this blood protein balance, such as the "lupus inhibitor" and other antibodies/influences can be analyzed. b. Monitoring of antithrombotic therapy. Coumadin anticoagulation affects multiple blood factors, but current methods of monitoring therapy do not consider the coincident affect of treatment on both the coagulation (e.g., thrombin) and the PC inhibitory systems. Similarly, heparin anti-thrombin or new anti-thrombin agents fail to evaluate intrinsic APC generation, the balance of which is important in effective prophylaxis and prevention of thrombotic complications. Novel anticoagulants can be studied by such a functional approach.

2. Laboratory research studies

The PL-TF complex is an ideal analytical tool for studies of genetics of coagulation, evaluation of recombinant mutants and protein mimetics of blood factors (for example, prothrombin, protein C, etc.), classical clotting biochemistry, drug or antibody influences, and development of new anti-thrombotic and hemostatic approaches.

It is to be understood that the description, specific examples, figures, tables and data, while indicating exemplary embodiments, are given by way of illustration and are not intended to limit the present invention in any way or manner. Various changes and modifications within the present invention will become apparent to the skilled artisan from the discussion, disclosure and data contained herein, and thus are considered part of the invention.

Claims

WE CLAIM:

1. A complex comprised of tissue factor bound to a phospholipid vesicle.

2. The complex according to claim 2, wherein the phospholipid vesicle comprises l,2-dioleyl-sn-glycero-3-phosphocholine (DOPC) and l-palmitoyl-2-oleyl-sn- glycero-3-phosphoserine (POPS).

3. The complex according to claim 2, wherein the phospholipid vesicles have a diameter of about 200 nm to about 300 nm.

4. The complex according to claim 1, wherein the phospholipid vesicles are unilamellar.

5. The complex according to claim 1, wherein the complex comprises at least 10 molecules of tissue factor bound to the phospholipid vesicle.

6. An assay kit for evaluating blood or plasma samples, wherein the kit comprises a complex comprised of tissue factor bound to a phospholipid vesicle.

7. The assay kit according to claim 6, wherein the phospholipid vesicle comprises l,2-dioleyl-sn-glycero-3-phosphocholine (DOPC) and l-palmitoyl-2-oleyl-sn- glycero-3-phosphoserine (POPS).

8. The assay kit according to claim 7, wherein the phospholipid vesicles have a diameter of about 200 nm to about 300 nm.

9. The assay kit according to claim 7, wherein the phospholipid vesicles are unilamellar.

10. The assay kit according to claim 6, wherein the complex comprises at least 10 molecules of tissue factor bound to the phospholipid vesicle.

11. The assay kit according to claim 6, further comprising thrombomodulin.

12. A method of assaying active blood proteins in a blood or plasma sample from a test subject, comprising: contacting the blood or plasma sample with a complex comprising tissue factor bound to a phospholipid vesicle, and determining the amount of active blood proteins in the sample, wherein the active blood proteins are at least one selected from the group consisting of activated protein C and thrombin.

13. The method according to claim 12, wherein the contacting step further comprises thrombomodulin, and the amount of activated protein C is determined.

14. The method according to claim 12, wherein thrombomodulin is not added during the contacting step, and only the amount of thrombin is determined.

15. The method according to claim 12, wherein the test subject possesses a lupus inhibitor.

16. A method of monitoring the effects of antithrombotic therapy in a patient, comprising contacting a blood or plasma sample from the patient with a complex comprising tissue factor bound to a phospholipid vesicle, and determining the amount of active blood proteins in the sample, wherein the active blood proteins are at least one selected from the group consisting of activated protein C and thrombin.

17. The method according to claim 16, wherein the contacting step further comprises thrombomodulin, and the amount of activated protein C is determined.

18 The method according to claim 16, wherein thrombomodulin is not added during the contacting step, and only the amount of thrombin is determined.

19. The method according to claim 16, wherein the antithrombotic therapy employs coumadin.

20. A method of assaying active blood proteins in a sample, comprising: contacting the blood or plasma sample with a complex comprising tissue factor bound to a phospholipid vesicle, and dete╧Çriining the amount of active blood proteins in the sample, wherein the active blood proteins are at least one selected from the group consisting of activated protein C and thrombin.

21. The method according to claim 20, wherein the active blood proteins are obtained from purified sources and comprise at least one protein selected from the group consisting of thrombin and activated protein C.

22. The method according to claim 21, wherein the contacting step further comprises thrombomodulin, and the level of activity of activated protein C is determined.

23. The method according to claim 20, wherein thrombomodulin is not added during the contacting step, and the level of activity of thrombin is determined.

24. A complex comprised of tissue factor and thrombomodulin bound to a phospholipid vesicle.

25. A method of assaying active blood proteins in a blood or plasma sample from a test subject, comprising: contacting the blood or plasma sample with a complex comprising tissue factor and thrombomodulin bound to a phospholipid vesicle, and determining the amount of active blood proteins in the sample, wherein the active blood proteins are at least one selected from the group consisting of activated protein C and thrombin.

26. A method of assaying active blood proteins in a sample, comprising: contacting the sample with a complex comprising tissue factor and thrombomodulin bound to a phospholipid vesicle, and determining the amount of active blood proteins in the sample, wherein the active blood proteins are at least one selected from the group consisting of activated protein C and thrombin.