T of Caz+, forms a 1:l stoichiometric complex with the

Transcription

1 Factor Xa Interacts With Two Sites on Monocytes With Different Functional Activities By Laura A. Worfolk, Royce A. Robinson, and Paula B. Tracy Studies were performed to elucidate the functional significance of factor Xa interactions at the monocyte membrane in the presence and absence of factor Va, with respect to prothrombin and factor IX cleavage. Factor Xa-catalyzed prothrombin activation at the monocyte surface was absolutely dependent on the addition of factor Va, indicating that thrombin was generated solely by a membrane-bound complex of factors Va and Xa. In contrast, in the absence of added factor Va. factor Xa bound to monocytes catalyzed the cleavage of factor IX to the nonenzymatic intermediate factor lxa through a reaction that was dependent on both monocyte and factor Xa concentration. At limiting factor Xa concentration, added factor Va inhibited the factor Xacatalyzed cleavage of factor IX, suggesting that a monocytebound complex of factors Va and Xa did not recognize factor IX as a substrate. These combined data suggest that factor Xa interacts with the monocyte through two sites which can be distinguished by their requirement for added factor Va and their expression of different functional activities. Both functional sites could be distinguished also by their differential susceptibility to inhibition by a monoclonal antibody directed against the light chain of factor Va (a-hfv1). At the HE SERINE PROTEASE factor Xa, in the presence T of Caz+, forms a 1:l stoichiometric complex with the nonenzymatic cofactor factor Va on a membrane surface to assemble the prothrombinase complex, which catalyzes the activation of prothrombin to The membrane surface required for this reaction can be provided by peripheral blood cells, including platelet^,^,^ monocytes,ls2 lymphocytes,',2 and neutrophils,2 as well as intact5 and perturbed vascular endothelial cek6 The extent to which these cells can participate in prothrombin activation appears to be dependent on the number of prothrombinase binding sites expressed at their membrane surface^.^.^.^.^ Functional binding interactions of factor Xa with a membrane surface occurring independently of factor Va have also been investigated. Factor Xa bound to synthetic phospholipid vesicles activates both factors V7 and VIII.s More recently, Lawson and Mann9 showed that factor Xa bound to phospholipid vesicles catalyzes the cleavage of factor IX to the nonenzymatic intermediate, factor IXa. This activation intermediate can then be converted to the active enzyme, factor IXaP, by the tissue factor/factor VIIa complex. Both factor Va-dependent and -independent binding interactions of factor Xa with various cells have been shown. Whereas the binding of factor Xa to platelets is absolutely dependent on factor Va,4J0J1 factor Xa binds to monocytes,'j2 vascular endothelial ~ells,~~j~ and some tumor cell^,^^.^^ both directly and through membrane-bound factor Va. The factor Va-dependent factor Xa binding appears to be most important in prothrombin activation, whereas the function of factor Va-independent binding of factor Xa is less well understood. Altieri and Edgington12J7 have reported that in the absence of added factor Va, factor Xa binds to as many as 150,000 sites expressed at the monocyte surface (kd, 10 to monocyte surface, the factor Va/Xa-catalyzed activation of prothrombin was maximally inhibited with 0.25 pmol/l a-hfv1, whereas 1.0 pmol/l a-hfv1 was required to effect 50% inhibition of the factor Xa-catalyzed cleavage of factor IX. The ability of factor Va to modulate factor Xa substrate specificity was investigated further. Factor Xa bound to thrombin-activated platelets either through platelet-released factor Va or added factor Va did not cleave factor IX. Consistent with this result, a plasma concentration of factor IX had no effect on thrombin generation catalyzed by a platelet-bound complex of factors Va and Xa. In marked contrast, factor Xa bound to phospholipid vesicles either independently or in complex with factor Va catalyzed factor IX cleavage with equal efficiency. These combined data indicate that factor Va bound to cell surfaces modulates factor Xa substrate specificity, whereas no discriminatory effect is conferred by factor Va bound to phospholipid vesicles. Thus, by providing two distinct sites at its membrane surface, the monocyte modulates factor Xa binding and the functional activity expressed by the bound enzyme, depending on the availability of factor Va. o 1992 by The American Society of Hematology. 30 nmol/l). This factor Xa-monocyte interaction appears to be mediated through a membrane protein that is similar to the light chain of factor Va, because monoclonal antibodies (MoAbs) directed against the factor Va light chain bound to monocyte~,~~j~ inhibited the subsequent binding of radiolabeled factor Xa,12J7 and precipitated an - 74-Kd protein from monocytes metabolically labeled with 35Smethionine.I2 This factor Va-like protein was named effector protease receptor-1 (EPR-1),I7 and was hypothesized to provide a cofactor-like effect in factor Xa-catalyzed prothrombin activation because factor Xa bound to monocytes in the absence of added factor Va effectively mediated thrombin generation.lz However, previous reports from our laboratory suggest that the addition of factor Va is required for the generation of thrombin at the monocyte surface.ls2 Therefore, the goal of this study was to further elucidate the functional consequences of factor Xa binding to monocytes in the presence and absence of factor Va, as related to both prothrombin and factor IX activation. From the Departments of Cell and Molecular Biology and Biochemistry, University of Vermont College of Medicine, Burlington, W. Submitted February 6, 1992; accepted June 22, Supporled by an American Heart Association Established Investigator Award (P.B. T.) and HL Address reprint requests to Paula B. Tracy, PhD, Department of Biochemistry, University of Vermont College of Medicine, Given Building C201, Burlington, W The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 I992 by The American Society of Hematology J92J $3. OOJO Blood, Vol80, No 8 (October 15). 1992: pp

2 1990 WORFOLK, ROBINSON, AND TRACY MATERIAL AND METHODS Cell isolation. Peripheral blood mononuclear cells (MNCs), devoid of platelets, were prepared daily from 1 U of citrate phosphate dextrose-adenine anticoagulated blood using a modification of the method described by Boyum.ls Briefly, the bu@ coat was diluted (1:4) with Hank s Balanced Salt Solution (HBSS; 137 mmol/l NaCI, 5.4 mmol/l KCI, 0.4 mmol/l KH2P04,0.3 mmol/l Na2HP04, 4.2 mmol/l NaHCO3, 5 mmol/l dextrose) containing 20 mmol/l EDTA and layered over Ficoll/Hypaque (Pharmacia, Piscataway, NJ). Separation of cells was achieved by centrifugation at 600g for 30 minutes at 4 C. The interface containing MNCs was collected and washed twice with 20 mmol/l EDTA-HBSS by centrifugation at 200g for 10 minutes. The MNC pellet was resuspended in 2 to 3 ml of HBSS and counted. Monocytes were purified from MNC suspensions using two protocols that use density gradient centrifugation. In one protocol that is a modificationly of a previously described method, T lymphocytes were removed from MNC suspensions by rosette formation with sheep red blood cells (Crane Laboratories, Syracuse, NY) treated with 2-aminoethylisothiouronium bromide hydrobromide20%21 (AET; Sigma, St Louis, MO), followed by centrifugation over FicoWHypaque. The interface containing monocytes and B cells was washed twice with 20 mmol/l EDTA-HBSS and monocytes were isolated by centrifugation through a continuous Percoll gradient (Pharmacia). The second isolation procedure was a modificationz2 of the method described by Re~aldC.~~ Isolated monocytes were identified by morphology and differentia1 staining using modified Wright s stain. Monocyte purity, using both isolation procedures, was typically greater than 80%, with the major contaminant being lymphocytes. Cell viability (> 98%) was determined by monitoring trypan blue dye exclusion. Human platelets were isolated from freshly drawn blood using acid-citrate-dextrose (ACD) as an anticoagulant (l:6 [vol/vol] anticoagulant to blood). Platelet-rich plasma was isolated from whole blood and platelets were washed as described24 with minor modifications. Apyrase was omitted from the wash solutions, which were acidified to -ph 6.8 with 2.9% (vol/vol) ACD. Isolated platelets were suspended in 5 mmol/l HEPES-buffered Tyrodes s solution (0.137 mol/l NaCI, 2 mmol/l CaC12, 2.7 mmol/l KCI, 12.0 mmol/l NaHC03, 0.36 mmol/l NaHZP04, 1 mmol/l MgC12, 5 mmol/l dextrose), ph 7.35, supplemented with 0.35% bovine serum albumin (ICN, Cleveland, OH). Isolation of coagulant proteins. Coagulation proteins were isolated from human fresh frozen plasma obtained from the American Red Cross (Burlington, VT). Factors IX and X and prothrombin were isolated as described by Bajaj et alz5 In some instances, purified prothrombin was adsorbed with an anti-factor V antibody, a-hfv1 (provided by Dr William Church, University of Vermont College of Medicine), to remove trace amounts of contaminating factor V. Factor X was applied to an or-human protein C immunoaffinity column (obtained from Dr W. Church) to remove trace protein C contamination undetectable by gel electrophoresis. Factor Xa was prepared as described by Jesty and NemersonZ6 using the factor X activator isolated from Russell s viper venom. Taipan snake venom (Sigma) was used to activate prothrombin to a-thrombin, as previously des~ribed.~ Factor V was isolated using immunoaffinity chromatography as described:s and activated with 3 NIH U/mL of thrombin for 10 minutes at 37 C. Protein purity was assessed by gel electrophoresis before and after reduction with 5% 2mercaptoethanol (vol/vol) in either 5% to 15% gradient or 10% slab gels according to the methods described by Laemmli.2y Gels were stained with Coomassie brilliant blue R-250 to visualize proteins. Molecular weights and extinction coefficients, 2ignm, of the various proteins were taken as follows: factor V, 330,000,9.630; factor Xa, 50,000, ; factor IX, 57,000, ; prothrombin, 72,000, ; and thrombin, 37,000, Preparation and characterization of Iz5I-factor IX. Factor IX was radioiodinated using the IODO-GEN (Pierce, Rockford, IL) transfer technique as previously described for radiolabeling of factor V. NalZ5I (0.08 to 1.2 mci/0.2 mg protein) was added to an IODO-GEN coated tube (1 pg of IODO-GEN/lO wg of protein) in 100 pl of 0.02 mol/l Tris, 0.15 mol/l NaCI, ph 7.4. After 5 minutes of gentle vortexing, the oxidized isotope was transferred to a separate tube containing factor IX (0.4 mg/ml) in 0.02 mol/l Tris, 0.15 mol/l NaCl containing 5 mmol/l benzamidine, ph 7.4, and incubated on ice for 5 minutes. Labeled protein was separated from free isotope by gel filtration with a Sephadex G column (Sigma), using 0.02 mol/l Tris, 0.15 mol/l NaCI, ph 7.4, to develop the column. Labeled protein fractions (> 95% precipitable with 10% trichloroacetic acid) were pooled and then dialyzed first against 0.02 mol/l HEPES, 0.15 mol/l NaCI, ph 7.4, to remove the benzamidine, and finally into 50% glycerol, and stored at -20 C. Specific radioactivities ranged from 4,460 to 9,677 cpm/ng of protein. To confirm that the labeled and unlabeled protein were identical substrates, an initial experiment was performed comparing the rate of cleavage of factor IX and lz5i-factor IX by factor Xa, as described below. No significant difference in cleavage was observed between the labeled and unlabeled protein (data not shown). Cleavage of 1251-factor IX by factor Xu. All assays were performed in 0.02 mol/l HEPES, 0.15 mol/l NaCI, 5 mmol/l CaC12, ph 7.4. Freshly isolated peripheral blood monocytes (0.01 to 2.0 x 107/mL) were incubated with factor Xa (10 to 200 nmol/l) for 10 minutes at 37T, followed by the addition of lz5i-factor IX (100 nmol/l) to initiate the reaction. Control reaction mixtures included 1251-factor IX alone, monocytes plus Iz51-factor IX (no factor Xa added), and lz5i-factor IX plus factor Xa (at the indicated concentrations, without monocytes added). At timed intervals, aliquots were removed and added to EDTA (25 mmol/l, final concentration) to quench the reaction. The subsamples were microcentrifuged at 13,OOOg for 30 seconds and the supernatant was removed, frozen, and lyophilized. The cleavage of factor IX was monitored by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using 10% to 20% polyacrylamide gels as described by Laemmli under reducing condition^.^^ After electrophoresis, dried gels were subjected to autoradiography at -70 C using Kodak XR-1 film (Eastman Kodak, Rochester, NY) and DuPont Lighting Plus intensifying screens (DuPont, Wilmington, DE). The activation of factor IX on the platelet surface was studied as follows. Platelets (2 x 10s/mL) were activated with 1 NIH U/mL thrombin for 2 minutes at room temperature. After activation, thrombin was inhibited by 3.0 pmol/l dansylarginine N,N-(3-ethyl- 1,5-pentanediyl)amide (DAPA; a gift from Dr K.G. Mann, University of Vermont College of Medicine). Factor Xa (5 nmol/l) was added and incubated for 2 minutes at 37 C to facilitate its binding to the activated platelet surface via platelet-released factor Va (- 2 nmol/l). In some reactions, factor Va (5 nmol/l) was added simultaneously with factor Xa. The reactions were initiated with 25I-factor IX (100 nmol/l) to assay for factor IX cleavage as detailed above. The activation of factor IX by factor Xa bound to synthetic phospholipid (PCPS) vesicles (200 pmol/l) was studied using protocols similar to those described above. PCPS vesicles composed of 75% (wt/wt) L-a-phosphatidylcholine (hen egg) and 25% (wt/wt) L-a-phosphatidylserine (bovine brain) (Sigma) were prepared as previously described.33 In some experiments, factor Va (50 to 100 nmol/l) was added to the monocyte suspensions or PCPS vesicles for 2 minutes before the addition of factor Xa.

3 FACTOR XA INTERACTIONS WITH MONOCYTES 1991 Residual thrombin used to activate factor V was inhibited by 3 pmol/l DAPA. In antibody inhibition experiments, 0.05 to 1.0 pmol/l a-hfv1, an MoAb directed against human factor V/Va,28 or a nonimmune mouse IgG (ICN, Costa Mesa, CA) was preincubated with the cells for 10 minutes at 37 C. The reactions were subsequently initiated by the simultaneous addition of factors Xa and lzi-factor IX. In control experiments, cells were incubated in the absence of antibody and treated in the same manner. Progress curves of the cleavage of factor IX by factor Xa bound to monocytes, PCPS vesicles, or platelets were generated by densitometric analyses of the autoradiographs using a Microscan 1000 scanning densitometer (TRI, Inc, Nashville, TN) as described previously.' Data were expressed as integrated volumes for each protein band using the arbitrary density units of the scanning system. Assessment of prothrombinase complex assembly and function at the monocyte surface. Prothrombinase complex assembly and function at the monocyte surface was monitored by assessing the rate of thrombin formation. Assay mixtures consisted of monocytes (5 x 106/mL), 2 mmol/l CaC12, 1.39 pmol/l prothrombin, and the indicated concentrations of factors Va and Xa in 0.02 mol/l HEPES, 0.15 mol/l NaC1, ph 7.4. Factors Va and Xa were added to the cells (2 minutes at 22 C) and the reactions were initiated by the addition of prothrombin. Aliquots of assay mixtures were removed at various timed intervals (0, 1, 2, 4, 7, and 10 minutes) and added to an equal volume of 0.02 mol/l HEPES, 0.15 mol/l NaCI, 25 mmol/l EDTA, ph 7.4, to quench the reaction. Thrombin concentration in each sample was determined using the chromogenic substrate, Spectrozyme TH (0.4 mmol/l; American Diagnostica, Inc, Greenwich, CT). The change in sample absorbance at 405 nm was monitored using a Molecular Devices (Palo Alto, CA) V, spectrophotometer, and compared with a thrombin standard curve (0 to 200 nmol/l) prepared daily using purified thrombin. The initial rate of thrombin generated in the various assay mixtures was calculated by linear regression analysis of the data obtained from the subsamples removed over time. RESULTS The requirement of factor Vu for thrombin formation at the monocyte surface. Peripheral blood monocytes, isolated using density gradient separation techniques (and maintained in suspension), were used to study the effect of factors Va and Xa on the rate of prothrombin activation. Analysis of kinetics of prothrombin activation had shown previously that the assembly of the prothrombinase complex at the monocyte surface was consistent with the formation of a 1:l stoichiometric complex of factors Va and Xa resulting in rapid thrombin generation (34 mol IIa/s/ site).* To investigate the possibility that factor Xa bound to the monocyte surface independently of added factor Va could function in this capacity, experiments were performed to determine the rate of thrombin formation at increasing concentrations of factor Xa in the presence or absence of a rate-saturating concentration of factor Va (20 nmol/l). Because the substrate prothrombin was present at a physiologic concentration (1.39 pmol/l), it was preadsorbed with a-hfvl, an MoAb directed against factor Va, to eliminate a potential source of contaminating cofactor. Thrombin generation at the monocyte surface was undetectable when factor Va was omitted from the reaction mixtures (Fig l), even though factor Xa concentrations as high as 50 0 AT a FACTOR Xa (M x 109) Fig 1. The rate of thrombin generation at the monocyte surface as a function of the nominal factor Xa concentration. Prothrombin activation reaction mixtures contained 1.39 pmol/l prothrombin, preadsorbed with an antihuman factor Va MoAb, a-hfvi, 3 pmol/l DAPA, 2 mmol/l CaCI,, 5 x 106 monocytes/ml, and the indicated concentrations of factor Xa in the presence (0) or absence (A) of 20 nmol/l factor Va in 20 mmol/l HEPES, 0.15 mol/l NaCI, ph 7.4. The rate of thrombin formation was determined from reaction subsamples removed at timed intervals and subsequently assayed for thrombin as described in Materials and Methods. nmol/l were used and an assay system was used in which as little as 1.5 nmol/l thrombin could be detected. In contrast, in the presence of factor Va, the rate of thrombin generation was dependent on the nominal factor Xa concentration (Fig l), as shown previously.* These data show that, under the conditions of these studies, factor Va is absolutely required for detectable prothrombin activation at the monocyte surface. The functional signijicance of factor Vu-independent factor Xa binding to the monocyte surface. Based on the recent studies of Lawson and Mann,9 we hypothesized that factor Xa bound to the monocyte surface in the absence of factor Va may function in factor IX cleavage. To investigate this possibility, factor Xa (100 nmol/l) was incubated with freshly isolated monocytes (0.01 to 2 x 107/mL) in the presence of 1251-factor IX (100 nmol/l). The factor Xa concentration was chosen to ensure saturation of all available binding sites at the highest monocyte concentration used. Data shown in Fig 2 show that the rate and extent of factor IX cleavage was proportional to cell concentration, indicating that factor Xa bound to monocytes, independently of factor Va, supported the time-dependent cleavage of factor IX to the activation intermediate factor IXa. Factor IX cleavage did not occur to any significant extent (less than 4%) in the absence of cells, indicating that factor Xa bound to the monocyte surface is the effective catalyst. In addition, a cell-associated proteolytic activity that occurred in the absence of added factor Xa was observed that resulted in the cleavage of factor IX to yield a peptide of -50 Kd, as well as the factor IXa heavy and light chains (Fig 2, lane C3). The extent to which these cleavages occurred was dependent on cell concentration, although cell-associated factor IX cleavage did not appear to increase upon extended incubation. The - 50-Kd peptide that resulted from this cleavage appeared to be a substrate for added factor Xa because it was consumed over time.

4 From by guest on October 26, For personal use only. WORFOLK, D $z J 30 b 20 ap 10 n '5 20 TIME (minutes) Fig 2. The time-dependent cleavage of V-factor IX catalyzed by factor Xa bound to monocytes. Varying concentrations of monocytes (as indicated) were incubated with factor Xa (100 nmol/l) for 10 minutesat 37 C. followed by the addition of a plasma concentrationof '=I-factor IX (100 nmol/lj. At 5, 10, and 30 minutes (A, 6, and C, respectively). aliquots were removed, quenched in EDTA, and subsequently subjected to SDS-PAGE under reducing conditions, followed by autoradiography and densitometric analysis, as described in Materials and Methods. Three control reactions were performed: C1, lnl-factor IX alone; C2, '=I-factor IX plus factor Xa; and C3, '%factor IX plus cells (2 x 107/mL). (D) Relative concentrations of factor IXa light chain formed and factor IX remaining (inset) over time, as determined by densitometricanalysis of the gels (A through C). Each band was plotted as a percentage of total density per lane. To determine if the cleavage of factor IX to factor IXa was dependent on factor Xa concentration, a fixed concentration of monocytes (2 x 107/mL) was incubated with increasing concentrations of factor Xa (10 to 200 nmol/l) ROBINSON,AND TRACY and 12sI-factorIX cleavage was assayed as described in Materials and Methods. As shown in Fig 3, the cleavage of factor IX by factor Xa bound to monocytes was dependent on the concentration of added factor Xa. The extent of cleavage appeared to saturate at -30 to 50 nmol/l factor Xa, presumably due to complete occupation of a limited number of factor Xa binding sites at the monocyte surface. Comparison of the data shown in Fig 3 with the data shown in Fig 1 indicates that, in the absence of added factor Va, the concentration of factor Xa (30 to 50 nmol/l) that effected near-maximal factor IX cleavage had no effect on prothrombin activation. These combined data suggest that factor Xa expresses two different functional interactions at the monocyte surface. Whereas factor Xa-catalyzed activation of prothrombin is dependent on factor Va, factor Xa-catalyzed cleavage of factor IX occurs independently of added factor Va. The effectof factor Va on the factor Xa-catalyzed cleavage of factor IX at the monocyte surface. To determine how factor Va might influence this reaction, the ability of varying concentrations of factor Xa to catalyze the formation of factor IXa at the monocyte surface was assessed in both the presence and absence of factor Va (50 nmol/l). This concentration of factor Va was chosen to ensure saturation of all available factor Va binding sites that would support the subsequent binding of factor Xa. The data shown in Fig 4 indicate that the factor Xa-catalyzed cleavage of factor IX was inhibited in the presence of factor Va. Analysis of the amount of factor IX consumed (or factor IXa formed) at 30 minutes indicated that 57% inhibition occurred with 20 nmol/l factor Xa, whereas 39% inhibition occurred with 30 nmol/l enzyme. This observation was interpreted to indicate that, as the factor Xa concentration was increased, factor Xa binding to the factor Va-independent sites was likewise increased, resulting in greater cleavage of factor IX. These data suggest that factor Xa in complex with bound factor Va at the monocyte surface does not recognize factor IX as a substrate or, at best, cleaves it less efficiently than factor Xa bound to the monocyte surface independcntly of factor Va. Because factor Xa does not bind independently to platelets and only interacts through membrane-bound factor Va,1"J1J4the ability of a platelet-associated complex of factors Va and Xa to cleave factor IX was examined. As shown in Fig5, factor Xa (5 nmol/l) bound to the activated platelet surface via platelet-released factor Va ( 2 nmol/l) and/or added plasma factor Va (5 nmol/l) did not catalyze the time-dependent cleavage of factor IX. However, a limited proteolytic activity was observed that was similar to the factor Xa-independent cleavage associated with monocytes. In additional experiments, a plasma concentration of factor IX (100 nmol/l) had no effect on the prothrombin activation catalyzed by a platelet-bound complex of factors Va and Xa (data not shown). These results suggest that factor Xa associated with the platelet surface (which only binds through interaction with factor Va), shows absolute substrate specificity for prothrombin, whereas the monocyteassociated complex of factors Va and Xa recognizes prothrombin as substrate and may also recognize factor IX, -

5 FACTOR Xn INTERACTIONS WITH MONOCYTES 1993 A c [x~], ~ ~ 1 Fig 3. The factor Xa-dependent cleavage of factor IX at the monocyte surface. The indicated concentrations of factor Xa were incubated for 10 minutes at 37 C in the presence (+) or absence I-) of monocytes (2 x 107/mL), followed by the addition of 'SI-factor IX (100 nmoi/l). Allquots were removed at various times and treated as described in Materials and Methods. (A) Cleavage obtained at 10 minutes; lane C represents 'SI-factor IX alone. (E) Relative concentrations of factor IX remaining (0) and factor IX light chain formed (0) at (-) 10 and (---) 30 minutes, respectively, as determined by scanning densitometry. although cleavage appears to proceed relatively slowly, if at all. Because factor Va bound to both the monocyte and platelet surface appeared to modulate factor Xa substrate specificity, it was of interest to determine if similar effects were observed with phospholipid vesicles. The addition of factor Va (100 nmol/l) at a concentration sufficient to cause all the factor Xa (80 nmol/l) to form a factor Va/Xa complex on the surface of synthetic phospholipid vesicles (200 pmol/l) was without effect on the rate of factor IX cleavage when compared with the rate obtained without factor Va (Fig 6). Therefore, factor Va bound to cell surfaces appears to modulate factor Xa substrate specificity, whereas factor Va bound to phospholipid vesicles does not provide this discriminatory effect. Inhibitoty effects of an antihuman factor VMoAb. Altieri and Edgington have proposed that the factor Va-independent factor Xa binding site on the monocyte surface is a molecule that is strikingly similar to the light chain of factor Va.12J7 Consequently, experiments were performed to determine if an MoAb specific for the light chain of factor Va, a-hfv1, would inhibit factor Xa-catalyzed cleavage of factor IX at the monocyte surface. A dose-dependent inhibition of factor IX cleavage occurred when monocytes were preincubated with a-hfv1, with 50% inhibition observed at the highest antibody concentration (1 pmol/l) E 401 % =e== FACTOR xo (M x 109) used relative to a nonspecific control antibody. These data support the concept that factor Xa interacts with the monocyte through a factor Va-like molecule (perhaps EPR-1) expressed at the surface membrane, and that membrane-bound factor Xa can catalyze the cleavage of factor IX to factor IXa. In contrast to factor IX cleavage, prothrombin cleavage catalyzed by factor Xa was exquisitely sensitive to a-hfvl concentration. Addition of a-hfv1 at a concentration as low as 0.25 pmol/l was sufficient to almost completely inhibit factor Xa-catalyzed thrombin generation at the monocyte surface (data not shown). As indicated in Fig 7, this concentration of antibody had minimal effect on the factor Xa-catalyzed cleavage of factor IX. The differential susceptibility of these two factor Xa interactions with the monocyte to inhibition by an a-human factor Va MoAb provide additional support to the hypothesis that the two sites are functionally distinct. DISCUSSION Studies were performed to distinguish the potential functional consequences of factor Xa interactions at the monocyte surface in the presence and absence of factor Va, as related to prothrombin and factor IX cleavage. The results obtained support the concept that factor Xa interacts with two distinct sites at the monocyte surface and, in 0

6 1994 WORFOLK, ROBINSON, AND TRACY a 100, I loo FACTOR Xo (nu) TIME (minutes) Fig 4. Factor Va inhibits the factor Xa-catalyzed cleavage of factor IX at the monocyte surface. Monocytes (2 x 107/mL) were preincubated in the (---) presence or (-1 absence of factor Va (50 nmolll) for 2 minutes at 37 C. Factor Xa (20 or 30 nmol/l) was added for 10 minutes, followed by '%factor IX (100 nmol/l) to initiate the reaction. Aliquots were removed at various times and treated as described in Materials and Methods. The relative concentrations of factor IX remaining (0) and factor IX light chain formed (0) at 30 minutes was assessed by scanning densitometry as described in Materials and Methods. The data are representative of five experiments with three different donors. so doing, expresses different functional activities. As modeled in Fig 8, the direct interaction of factor Xa with the monocyte (perhaps mediated through a factor Va-like protein, EPR-1) did not serve as an effective catalyst for prothrombin activation; however, it effectively catalyzed the cleavage of factor IX to the activation intermediate factor IXa. In contrast, the association of factor Xa with monocyte-bound factor Va resulted in prothrombin activation, whereas factor IX cleavage was inhibited. Additional support for the existence of two sites was provided by their different susceptibility to inhibition by an a-human factor A 8 Fig 6. The factor Va/factor Xa complex bound to PCPS vesicles supports the cleavage of factor IX to factor IXa The factor Xacatalyzed (80 nmol/l) cleavage of 'Wfactor IX (100 nmol/l) on PCPS vesicles (200 pmol/l) was assessed in the (-1 presence or (-) absence of factor Va (100 nmol/l) as described in Materials and Methods. The relative concentrations of factor IX remaining (0) and factor IX light chain formed (0) were determined by densitometric analysis as described in Materials and Methods. Va MoAb, a-hfvl. Antibody concentrations that completely inhibited factor Va-dependent prothrombin activation had minimal effect on factor Va-independent factor IX cleavage. Factor Va is required for prothrombin activation at the monocyte surface. The ability of a-hfvl to inhibit the factor Xa-catalyzed cleavage of factor IX supports the observations of Altieri and Edgingt~n,~*J~ who showed that the direct interaction of factor Xa with the monocyte is mediated through a membrane protein similar to the factor Va light chain. Complex formation between this protein and factor Xa led to prothrombin activation at the monocyte surface, suggesting that the monocyte-associated factor C Fig 5. A platelet-associated complex of factors Va and Xa does not support the cleavage of factor IX to factor IXa Platelets (2 x 10"mL) suspended in 5 mmol/l HEPES-buffered fyrode's solution containing albumin (0.35%) were activated with thrombin (1 NIH U/mL) for 2 minutes at room temperature, followed by the addition of 3.0 pmol/l DAPA to inhibk thrombin. Activated platelets were incubated with '%factor IX alone, '"1-factor IX plus factor Xa (5 nmol/l), or '%factor IX with equimolar concentrations of factor Xa and factor Va (5 nmol/l) At the indicated times, aliquots were removed and treated as described in Materials and Meth- C +IX Xa+IX Va+Xa+IX ods.

7 From by guest on October 26, For personal use only. FACTOR XA INTERACTIONS WITH MONOCYTES 1995 A Om25 O m 1 00 Fig 7. An MoAb directed against the light chain of human factor Va, crhfv1, inhibits the factor Xa-catalyzed cleavage of factor IX to factor IXa at the monocytesurface. (A) Varying concentrations of crhnl(o.05 to 1.0 pmol/l) were preincubated with monocytes (2 x lo /ml) for 10 minutes at 37 C. The reactions were initiated by the simultaneous addition of factor Xa (50 nmol/l) and l=i-factor IX (100 nmol/l) and cleavage assessed at 30 minutes as described in Materials and Methods. (B) Relative concentrations of factor IX remaining( 0 )and factor IX light chain formed ( 0 )as determined by scanning densitometry. Lanes 1 through 6, cells plus the indicated concentrations of a-hfv1; lane 7, nonimmune mouse IgG (1.0 pmol/l) plus cells. Va-like molecule exerted a cofactor-like effect relative to thrombin generation. Therefore, we investigated the hypothesis that factor Xa bound independently of factor Va to the monocyte could catalyze prothrombin activation. To ensure that we were studying a monocyte-associated cofactor effect, the substrate present at micromolar concentration was preadsorbed with a-hfvl to eliminate a potential source of contaminating cofactor (although we had no evidence that our prothrombin preparation contained any contaminating factor Va). In contrast to the work published by Altieri and Edgington,? our data show that thrombin generation at the monocyte surface was undetectable in the absence of added factor Va, even though factor Xa concentrations ( nmol/l) were chosen to saturate all its documented factor Va-independent binding sites (n = 150,000; kd, 10 to 30 n ~ n o l / L ). ~This ~ J ~observation was not due to a lack of assay sensitivity, because the method used was capable of detecting the minimal amount of thrombin that would have been generated in the absence of any cofactor effect. Our experimental conditions and assay system were established assuming that the monocytes were expressing the reported number of factor Xa binding sites. Therefore, our inability to detect thrombin generation in the absence of added factor Va suggested, as one possibility, that the monocytes used in our studies may express significantly fewer factor Xa binding sites. In preliminary experi a 0 B,- lot [ANTIBODY]. M X 1.oo 1.25 lo6 ments, measurements of the direct binding of lsi-factor Xa to monocytes indicated that factor Xa bound to less than 5,000 factor Va-independent sites, a number significantly lower than previously reported. However, even if only 5,000 factor Xa binding sites were occupied, our assay system would have detected a mere 20-fold increase in the rate of thrombin generated due to a monocyte-associated cofactor effect. The cofactor effect provided by added factor Va has been shown to increase the rate of thrombin generation by four orders of magnitude. s Therefore, our inability to detect thrombin strongly suggests that factor Xa bound to the monocyte surface independently of factor Va does not function in prothrombin activation any more efficiently than factor Xa bound to a synthetic phospholipid m~mbrane.- ~ This observation is consistent with the current notion of how factor Va functions in the prothrombinase complex. By virtue of its lipid-binding characteristics, the light chain of the factor Va molecule forms at least part of the binding site for factor Xa at the membrane s~rface,~. ~ whereas the heavy chain is required to bind the substrate prothrombin.3xthus, both subunits are required to coconcentrate enzyme and substrate at the membrane surface, resulting in effective thrombin generation.3y Our inability to detect factor Xa-catalyzed thrombin generation and to demonstrate less than 5,000 factor Xa binding sites at the monocyte surface are in contrast to the studies by Altieri and Edgington.12J7These disparities may

8 1996 WORFOLK, ROBINSON, AND TRACY VaaXa Xa Fig 8. Factor Va-dependent and -independent factor Xa binding interactions at the monocyte surface express different functional activities. Factor Xa (Xa) in complex with factor Va [Va) at the monocyte surface catalyzes the activation of prothrombin (11) to thrombin (Ma) in a Ca2+-dependent reaction. The formation of thrombin is absolutely dependent on the cofactor, factor V; therefore, factor Xa bound independently of factor Va does not participate in effective thrombin generation. The cleavage of factor IX (1x1 to factor IXa (IXa) is mediated by factor Xa bound to the cell membrane independently of factor Va, and perhaps to a much lesser extent by the factor Va/Xa complex. It has been proposed that factor Xa bound independently of factor Va is mediated by a factor Va-like molecule, EPR-1.'2.'7 The inhibition of the factor Xa-catalyzed cleavage of factor IX by a-hfvi supports this hypothesis. Our results support the notion that factor Xa bound to monocytes in the presence or absence of factor Va expresses two different functional activities, prothrombin activation and factor IX cleavage, respectively. result from the different protocols used for monocyte isolation. Our protocols rely solely on density gradient centrifugation techniques, whereas those used by Altieri and Edgington use adherence for cell i~olation.~~j~ Adherence techniques may preferentially select for monocytes that are enriched in EPR-1 or, alternatively, may expose EPR-1 in a configuration capable of binding more factor Xa. Factor Xa bound to the monocyte independently of factor Va catalyzes the cleavage of factor IXto factor IXa Because the direct binding of factor Xa to monocytes did not appear to function in prothrombin activation (Fig l), other potential functional activities were investigated. Studies detailed in this report show that the monocyte supports the factor Xa-catalyzed cleavage of factor IX to the activation intermediate factor IXa. The reaction was observed to depend both on factor Xa (Fig 2) and cell (Fig 3) concentration, indicating that the membrane-bound enzyme is the effective catalyst. These observations are consistent with those of Lawson and Mann? who showed that factor Xa bound to synthetic phospholipid vesicles catalyzes the formation of factor IXa, which could be converted to the serine protease factor IXaP by the tissue factor/factor VIIa complex. Stimulation of monocytes by a variety of agonists results in tissue factor expression at their surface.40 The effect of these same agonists on the direct binding of factor Xa to the monocyte surface (perhaps by influencing EPR-1 expression) has not been reported. However, it is reasonable to speculate that the factor Xa-mediated formation of factor IXa may facilitate the tissue factor/factor VIIa-catalyzed formation of factor IXaP at the monocyte surface. Our results also show that monocytes express a proteolytic activity (in the absence of factor Xa) that resulted in the generation of a - 50-Kd peptide, as well as the factor IXa heavy and light chains. The -50-Kd peptide was a substrate for factor Xa, as evidenced by its consumption over time (in experiments in which factor Xa was added). A similar activity was observed with thrombin-activated platelets. This proteolytic activity was unique to cells and was not observed in similar studies using synthetic phospholipid vesicles, ruling out the possibility of a contaminating protease in our reagents. The nature and expression of this protease warrants future investigation. Factor Va and the membrane surface to which it is bound alter factor Xa substrate specificity. Factor Va markedly enhanced factor Xa-catalyzed thrombin generation at the monocyte surface. However, when factor IX was used as the substrate, factor IXa formation was inhibited by the addition of factor Va, indicating that the factor Va/Xa complex assembled on the monocyte surface did not cleave factor IX as efficiently as factor Xa alone. Most likely, added factor Va, at limiting factor Xa concentrations, caused factor Xa to partition between factor Va-dependent and -independent binding sites. Greater inhibition was observed with low concentrations of factor Xa, consistent with the published dissociation constants governing the direct binding of factor Xa to the monocyte surfacelzj7 versus its binding to monocyte-associated factor Va.2 However, as the factor Xa concentration increased, the amount of factor Xa associated with factor Va remained saturated, while factor Xa bound to the factor Va-independent site increased, thus resulting in greater factor IX cleavage. These results suggest that factor Xa substrate specificity is altered by complex formation with factor Va at the monocyte surface. Likewise, factor Xa bound to thrombin-activated platelets via platelet-released factor Va or added plasma factor Va showed absolute specificity for prothrombin. In marked contrast, no substrate discrimination was observed when factor IX cleavage catalyzed by either factor Xa or a complex of factor Va and Xa bound to PCPS vesicles was assessed. These combined results suggest that factor Xa substrate specificity is mediated by the presence of both the cofactor and the cell membrane surface to which the enzyme complex is bound. With the onset of a coagulant response, both factors Va and Xa would be generated. Therefore, factor Xa could partition between both the factor Va-dependent and -independent sites. Because factor Xa can interact with - 16,000 factor Va-dependent sites with high affinity (kd, -1O-l0 mol/l),2 these sites would be saturated at relatively low concentrations of both cofactor and enzyme. Factor Xa complex formation with monocyte-associated factor Va would dictate that it functions in prothrombin activation. In contrast, factor Xa bound to the monocyte surface (kd, 2 10 nm~l/l)'~j~ would not function in prothrombin activation, but rather would participate in other coagulant reactions, one being the generation of the factor IXaP activation intermediate, factor IXa. ACKNOWLEDGMENT We thank Dr Russell P. Tracy for his technical expertise with the scanning densitometer and his critical review of the manuscript. The Blood Drawing Services of the General Clinical Research Center of the Medical Center Hospital of Vermont are gratefully acknowledged.

9 FACTOR XA INTERACTIONS WITH MONOCYTES Tracy PB, Rohrbach MS, Mann KG: Functional prothrombinase complex assembly on isolated monocytes and lymphocytes. J Biol Chem 258:7264, Tracy PB, Eide LL, Mann KG: Human prothrombinase complex assembly and function on isolated peripheral blood cell populations. J Biol Chem 260:2119, Mann KG: The assembly of blood clotting complexes on membranes. Trends Biochem Sci 12:229, Miletich JP, Jackson CM, Majerus PW: Properties of the factor Xa binding site on human platelets. J Biol Chem 253:6908, Tracy PB: Cellular involvement in coagulation, in Cunningham DD, Long GL (eds): Workshop Report in: Proteases in Biological Control and Biotechnology. New York, NY, Liss, 1987, p Visser MR, Tracy PB, Vercellotti GM, Goodman JL, White JG, Jacob HS: Enhanced thrombin generation and platelet binding on herpes simplex virus-infected endothelium. Proc Natl Acad Sci USA 85:8227, Monkovic DD, Tracy PB: Activation of human factor V by factor Xa and thrombin. Biochemistry 29:1118, Neuenschwander P, Jesty J: A comparison of phospholipid and platelets in the activation of human factor VI11 by thrombin and factor Xa, and in the activation of factor X. Blood 72:1761, Lawson JH, Mann KG: Cooperative activation of human factor IX by the human extrinsic pathway of blood coagulation. J Biol Chem 266:11317, Kane WH, Lindhout MJ, Jackson CM, Majerus PW: Factor Va-dependent binding of factor Xa to human platelets. J Biol Chem255:1170, Tracy PB, Peterson JM, Nesheim ME, McDuffie FC, Mann KG: Interaction of coagulation factor V and factor Va with platelets. J Biol Chem 254:10354, Altieri DC, Edgington TS: Sequential receptor cascade for coagulation proteins on monocytes. Constitutive biosynthesis and functional prothrombinase activity of a membrane form of factor VIVa. J Biol Chem 264:2969, Rogers GM, Shuman MA: Characterization of the interaction between factor Xa and bovine aortic endothelial cells. Biochim Biophys Acta 844:320, Stern DM, Nawroth PP, Handley D, Kisiel W: An endothelial cell-dependent pathway of coagulation. Proc Natl Acad Sci USA 82:2523, Sakai T, Kisiel W: Binding of human factors X and Xa to HepG2 and 582 human tumor cell lines. J Biol Chem 265:9105, Van De Water L, Tracy PB, Aronson D, Mann KG, Dvorak HF: Tumor cell generation of thrombin via functional prothrombinase assembly. Cancer Res , Altieri DC, Edgington TS: Identification of effector cell protease receptor-1. A leukocyte-distributed receptor for the serine protease factor Xa. J Immunol145:246, Boyum A: Isolation and removal of lymphocyte from bone marrow of rats and guinea-pigs. Scand J Clin Lab Invest 21:91, Robinson RA, Worfolk L, Tracy PB: Endotoxin enhances the expression of monocyte prothrombinase activity. Blood 79:406, 1992 REFERENCES 20. Pellegrino MA, Ferrone S, Dierich MP, Reisfeld RA Enhancement of sheep red blood cell human lymphocyte rosette formation by the sulfhydryl compound 2-amino ethylisothiouronium bromide. Clin Immunol Immunopathol3:324, Pellegrino MA, Ferrone S, Theofilopoulos AN: Isolation of human T and B lymphocytes by rosette formation with 2-amino ethylisothiouronium bromide (AET)-treated sheep red blood cells and with monkey red blood cells. J Immunol Methods 11:273, DiCorleto PE, De La Motte C A Thrombin causes increased monocytic cell adhesion to endothelial cells through a protein kinase C-dependent pathway. Biochem J 264:71, Recalde HR: A simple method of obtaining monocytes in suspension. J Immunol Methods 69:71, Mustard JF, Perry DW, Ardlie NG, Packham MA: Preparation of suspension of washed platelets from humans. Br J Haematol 22:193, Bajaj SP, Rappaport SI, Prodanos C: A simplified procedure for purification of human prothrombin, factor IX and factor X. Prep Biochem 11:394, Jesty J, Nemerson J: The activation of bovine coagulation factor X. Methods Enzymol45:95, Owen WG, Jackson CM: Activation of prothrombin with oxyuranus scutellatus (Taipan snake) venom. Thromb Res 3:705, Katzmann JA, Nesheim ME, Hibbard LS, Mann KG: Isolation of functional human coagulation factor V by using a hybridoma antibody. Proc Natl Acad Sci USA 78:162, Laemmli UK Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680, Tracy PB, Eide LL, Bowie ET, Mann KG: Radioimmunoassay of factor V in human plasma and platelets. Blood 60:59, Di Scipio RG, Hermodson MA, Yates SG, Davie EW A comparison of human prothrombin, factor IX (Christmas factor), factor X (Stuart factor), and protein S. Biochemistry 16:698, Mann KG: Prothrombin. Methods Enzymol45:123, Barenholz Y, Gibbs D, Litmann BJ, Go11 J, Thompson E, Carlson FD: A simple method for the preparation of homogenous phospholipid vesicles. Biochemistry 16:2806, Kane WH, Majerus PW: The interaction of human coagulation factor Va with platelets. J Biol Chem 257:3963, Nesheim ME, Taswell JB, Mann KG: Contribution of bovine factor V and factor Va to the activity of prothrombinase. J Biol Chem 254:10952, Tracy PB, Mann KG: Prothrombinase complex assembly on the platelet surface is mediated through the 74,000-dalton component of factor Va. Proc Natl Acad Sci USA 80:2380, Higgins DL, Mann KG: The interaction of bovine factor V and factor V-derived peptides with phospholipid vesicles. J Biol Chem 258:6503, Guinto ER, Esmon CT Loss of prothrombin and of factor Xa-factor Va interactions upon inactivation of factor Va by activated protein C. J Biol Chem 259:13986, Nesheim ME, Tracy RP, Mann KG: Clotspeed, a mathematical simulation of the functional properties of prothrombinase. J Biol Chem 259:1447, Edwards RL, Rickles FR: Macrophage procoagulants, in Spaet TH (ed): In Progress in Hemostasis and Thrombosis. Orlando, FL, Grune and Stratton, 1984, p 183

10 : Factor Xa interacts with two sites on monocytes with different functional activities LA Worfolk, RA Robinson and PB Tracy Updated information and services can be found at: Articles on similar topics can be found in the following Blood collections Information about reproducing this article in parts or in its entirety may be found online at: Information about ordering reprints may be found online at: Information about subscriptions and ASH membership may be found online at: Blood (print ISSN , online ISSN ), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC Copyright 2011 by The American Society of Hematology; all rights reserved.