Plasma Lipoproteins Support Prothrombinase and Other Procoagulant Enzymatic Complexes

From the Departments of Biochemistry (M.P.M., R.P.T., P.B.T., C.v.V., K.G.M.) and Pathology (R.P.T.), University of Vermont, Burlington; and the Department of Pathology (C.E.S.), University of Rochester, Rochester, NY.

Correspondence to Russell P. Tracy, PhD, Laboratory for Clinical Biochemistry Research, University of Vermont, 55A South Park Dr, Colchester, VT 05446. E-mail rtracy{at}salus.uvm.edu

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Abstract—The prothrombinase complex (factor [F]Xa, FVa, calcium ions, and lipid membrane) converts prothrombin to thrombin (FIIa). To determine whether plasma lipoproteins could provide a physiologically relevant surface, we determined the rates of FIIa production by using purified human coagulation factors, and isolated fasting plasma lipoproteins from healthy donors. In the presence of 5 nmol/L FVa, 5 nmol/L FXa, and 1.4 µmol/L prothrombin, physiological levels of very low density lipoprotein (VLDL) (0.45 to 0.9 mmol/L triglyceride, or 100 to 200 µmol/L phospholipid) yielded rates of 2 to 8 nmol FIIa · L^-1 · s^-1 in a donor-dependent manner. Low density lipoprotein (LDL) and high density lipoprotein (HDL) also supported prothrombinase but at much lower rates (1.0 nmol FIIa · L^-1 · s^-1). For comparison, VLDL at 2 mmol/L triglyceride yielded 50% the activity of 2x10⁸ thrombin-activated platelets per milliliter. Although the FIIa production rate was slower on VLDL than on synthetic phosphatidylcholine/phosphatidylserine vesicles (50 nmol FIIa · L^-1 · s^-1), the prothrombin K_m values were similar, 0.8 and 0.5 µmol/L, respectively. Extracted VLDL lipids supported rates approaching those of phosphatidylcholine/phosphatidylserine vesicles, indicating the importance of the intact VLDL conformation. However, the presence of VLDL-associated, factor-specific inhibitors was ruled out by titration experiments, suggesting a key role for lipid organization. VLDL also supported FIIa generation in an assay system comprising 0.1 nmol/L FVIIa; 0.55 nmol/L tissue factor; physiological levels of FV, FVIII, FIX, and FX; and prothrombin (3 nmol/L FIIa · L^-1 · s^-1). These results indicate that isolated human VLDL can support all the components of the extrinsic coagulation pathway, yielding physiologically relevant rates of thrombin generation in a donor-dependent manner. This support is dependent on the intact lipoprotein structure and does not appear to be regulated by specific VLDL-associated inhibitors. Further studies are needed to determine the extent of this activity in vivo.

Key Words: lipoproteins • blood coagulation • prothrombinase • thrombin

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
The coagulation cascade is a complex set of enzymatic reactions that results in the activation of prothrombin to thrombin. For the enzymatic reactions in this cascade to occur at physiologically relevant rates, the components must be localized to an appropriate surface.¹ This surface has been provided by PL vesicles in vitro² and is presumed to be supplied by platelets^{3 4} and, to a lesser degree, by mononuclear cells⁵ in vivo. The PL component of these biological surfaces is important in complex assembly and function, although specific receptors have not been ruled out.^{6 7} Lipoproteins are also a source of plasma PLs and may provide another surface that will support coagulation reactions.

Plasma lipoproteins have been shown to be risk factors for CHD.⁸ High levels of LDL-C are associated with CHD, whereas high levels of HDL-C are inversely associated to CHD. The relationship of VLDL-C, as estimated by fasting TG levels, to CHD is less clear,^{9 10 11 12 13} but some recent data suggest that these larger lipoproteins also may be associated with the atherothrombotic process, especially in postprandial states.^{14 15 16 17 18} There is evidence from population studies that plasma levels of vitamin K–dependent coagulation factors (prothrombin; FVII, FIX, and FX; and proteins C and S) are correlated with fasting cholesterol and TG levels, suggesting some form of coordinate regulation.^{19 20} If lipoproteins were shown to support coagulation reactions under physiologically relevant conditions, then this would support the hypothesis that they are thrombotic risk factors as well as atherogenic risk factors and would provide a possible mechanism, through direct binding, for the epidemiological association of lipoprotein levels with factor levels.

To explore this relationship, we characterized the support of surface-dependent coagulation reactions by lipoproteins in a purified human system. We determined the degree of support provided by VLDL, LDL, and HDL for the prothrombinase complex and determined the kinetic parameters of the prothrombinase reaction on VLDL and LDL. We also determined the ability of lipoproteins to support thrombin generation in a complete vitamin K–dependent protein procoagulant assay system.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Materials and Reagents
Trizma (Tris base), Tris HCl, PS (from bovine brain), PC (from egg yolk), EDTA, benzamidine, NaN₃, HEPES, NaCl, and chloroform were purchased from Sigma Chemical Co. NaBr and diethyl ether were purchased from Krackler. BSA was purchased from GIBCO BRL. The -thrombin inhibitor DAPA was purchased from Haematologic Technologies Inc. PL vesicles composed of 75% (wt/wt) PC and 25% (wt/wt) PS were prepared as described previously.^{21 22} The concentration of the PL vesicles was determined by a phosphorus assay.²³ The chromogenic thrombin substrate S-2238 was purchased from Pharmacia. Polyclonal anti-TFPI IgG was made in rabbits by Cocalico Biologicals Inc.

Proteins
Proteins were purified from human, fresh frozen plasma. FV was isolated by immunoaffinity chromatography as described and was activated to FVa with 2 NIH U/mL of -thrombin for 15 minutes at 37°C.^{24 25} FIX, FX, and prothrombin were purified by the method of Bajaj et al.²⁶ FX was activated with the FX-activating factor purified from Russell's viper venom.²⁷ -Thrombin was prepared by activation of prothrombin with Taipan snake venom, as described by Owen and Jackson.²⁸ Recombinant FVIII and recombinant TF (residues 1 to 242) were gifts from Drs Shu Len Liu and R. Lundblad, Hyland division, Baxter Healthcare Corp. Recombinant FVIIa was purchased from Novo Pharmaceuticals. In some cases FV, FXa, prothrombin, and -thrombin were purchased from Haematologic Technologies Inc. Protein purity was routinely assessed by SDS–polyacrylamide gel electrophoresis before and after disulfide bond reduction according to the method of Laemmli.²⁹ Proteins were visualized by Coomassie brilliant blue R-250 staining. Fig 1 illustrates typical electrophoretic results for purified VLDL and LDL, prothrombin, FXa, and FV.

View larger version (67K): [in this window] [in a new window]   Figure 1. Reduced SDS–polycrylamide gel electrophoresis of purified VLDL and LDL lipoproteins and coagulation FII, FV, and FXa. This figure is representative of routine quality assessment. Lanes 1 to 4, 3% to 10% SDS–polycrylamide gel electrophoresis of delipidated VLDL (lanes 1 and 2) and delipidated LDL (lanes 3 and 4) from two different donors. Lipoproteins were isolated by sequential flotation and lipids were extracted with ethanol/ether as described in "Methods." The bands at 97 and 68 KDa in lane 4 are from the molecular weight standards in the adjacent lane. Lanes 5 to 7, 5% to 15% SDS–polycrylamide gel electrophoresis of prothrombin (5 µg, lane 5), FXa (2 µg, lane 6), and FV (0.5 µg, lane 7) isolated as described in "Methods."

Isolation and Characterization of Lipoproteins
Blood for lipoprotein isolation, collected under institutionally approved protocols for the use of human subjects, was collected after an overnight fast into a preservative cocktail of EDTA, benzamidine, and NaN₃, so that the final concentrations of each were 5 mmol/L, 1 µmol/L, and 1.5 mmol/L, respectively. Donors were healthy, young adult, laboratory personnel. Lipoproteins were isolated from fresh human plasma by sequential flotation ultracentrifugation as described³⁰ using an SW-41 rotor from Beckman. In one experiment, lipoproteins were isolated by gradient centrifugation using Iodixinol (Accurate Chemical) according to the manufacturer's recommendations.

After isolation, lipoproteins were dialyzed into HBS, pH 7.4, and stored at 4°C. After sequential floatation, the HDL preparation was purified further by magnesium/phosphotungstic acid precipitation³¹ to remove contaminating dense LDL and lipoprotein(a). The quality and purity of isolated lipoproteins were determined by agarose electrophoresis. Purity and quality were also assessed by 3% to 10% SDS–polyacrylamide gel electrophoresis after delipidation of the VLDL and LDL samples by ethanol/ether extraction as described in Fig 1 of Havel et al.³² Quantification of the lipoprotein samples was done using enzymatic cholesterol and TG assays purchased from Sigma. PL concentrations were assessed with ammonium ferrothiocyanate and chloroform by the method of Stewart.³³ Protein concentrations were determined by using the BCA protein assay reagent from Pierce Chemical Co. Isolated lipoproteins were used within 5 days.

For experiments using extracted VLDL lipids reconstituted in aqueous buffer, we used a method similar to that of Bajaj et al.³⁴ One volume of VLDL in aqueous solution was extracted with five volumes of 2:1 chloroform/methanol, and the organic layer was removed and dried under a steady N₂ stream. The lipids were then dissolved in HBS and quantified by PL assay.

To determine the possible extent of microparticle contamination in our lipoprotein samples, we analyzed the plasma by flow cytometry. A 1:12 dilution of plasma was incubated with 0.33 µg/mL of FITC-conjugated HP1–1D, which recognizes glycoprotein IIb/IIIa as described elsewhere.³⁵ Samples were analyzed for forward- and right-angle scatter and for green fluorescence with a Coulter Elite fluorescence activated cell sorter using platelet and microparticle gates as described.³⁵

Prothrombinase Activity on Lipoproteins
To perform the prothrombinase experiments, FVa and DAPA were incubated with a solution of lipoproteins or PCPS vesicles for 1 minute. The reactions were carried out in HBS (pH 7.4) with 5 mmol/L CaCl₂ and 0.1% BSA at room temperature. DAPA was required to inhibit thrombin from converting prothrombin to prethrombin 1, a species that is not a suitable substrate for prothrombinase. Following that incubation, prothrombin was added and incubated with the reaction mixture for 1 minute. The reaction was initiated by adding 25 µL of FXa to 175 µL of the lipoprotein/FVa/DAPA/prothrombin mixture. In some experiments, lipoprotein and PCPS were titrated. The final concentrations of the reactants (other than PL) were 5 nmol/L FVa, 5 nmol/L FXa, 1.4 µmol/L prothrombin, and 3 µmol/L DAPA. In some cases, the concentrations of FVa, FXa, and prothrombin were varied. The concentrations of PCPS typically ranged from 1 to 100 µmol/L phosphate, and the concentrations of lipoprotein, though dependent on the concentration of the isolated sample, typically ranged from 0 to 500 µmol/L PL (0 to 2.25 mmol/L TG) for VLDL, 1 to 1000 µmol/L PL (0 to 5.15 mmol/L cholesterol) for LDL, and 0 to 650 µmol/L PL (0 to 1.5 mmol/L cholesterol) for HDL. After initiation of the reactions, 25-µL aliquots were quenched at various times in 75 µL of HBS with 50 mmol/L EDTA and then assayed for thrombin concentration by assessing the rate of S-2238 (0.4 mmol/L) conversion by thrombin in a Spectra Max 250 spectrophotometer from Molecular Devices. Thrombin generation was calculated from a standard curve prepared by using various concentrations of purified -thrombin.

Experiments to determine kinetic parameters were done using prothrombin concentrations from 0.067 to 3.195 µmol/L. Data were analyzed using curve-fitting software to determine K_m and V_max (Enzfitter, Elsevier-Biosoft). For experiments using rabbit IgG, the lipoprotein samples were incubated for 30 minutes at room temperature with 0.15 mg/mL rabbit anti-TFPI IgG, 0.15 mg/mL control rabbit IgG, or diluent. After the incubation was complete, the prothrombinase reaction was carried out as described above.

FVIIa/TF-Dependent Procoagulant Assay System
TF, 0.55 nmol/L, was incubated with lipoproteins or PCPS vesicles for 30 minutes at 37°C in HBS (pH 7.4) with 2 mmol/L CaCl₂. FVIIa was then added at 0.1 nmol/L and incubated for 20 minutes to allow complex formation. To initiate the reaction, 50 µL of the lipoprotein/TF/FVIIa solution was mixed with 50 µL of a solution of FV, FVIII, FIX, FX, and prothrombin. The final concentrations were 20 nmol/L FV, 0.7 nmol/L FVIII, 90 nmol/L FIX, 170 nmol/L FX, and 1.4 µmol/L prothrombin, which were chosen to reflect typical plasma concentrations. After initiation of the reaction, 5-µL aliquots were quenched in 100 µL of HBS with 20 mmol/L EDTA, and the concentration of thrombin was determined as described above.

Platelet Isolation
Platelets were isolated by the method of Mustard et al³⁶ using venous blood from healthy, nonmedicated individuals as described above. Modifications to this procedure included omission of apyrase from all washing steps and the use of 5 mmol/L HEPES/Tyrode's solution, pH 7.4, as the final platelet suspension buffer. Platelets were counted on a Coulter counter (Coulter Electronics). Platelet activation was accomplished by incubation of platelets with 20 nmol/L thrombin (2 NIH U/mL) for 5 minutes at room temperature immediately prior to the prothrombinase assays. Platelet-dependent prothrombinase assays were performed as described.³⁷

	Results

Top Abstract Introduction Methods Results Discussion References

 
Prothrombinase Activity on Lipoproteins and PCPS Vesicles
The relationships between the rate of thrombin generation and the amount of PL (as either lipoprotein or PCPS vesicles) are shown in Fig 2. We observed that although all three lipoprotein classes supported prothrombinase activity, VLDL consistently supported the highest rate. In all cases the maximum rates observed were lower than those obtained with PCPS vesicles. We observed a range in rate with different VLDL samples, 2 to 10 nmol FIIa · L^-1 · s^-1. The range we observed with different samples of LDL was between 0.4 and 1.0 nmol FIIa · L^-1 · s^-1. HDL supported prothrombinase activity at very slow rates, consistently <0.1 nmol FIIa · L^-1 · s^-1.

View larger version (21K): [in this window] [in a new window]   Figure 2. Prothrombinase activity using isolated fasting lipoproteins. VLDL (-), LDL (-), and HDL (-) from one donor and PCPS (-) vesicles were used in a prothrombinase assay. Results are expressed as thrombin (FIIa) generation rates in nmol/L of thrombin generated per second with respect to the amount of lipoprotein or synthetic vesicle PL used. The scale for PCPS vesicles is on the right and the scale for lipoproteins on the left.

Using flow cytometry, we determined that there were 4x10⁵ microparticles per milliliter of plasma as prepared for lipoprotein isolation. If one assumes that all microparticles were concentrated in the VLDL preparation (unlikely, since even mild centrifugation pellets the majority of microparticles³⁸ ), this would yield 4x10⁶ microparticles per milliliter of VLDL. Since 40% of the prothrombinase activity of activated platelets is in the microparticle fraction³⁹ and approximately one or two microparticles are formed for every two thrombin-activated platelets (our observations and Reference 40⁴⁰ ), then on the basis of our observed thrombin generation rates for platelets (eg, Fig 6), the maximum contribution of microparticles to the rate of thrombin generation would be 0.08 nmol FIIa · L^-1 · s^-1, or <3% of the typical rates we observed with VLDL. We conclude that platelet microparticles cannot be a major contributor to prothrombinase rates under the conditions described here.

View larger version (15K): [in this window] [in a new window]   Figure 6. Comparison of platelet-dependent and VLDL-dependent prothrombinase activity from 2 healthy, nonmedicated donors. Blood was drawn for VLDL isolation [donor 1 (-); donor 2 (-)] 1 day before the blood draw for platelet isolation so that all prothrombinase assays could be performed on the day of platelet isolation. Results are expressed as in Fig 2, with the exception that VLDL concentrations are given as VLDL TG to emphasize physiological relevance. For VLDL, 100 µmol/L PL<202>0.45 mmol/L TG.

The titration curve for PCPS vesicles reached a maximum and then declined because the maximum represents the point at which all components (FVa, FXa, and prothrombin) are optimally bound to the surface; additional surface "dilutes" the components and reduces the rate.⁴¹ In this experiment, VLDL (and HDL) had titration profiles similar to that of PCPS vesicles: a maximum followed by a decline. Although LDL did not reach a maximum in this experiment, we have performed this experiment five times with generally similar results and have seen LDL reach a maximum in other experiments.

Kinetic Parameters of Prothrombinase on VLDL, LDL, and PCPS Vesicles
We determined the K_m values for prothrombin by using two different concentrations of VLDL, two different concentrations of LDL, and PCPS vesicles (the Table). The concentrations of lipid were chosen to represent points on the ascending side of the curves shown in Fig 2. The K_m values obtained with lipoproteins (0.74 to 0.96 µmol/L) were similar to those obtained with PCPS vesicles (0.5 µmol/L), though consistently slightly higher. The V_max values observed with lipoproteins reflected the lower rates that we observed in the titration experiments.

Variability in Prothrombinase Activity With VLDL From Different Donors
As mentioned previously, we have observed a range of rates for prothrombinase on VLDL between plasma from different donors. Fig 3 is an example of what we have typically seen, illustrating data for six different donors, with lipoprotein preparations and enzyme assays done simultaneously. The maximum rates from these donors varied from 1.8 to 6.0 nmol FIIa · L^-1 · s^-1 . Because of our concern about variability due to phlebotomy or sample preparation, we obtained two blood samples simultaneously from 1 donor, one from each arm, and treated these samples separately. We performed this experiment twice on 2 separate days. The thrombin generation rates observed with 0.55 mmol/L VLDL TG were as follows: day 1, 4.5 and 4.2 nmol FIIa · L^-1 · s^-1; and day 2, 4.3 and 4.3 nmol FIIa · L^-1 · s^-1. These results indicate excellent reproducibility for the method.

View larger version (26K): [in this window] [in a new window]   Figure 3. VLDL-dependent prothrombinase rates from 6 different healthy, fasting donors. VLDLs were isolated simultaneously from 6 volunteers and used in a prothrombinase assay. Results are expressed as in Fig 2.

Effect of TFPI in the Prothrombinase Assay System
TFPI is known to inhibit prothrombinase, but very slowly. We determined the rate of prothrombinase on lipoproteins in the presence and absence of an inhibitory polyclonal anti-TFPI IgG. VLDL-supported prothrombinase rates (1 mmol/L VLDL TG) were 4.98 nmol FIIa · L^-1 · s^-1 without IgG, 5.31 nmol FIIa · L^-1 · s^-1 with 0.15 mg/mL rabbit anti-TFPI IgG, and 5.51 nmol FIIa · L^-1 · s^-1 with 0.15 mg/mL control rabbit IgG. LDL-supported prothrombinase rates (1 mmol/L LDL-C) were 0.96 nmol FIIa · L^-1 · s^-1 without IgG, 0.96 nmol FIIa · L^-1 · s^-1 with 0.15 mg/mL rabbit anti-TFPI IgG, and 1.18 nmol FIIa · L^-1 · s^-1 with 0.15 mg/mL of control rabbit IgG. Because this anti-TFPI is known to inhibit TFPI activity in systems using PL vesicles (R.P. Tracey et al, unpublished data, 1997), we think that these data are consistent with TFPI's having no effect in our assays.

Tests for Specific Inhibitory Activity of FVa or FXa by VLDL
To further characterize the support of prothrombinase by VLDL, we investigated prothrombinase activity on extracted VLDL lipid reconstituted in aqueous buffer. This procedure (1) removed any protein from the lipid particles and (2) allowed for major reorganization of the lipid components. Fig 4 shows that VLDL lipid supported prothrombinase activity to a greater degree than did isolated VLDL particles, with maximal rates of thrombin generation approaching those observed with PCPS vesicles.

View larger version (20K): [in this window] [in a new window]   Figure 4. Comparison of prothrombinase activity from intact VLDL and extracted VLDL lipids. VLDL (-) and extracted VLDL lipid (-) from one donor and PCPS vesicles (-) were used in a prothrombinase assay. Results are expressed as in Fig 2. Rates achieved with extracted lipids were similar to those observed with PCPS vesicles.

To determine whether or not the variability in VLDL-dependent thrombin generation was due to an inhibitory activity toward either FVa or FXa, we titrated PCPS vesicles, intact VLDL, and extracted VLDL lipid from a single donor in prothrombinase assays using three different combinations of FVa and FXa concentrations. The three combinations used were as follows: (1) 5 nmol/L FVa, 5 nmol/L FXa; (2) 0.5 nmol/L FVa, 5 nmol/L FXa; and, (3) 5 nmol/L FVa, 0.5 nmol/L FXa. If intact VLDL contains specific inhibitory activity toward one of the components (FVa or FXa) when the concentration of that component is decreased, then we would expect to see a greater decline in activity compared with the decline observed with PCPS vesicles, where there is no specific inhibitory activity. Fig 5 shows the results of this experiment. There were no differences in activity changes seen in any of the experiments when compared with PCPS vesicles. The maximum rates observed with 0.5 nmol/L FVa were 10% of those seen with 5 nmol/L FVa in all cases, and the maximum rates observed with 0.5 nmol/L FXa were 20% of those seen with 5 nmol/L FXa in all cases. This observation is consistent with the idea that there is no specific inhibitory activity of either FVa or FXa associated with intact VLDL.

View larger version (17K): [in this window] [in a new window]   Figure 5. Lipoprotein-dependent prothrombinase activity with reduced enzyme or cofactor levels. The concentrations of FVa or FXa were decreased 10-fold and the effect on the rate of thrombin generation measured for PCPS-dependent (top), VLDL-dependent (middle), and extracted VLDL lipid-dependent (bottom) prothrombinase reactions. Concentrations used in the assays were 5 nmol/L FVa, 5 nmol/L FXa (-), 5 nmol/L FVa, 0.5 nmol/L FXa (-), 0.5 nmol/L FVa, and 5 nmol/L FXa (-). Results are expressed as in Fig 2.

Prothrombinase on Platelets Compared With Isolated VLDL
Activated platelets are known to be a major source of procoagulant surfaces in vivo. To test the physiological significance of VLDL-dependent prothrombinase rates, we determined the rate of thrombin generation on VLDL and activated platelets from the same donor. The blood utilized for VLDL isolation was drawn 1 day before platelet isolation so that all prothrombinase assays could be performed on the same day. Fig 6 shows that VLDL at 2 mmol/L (175 mg/dL) TG gave rates of thrombin generation that were at least one half of those seen with 2x10⁸ platelets per milliliter from these same donors.

FVIIa/TF-Dependent Procoagulant Assays Using Lipoproteins and PCPS Vesicles
To determine whether isolated lipoproteins could support other surface-dependent coagulation reactions besides prothrombinase, we used a more complex procoagulant assay system. Physiological levels of FV, FVIII, FIX, FX, and prothrombin were combined with lipoproteins, TF, and FVIIa, and the rate of thrombin generation was measured over time (Fig 7). In this experiment PCPS vesicles were used at 200 µmol/L phosphate, whereas VLDL from three different donors and one sample of LDL were used at their fasting levels. As in the simpler prothrombinase assay system, VLDL was more effective a surface than LDL but less so than PCPS vesicles. Donor variability in VLDL-dependent rates was reflected in differences in the observed lag time, not in the actual rate of thrombin generation, which was 3 nmol FIIa · L^-1 · s^-1 in each case.

View larger version (24K): [in this window] [in a new window]   Figure 7. Thrombin generation over time using the procoagulant assay system initiated with TF/FVIIa. PCPS vesicles (–) 142), VLDL isolated from 3 healthy, fasting donors [donor 1 (-); donor 2 (-); donor 3 (-)] and fasting LDL isolated from donor 3 (-) were used in the assay system described in "Methods," initiated with TF/FVIIa, and comprising FIX, FVIII, FX, FV, and prothrombin. Results are expressed as micromoles of FIIa assessed at different time points during the reactions. The concentrations of lipoproteins corresponded to the fasting levels from these donors are as follows: donor 1 VLDL, 0.6 mmol/L TG (133 µmol/L PL); donor 2 VLDL, 0.9 mmol/L TG (200 µmol/L PL); donor 3 VLDL, 2.3 mmol/L TG (511 µmol/L PL); and donor 3 LDL, 5.2 mmol/L cholesterol. PCPS vesicles were used at 200 µmol/L PL.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
In this article we present the first kinetic data, using exclusively human components, which demonstrate plasma lipoprotein support for prothrombinase complex activity and, in fact, all reactions of the tissue factor pathway. The maximal VLDL-dependent rates were similar to those seen using physiological levels of activated platelets, suggesting that under certain conditions, lipoproteins, especially VLDL, might play a significant role supporting thrombin generation in vivo.

Purified VLDL particles have the ability to support prothrombinase at a level near that of activated human platelets. These rates are achieved with physiological levels of VLDL, 0.3 to 1.7 mmol/L TG. The results from the procoagulant assay system also indicate that VLDL can support generation of thrombin in a reaction initiated with FVIIa/TF at a significant rate. This further supports the concept that lipoproteins may provide a suitable surface for coagulation reactions and do not require activation, as in the case of platelets.

Because lipoproteins accumulate in atherosclerotic plaque,^{42 43 44 45 46} their presence may have significance for plaque-associated thrombin generation, especially in relatively minor situations as proposed by Harker et al,⁴⁷ wherein small amounts of thrombin generation have implications for plaque progression. Regarding venous thrombosis and thromboembolic disease, conditions of blood stasis may be particularly relevant, where the components may have the opportunity to form complexes, at least transiently. It has also been suggested that postprandial VLDL increases may be associated with a hypercoagulable state.^{14 48} Our observations suggest that, if this is true, VLDL-mediated procoagulant complex assembly may be a significant contributor to this process.

Our observed K_m values for prothrombin experiments using lipoproteins are similar to those determined with PCPS vesicles. They also agree with previous reports of prothrombinase kinetics on PL surfaces.⁴⁹ This is consistent with the notion that prothrombinase has the same mechanism of action on lipoproteins as on PCPS vesicles. The lower V_max values suggest that substrate delivery and incorporation are not significantly affected on the lipoprotein surface when compared with PCPS vesicles but that there is an effect on the turnover rate of the enzyme.

We observed donor variability in VLDL-dependent thrombin generation. In the prothrombinase assay, different individuals exhibited consistently different lipid titration profiles, regarding both the level of lipoprotein required for the maximum rate and the maximum rate achieved. Reproducibility studies indicated that donor variability was most likely not a result of sample preparation artifacts.

We considered the concentration of negatively charged PLs on the surface of the VLDL particles from different donors as a source of variability. It is known that the concentration of negatively charged phosphate groups on a surface will influence the degree of protein binding and the reaction rate.⁵⁰ Because lipoprotein particles consist of 5% negatively charged PLs,⁵¹ whereas PCPS vesicles as we prepared them are 25% negatively charged PLs, this is a possible explanation for the difference between the rates on PCPS vesicles and the rates on lipoproteins. However, because the VLDL-dependent rates could be dramatically increased to approximately the same values as the PCPS-dependent rates by extracting the lipids, this suggests that rate differences cannot be explained simply by the different percentages of negatively charged PLs and that lipoprotein conformation has a major influence on prothrombinase rates.

TFPI is known to be associated with lipoproteins. TFPI has been shown to inhibit FXa in the prothrombinase complex.⁵² Mast and Broze⁵³ observed a 50% drop in activity in an FXa-initiated assay when 8 nmol/L TFPI (approximately threefold the physiological concentration) was added to a reaction with 3 nmol/L FVa, 0.1 nmol/L FXa, and 1.4 µmol/L prothrombin.⁵³ On the basis of their results, we would anticipate a minimal effect in our assays, as we used an 50-fold higher enzyme concentration. Our immunological and titration-based data support this position.

VLDL contains apo B-100 as one of its apolipoprotein components. The effect of apo B-100 on prothrombinase activity is not well understood. Although our data with FVa and FXa titrations suggest that apo B-100 is not specifically inhibitory towards either FVa or FXa, the possibility that apo B-100 is inhibitory to the prothrombinase complex as a whole has not been ruled out. One interpretation of the data regarding extracted lipids might be that apo B does have some inhibitory effect, since its removal results in increased activity. However, since there are likely to be enormous changes in lipid organization during extraction, one must be cautious about the implications of the experiment regarding apo B-100. Nonetheless, it remains possible that apo B could have effects on the surface, including the masking of PL binding sites, alterations in fluidity of the PL surface, or direct protein-protein interactions with prothrombinase components. VLDL also contains exchangeable apolipoproteins of the C and E groups, which are absent on LDL. These apolipoproteins may play a role in the surface complexes involved in the procoagulant enzymatic activity by supporting protein-protein or protein-lipid interactions.

Another factor known to affect prothrombinase rates on surfaces is the composition of the fatty acid chains on the surface PLs. For example, vesicles composed of PLs containing predominately stearic acid (18:0) had rates that were 5% of those containing oleic acid (18:1).⁵⁴ Although we did not directly assess the fatty acid composition of our lipoprotein preparations, our data with extracted VLDL lipids again suggest that the fatty acid side-chain composition does not play an important role per se. However, we do note that although the extracted VLDL–dependent rates of thrombin generation approached those seen on PCPS vesicles, the level of PL needed to achieve these rates was higher. The reason for this is unclear at this time.

The presence of cholesterol in the lipoproteins might cause a loss of activity. However, vesicles composed of 60% PC, 20% PS, and 20% cholesterol have been shown to support prothrombinase to the same degree as vesicles composed of 80% PC and 20% PS.⁴⁹ The experiments with the procoagulant assay system indicate that VLDL and LDL can support FVIIa/TF-initiated thrombin generation in an assay with physiological levels of FV, FVIII, FIX, FX, and prothrombin. VLDL-mediated reactions supported rates of 3 nmol FIIa · L^-1 · s^-1 after a lag of 1 minute, with little donor variability in the rate of thrombin generation after the initial lag. The initial lag time may represent the time that it takes for small amounts of thrombin to be formed and subsequently convert FV and FVIII to FVa and FVIIIa, respectively. TFPI is known to be a potent inhibitor of the TF/FVIIa complex,⁵⁵ and while we believe it is unlikely that TFPI inhibits FXa, we cannot rule out VLDL-associated inhibition of TF/FVIIa at this time.

Recently Ettelaie et al⁵⁶ have reported inhibition of TF activity by apo B-100. However, it is difficult to apply their findings to our procoagulant assay system, since they used apo B-100 reconstituted in soybean PC vesicles and not native lipoprotein particles. Furthermore, they measured the inhibition of rabbit brain TF, and it is unclear whether apo B-100 would have the same effect on human TF reconstituted into purified lipoproteins or PCPS vesicles.

In conclusion, we have observed physiologically relevant, donor-dependent rates of VLDL-mediated thrombin generation in assay systems using appropriately isolated human, fasting lipoproteins and purified human coagulation factors. Lipid organization appears to be a key regulatory factor. We found no evidence for VLDL-mediated factor-specific inhibition.

	Selected Abbreviations and Acronyms

 

-C = cholesterol CHD = coronary heart disease DAPA = dansylarginine-N-(3-ethyl-1,5-pentanediyl)amide F = (coagulation) factor HBS = HEPES-buffered saline IgG = immunoglobulin G PC = L--phosphatidylcholine PL = phospholipid PS = l--phosphatidyl-L-serine TF(PI) = tissue factor (pathway inhibitor) TG = triglyceride

	Acknowledgments

 
This work was supported by Public Health Service awards RO1 HL46696 (to R.P.T.), PO1 HL46703 project 4 (to P.B.T.), PO1 HL46703 project 1 (to K.G.M.), and Environmental Pathology Training Award T3207122 (to M.P.M.).We would like to express our appreciation to Kelly Hayes, MD, PhD for help with blood collection and Jay Silveira for help with platelet isolation.

Received June 10, 1997; accepted November 20, 1997.

	References

Top Abstract Introduction Methods Results Discussion References

 
1. Mann K, Nesheim M, Church W, Haley P, Krishnaswamy S. Surface dependent reactions of the vitamin K dependent enzyme complexes. Blood. 1990;76:1–16.[Abstract/Free Full Text]

2. Nesheim M, Taswell J, Mann K. The contribution of bovine factor V and factor Va to the activity of prothrombinase. J Biol Chem. 1979;254:10952–10962.[Abstract/Free Full Text]

3. Chandler A. An overview of thrombosis and platelet involvement in the development of the human atherosclerotic plaque. In: Glagov S, Newman W, Schaffer S, eds. Pathobiology of the Human Atherosclerotic Plaque. New York, NY: Springer-Verlag; 1990:359–378.

4. White J. Anatomy and structural organization of the platelet. In: Coman R, Hirsh J, Marder V, Salzman E, eds. Hemostasis and Thrombosis: Basic Principles and Clinical Practice Philadelphia, Pa: JB Lippincott Co; 1994:397–413.

5. Tracy P, Allen D, Worfolk L, Robinson-Lawler R. Monocyte/macrophage regulation of coagulation events. Hemostasis. 1996;26(suppl 1):6–11.

6. Bouchard B, Catcher C, Thrash B, Adida C, Tracy P. Effector cell protease receptor-1, a platelet activation-dependent membrane protein, regulates prothrombinase-catalyzed thrombin generation. J Biol Chem. In press.

7. Bouchard B, Shatos M, Tracy P. Human brain pericytes differentially regulate expression of procoagulant enzyme complexes comprising the extrinsic pathway of blood coagulation. Arterioscler Thromb Vasc Biol. 1997;17:1–9.[Abstract/Free Full Text]

8. Gotto A, Farmer J. Risk factors for coronary artery disease and their management. In: Braunwald E, ed. Heart Disease: A Textbook of Cardiovascular Medicine. Philadelphia, Pa: WB Saunders Co; 1988:1153–1187.

9. National Heart, Lung, and Blood Institute. National Heart, Lung, and Blood Institute consensus development panel: treatment of hypertriglyceridemia. JAMA. 1984;251:1196–1200.[Abstract/Free Full Text]

10. Austin M. Epidemiologic associations between hypertriglyceridemia and coronary heart disease. Sem Thromb Hemost. 1988;14:137–142.

11. Menotti A, Scanga M, Morisi G. Serum triglycerides in the prediction of coronary artery disease. Am J Cardiol. 1994;73:29–32.[Medline] [Order article via Infotrieve]

12. Mann K. Factor VII assays, plasma triglyceride levels and cardiovascular disease risk. Arteriosclerosis. 1989;9:783–784.[Free Full Text]

13. Schaeffer E, McNamara J, Genest J, Ordovas J. Clinical significance of hypertriglyceridemia. Semin Thromb Hemost. 1988;14:143–148.[Medline] [Order article via Infotrieve]

14. Miller G, Martin J, Mitropoulos K, Reeves B, Thompson R, Meade T, Cooper J, Cruickshank J. Plasma factor VII is activated by postprandial triglyceridemia, irrespective of dietary fat composition. Atherosclerosis. 1991;86:163–171.[Medline] [Order article via Infotrieve]

15. Juhan-Vague I, Vague P, Alessi M, Badier C, Valadier J, Aillaud M, Atlan C. Relationships between plasma insulin, triglyceride, body mass index, and plasminogen activator inhibitor 1. Diabetes Metab. 1987;13:331–336.

16. Tracy R, Losito R, Williams S, Fried L, Furberg C, Kuller L, Robbins J. Plasma levels of coagulation factors II, IX and X, in addition to factor VII, are associated with plasma levels of cholesterol and triglyceride. Thromb Haemostat.. 1993;69:1081. Abstract.

17. Simpson H, Meade T, Stirling Y, Mann J, Chakrabarti R, Woolf L. Hypertriglyceridaemia and hypercoagulability. Lancet. 1983;1:786–790.[Medline] [Order article via Infotrieve]

18. Mussoni L, Mannucci L, Sirtori M, Camera M, Maderna P, Sironi L, Tremoli E. Hypertriglyceridemia and regulation of fibrinolytic activity. Arterioscler Thromb. 1992;12:19–27.[Abstract/Free Full Text]

19. Hoffman C, Lawson W, Miller R, Hultin M. Correlation of vitamin K–dependent clotting factors with cholesterol and triglycerides in healthy young adults. Arterioscler Thromb. 1994;14:1737–1740.[Abstract/Free Full Text]

20. Tracy R, Losito R, Howard P, Cushman M, Bovill E. The association of plasma lipids and vitamin K-dependent coagulation factors. In: Bray G, Ryan D, eds. Nutrition, Genetics and Heart Disease. Baton Rouge, La: LSU Press; 1996:29–49.

21. Bloom J, Nesheim M, Mann K. Phospholipid-binding properties of bovine factor V and factor Va. Biochemistry. 1979;18:4419–4425.[Medline] [Order article via Infotrieve]

22. Barenholz Y, Gibbs D, Litmann B, Goll J, Thompson E, Carlson F. A simple method for the preparation of homogeneous phospholipid vesicles. Biochemistry. 1977;16:2806–2810.[Medline] [Order article via Infotrieve]

23. Gomori G. A modification of the colorimetric phosphorous determination for use with a photoelectric colorimeter. J Lab Clin Med. 1942;27:955–960.

24. Katzmann J, Nesheim M, Hibbard L, Mann K. Isolation of functional human coagulation factor V by using a hybridoma antibody. Proc Natl Acad Sci U S A. 1981;78:1622–1627.

25. Nesheim M, Katzmann J, Tracy P, Mann K. Factor V. Methods Enzymol. 1981;80:249–274.

26. Bajaj S, Rapaport S, Prodanos C. A simplified procedure for purification of human prothrombin, factor IX and factor X. Prep Biochem. 1981;11:397–405.[Medline] [Order article via Infotrieve]

27. Jesty J, Nemerson J. The activation of bovine coagulation factor X. Methods Enzymol. 1976;45:95–107.[Medline] [Order article via Infotrieve]

28. Owen W, Jackson C. Activation of prothrombin with Oxyuranus scutellatus scutellatus (Taipan snake) venom. Thromb Res. 1973;3:705–710.

29. Laemmli U. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–685.[Medline] [Order article via Infotrieve]

30. Schumaker V, Puppione D. Sequential floatation ultracentrifugation. Methods Enzymol. 1986;128:155–170.[Medline] [Order article via Infotrieve]

31. Mackness M, Durrington P. Lipoprotein Separation and Analysis for Clinical Studies. New York, NY: Oxford University Press; 1992.

32. Havel R, Eder H, Bragdon J. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955;34:1345–1354.

33. Stewart J. Colorimetric determination of phospholipids with ammonium ferrothiocyanate. Anal Biochem. 1980;104:10–14.[Medline] [Order article via Infotrieve]

34. Bajaj S, Harmony J, Martinez-Carrion M, Castellino F. Human plasma lipoproteins as accelerators of prothrombin activation. J Biol Chem. 1976;251:5233–5236.[Abstract/Free Full Text]

35. Swords N, Tracy P, Mann K. Intact platelet membranes, not platelet-released microvesicles, support the procoagulant activity of adherent platelets. Arterioscler Thromb. 1993;13:1613–1622.[Abstract/Free Full Text]

36. Mustard J, Perry D, Ardlie N, Packham M. Preparation of suspensions of washed platelets from humans. Br J Haematol. 1972;22:193–204.[Medline] [Order article via Infotrieve]

37. Tracy P, Nesheim M, Mann K. The platelet factor Xa receptor. Methods Enzymol. 1992;215:329–360.[Medline] [Order article via Infotrieve]

38. George J, Thio L, McManus L, Reimann T. Isolation of platelet membrane microparticles from plasma and serum. Blood. 1982;60:834–840.[Abstract/Free Full Text]

39. Sims P, Faioni E, Wiedmer T, Shattil S. Complement proteins C5–b9 cause release of membrane vesicles from the platelet surface that are enriched in the membrane receptor for coagulation factor Va and express prothrombinase activity. J Biol Chem. 1988;263:18205–18212.[Abstract/Free Full Text]

40. Wiedmer T, Shattil S, Cunningham M, Sims P. Role of calcium and calpain in complement-induced vesiculation of the platelet plasma membrane and in the exposure of the platelet factor Va receptor. Biochemistry. 1990;29:623–632.[Medline] [Order article via Infotrieve]

41. Nesheim M, Tracy R, Tracy P, Boskovic D, Mann K. Mathematical simulation of prothrombinase. Methods Enzymol. 1992;215:316–328.[Medline] [Order article via Infotrieve]

42. Yla-Herttuala S, Palinski W, Rosenfeld M, Parthasarathy S, Carew T, Butler S, Witztum J, Steinberg D. Evidence for the presence of oxidatively modified low density lipoprotein in atherosclerotic lesions of rabbit and man. J Clin Invest. 1989;84:1086–1095.

43. Smith E, Staples E, Dietz H. Role of endothelium in sequestration of lipoprotein and fibrinogen in aortic lesions, thrombi, and graft pseudo-intimas. Lancet. 1979;2:812–816.[Medline] [Order article via Infotrieve]

44. Kunz F, Pechlaner C, Erhart R, Zwierzina W, Kemmler G. Increased lipid binding to thrombi in coronary artery disease. Arterioscler Thromb. 1992;12:1516–1521.[Abstract/Free Full Text]

45. Quinn M, Parthasarathy S, Fong L, Steinberg D. Oxidatively modified low density lipoproteins: a potential role in recruitment and retention of monocyte/macrophages during atherogenesis. Proc Natl Acad Sci U S A. 1987;84:2995–2998.[Abstract/Free Full Text]

46. Shaikh M, Martini S, Quiney J, Baskerville P, La Ville A, Browse N, Duffield R, Turner P, Lewis B. Modified plasma-derived lipoproteins in human atherosclerotic plaques. Atherosclerosis. 1988;69:165–172.[Medline] [Order article via Infotrieve]

47. Harker LA, Hanson SR, Runge MS. Thrombin hypothesis of thrombus generation and vascular lesion formation. Am J Cardiol. 1995;75:12B–17B.[Medline] [Order article via Infotrieve]

48. Miller G, Martin J, Mitropoulos M, Esnouf M, Cooper J, Morrissey J, Howarth D, Tuddenham E. Activation of factor VII during alimentary lipemia occurs in healthy adults and patients with congenital factor XII or factor IX deficiency, but not in patients with factor IX deficiency. Blood. 1996;87:4187–4196.[Abstract/Free Full Text]

49. Higgins D, Callahan P, Prendergast F, Nesheim M, Mann K. Lipid mobility in the assembly and expression of the activity of the prothrombinase complex. J Biol Chem. 1985;260:3604–3612.[Abstract/Free Full Text]

50. Jones M, Lentz B, Dombrose F, Sandberg H. Comparison of the abilities of synthetic and platelet-derived membranes to enhance thrombin formation. Thromb Res. 1985;39:711–724.[Medline] [Order article via Infotrieve]

51. Chapman M. Comparative analysis of mammalian plasma lipoproteins. Methods Enzymol. 1986;128:70–143.[Medline] [Order article via Infotrieve]

52. Broze G, Warren L, Novotny W, Higuchi D, Girad J, Miletich J. The lipoprotein-associated coagulation inhibitor that inhibits the factor VII-tissue factor complex also inhibits factor Xa: insight into possible mechanism of action. Blood. 1988;71:335–343.[Abstract/Free Full Text]

53. Mast A, Broze G. Physiologic concentration of tissue factor pathway inhibitor do not inhibit prothrombinase. Blood. 1996;87:1845–1850.[Abstract/Free Full Text]

54. Kung C, Hayes E, Mann K. A membrane-mediated catalytic event in prothrombin activation. J Biol Chem. 1994;269:25838–25848.[Abstract/Free Full Text]

55. Broze G. Tissue factor pathway inhibitor. Thromb Haemost. 1993;74:90–93.

56. Ettelaie C, James N, Wilbourn B, Adam J, Naseem K, Bruckdorfer K. The mechanism of inhibition of factor III (thromboplastin) activity by apolipoprotein B-100. Arterioscler Thromb Vasc Biol. 1996;16:639–647.[Abstract/Free Full Text]

This article has been cited by other articles:

				C. Oslakovic, M. J. Krisinger, A. Andersson, M. Jauhiainen, C. Ehnholm, and B. Dahlback Anionic Phospholipids Lose Their Procoagulant Properties When Incorporated into High Density Lipoproteins J. Biol. Chem., February 27, 2009; 284(9): 5896 - 5904. [Abstract] [Full Text] [PDF]

				N. Cheung, R. Klein, J. J. Wang, M. F. Cotch, A. F. M. Islam, B. E. K. Klein, M. Cushman, and T. Y. Wong Traditional and Novel Cardiovascular Risk Factors for Retinal Vein Occlusion: The Multiethnic Study of Atherosclerosis Invest. Ophthalmol. Vis. Sci., October 1, 2008; 49(10): 4297 - 4302. [Abstract] [Full Text] [PDF]

				M Leiba, U Seligsohn, Y Sidi, D Harats, B A Sela, J H Griffin, A Livneh, N Rosenberg, I Gelernter, H Gur, et al. Thrombophilic factors are not the leading cause of thrombosis in Behcet's disease Ann Rheum Dis, November 1, 2004; 63(11): 1445 - 1449. [Abstract] [Full Text] [PDF]

				L. H. Kuller Hormone Replacement Therapy and Risk of Cardiovascular Disease: Implications of the Results of the Women's Health Initiative Arterioscler Thromb Vasc Biol, January 1, 2003; 23(1): 11 - 16. [Abstract] [Full Text] [PDF]

				C. H. Toh, J. Samis, C. Downey, J. Walker, L. Becker, N. Brufatto, L. Tejidor, G. Jones, W. Houdijk, A. Giles, et al. Biphasic transmittance waveform in the APTT coagulation assay is due to the formation of a Ca++-dependent complex of C-reactive protein with very-low-density lipoprotein and is a novel marker of impending disseminated intravascular coagulation Blood, September 18, 2002; 100(7): 2522 - 2529. [Abstract] [Full Text] [PDF]

				S. R. Lentz, F. J. Miller Jr, D. J. Piegors, R. A. Erger, J. A. Fernandez, J. H. Griffin, and D. D. Heistad Anticoagulant Responses to Thrombin Are Enhanced During Regression of Atherosclerosis in Monkeys Circulation, August 13, 2002; 106(7): 842 - 846. [Abstract] [Full Text] [PDF]

				S. Ilveskero, P. Siljander, and R. Lassila Procoagulant Activity on Platelets Adhered to Collagen or Plasma Clot Arterioscler Thromb Vasc Biol, April 1, 2001; 21(4): 628 - 635. [Abstract] [Full Text] [PDF]

				H. Deguchi, J. A. Fernandez, I. Pabinger, J. A. Heit, and J. H. Griffin Plasma glucosylceramide deficiency as potential risk factor for venous thrombosis and modulator of anticoagulant protein C pathway Blood, April 1, 2001; 97(7): 1907 - 1914. [Abstract] [Full Text] [PDF]

				P. A. Sakkinen, P. Wahl, M. Cushman, M. R. Lewis, and R. P. Tracy Clustering of Procoagulation, Inflammation, and Fibrinolysis Variables with Metabolic Factors in Insulin Resistance Syndrome Am. J. Epidemiol., November 15, 2000; 152(10): 897 - 907. [Abstract] [Full Text] [PDF]

				M. Kjalke, A. Silveira, A. Hamsten, U. Hedner, and M. Ezban Plasma Lipoproteins Enhance Tissue Factor-Independent Factor VII Activation Arterioscler Thromb Vasc Biol, July 1, 2000; 20(7): 1835 - 1841. [Abstract] [Full Text] [PDF]

				M. C. Minnema, M. E. Wittekoek, N. Schoonenboom, J. J. P. Kastelein, C. E. Hack, and H. t. Cate Activation of the Contact System of Coagulation Does Not Contribute to the Hemostatic Imbalance in Hypertriglyceridemia Arterioscler Thromb Vasc Biol, October 1, 1999; 19(10): 2548 - 2553. [Abstract] [Full Text] [PDF]

				O. Safa, K. Hensley, M. D. Smirnov, C. T. Esmon, and N. L. Esmon Lipid Oxidation Enhances the Function of Activated Protein C J. Biol. Chem., January 12, 2001; 276(3): 1829 - 1836. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a response 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 Moyer, M. P. Articles by Mann, K. G. Search for Related Content PubMed PubMed Citation Articles by Moyer, M. P. Articles by Mann, K. G.