- Email: [email protected]
Effects of lipid-binding proteins APO A-I, APO A-II, β2-glycoprotein I, and C-reactive protein on activation of factor X by tissue factor - factor VIIa
THROMBOSIS RESEARCH 50; 669-678, 1988 0049-3848/88 $3.00 t .OO Printed in the USA. Copyright (c) 1988 Pergamon Press plc. All rights reserved.
EFFECTS OF LIPID-BINDING PROTEINS APO A-I, APO A-II, 62-GLYCOPROTEIN I, AND C-REACTIVE PROTEIN ON ACTIVATION OF FACTOR X BY TISSUE FACTOR - FACTOR VIIa
Steven D. Carson and Sherman E. Ross Department of Pathology University of Colorado Health Sciences Center Denver, Colorado USA (Received 21.10.1987;
Accepted in revised form 10.12.1987 by Editor C.T. Esmon) (Received in final form by Executive Editorial Office 23.3.1988) ABSTRACT Tissue factor is the membrane-associated protein which mediates activation of factors IX and X by factor VII. In a purified, reconstituted bovine system, factor X activation by the tissue factor-factor VIIa complex is inhibited by the mixed apoproteins from human high density lipoprotein (HDL) and by isolated apolipoprotein A-II (apo A-II). Other proteins found associated with plasma lipoproteins, apolipoprotein A-I (apo A-I), C-reactive protein (CRP), and 62-glycoprote-in I (62GPI), have been examined for effects on the activation of factor X by tissue factor-factor In these experiments, bovine tissue factor, reconstituted VIIa. into phosphatidylserine-phosphatidylcholine (PS/PC; 30/70) vesicles, was used at a single concentration while factor X (the substrate), factor VIIa (the enzyme), and the potentially inhibitory proteins were varied in a continuous chromogenic assay. Apo A-II and CRP clearly inhibit tissue factor-factor VIIa activation of factor X, while apo A-I and 62 GPI have little or no effect. These results demonstrate that different lipid binding proteins vary in their effects on tissue factor activity.
INTRODUCTION Observation of components in the lipoprotein fraction of human plasma which inhibit tissue factor activity in a system reconstituted from isolated components (1) has been confirmed by several studies using contemporary methodolo:y (2-5). A detailed kinetic analysis of the effect of apolipoprotein A-II demonstrated a loss of tissue factor activity and partial protection by factor VIIa, indicating that inhibition was manifested through interference with functional association of factor VIIa and tissue factor (4). Subsequent experiments indicate that the apo A-II selectively Key Words:
Tissue factor, inhibitors,
apolipoproteins
669
670
Vol. 50, No. 5
INHIBITION OF TISSUE FACTOR
inhibits factor X activation, and does not affect factor IX activation by Other investigators have identified at tissue factor - factor VIIa (6). least two other lipid-binding inhibitors of tissue factor activity. One of these lipoprotein-associated coagulation inhibitors (LACI or extrinsic pathway inhibitor, EPI) has been identified by at least three laboratories (2, 3, 5) and is dependent on factor Xa for its effect. This inhibitor is capable of inhibiting factor Xa, is calcium-deoendent. and aooears to inhibit tissue factor activity by format on of an inactive tissue factorfactor VIIa-factor Xa-LACI complex (7, 8 . An additional lipid-binding tissue factor inhibitor, isolated from p i acenta, is structurally related to lipocortin and may have an inhibitory mechanism similar to that of apo A-II (9). I
I
In this study, three additional lip d-binding proteins have been isolated for study with regard to tissue factor-factor VIIa activation of s a major apoprotein of high factor X. Apolipoprotein A-I (apo A-I) density lipoprotein (HDL) and therefore 0 f interest due to the previous reports of inhibition of tissue factor activity by HDL and its mixed apolipoproteins (1). i-32-glycoprotein I (pzGP1) inhibits the activation of Because of its association with plasma lipoproteins (11) factor XII (10). and its effect on the intrinsic pathway of coagulation, it is potentially interesting with regard to tissue factor regulation. C-reactive protein (CRP) is an acute phase protein which has calcium-dependent abilities to interact with carbohydrate and phosphorylcholine moieties (12). In vitro, it prolongs the activated partial thromboplastin time in the presence of heparin (13). Pathologically, CRP is observed in necrotic lesions, and is apparently bound to damaged cell membranes (14). In vitro, CRP has been demonstrated to bind "traumatized" membranes but not their healthy counterparts (14). These features make CRP extremely interesting with regard to its potential effects on tissue factor activity since tissue factor is expressed in "damaged" cell membranes to a much greater extent than in healthy cell membranes (15, 16). MATERIALS AND METHODS Bovine brain phosphatidylserine (PS) and egg yolk phosphatidylcholine (PC) were obtained from Sigma. Bovine brain tissue factor was purified by immunoaffinity chromatography using monoclonal antibody TFl-F7 (17) and was kindly provided by Dr. Ron Bach (Mt. Sinai School of Medicine, New York). Factor X was purified from bovine plasma using established methods (18, 19) its concentration determined spectrophotometrically at 280 nm using 9.6 (19). Factor VII was kindly provided by Drs. Arabinda Guha and Yale Nemerson (Mt. Sinai School of Medicine, New York). Factor VII was activated with factor Xa, which had been activated with the factor X act lvating enzyme from Russell's viper venom (RVV-XAE Sigma) as described by Jesty et al. (20) and Bach et al. (21). Working stocks of factors X (400 PM) and VIIa (3.4 0) were stored at -20" in tris-saline (0.05 M tr is, 0.1 M NaCl, 0.01% NaNs, pH 7.6) with 50% glycerol and 5 mM benzamidine. Spectrozyme FXa was from American Diagnostica.
Erq =
Apolipoproteins A-I and A-II were chromatographically prepared from human plasma (22). fi2GPI was isolated from human plasma according to Finlayson and Mushinski (23), and its identity was confirmed by immunodiffusion against specific rabbit antiserum toward @2GPI (Calbiochem). p2GPI concentration was estimated using the dye-binding assay of Bradford (24) with bovine serum albumin serving as reference. Human CRP was isolated from pleural fluid obtained at autopsy using affinity chromatography on
Vol. 50, No. 5
671
INHIBITION OF TISSUE FACTOR
p-aminophenylphosphorylcholine agarose from Pierce Chemical Co 5 . CRP concentration was estimated by absorbance at 280 nm,using E le ) = 19.5 (26) and pentameric molecular weight of 110,000 (27).
The
Tissue factor was reconstituted into lipid vesicles using deoxycholate and CdClz with subsequent dialysis against tris-saline (28). The final reconstituted mixture contained 24.1 pM tissue factor, 138 #i phospholipid (PS/PC = 30/70 w/w), and 1 mg/ml bovine serum albumin (Sigma, fatty acid free). Tissue factor and phospholipid were constant throughout these experiments and were used at final reaction concentrations of 0.38 pM and 2.16 PM respectively. Tissue factor activity was measured using a continuous chromogenic assay for factor X activation and a Bio-Tek automated EIA reader (15, 29). Reagents added to the wells of a 96-well plate were, in order, 10 ul tissue factor, test protein and/or tris-saline with 1 mg/ml bovine serum albumin to give 50 ~1, 20 ~1 factor VIIa, 20 ~1 factor X, and to initiate the reaction, 60 ~1 of Spectrozyme FXa and CaClz (the 60 ~1 contained 10 ~1 of 5 mM Spectrozyme and 50 ~1 of 25 mM CaC12). The reactions were monitored at 405 nm every min for 80 min and the results were analyzed on a TI99/4A computer (29). Experiments were conducted so that data sets contained a single set of velocity curves obtained with a single inhibitor at multiple dilutions, with each experiment conducted on a separate day using unique dilutions of the enzymes, substrates and reagents (except for apo A-II, dilutions of which were distributed among all of the experiments). The test proteins were examined for potential effects on the coupled reaction (factor Xa hydrolysis of the chromogenic substrate) which could complicate interpretation with regard to the tissue factor component of the Factor X was activated by combining 3 ul of 400 $i faccoupled systems. tor X with 1 ml tris-saline (containing 1 mg/ml bovine serum albumin), 50 ~1 of RVV-XAE, and 100 ~1 25 mM CaC12. Factor X activation was monitored by combining 20 ul of the reaction mixture with 100 ~1 of plasma deficient in factors VII and X (Sigma), 100 ul of 25 mM CaC12, and 100 ~1 tris-saline containing albumin, and determining the clot-time with a fibroClot-times typically plateaued after 20 to 30 min during which the meter. clot-time dropped from greater than 70 set to near 30 sec. The chromogenic assay was conducted using PS/PC vesicles formed without tissue factor, 8.1 nM factor Xb, varied concentrations of the test proteins (apo A-I, apo A-II, fl7_GPI, CRP) and Spectrozyme FXa. The computer program (29) was modified to determine the slope of the linear A405 vs min reaction plot. None Xa hydrolysis of of these proteins altered the Vmax or Km of theTactor the Spectrozyme FXa, and any observed effects on chromogen production must therefore originate in the first reaction of the test system (the tissue factor-factor VIIa activation of factor X). Where the equation of a rectangular hyperbola was fit to the data, the direct-plot curve fitting method and algorithm (4, 30) was used to provide nonparametric estimates of curve parameters. Other data was analyzed by analysis of covariance (31). RESULTS AND DISCUSSION Comparison of apo A-I, apo A-II, @2GPI and CRP for their effects on factor X activation by tissue factor-factor VIIa (Fig. 1) revealed that
INHIBITION OF TISSUE FACTOR
672
Vol. 50, No. 5
these proteins could be paired into two categories. Apo A-I and @2GPI were virtually without effect while apo A-II and CRP both inhibited factor X activation.
50
- 40 ,\"
IO
0.2
0.4
06
08
InhIbItor(uM)
Fig. 1. Inhibition of factor Xa activation as a function of test protein Tissue factor and phospholipid were present at 0.38 pM and concentration. Factor VIIa was present at 0.86 nM and the concen2.16 PM, respectively. tration of factor X was 100 nM. The symbols represent apo A-II (v), CRP Only the solid symbols were significant(a), apo A-I (0), and ppGP1 (A). ly greater than zero by the t test (~~0.05). Error bars represent one standard deviation of the experimental values from the mean.
Since the inhibition of factor X activation by apo A-II had been shown to be influenced by factor VIIa concentration (4), inhibition was reexamined for each of the lipid-binding proteins as functions of either factor VIIa concentration or factor X concentration. The effect of factor VIIa concentration was notable only with respect to the inhibition observed with apo A-II and CRP, and these results are shown in Fig. 2.
Vol.
50,
No.
673
INHIBITION OF TISSUE FACTOR
5
loo-
IOO-
A-II
CRP
60-
3 - 60‘0 ._ 2 I5 = 40-
1
I I .o Factor IZlIo (nM)
* . 2. P-
Effect of ition of factor X were 0.103 pM (m), 0.17 uM (a), 0.35 standard deviation
I 2.0
1.0 Factor PlIo (nM)
2.0
factor VIIa concentration on the apo A-II and CRP inhiThe concentrations of apo A-II examined activation. 0.206 @i (O), and 0.413 @l (A). CRP was tested at uM (O), and 0.70 uM (A). Error bars represent one of experimental values either above or below the mean.
Clearly, increasing factor VIIa diminished the inhibitory effect of these proteins, but in no case was inhibition completely overcome. As presented in Fig. 1, neither apo A-I nor ,¶2GPI caused a notable decrease in factor X activation, and the level of inhibition observed was not significantly affected by changes in factor VIIa concentration (analysis of covariance). The effects of factor X concentration on inhibition by these proteins were small (Fig. 3), but as shown in Table 1, were significant in three cases by analysis of covariance. The regression coefficients show that 100 nM increase in factor X concentration resulted in only 6.3% and 4.9% decreases in inhibition by apo A-II and CRP, respectively, while increasing the effect of apo A-I by
2.4%. These interactions are extremely small in comparison to those of factor VIIa with apo A-II and CRP revealed in Fig. 2, even though factor X was tested up to 200 nM, which is about twice the concentration of factor X in human plasma (35). The other proteins were examined over ranges well below their normal plasma concentrations, except CRP, which is an acute phase While CRP is a trace protein in normal plasma, it may exceed reactant. 1 uM in inflammatory states (36). The normal human plasma concentrations of the other proteins used in these studies are 12.3 nM (factor VII),
Vol. 50, No. 5
INHIBITION OF TISSUE FACTOR
674
IOO-
CRP
60-
2
60-
z ._ I E 405
FOG(“M)
2%
1_
loo FactorX (nh4)
Fig. 3. Effect of factor X concentration factor VIIa activity. The concentrations to the symbols are as for Fig. 2.
200
on inhibition of tissue factorof apo A-II and CRP corresponding
TABLE I Analysis of Covariance: Inhibition (X) versus Factor X (nM)
Test Protein apo A-II CRP
apo A-I t32'3'1
Overall Coefficient of Regression* (Inhibition (X)/fX (nM))* -Il.063 (p<.fll)t -0.049 (p<.Ol)
0.024 (pai) 0.015 (N.S.)
* Coefficients of regression did not differ significantly with concentration of a given test protein (F-test). t P value determined by t test. N.S.=not significant.
41.4 uM (apo A-I), 17.1 fl (apo A-II), and 6 fl (&GPI) (37-39). The physiological importance of apo A-II and CRP as participants in the hemostatic system clearly remains to be established. Now that these proteins have been identified for their potential to regulate tissue factor initiation of coagulation, it will be possible to prepare plasma both depleted and augmented with respect to these proteins as reagents with which to address this question.
Vol.
50,
No. 5
INHIBITION
OF TISSUE FACTOR
675
It is intriguing that apo A-II and CRP produced remarkably greater effects than apo A-I and pzGP1, since all four proteins are known to bind lipid surfaces (11, 12, 34). Even though B2GPI and CRP have both been shown to interfere with the "intrinsic" pathway of coagulation (10, 13), CRP had a noteworthy effect in these experiments with tissue factor while B2GPI did not. Clearly, the mechanisms of their individual effects must not lie simply in their abilities to bind lipids. In this regard, the lipid-binding inhibitor identified by Rapaport's group and by Braze and Miletich, must not exert its effect simply through binding to the tissue factor-associated lipids since it appears to be dependent on factor Xa for its inhibitory activity (2, 5, 7, 8). Like apo A-II, CRP inhibition was diminished by increasing factor VIIa concentration, suggesting that it, too, alters the ability of tissue factor to properly interact with the factor VIIa (4). Studies with CRP will continue with added attention to the calcium dependence of its association with membranes (12). Because of its selectivity for binding "damaged" membranes in preference to intact membranes, and the expression of tissue factor by "damaged" but not healthy cells in culture (15, lf;), CRP is of paramount interest for futher studies of its effect on tissue factor activity. The comparison of apo A-II and CRP effects on tissue factor activity should provide new information pertinent to tissue factor expression and regulation. ACKNOWLEDGEMENTS Our continued interactions with Drs. Ron Bach, Arabinda Guha, William Konigsberg, and Yale Nemerson are gratefully acknowledged. This work was supported by grant HL-31408 from NIH, a research grant-in-aid from the Colorado Heart Association, and a gift from R.J. Reynolds Industries, Inc. REFERENCES
1.
CARSON, S.D. Plasma high density lipoproteins inhibit the activation of coagulation factor X by factor VIIa and tissue factor. FEBS Lett. 132, 37-40, 1981.
2.
SANDERS, N.L., BAJAJ, S.P., ZIVELIN, A., and RAPAPORT, S.I. Inhibition of tissue factor/factor VIIa activity in plasma requires factor X and an additional plasma component. -Blood 66, 204-212, 1485.
3.
HUBBARD, A.R., and JENNINGS, C.A. Inhibition of tissue thromboplastin-mediated blood coagulation. Thrombosis Res. 42, 489-498, 1986.
4.
CARSON, S.D. Tissue factor (coagulation factor III) inhibition by J. Biol. Chem. 262, 718-721, 1987. apolipoprotein A-II.
5.
BROZE, G.J., Jr., and MILETICH, J.P. Characterization of the inhibition of tissue factor in serum. Blood 69, 150-155, 1987.
6.
KONDO, S., and KISIEL, W. Evidence that plasma lipoproteins inhibit the factor VIIa-tissue factor complex by a different mechanism from extrinsic pathway inhibitor. Blood, in press.
7.
RAO, L.V.M., and RAPAPORT, S.I. Studies of a mechanism inhibiting the initiation of the extrinsic pathway of coagulation. Blood 69, 645-651, 1987.
INHIBITION OF TISSUE FACTOR
676
Vol. 50, No. 5
8.
BROZE, G.J., Jr. The tissue factor inhibitor is also a factor Xa Clin. Res. 35, 597A, 1987 inhibitor.
9.
FUNAKDSHI, T., HEIYARK, R.L., HENDRICKSON, L.E., MCMULLEN, B.A., and Isolation and FUJIKAWA, K. Human placental anticoagulant protein: Riochemistry 26, 5572-5578, 1987. characterization.
10.
SCHOUSBOE, I. @2-glycoprotein I: A plasma inhibitor of the contact activation of the intrinsic hlood coagulation pathway. Blood 66, 1086-1091, 1985.
11.
POLZ, E., and KOSTNER, G.M. The binding of p2-glycoprotein Distribution among density fractions. serum lipoproteins. 102 183-186, 1979. -'
12.
GOTSCHLICH, E.C., LIU, T.-Y., and OLIVEIRA, E. Binding of C-reactive protein to C-carbohydrate and PC-suhstituted protein. Ann. N.Y. Acad. Sci. U.S.A. 389, 163-169, 1982.
13.
Further studies on the modulation of FIEDEL, B.A., and KU, C.S.L. blood coagulation by human serum amyloid P component and its acute phase homologue C-reactive protein. Thrombos. Haemostas. 55, 406-4119, 1986.
14.
NARKATES, A.J., and VOLANAKIS, J.E. C-reactive protein binding specificities: Artificial and natural phospholipid hilayers. Ann. N.Y. -Acad. Sci. U.S.A. 389, 172-182, 1982.
15.
CARSON, S.D., and ARCHER, P.G. Tissue factor activity in HeLa cells measured with a continuous chromogenic assay and ELISA reader. Thrombos. Res. 41, 185-195, 1986.
16.
MAYNARD, J.R., HECKMAN, C.A., PITLICK, F.A., and NEMERSON, Y. Association of tissue factor activity with the surface of cultured cells. J. Clin. Invest. 55, 814-824, 1975.
17.
CARSON, S.D., BACH, R., and CARSON, S.M. Monoclonal antihodies against bovine tissue factor which block interaction with factor VIIa. Blood 66, 152-156, 1985.
18.
JESTY, J., and NEMERSON, Y. The activation of bovine coagulation factor X. Meth. Enzymol. -45, 95-107, 1976
19.
ESNOUF, M.P., LLOYD, P.H., and JESTY, J. A method for the simultaneous isolation of factor X and prothrombin from bovine plasma. Biochem. J. 131, 781-789, 1973.
20.
JESTY, J., SPENCER, A.K., and NEMERSON, Y. The mechanism of activation of factor X. Kinetic control of alternative pathways leading to the formation of activated factor X. J. Biol. Chem. 249, 5614-5622, 1974.
21.
BACH, R., OBERDICK, J., and NEMERSON, Y. Immunoaffinity of bovine factor VII. Blood 63, 393-398, 1984.
22.
ROSS, S.E., and CARSON, S.D. Rapid chromatographic purification apolipoproteins A-I and A-II from human plasma. Anal. Biochem. 149 166-168, 1985. -'
I to human FEBS Lett.
purification of
Vol.
50,
No. 5
INHIBITION
677
OF TISSUE FACTOR
23.
Separation of subfractions of FINLAYSON, J.S., and MUSHINSKI, J.F. human B2-glycoprotein I. Biochim. Biophys. Acta. 147, 413-420, 1967.
24.
BRADFORD, M.M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254, 1976.
25.
VOLANAKIS, J.E., CLEMENTS, W.L., and SCHROHENLOHER, R.E. C-reactive protein: Purification by affinity chromatography and physicochemical J. Immunol. Meth. 23, 285-295, 1978. characterization.
26.
WOOD, H.F., and MCCARTY, M. The measurement of CRP in human sera. Comparison of the clinical tests on the basis of a quantitative J. Clin. Invest. 30, 616-622, 1951. method.
27.
OLIVEIRA, E.R., GOTSCHLICH, E.C., and LIU, T.-Y. Primary structure of human C-reactive protein. J. Biol. Chem. 254, 489-502, 1979.
28.
CARSON, S.D., and KONIGSBERG, W.H. Cadmium increases tissue factor (coagulation factor III) activity by facilitating its reassociation with lipids. Science 208, 307-309, 1980.
29.
CARSON, S.D. Computerized analysis of enzyme cascade reactions using continuous rate data obtained with a ELISA reader. Computer Programs Biomed. 19, 151-157, 1985.
30.
CORNISH-BOWDEN, A. In: Fundamentals Rutterworths, 1979, pp. 2‘03-207.
31.
EDWARDS, A.L. Multiple Regression and the Analysis of Variance and Covariance. WH Freeman and Co, New York, 1985.
32.
BACH, R., GENTRY, R., and NEMERSON, Y. Factor VII binding to tissue factor in reconstituted phospholipid vesicles: Induction of cooperaRiochemistry 25, 4007-4020, 1986. tivity by phosphatidylserine.
33.
FORMAN, S.D., and NEMERSON, Y. Membrane-dependent coagulation reaction is independent of the concentration of phospholipid-hound suhFluid phase factor X regulates the extrinsic system. Proc. strate: Nat1 . Acad. Sci . U.S.A. 83, 4675-4679, 1986.
34.
VAUGHAN, D.J., BRECKENRIDGE, W.C., and STANACEV, N.Z. Reconstitution of lipoproteins. I. Lipid-protein interaction of high density apoproteins, purified apoA-I and apoA-II with dimyristoyl-lecithin and dimyristoyllecithin:cholesterol vesicles studied by isomeric spinCan. J. Biochem. 58, 581-591, 1980. labelled lecithins.
35.
MILETICH, J.P., BROZE, G.J., Jr., and MAJERUS, P.W. Purification of human coagulation factors II, IX, and X using sulfated dextran beads. Meth. Enzymol. 80, 221-228, 1981.
36.
MACINTYRE, S.S., SCHULTZ, D., and KUSHNER, I. c-reactive protein. Ann. N.Y. Acad. Sci. 389,
37.
BROZE, G.J., Jr., and MAJERUS, P.W. Enzymol. 80, 228-237, 1981.
of Enzyme Kinetics.
Boston,
Biosynthesis of 76-87, 1982.
Human factor VII.
Met_h.~
678
INHIBITION OF TISSUE FACTOR
Vol. 50, No. 5
38.
ALRERS, J.J., CHEUNG, M.C., and WAHL, P.W. measurement of apolipoproteins A-I and A-II 874-878, 1980. fusion. ,I. Lip. Res. 2,
Effect of storage on the by radial immunodif-
39.
KOPPE, A.L., WALTER, H., CHOPRA, V.P., and BAJATZAOEH, M. Investigations on the genetics and population genetics of the @2-glycoprotein I Humangenetik 2, 164-171, 1970. polymorphism.
EFFECTS OF LIPID-BINDING PROTEINS APO A-I, APO A-II, 62-GLYCOPROTEIN I, AND C-REACTIVE PROTEIN ON ACTIVATION OF FACTOR X BY TISSUE FACTOR - FACTOR VIIa
Steven D. Carson and Sherman E. Ross Department of Pathology University of Colorado Health Sciences Center Denver, Colorado USA (Received 21.10.1987;
Accepted in revised form 10.12.1987 by Editor C.T. Esmon) (Received in final form by Executive Editorial Office 23.3.1988) ABSTRACT Tissue factor is the membrane-associated protein which mediates activation of factors IX and X by factor VII. In a purified, reconstituted bovine system, factor X activation by the tissue factor-factor VIIa complex is inhibited by the mixed apoproteins from human high density lipoprotein (HDL) and by isolated apolipoprotein A-II (apo A-II). Other proteins found associated with plasma lipoproteins, apolipoprotein A-I (apo A-I), C-reactive protein (CRP), and 62-glycoprote-in I (62GPI), have been examined for effects on the activation of factor X by tissue factor-factor In these experiments, bovine tissue factor, reconstituted VIIa. into phosphatidylserine-phosphatidylcholine (PS/PC; 30/70) vesicles, was used at a single concentration while factor X (the substrate), factor VIIa (the enzyme), and the potentially inhibitory proteins were varied in a continuous chromogenic assay. Apo A-II and CRP clearly inhibit tissue factor-factor VIIa activation of factor X, while apo A-I and 62 GPI have little or no effect. These results demonstrate that different lipid binding proteins vary in their effects on tissue factor activity.
INTRODUCTION Observation of components in the lipoprotein fraction of human plasma which inhibit tissue factor activity in a system reconstituted from isolated components (1) has been confirmed by several studies using contemporary methodolo:y (2-5). A detailed kinetic analysis of the effect of apolipoprotein A-II demonstrated a loss of tissue factor activity and partial protection by factor VIIa, indicating that inhibition was manifested through interference with functional association of factor VIIa and tissue factor (4). Subsequent experiments indicate that the apo A-II selectively Key Words:
Tissue factor, inhibitors,
apolipoproteins
669
670
Vol. 50, No. 5
INHIBITION OF TISSUE FACTOR
inhibits factor X activation, and does not affect factor IX activation by Other investigators have identified at tissue factor - factor VIIa (6). least two other lipid-binding inhibitors of tissue factor activity. One of these lipoprotein-associated coagulation inhibitors (LACI or extrinsic pathway inhibitor, EPI) has been identified by at least three laboratories (2, 3, 5) and is dependent on factor Xa for its effect. This inhibitor is capable of inhibiting factor Xa, is calcium-deoendent. and aooears to inhibit tissue factor activity by format on of an inactive tissue factorfactor VIIa-factor Xa-LACI complex (7, 8 . An additional lipid-binding tissue factor inhibitor, isolated from p i acenta, is structurally related to lipocortin and may have an inhibitory mechanism similar to that of apo A-II (9). I
I
In this study, three additional lip d-binding proteins have been isolated for study with regard to tissue factor-factor VIIa activation of s a major apoprotein of high factor X. Apolipoprotein A-I (apo A-I) density lipoprotein (HDL) and therefore 0 f interest due to the previous reports of inhibition of tissue factor activity by HDL and its mixed apolipoproteins (1). i-32-glycoprotein I (pzGP1) inhibits the activation of Because of its association with plasma lipoproteins (11) factor XII (10). and its effect on the intrinsic pathway of coagulation, it is potentially interesting with regard to tissue factor regulation. C-reactive protein (CRP) is an acute phase protein which has calcium-dependent abilities to interact with carbohydrate and phosphorylcholine moieties (12). In vitro, it prolongs the activated partial thromboplastin time in the presence of heparin (13). Pathologically, CRP is observed in necrotic lesions, and is apparently bound to damaged cell membranes (14). In vitro, CRP has been demonstrated to bind "traumatized" membranes but not their healthy counterparts (14). These features make CRP extremely interesting with regard to its potential effects on tissue factor activity since tissue factor is expressed in "damaged" cell membranes to a much greater extent than in healthy cell membranes (15, 16). MATERIALS AND METHODS Bovine brain phosphatidylserine (PS) and egg yolk phosphatidylcholine (PC) were obtained from Sigma. Bovine brain tissue factor was purified by immunoaffinity chromatography using monoclonal antibody TFl-F7 (17) and was kindly provided by Dr. Ron Bach (Mt. Sinai School of Medicine, New York). Factor X was purified from bovine plasma using established methods (18, 19) its concentration determined spectrophotometrically at 280 nm using 9.6 (19). Factor VII was kindly provided by Drs. Arabinda Guha and Yale Nemerson (Mt. Sinai School of Medicine, New York). Factor VII was activated with factor Xa, which had been activated with the factor X act lvating enzyme from Russell's viper venom (RVV-XAE Sigma) as described by Jesty et al. (20) and Bach et al. (21). Working stocks of factors X (400 PM) and VIIa (3.4 0) were stored at -20" in tris-saline (0.05 M tr is, 0.1 M NaCl, 0.01% NaNs, pH 7.6) with 50% glycerol and 5 mM benzamidine. Spectrozyme FXa was from American Diagnostica.
Erq =
Apolipoproteins A-I and A-II were chromatographically prepared from human plasma (22). fi2GPI was isolated from human plasma according to Finlayson and Mushinski (23), and its identity was confirmed by immunodiffusion against specific rabbit antiserum toward @2GPI (Calbiochem). p2GPI concentration was estimated using the dye-binding assay of Bradford (24) with bovine serum albumin serving as reference. Human CRP was isolated from pleural fluid obtained at autopsy using affinity chromatography on
Vol. 50, No. 5
671
INHIBITION OF TISSUE FACTOR
p-aminophenylphosphorylcholine agarose from Pierce Chemical Co 5 . CRP concentration was estimated by absorbance at 280 nm,using E le ) = 19.5 (26) and pentameric molecular weight of 110,000 (27).
The
Tissue factor was reconstituted into lipid vesicles using deoxycholate and CdClz with subsequent dialysis against tris-saline (28). The final reconstituted mixture contained 24.1 pM tissue factor, 138 #i phospholipid (PS/PC = 30/70 w/w), and 1 mg/ml bovine serum albumin (Sigma, fatty acid free). Tissue factor and phospholipid were constant throughout these experiments and were used at final reaction concentrations of 0.38 pM and 2.16 PM respectively. Tissue factor activity was measured using a continuous chromogenic assay for factor X activation and a Bio-Tek automated EIA reader (15, 29). Reagents added to the wells of a 96-well plate were, in order, 10 ul tissue factor, test protein and/or tris-saline with 1 mg/ml bovine serum albumin to give 50 ~1, 20 ~1 factor VIIa, 20 ~1 factor X, and to initiate the reaction, 60 ~1 of Spectrozyme FXa and CaClz (the 60 ~1 contained 10 ~1 of 5 mM Spectrozyme and 50 ~1 of 25 mM CaC12). The reactions were monitored at 405 nm every min for 80 min and the results were analyzed on a TI99/4A computer (29). Experiments were conducted so that data sets contained a single set of velocity curves obtained with a single inhibitor at multiple dilutions, with each experiment conducted on a separate day using unique dilutions of the enzymes, substrates and reagents (except for apo A-II, dilutions of which were distributed among all of the experiments). The test proteins were examined for potential effects on the coupled reaction (factor Xa hydrolysis of the chromogenic substrate) which could complicate interpretation with regard to the tissue factor component of the Factor X was activated by combining 3 ul of 400 $i faccoupled systems. tor X with 1 ml tris-saline (containing 1 mg/ml bovine serum albumin), 50 ~1 of RVV-XAE, and 100 ~1 25 mM CaC12. Factor X activation was monitored by combining 20 ul of the reaction mixture with 100 ~1 of plasma deficient in factors VII and X (Sigma), 100 ul of 25 mM CaC12, and 100 ~1 tris-saline containing albumin, and determining the clot-time with a fibroClot-times typically plateaued after 20 to 30 min during which the meter. clot-time dropped from greater than 70 set to near 30 sec. The chromogenic assay was conducted using PS/PC vesicles formed without tissue factor, 8.1 nM factor Xb, varied concentrations of the test proteins (apo A-I, apo A-II, fl7_GPI, CRP) and Spectrozyme FXa. The computer program (29) was modified to determine the slope of the linear A405 vs min reaction plot. None Xa hydrolysis of of these proteins altered the Vmax or Km of theTactor the Spectrozyme FXa, and any observed effects on chromogen production must therefore originate in the first reaction of the test system (the tissue factor-factor VIIa activation of factor X). Where the equation of a rectangular hyperbola was fit to the data, the direct-plot curve fitting method and algorithm (4, 30) was used to provide nonparametric estimates of curve parameters. Other data was analyzed by analysis of covariance (31). RESULTS AND DISCUSSION Comparison of apo A-I, apo A-II, @2GPI and CRP for their effects on factor X activation by tissue factor-factor VIIa (Fig. 1) revealed that
INHIBITION OF TISSUE FACTOR
672
Vol. 50, No. 5
these proteins could be paired into two categories. Apo A-I and @2GPI were virtually without effect while apo A-II and CRP both inhibited factor X activation.
50
- 40 ,\"
IO
0.2
0.4
06
08
InhIbItor(uM)
Fig. 1. Inhibition of factor Xa activation as a function of test protein Tissue factor and phospholipid were present at 0.38 pM and concentration. Factor VIIa was present at 0.86 nM and the concen2.16 PM, respectively. tration of factor X was 100 nM. The symbols represent apo A-II (v), CRP Only the solid symbols were significant(a), apo A-I (0), and ppGP1 (A). ly greater than zero by the t test (~~0.05). Error bars represent one standard deviation of the experimental values from the mean.
Since the inhibition of factor X activation by apo A-II had been shown to be influenced by factor VIIa concentration (4), inhibition was reexamined for each of the lipid-binding proteins as functions of either factor VIIa concentration or factor X concentration. The effect of factor VIIa concentration was notable only with respect to the inhibition observed with apo A-II and CRP, and these results are shown in Fig. 2.
Vol.
50,
No.
673
INHIBITION OF TISSUE FACTOR
5
loo-
IOO-
A-II
CRP
60-
3 - 60‘0 ._ 2 I5 = 40-
1
I I .o Factor IZlIo (nM)
* . 2. P-
Effect of ition of factor X were 0.103 pM (m), 0.17 uM (a), 0.35 standard deviation
I 2.0
1.0 Factor PlIo (nM)
2.0
factor VIIa concentration on the apo A-II and CRP inhiThe concentrations of apo A-II examined activation. 0.206 @i (O), and 0.413 @l (A). CRP was tested at uM (O), and 0.70 uM (A). Error bars represent one of experimental values either above or below the mean.
Clearly, increasing factor VIIa diminished the inhibitory effect of these proteins, but in no case was inhibition completely overcome. As presented in Fig. 1, neither apo A-I nor ,¶2GPI caused a notable decrease in factor X activation, and the level of inhibition observed was not significantly affected by changes in factor VIIa concentration (analysis of covariance). The effects of factor X concentration on inhibition by these proteins were small (Fig. 3), but as shown in Table 1, were significant in three cases by analysis of covariance. The regression coefficients show that 100 nM increase in factor X concentration resulted in only 6.3% and 4.9% decreases in inhibition by apo A-II and CRP, respectively, while increasing the effect of apo A-I by
2.4%. These interactions are extremely small in comparison to those of factor VIIa with apo A-II and CRP revealed in Fig. 2, even though factor X was tested up to 200 nM, which is about twice the concentration of factor X in human plasma (35). The other proteins were examined over ranges well below their normal plasma concentrations, except CRP, which is an acute phase While CRP is a trace protein in normal plasma, it may exceed reactant. 1 uM in inflammatory states (36). The normal human plasma concentrations of the other proteins used in these studies are 12.3 nM (factor VII),
Vol. 50, No. 5
INHIBITION OF TISSUE FACTOR
674
IOO-
CRP
60-
2
60-
z ._ I E 405
FOG(“M)
2%
1_
loo FactorX (nh4)
Fig. 3. Effect of factor X concentration factor VIIa activity. The concentrations to the symbols are as for Fig. 2.
200
on inhibition of tissue factorof apo A-II and CRP corresponding
TABLE I Analysis of Covariance: Inhibition (X) versus Factor X (nM)
Test Protein apo A-II CRP
apo A-I t32'3'1
Overall Coefficient of Regression* (Inhibition (X)/fX (nM))* -Il.063 (p<.fll)t -0.049 (p<.Ol)
0.024 (pai) 0.015 (N.S.)
* Coefficients of regression did not differ significantly with concentration of a given test protein (F-test). t P value determined by t test. N.S.=not significant.
41.4 uM (apo A-I), 17.1 fl (apo A-II), and 6 fl (&GPI) (37-39). The physiological importance of apo A-II and CRP as participants in the hemostatic system clearly remains to be established. Now that these proteins have been identified for their potential to regulate tissue factor initiation of coagulation, it will be possible to prepare plasma both depleted and augmented with respect to these proteins as reagents with which to address this question.
Vol.
50,
No. 5
INHIBITION
OF TISSUE FACTOR
675
It is intriguing that apo A-II and CRP produced remarkably greater effects than apo A-I and pzGP1, since all four proteins are known to bind lipid surfaces (11, 12, 34). Even though B2GPI and CRP have both been shown to interfere with the "intrinsic" pathway of coagulation (10, 13), CRP had a noteworthy effect in these experiments with tissue factor while B2GPI did not. Clearly, the mechanisms of their individual effects must not lie simply in their abilities to bind lipids. In this regard, the lipid-binding inhibitor identified by Rapaport's group and by Braze and Miletich, must not exert its effect simply through binding to the tissue factor-associated lipids since it appears to be dependent on factor Xa for its inhibitory activity (2, 5, 7, 8). Like apo A-II, CRP inhibition was diminished by increasing factor VIIa concentration, suggesting that it, too, alters the ability of tissue factor to properly interact with the factor VIIa (4). Studies with CRP will continue with added attention to the calcium dependence of its association with membranes (12). Because of its selectivity for binding "damaged" membranes in preference to intact membranes, and the expression of tissue factor by "damaged" but not healthy cells in culture (15, lf;), CRP is of paramount interest for futher studies of its effect on tissue factor activity. The comparison of apo A-II and CRP effects on tissue factor activity should provide new information pertinent to tissue factor expression and regulation. ACKNOWLEDGEMENTS Our continued interactions with Drs. Ron Bach, Arabinda Guha, William Konigsberg, and Yale Nemerson are gratefully acknowledged. This work was supported by grant HL-31408 from NIH, a research grant-in-aid from the Colorado Heart Association, and a gift from R.J. Reynolds Industries, Inc. REFERENCES
1.
CARSON, S.D. Plasma high density lipoproteins inhibit the activation of coagulation factor X by factor VIIa and tissue factor. FEBS Lett. 132, 37-40, 1981.
2.
SANDERS, N.L., BAJAJ, S.P., ZIVELIN, A., and RAPAPORT, S.I. Inhibition of tissue factor/factor VIIa activity in plasma requires factor X and an additional plasma component. -Blood 66, 204-212, 1485.
3.
HUBBARD, A.R., and JENNINGS, C.A. Inhibition of tissue thromboplastin-mediated blood coagulation. Thrombosis Res. 42, 489-498, 1986.
4.
CARSON, S.D. Tissue factor (coagulation factor III) inhibition by J. Biol. Chem. 262, 718-721, 1987. apolipoprotein A-II.
5.
BROZE, G.J., Jr., and MILETICH, J.P. Characterization of the inhibition of tissue factor in serum. Blood 69, 150-155, 1987.
6.
KONDO, S., and KISIEL, W. Evidence that plasma lipoproteins inhibit the factor VIIa-tissue factor complex by a different mechanism from extrinsic pathway inhibitor. Blood, in press.
7.
RAO, L.V.M., and RAPAPORT, S.I. Studies of a mechanism inhibiting the initiation of the extrinsic pathway of coagulation. Blood 69, 645-651, 1987.
INHIBITION OF TISSUE FACTOR
676
Vol. 50, No. 5
8.
BROZE, G.J., Jr. The tissue factor inhibitor is also a factor Xa Clin. Res. 35, 597A, 1987 inhibitor.
9.
FUNAKDSHI, T., HEIYARK, R.L., HENDRICKSON, L.E., MCMULLEN, B.A., and Isolation and FUJIKAWA, K. Human placental anticoagulant protein: Riochemistry 26, 5572-5578, 1987. characterization.
10.
SCHOUSBOE, I. @2-glycoprotein I: A plasma inhibitor of the contact activation of the intrinsic hlood coagulation pathway. Blood 66, 1086-1091, 1985.
11.
POLZ, E., and KOSTNER, G.M. The binding of p2-glycoprotein Distribution among density fractions. serum lipoproteins. 102 183-186, 1979. -'
12.
GOTSCHLICH, E.C., LIU, T.-Y., and OLIVEIRA, E. Binding of C-reactive protein to C-carbohydrate and PC-suhstituted protein. Ann. N.Y. Acad. Sci. U.S.A. 389, 163-169, 1982.
13.
Further studies on the modulation of FIEDEL, B.A., and KU, C.S.L. blood coagulation by human serum amyloid P component and its acute phase homologue C-reactive protein. Thrombos. Haemostas. 55, 406-4119, 1986.
14.
NARKATES, A.J., and VOLANAKIS, J.E. C-reactive protein binding specificities: Artificial and natural phospholipid hilayers. Ann. N.Y. -Acad. Sci. U.S.A. 389, 172-182, 1982.
15.
CARSON, S.D., and ARCHER, P.G. Tissue factor activity in HeLa cells measured with a continuous chromogenic assay and ELISA reader. Thrombos. Res. 41, 185-195, 1986.
16.
MAYNARD, J.R., HECKMAN, C.A., PITLICK, F.A., and NEMERSON, Y. Association of tissue factor activity with the surface of cultured cells. J. Clin. Invest. 55, 814-824, 1975.
17.
CARSON, S.D., BACH, R., and CARSON, S.M. Monoclonal antihodies against bovine tissue factor which block interaction with factor VIIa. Blood 66, 152-156, 1985.
18.
JESTY, J., and NEMERSON, Y. The activation of bovine coagulation factor X. Meth. Enzymol. -45, 95-107, 1976
19.
ESNOUF, M.P., LLOYD, P.H., and JESTY, J. A method for the simultaneous isolation of factor X and prothrombin from bovine plasma. Biochem. J. 131, 781-789, 1973.
20.
JESTY, J., SPENCER, A.K., and NEMERSON, Y. The mechanism of activation of factor X. Kinetic control of alternative pathways leading to the formation of activated factor X. J. Biol. Chem. 249, 5614-5622, 1974.
21.
BACH, R., OBERDICK, J., and NEMERSON, Y. Immunoaffinity of bovine factor VII. Blood 63, 393-398, 1984.
22.
ROSS, S.E., and CARSON, S.D. Rapid chromatographic purification apolipoproteins A-I and A-II from human plasma. Anal. Biochem. 149 166-168, 1985. -'
I to human FEBS Lett.
purification of
Vol.
50,
No. 5
INHIBITION
677
OF TISSUE FACTOR
23.
Separation of subfractions of FINLAYSON, J.S., and MUSHINSKI, J.F. human B2-glycoprotein I. Biochim. Biophys. Acta. 147, 413-420, 1967.
24.
BRADFORD, M.M. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254, 1976.
25.
VOLANAKIS, J.E., CLEMENTS, W.L., and SCHROHENLOHER, R.E. C-reactive protein: Purification by affinity chromatography and physicochemical J. Immunol. Meth. 23, 285-295, 1978. characterization.
26.
WOOD, H.F., and MCCARTY, M. The measurement of CRP in human sera. Comparison of the clinical tests on the basis of a quantitative J. Clin. Invest. 30, 616-622, 1951. method.
27.
OLIVEIRA, E.R., GOTSCHLICH, E.C., and LIU, T.-Y. Primary structure of human C-reactive protein. J. Biol. Chem. 254, 489-502, 1979.
28.
CARSON, S.D., and KONIGSBERG, W.H. Cadmium increases tissue factor (coagulation factor III) activity by facilitating its reassociation with lipids. Science 208, 307-309, 1980.
29.
CARSON, S.D. Computerized analysis of enzyme cascade reactions using continuous rate data obtained with a ELISA reader. Computer Programs Biomed. 19, 151-157, 1985.
30.
CORNISH-BOWDEN, A. In: Fundamentals Rutterworths, 1979, pp. 2‘03-207.
31.
EDWARDS, A.L. Multiple Regression and the Analysis of Variance and Covariance. WH Freeman and Co, New York, 1985.
32.
BACH, R., GENTRY, R., and NEMERSON, Y. Factor VII binding to tissue factor in reconstituted phospholipid vesicles: Induction of cooperaRiochemistry 25, 4007-4020, 1986. tivity by phosphatidylserine.
33.
FORMAN, S.D., and NEMERSON, Y. Membrane-dependent coagulation reaction is independent of the concentration of phospholipid-hound suhFluid phase factor X regulates the extrinsic system. Proc. strate: Nat1 . Acad. Sci . U.S.A. 83, 4675-4679, 1986.
34.
VAUGHAN, D.J., BRECKENRIDGE, W.C., and STANACEV, N.Z. Reconstitution of lipoproteins. I. Lipid-protein interaction of high density apoproteins, purified apoA-I and apoA-II with dimyristoyl-lecithin and dimyristoyllecithin:cholesterol vesicles studied by isomeric spinCan. J. Biochem. 58, 581-591, 1980. labelled lecithins.
35.
MILETICH, J.P., BROZE, G.J., Jr., and MAJERUS, P.W. Purification of human coagulation factors II, IX, and X using sulfated dextran beads. Meth. Enzymol. 80, 221-228, 1981.
36.
MACINTYRE, S.S., SCHULTZ, D., and KUSHNER, I. c-reactive protein. Ann. N.Y. Acad. Sci. 389,
37.
BROZE, G.J., Jr., and MAJERUS, P.W. Enzymol. 80, 228-237, 1981.
of Enzyme Kinetics.
Boston,
Biosynthesis of 76-87, 1982.
Human factor VII.
Met_h.~
678
INHIBITION OF TISSUE FACTOR
Vol. 50, No. 5
38.
ALRERS, J.J., CHEUNG, M.C., and WAHL, P.W. measurement of apolipoproteins A-I and A-II 874-878, 1980. fusion. ,I. Lip. Res. 2,
Effect of storage on the by radial immunodif-
39.
KOPPE, A.L., WALTER, H., CHOPRA, V.P., and BAJATZAOEH, M. Investigations on the genetics and population genetics of the @2-glycoprotein I Humangenetik 2, 164-171, 1970. polymorphism.