Proc. Nati. Acad. Sci. USA Vol. 80, pp. 6086-6090, October 1983 Medical Sciences

Inhibition of thromboxane A2 synthesis in human platelets by coagulation factor Xa (arachidonic acid/factor V antibody/thrombin) A. KUMAR SINHA, A. KONETI RAO, JANET WILLIS, AND ROBERT W. COLMAN Thrombosis Research Center and Hematology-Oncology Section of the Department of Medicine, Temple University School of Medicine, Philadelphia, PA 19140 Communicated by Oscar D. Ratnoff, June 23, 1983 ABSTRACT Factor X. binds to platelets provided that factor with platelets that leads to the efficient formation of thrombin Va is present on the platelet surface, an interaction that results in from prothrombin, a concomitant effect of this binding of factor a striking acceleration of the conversion of prothrombin to throm- Xa is the inhibition of TXA2 synthesis in these cells. Because bin. Thrombin then initiates fibrin formation, induces platelet ag- TXA2 is a platelet-aggregating agent and a potent vasoconstric- gregation, and stimulates the intraplatelet synthesis of throm- tor, the inhibition of TXA2 synthesis by factor Xa would provide boxane A2 (TXA2). Addition of thrombin (2.4-14.4 nM) to platelet- negative feedback control of the synthesis of TXA2 in platelets. rich plasma increased the basal level of TXA2, measured as thromboxane B2, from <0.5 pmol per 108 platelets to (mean + MATERIALS AND METHODS SEM) 100 ± 22 and 250 ± 10 pmol per 108 platelets, respectively. Treatment of platelet-rich plasma with increasing concentrations Chemicals and Reagents. [1-`4C]Arachidonic acid (55.8 mCi/ of factor X. (1-12 nM) prior to the addition of thrombin pro- mmol; 1 Ci = 3.7 x 1010 Bq) and [5,6,8,9,11,12,14,15-3H(N)] gressively inhibited the production of TXA2. Thrombin (9.6 nM), thromboxane B2 (TXB2) (100 Ci/mmol) were purchased from which produced 93% of the maximal formation of TXA2, was in- New England Nuclear. [5,6,8,11,12,14,15-3H(N)]Prostaglan- hibited 70% by factor Xa (10 nM). To identify which of these steps ,din E2 (PGE2) (160 Ci/mmol) was bought from Amersham. Di- in thromboxane synthesis was inhibited by factor Xa, platelets la- isopropyl fluorophosphate (iPr2P-F) was bought from Sigma. beled with [14C]arachidonic acid were exposed to thrombin and Antisera to PGE2 was obtained from Miles. All other chemicals products of prostaglandin synthesis were separated by thin-layer used were of analytical grades. chromatography. In contrast to the inhibition of TXA2 synthesis, Coagulation Factors. Bovine factor X was prepared by the prostaglandin E2 and prostaglandin F2a synthesis were not inhib- method of Bajaj and Mann (5). The purified factor X was ac- ited suggesting that neither phospholipase(s) nor cycloxygenase was tivated by using the Russell's viper venom and the resulting involved. The inhibition of TXA2 formation by factor Xa could be factor Xa purified according to Jesty and Esnouf (6). Polyacryl- reversed by increasing the molar ratio of thrombin to factor Xa amide gel (8.6%) electrophoresis of factor Xa in the presence to 5.5. Incubation of platelets with an IgG fraction of a human of dithiothreitol showed the presence of three bands corre- monoclonal antifactor V antibody, previously shown to inhibit fac- sponding to heavy chain of aXa (Mr = 34,500), heavy chain of tor X. binding, was found to block factor Xa inhibition of TXA2 synthesis. The inhibition of TXA2 synthesis requires the presence (3Xa (Mr = 29,500), and light chain of I3Xa (Mr = 16,000) (7), of the active site serine of factor Xa and is not specific for TXA2 the major component accounting for 95%. The prepared factor formation induced by thrombin because it is also demonstrable Xa had a specific activity of 5,500 units per mg of protein de- when the agonist is ADP. Further, factor Xa does not require ad- termined according to Cole et al. (8). The molarity of factor Xa ditional plasma components for its action because its inhibitory was obtained by adding the molecular weights of heavy chain effects are detected in gel-filtered platelets. The effect of factor of PXa and the light chain (Mr = 45,000). Xa was evident at physiological (1.3 mM) calcium concentrations. Both bovine factors X and Xa were also obtained from Sigma. These results indicate that factor X. binding to platelets through Factor X was found to be a single component (Mr = 54,000) on factor Va not only stimulates thrombin formation but also has a 8.6% polyacrylamide gel electrophoresis in NaDodSO4 (0.1%) countervailing effect by inhibiting TXA2 formation. but in the absence of reducing agent. The specific activity of factor X was 100-150 units/mg of protein. Factor Xa used had The interaction of blood coagulation factor Xa with the human a specific activity of 4,000-5,000 units/mg of protein. Poly- platelet surface in the presence of Ca2" accelerates the gen- acrylamide gel electrophoresis of factor Xa in the presence of eration of thrombin from prothrombin at a rate 300,000 times NaDodSO4 and dithiothreitol showed the presence ofthree bands faster than the rate observed with factor Xa in solution (1). Fac- corresponding to the heavy chain of aXa (Mr = 34,500, 10%), tor Xa binding to platelets requires the presence of factor Va, heavy chain of /3Xa (Mr = 29,500, 90%), and light chain (Mr = which may be derived by release from platelets (1) from its in- 16,000). Except that the commercial factor Xa was somewhat tracellular site, the a granules (2), by agonists such as thrombin. less active than the proteinase prepared as described above, the Alternatively, exogenous factor Va produced by the action of enzyme was essentially identical in both preparations. thrombin on plasma factor V (3) facilitates factor Xa binding in Human thrombin was purified by the method of Fenton et a coordinate manner. In addition to these two actions of throm- al. (9) by using commercial thrombin as the starting material bin on platelets and factor V, respectively, thrombin is also a from Sigma. The purified material was a homogenous protein potent inducer of thromboxane A2 (TXA2) synthesis in these in gel electrophoresis and had a specific activity of 2,480 units/ cells (4). mg. The molarity of the thrombin preparation was calculated We now report that in addition to the interaction of factor Xa from the molecular weight of the enzyme, Mr 33,500. The publication costs of this article were defrayed in part by page charge Abbreviations: TXA2, thromboxane A2; TXB2, thromboxane B2; PGF2a, payment. This article must therefore be hereby marked "advertise- prostaglandin F2a; PGE2, prostaglandin E2; iPr2P-F, diisopropyl fluo- ment" in accordance with 18 U.S.C. §1734 solely to indicate this fact. rophosphate; RIA, radioimmunoassay.

6086 Downloaded by guest on September 26, 2021 Medical Sciences: Sinha et al. Proc. Natl. Acad. Sci. USA 80 (1983) 6087 iPr2P-F-inactivated factor Xa was prepared by incubating 1.0 Antifactor V Antibody. Antifactor V antibody used in these mg of factor Xa, prepared as above, in 25 mM Tris'HCl buffer studies was a IgG fraction of a human monoclonal IgG4 A an- (pH 8.0) with 15 mM iPr2P-F in a total volume of 0.5 ml for 2.5 tifactor V antibody (19) kindly supplied by Helen Glueck (Uni- hr at 370C as described by Leveson and Esnouf (10). Treatment versity of Cincinnati). An IgG fraction prepared by (NH4)2SO4 of factor Xa by iPr2P-F does not result in total loss of activity, precipitation and DEAE-Sephadex chromatography of the an- but under these conditions we found that factor Xa lost >85% tibody was prepared and furnished by H. C. Chiu (Thrombosis of its original activity. The reaction mixture was then dialyzed Research Center). Approximately 40 Mg of the IgG fraction per against 0.15 M NaCl for 16 hr at 40C. ml neutralized >99% offactorV activity in 1 ml of normal plasma. Human factor X was prepared according to DiScipio et al. (11), was activated and purified as described above (6), and had RESULTS a specific activity of 4,000 units/mg of protein. Stimulation of TXA2 Formation by Thrombin. Addition of Platelet Preparation. All blood donors had not taken any increasing concentrations of thrombin to platelet-rich plasma medication at least 2 wk before blood donation. Venous blood induced the synthesis of higher levels of platelet TXA2, as de- was collected through siliconized needles into plastic syringes; termined by RIA of TXB2 (Fig. 1). Although as little as 0.2 unit coagulation was prevented by mixing 9 vol of blood with 1 vol of thrombin per ml (2.4 nM) would aggregate 80% of the plate- of trisodium citrate (final concentration, 13 mM). Platelet-rich lets (not shown), the formation of TXB2 continued to increase, plasma was obtained by centrifugation of samples for 10 min at up to a thrombin concentration of 14.4 nM. The basal level of 170 x g. The supernatant platelet-rich plasma (2-3 X 10' cells TXB2 (<0.5 pmol per 108 platelets) increased -300-fold (100 per ml) was collected. Platelet counts were performed in a ± 22 pmol per 108 cells; mean ± SEM) and 800-fold (250 ± 10 Coulter Counter model ZB equipped with a 50-Mum aperture pmol per 108 cells) in the presence of 2.4 nM and 14.4 nM tube. Gel-filtered platelets were prepared as described (12). thrombin, respectively. The labeling of gel-filtered platelets with [I4C]serotonin and its Inhibition of TXB2 Production by Factor Xa. When platelet- release were studied according to Walsh et al. (13). rich plasma was treated with 9.6 nM thrombin, the TXA2 level Platelet Aggregation Studies. Platelet-rich plasma (0.5 ml) increased from the basal level of <0.5 pmol per 108 platelets was placed in a cylindrical cuvette 8 mm in diameter containing to (mean ± SEM) 224 ± 15.6 pmol per 108 platelets. Treatment a silicone-coated stirring bar. Aggregation was studied by add- of platelet-rich plasma with increasing concentrations of factor ing thrombin or ADP and measuring the increase in light trans- Xa (1-12 nM) prior to the addition of thrombin progressively mittance as platelet aggregates formed in an aggregometer inhibited the production of TXA2 in these cells, to a value 20% (Chronolog, Broomall, PA) with a stirring rate of 1,200 rpm at of the level with thrombin alone (Fig. 2). 37°C. The apparatus was calibrated so that the difference in In a separate experiment ['4C]arachidonic acid-labeled light transmittance between platelet-rich and platelet-poor plasma platelets were treated with thrombin in the presence and ab- was defined as 100% (14). sence of factor Xa. The platelet lipids were extracted and the Assay of TXA2. TXA2 is a very unstable compound and is presence of ['4C]TXB2 in the extract was determined by TLC. rapidly converted into more stable derivative TXB2 (4). In the As in the case of the RIA, the conversion of ['4C]arachidonic present study TXA2 was determined by TLC or radioimmu- to radiolabeled TXA2 by thrombin was inhibited by factor Xa noassay (RIA) in its TXB2 form. because the peak of radioactivity corresponding to TXB2 is al- TLC of TXB2. Platelet-rich plasma was incubated with most completely ablated (Fig. 3). Comparison of the TLC of [I4C]arachidonic acid (5 ,Ci/1O ml of platelet-rich plasma) for extracts after thrombin treatment alone and in the presence of 30 min at 37C. Ten percent of the [14C]arachidonic acid was factor Xa indicated that the formation of radiolabeled TXA2 was incorporated under these conditions. The labeled platelet-rich specifically decreased in the presence of factor Xa, whereas the plasma was then treated with thrombin in the presence or ab- conversion to PGE2 or PGF2, was unaffected. The latter ob- sence of factor Xa. After aggregation, the plasma was acidified servation was quantified by using a RIA for PGE2. In three ex- with cold (0°C) formic acid (final concentration, 1%) and the periments, platelets in plasma contained (mean ± SEM) 14.6 lipids were extracted twice with 3 vol of cold (0°C) ethyl ace- ± 4.5 pg per 109 cells. In the presence of thrombin (2.4 nM) tate. The extract was concentrated and analyzed by TLC on a the PGE2 concentration rose to 55.6 ± 10.6 pg per 109 plate- silica gel G plate by using CHCl3/CH3OH/CH3COOH/H20, 90:8:1:0.8 (vol/vol) (15). The plates were subsequently scanned in a Vanguard Scanner for the radioactivity. The arachidonic w 320 acid metabolites representing 19% of the incorporated fatty acid 7u were identified by using authentic samples of prostaglandin F2, X 240 (PGF2), TXB2, PGE2, and PGA2 in the silica gel platelets. These compounds were visualized in the chromatographic plate by a, staining with iodine vapor. a 160 RIA of TXB2 and PGE2. RIA of TXB2 was performed by us- E ing the antiserum against TXB2 as described by Lewy et al. (16) 80 m~ with antiserum provided by J. Bryan Smith (Thrombosis Re- x search Each is in I-- Center). sample assayed triplicate and the mean 0 value reported. The RIA of PGE2 was carried out by using com- 0 2.4 4.8 7.2 9.6 12.0 14.4 mercial antibody following the instruction of the manufacturer. Thrombin, nM Assay of Coagulation Factors. Factor Xa was assayed ac- cording to Denson (17) by using congenitally factor X-deficient FIG. 1. Thrombin stimulation of thromboxane formation by plate- plasma, except that the activation step with Russell's venom lets. Platelets were aggregated by adding different concentration of viper thrombin to platelet-rich plasma (0.5 ml, 2.5 x 108 cells per ml). After was omitted. Factor V was assayed by the one-stage method of aggregation, the platelet preparations were immediately centrifuged Lewis and Ware (18). One unit of factor V is defined as the and the plasma was collected. Concentration ofTXB2 in the plasma was amount in 1 ml of normal human plasma. determined by RIA. Values are mean ± SEM for 5-10 different donors. Downloaded by guest on September 26, 2021 6088 Medical Sciences: Sinha et al. Proc. Natl. Acad. Sci. USA 80 (1983) 240 Reversal of Factor Xa Inhibition of TXB2 Formation by q)0 Thrombin. The inhibition of TXA2 formation by factor Xa was i 200 found to be reversible. Thrombin at 9.6 nM, a concentration that produced 93% of a maximal response, was inhibited 70% 160 a by factor Xa (10 nM) (Fig. 4). As the concentration of thrombin was increased in the presence of a constant amount of factor Xa, 0 120 the inhibitory effect of the protease was overcome. At a con- centration of 52.8 nM thrombin, the concentration of TXA2 was 04E 80 not different from that produced by 9.6 nM thrombin in the absence of factor Xa. xm 40 Although the addition of factor Xa to platelet-rich plasma in- O .I I I hibited thrombin-induced TXA2 synthesis in platelets, the fac- 0 2 4 6 8 10 12 tor Xa at the above concentration has no effect on the aggre- Factor Xa, nM gation of these cells induced by thrombin. However, by increasing the factor Xa concentration by 10-fold (24 nM) over FIG. 2. Inhibition ofthrombin-inducedthromboxaneproductionby thrombin (2.4 nM), inhibition of platelet aggregation could be factor Xa. Platelet-rich plasma (0.5 ml, 2.5 x 108 cells per ml) was in- in plasma (not shown). cubated with increasing concentrations offactor Xa as indicated for 30 shown platelet-rich sec at 37°C before addition of9.6 nM thrombin (0.8 unit/ml). After ag- Effect of Antifactor V Antibody on the Inhibition of TXB2 gregation TXB2 concentration was determined by RIA. Values are mean Formation by Factor Xa. Factor Xa binding to the platelet sur- ± SEM of experiments on three different donors. face requires the presence of factor Va molecules, which are thought to behave as receptors for such interactions (1). If the lets. However, factor Xa (4.0 nM) failed to inhibit the thrombin- effects of factor Xa on the inhibition of TXA2 were mediated stimulated formation of PGE2 because the concentration was through the interaction of the coagulant enzyme with factor Va, 58.2 ± 12.8 pg per 109 platelets. it might be expected that the addition of antibody against factor V to the platelet suspension would be able to counteract the 45,000 r A inhibitory effect of factor Xa. As shown in Fig. 5, incubation of plasma with increasing concentrations of the IgG fraction of a human monoclonal antibody against factor V caused progressive inhibition of factor V coagulant activity until, at 40 ,g/ml, <10% 1,200 of the original activity was present. The antibody had no effect on the formation of TXA2 in platelet by thrombin. However, 1,000 the antibody was able to reverse the inhibition of thrombin- -800 stimulated TXA2 formation by factor Xa. The neutralization of factor V coagulant activity by the antibody closely parallels the 600 ability of the antibody to reverse the inhibitory effects of factor 400 Xa on thrombin-stimulated TXB2 formation. Specificity of Factor Xa on the Inhibition of TXA2 For- E 200 mation. The inhibition of thrombin-stimulated TXA2 formation CL0. 0 by factor Xa (80%) was not demonstrable after factor Xa was in- activated by iPr2P-F (9%) (Table 1). Factor X in high concen- . 0 How- 0 tration showed slight inhibition of TXA2 synthesis (18%). 0 47,500 0 B ever, the minimal inhibition was probably due to the presence 'ID 0 of a small amount of factor Xa in the factor X preparation. When 1,400 0 1,200 4-0 0) 1,000 4-0 800 Q 600 0o 400 200 L-0 0 x 0 Ll- FIG. 3. Thin-layer radiochromatogram scan of["4C]arachidonic acid- 0 9.6 19.2 28.8 38.4 48.0 57.6 labeled platelet lipid extract in the presence of thrombin with or with- Thrombin, nM out factor Xa. Platelet-rich plasma was incubated with ['4C]arachi- donic acid (0.5 ,Ci/ml) for 60 min at 37°C. The labeled platelets were FIG. 4. Reversal of inhibitory effect offactor Xa by increasing con- then treated with either 9.6 nM thrombin or 10 nM factor Xa followed centration ofthrombin. Platelet-rich plasma (0.5 ml, 2.5 x 10i cells per by 9.6 nM thrombin. After the aggregation ofthe platelets, lipids were ml) was treated with a constant quantity of factor Xa (10 nM) and in- extracted and analyzed by TLC by using a CHC13/CH3OH/CH3COOH/ creasing amounts of thrombin (o). The value was compared with the H20, 90:8:1:0.8 (vol/vol), solventsystem. (A) Platelet-richplasmaand function of TXA2 by thrombin (9.6 nM) alone (o). After platelet ag- thrombin. (B) Platelet-rich plasma and factor Xa with thrombin. The gregation was completed, TXB2 in the supernatant fraction was mea- arrows indicate the origin. suredby RIA. Values are mean ± SEM for three different experiments. Downloaded by guest on September 26, 2021 Medical Sciences: Sinha et al. Proc. Natl. Acad. Sci. USA 80 (1983) 6089

* 240 Table 2. Effects of factor Xa on ADP- and thrombin-induced gel- filtered platelet aggregation and release of ["4C]serotonin 200 T 100 .> +O0 i Thrombin, ADP, Factor Xa, Aggregation, [14ClSerotonin T- ,.' nM % % " AM nM release, 160 F 80 0co 'a 0.5 - 80 39.0 120 F 60 o 0.5 4 4 0.0 h- 0 - cU 5 80 27.4 a 80 40 <, 5 4 72 6.8 F 5 16 10 1.1 E a 40 20 10 - 80 27.4 Im 10 4 76 9.2 >< 0 0S 0 20 40 60 80 Gel-filtered platelets suspended in Tyrode's buffer without added Ca2+ (1.5 x 10" cells per ml) were labeled with [14C]serotonin. Platelet ag- Antifactor V IgG fraction, ,tg/ml gregation was initiated with thrombin orADPas indicated. The results are representative of three experiments with three different donors. FIG. 5. Effect of antifactor V antibody on the inhibition of throm- bin-induced thromboxane formation by factor X.. Platelet-rich plasma (0.5 ml, 2.5 x 10' cells per ml) was incubated at various concentrations Effect of Calcium on the Inhibition of Thrombin-Induced with IgG fractions of monoclonal antifactor V antibody for 15 min. After TXA2 Formation by Factor Xa. The described ex- incubation platelets were treated with factor Xa (10 nM) followed by previously thrombin (9.6 nM) (A). In the control experiment (A) addition of factor periments were performed in either citrated platelet-rich plasma Xa to platelet-rich plasma was omitted. After platelet aggregation the (estimated Ca2+ = 30 AM) or Tyrode's buffer with no added plasma supernatant was analyzed for the determination ofTXB2 by RIA. calcium, in which Ca2+ measured by atomic absorption spec- Parallel experiments were run to determine the neutralization offactor troscopy is 40 (21). Therefore, it is possible that at the cal- V by the antibody (o). Values are mean ± SEM for three different ex- AtM periments. cium concentration in blood, 1.3 mM (22), the inhibition ob- served at depressed ionic calcium concentration might not obtain. Therefore, the effect of added calcium on the formation of TXA2 ADP is added at low concentrations (4 ,uM), little TXA2 is formed stimulated by thrombin (2.4 nM) in the presence and absence (TXB2 = 8.0 pmol/ml, where 1 ml contains 2.5 X 108 plate- of factor Xa (4 nM) was studied (Fig. 6). No effect on thrombin lets). However, at high concentrations of ADP (0.5 mM), more stimulation of TXA2 formation was noted when calcium extensive TXA2 formation is observed (Table 1). Substantial in- (1 mM) hibition of the thromboxane synthesis is observed when factor was added and even at 2 mM added calcium only 11% decrease Xa is added and this increases at higher concentrations of factor was noted. Similarly, the inhibitory effect of factor Xa was un- Xa. changed at 1 mM added calcium and only decreased 7% at 2 Effect of Factor Xa on Thrombin- and ADP-Induced Ag- mM. Thus, at 1.3 mM calcium added (Fig. 6, arrows) the in- gregation in Gel-Filtered Platelets and Serotonin Release. Fac- hibitory effect of factor Xa on thrombin-stimulated TXA2 was tor Xa inhibited thrombin-induced aggregation and ['4C]sero- essentially unchanged. Only at the unphysiologic concentration tonin release in gel-filtered platelets (Table 2). Concentrations of 3 mM added calcium was there a significant alteration of the of both thrombin and factor Xa needed to induce and inhibit inhibitory effect of factor Xa. Addition of EDTA (1.0 mM) either platelet aggregation are much lower than that in the case of to platelet-rich plasma or to gel-filtered platelets in Tyrode's platelet-rich plasma. This discrepancy is probably due to the buffer completely inhibited the formation of TXA2 (<0.5 pmol presence of antithrombin in the plasma, which inactivates both per 10' cells) in the presence of 2.4 nM thrombin, indicating thrombin and factor Xa (20). Factor Xa also inhibited ADP-in- the necessity of divalent cations in the synthesis of TXA2. duced aggregation and release, although the effect was much more marked on serotonin release. 1,600 Table 1. Specificity of factor Xa-induced inhibition of TXA2 CU 4-a 1,400 formation in platelets Co T --4 0 1,200 -I TXB2, 0 Addition to platelet-rich plasma pmol/ml 00m 1,000 None <0.5 Q 800 Bovine X., 10 nM <0.5 I Thrombin, 9.6 nM 590 ± 70 E 600 --1 -1 + Bovine X., 10 nM 120 ± 10 m 400 - + Bovine X, 100 nM 485 ± 25 x + iPr2P-F bovine Xa, 100 nM 538 ± 30 200 + Human Xa, 10 nM 230 ± 25 0 I ADP, 0.5 mM 89.0 ± 10.2 0 1 2 3 + Bovine Xa, 4 nM 33.6 ± 5.9 CaCI2 added, mM + Bovine Xa, 10 nM 18.2 ± 6.5 FIG. 6. Effect of addition of CaCl2 on the inhibition of TXA2 for- Platelet-rich plasma (0.5 ml; 2.5 x 108 cells per ml) was treated with mation in gel-filtered platelets. Varying amounts of CaC12 were added factor X, with factor Xa, or with iPr2P-F factor Xa and the platelet ag- to the gel-filtered platelets before the addition of factor Xa (4 nM) fol- gregation was initiated with thrombin or ADP as indicated. Factor X lowed by the addition of thrombin (2.4 nM). After aggregation of the andfactorXawere preparedasdescribed in the text. Afteraggregation, platelets (3.0 min), at 1,200 rpm at 37°C, TXA2 was measured in the TXB2 concentration in the plasma supernatant was determined by RIA. supernatant. Values are mean ± SEM for three different experiments. Values are mean ± SEM for three different experiments in different *, Thrombin; A, thrombin and factor Xa. The arrows indicate the for- donors. mation of TXA2 at plasma Ca2+ concentration (1.33 mM). Downloaded by guest on September 26, 2021 6090 Medical Sciences: Sinha et al. Proc. Natl. Acad. Sci. USA 80 (1983) DISCUSSION contact does not induce TXA2 formation. The lack of effect of This study demonstrates that thrombin-induced formation of factor Xa on the inhibition of TXA2 synthesis in platelet ho- TXA2 in human platelets (Fig. 1) is inhibited by factor Xa. This mogenate indicates the participation of a common mediator for conclusion is supported by further experiments using both RIA the stimulation of thrombin and ADP of TXA2 formation and (Fig. 2) and measurements of arachidonic acid incorporation inhibition of that process by factor Xa. However, the nature of (Fig. 3) into TXB2 by TLC. The inhibition by factor Xa of TXA2 the mediator is currently unknown. synthesis is reversed by high concentrations of thrombin (Fig. 4) or by inhibiting factor Va, the putative receptor of factor Xa We thank Ms. Cheryl Beckett for her assistance, J. B. Smith for sup- plying us with the TXB2 antibody, H. Glueck for the human monoclonal on platelets, with a human monoclonal antibody (Fig. 5). The antibody to factor V, and R. Rawala for supplying us with purified bo- inhibition of TXA2 synthesis requires the presence of the active vine factor Xa. Human factor Xa was kindly donated by the late J. Pi- site serine of factor Xa and is not specific for thrombin-induced perno. This work was supported by National Institutes of Health Grants TXA2 formation (Table 1) because TXA2 synthesis stimulated HL24674 and HL27189 and by a grant for a Specialized Center of Re- by high concentrations of ADP is inhibited by factor Xa. The search in Thrombosis (HL14217). inhibition by factor Xa is not mediated by other components of plasma because it is also demonstrated in gel-filtered platelets 1. Miletich, J. P., Jackson, C. M. & Majerus, P. W. (1978)J. Biol. (Table 2). Although TXA2 formation can be stimulated by close Chem. 253, 6908-6916. 2. Chesney, C. M., Pifer, D. & Colman, R. W. (1981) Proc. Natl. Acad. platelet contact, as in centrifugation at low ionic calcium levels, Sci. USA 78, 5180-5184. the inhibition by factor Xa appears to be similar at physiologic 3. Tracy, P. B., Peterson, J. M., Hosherm, M. E., McDuffie, F. C. calcium concentrations as at the ionic calcium concentration ex- & Mann, K. G. (1979)J. Biol. Chem. 254, 10354-10361. isting in citrated plasma (Fig. 6). 4. Hamberg, M., Svensson, J. & Samuelsson, B. (1975) Proc. Natl. TXA2 that is synthesized from arachidonic acid during plate- Acad. Sci. USA 72, 2994-2998. let aggregation is believed to be an important mediator of sec- 5. Bajaj, S. P. & Mann, K. G. (1973)J. Biol. Chem. 248, 7729-7741. and release reaction This arachi- 6. Jesty, J. & Esnouf, M. P. (1973) Biochem. J. 131, 791-799. ondary aggregation the (4). 7. Jesty, J., Spencer, A. K. & Nemerson, Y. (1974)J. Biol. Chem. 249, donic acid metabolite, in addition to being an inducer of platelet 5614-5622. aggregation, is a potent vasoactive agent, which causes con- 8. Cole, E. R., Marciniak, E. & Seegers, W. H. (1962) Thromb. Diath. traction of a wide variety of vascular and smooth muscles (23). Haemorrh. 8, 434-441. The factors regulating synthesis of TXA2 are incompletely 9. Fenton, J. W., II, Fasco, M. J., Stackrow, A. B., Aaronson, D. understood. The present study indicates that the synthesis of L., Young, A. M. & Finlayson, J. S. (1977)J. Biol. Chem. 252, 3587- 3598. TXA2 stimulated by thrombin is inhibited by a component (fac- 10. Leveson, J. E. & Esnouf, M. P. (1969) Br. J. Haematol. 17, 173- tor Xa) of the enzyme complex that catalyzes the generation of 178. the agonist itself. Thus, this inhibitory effect of factor Xa could 11. DiScipio, R. G., Hermodson, M. A., Yates, S. G. & Davie, E. W. control the synthesis of the vasoactive compound TXA2 without (1977) Biochemistry 16, 698-706. affecting the generation of thrombin or the aggregation of 12. Leges, B., Scrutton, M. C. & Holmsen, H. (1975)J. Lab. Clin. Med. platelets induced by high concentrations of that enzyme. 85, 811-822. 13. Walsh, P. N., Pareti, F. I. & Corbett, J. J. (1975) N. Engl.J. Med. The conversion of ['4C]arachidonic acid to various prosta- 295, 854-858. glandin metabolites in the labeled platelets induced by throm- 14. Sinha, A. K., Shattil, S. J. & Colman, R. W. (1977)J. Biol. Chem. bin was inhibited by factor Xa specifically at the step of TXA2 252, 3310-3314. formation without affecting the production of either PGF2. or 15. Nugteren, D. H. & Hazelhof, E. (1973) Biochim. Biophys. Acta PGE2. Because these stable prostaglandins require both phos- 326, 448-461. pholipase(s) and cyclooxygenase for their syntheses (24) it ap- 16. Lewy, R. I., Smith, J. B., Silver, M. J., Saia, J., Walinski, P. & Wiener, L. (1979) Prostaglandin Med. 2, 243-248. pears that neither of these enzymes is inhibited by factor Xa. 17. Denson, K. W. E. (1972) in Human Blood Coagulation, Haemo- However, the inhibitory effect of factor Xa is limited to the in- stasis and Thrombosis, ed. Biggs, R. (Blackwell Scientific, Ox- tact platelet. The coagulant enzyme has no effect on the for- ford), pp. 631-634. mation of TXA2 from arachidonic acid by a platelet homogenate 18. Lewis, M. L. & Ware, A. G. (1953) Proc. Soc. Exp. Biol. Med. 84, in the presence or absence of added factor Va (unpublished data). 640-643. Previous investigators have suggested that the factor Xa 19. Hurtubise, P., Coots, M. C., Jacob, D. J., Muhleman, A. F. & Glueck, H. I. (1979)J. Immunol. 122, 2119-2121. binding sites in platelets are distinct from the thrombin binding 20. Rosenberg, R. D. (1982) in Hemostasis and Thrombosis, eds. Col- sites, because no competition was observed for binding to man, R. W., Hirsh, J., Marder, V. J. & Salzman, E. W. (Lippin- platelets (25). The results of our investigation indicate that the cott, Philadelphia), pp. 962-985. inhibitory effect of factor Xa could be overcome by increasing 21. Greenquist, A. C. & Colman, R. W. (1975) Blood 46, 769-782. the concentration of the agonist. Further evidence that this ap- 22. Jackson, C. M. (1982) in Hemostasis and Thrombosis, eds. Col- parent competition between thrombin and factor Xa is not at man, R. W., Hirsh, J., Marder, V. J. & Salzman, E. W. (Lippin- cott, Philadelphia), pp. 100-111. the receptor level is provided by the finding that factor Xa can 23. Moncada, S. & Vane, J. R. (1979) Pharmacol. Rev. 30, 293-331. inhibit the stimulation of TXA2 production by high concentra- 24. Marcus, A. J. (1978) J. Lipid Res. 19, 793-826. tions of ADP (Table 1). It should be noted that at physiological 25. Miletich, J. P., Jackson, C. M. & Majerus, P. W. (1977) Proc. Natl. calcium concentrations in the presence of ADP even close platelet Acad. Sci. USA 74, 4033-4036. Downloaded by guest on September 26, 2021