Regulation of Activated Protein C by Thrombin-Modified Protein S1

Transcription

1 J. Biochem. 94, (1983) Regulation of Activated Protein C by Thrombin-Modified Protein S1 Koji SUZUKI,2 Junji NISHIOKA, and Senichiro HASHIMOTO Department of Laboratory Medicine of Medicine, Tsu, Mie 514, Mie University School Received for publication, February 19, 1983 Protein S, a vitamin K-dependent plasma protein having Gla-residues, increases the rate of inactivation of Factor Va by activated protein C by enhancing the binding of activated protein C to phospholipid [Walker, J.F. (1981) J. Biol. Chem. 256, ]. The present study aimed at elucidating the effect of thrombin modified protein S on Factor Va inactivation by activated protein C. Nondigested protein S consisted 81 % of intact form and 19% of modified form, and thrombin digested protein S had 96% modified form. Protein S, both nondigested and digested, did not show any effects on the amidolytic activity of activated protein C towards synthetic peptide substrate. Nondigested protein S stimulated the Factor Va inactivation by activated protein C, whereas the digested protein appeared to suppress the inactivation. Protein-phospholipid binding experiments showed that although nondigested protein S enhanced the binding of activated protein C to phospholipid stoichiometrically, digested protein S appeared to not only suppress the complex formation, but also dissociate the complex. This evidence suggested that protein S modified by thrombin regulates the action of activated protein C towards Factor Va on phospholipid. Activated protein C, a vitamin K-dependent serine protease, appears to be a new regulatory enzyme on blood coagulation, selectively inactivating co agulant Factor V (Va) and Factor VIII (Villa) (1-4), and also on fibrinolysis, presumably by elevating plasma plasminogen activator (5, 6). 1 This study was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan, and the Clinical Pathology Research Foundation of Japan. 2 To whom correspondence should be addressed. Abbreviations: SDS, sodium dodecyl sulfate; Boc, t-butyloxycarbonyl; MCA, 4-methylcoumaryl-7-amide; AMC, 7-amino-4-methylcoumarin; MES, 2-(N-mor pholino)-ethanesulfonic acid; Mr, molecular weight. Recently it has been shown that the inactivation of human Factor Va by activated protein C is due to the digestion of high and low molecular weight (Mr) subunits of Factor Va into several degradation products on the presence of calcium ions and phospholipid (2, 7). Moreover, this inactivation is protected by Factor Xa, partly due to the protection of the digestion of Factor Vahigh Mr subunit (7). Therefore the binding site of Factor Xa has been suggested to be localized on the high Mr subunit of Factor Va. Recently, Walker (8, 9) reported that protein S, another vitamin K-dependent plasma protein whose physiological function is not yet established, can stimulate the rate of the inactivation of Factor Vol. 94, No. 3,

2 700 K. SUZUKI, J. NISHIOKA, and S. HASHIMOTO Va by activated protein C by enhancing the binding of the enzyme to phospholipid vesicles. He proposed that the equimolar binding of protein S to activated protein C on the surface of phospho lipid is required to form a Factor Va inactivation complex. Quite recently, Dahlback (10) has shown the limited proteolysis of human protein S by thrombin, and independently Morita et al. (11) have clarified that bovine protein S has a site, whose peptide bond is situated at the carboxyl-terminal side of Arg-54, very susceptible to thrombin. According to the latter authors, the amino-terminal fragment (1-54 residues), having a Gla-domain, still connects with another large carboxyl-terminal polypeptide by an intermolecular disulfide bridge after being separated from the mother molecule by thrombin. This fact is uniquely different from the characteristics observed in other vitamin K- dependent proteins (protein C, prothrombin, Factor IX, and Factor X) which have also a similar Gla-domain, but have no disulfide bridge connecting the Gla-domain with another carboxyl-terminal region (11). Based on these observations, we infer that when thrombin is formed accompanied with the activation of prothrombinase complex, consisting of Factor Xa, Factor Va, calcium ions, and phos pholipid (12), thrombin may cleave free and bound protein S on phospholipid vesicles. In this case, protein S modified by thrombin may bind to the phospholipid; however, it is unclear whether the modified protein S can still stimulate Factor Va inactivation by activated protein C. The present article attempts to elucidate the effect of the thrombin-modified human protein S on the inactivation of Factor Va by activated protein C. EXPERIMENTAL PROCEDURES Materials-All chemicals from commercial sources were the best grade available. SP-Sepha dex C-50 and QAE-Sephadex A-50 were purchased from Pharmacia Fine Chem., Tokyo. 2-(N-Morpholino)-ethanesulfonic acid (MES), 2-mercapto ethanol, and sodium dodecyl sulfate (SDS) were from Nakarai Chem., Kyoto. The synthetic fluo rogenic peptide substrate for activated protein C, Boc-Leu-Ser-Thr-Arg-MCA (13), and substrate for thrombin, Boc-Val-Pro-Arg-MCA (14), were from Protein Research Foundation, Osaka. Proteins-Protein S was prepared from freshly frozen human plasma according to Stenflo and Jonsson (15). Human protein C was purified from freshly frozen plasma according to the method of Suzuki et al. (7) and activated by thrombin (mole ratio of substrate to enzyme was 20 to 1) at 37 C for 6h. Thereafter throbmin was removed from acti vated protein C by passing it through a SP-Sephadex C-50 column (2 x 10 cm) equilibrated with 0.02 M MES-Tris, 0.05 M NaCl, ph 6.0, and subsequently activated protein C was concentrated on a QAE- Sephadex A-50 column (1 x 3 cm) equilibrated with the same buffer and then eluted out from the column with 1 M NaCl in the same buffer. Human Factor V was purified and activated by thrombin as previously described (16). Human prothrom bin and Factor X were purified according to the method of Miletich et al. (17). Factor X was activated by Russell's viper venom Factor X ac tivator purified by the method of Kisiel et al. (18). The concentration of each protein was determined by absorbance at 280 nm. The Mr and extinction coefficients used for the respective proteins were as follows. Activated protein C, 61,000, E1%E1cm gym= 13.7 (19); protein S, 70,000, El = 10.0 (20); Factor Xa, 45,000, E1%E1cm = 12.4 (21); Factor Va, c 175,000, El = 10.0 (16) and prothrobmin, 72,000, E1%E1cm=15.5 (21). Phospholipid Preparation-Phospholipid sus pension was prepared as follows from bovine brain extract (type III; Sigma Chem., St. Louis, Mo.) containing % phosphatidyl serine, 5-10% cerebrosides, and 5 % phosphatidic acid according to the instructions of the manufacturer. Exactly 500 mg of brain extract was dissolved in 5 ml of a mixture of chloroform and methanol (v/v, 2 : 1) at 4 C and then evaporated under vacuum using a rotary evaporator. Thereafter the brain extract pellet was suspended in 25 ml of 0.02 M Tris-HCl, 0.1 M NaCl, ph 7.35, followed by sonication at 4 C for 6 h with N2 gas flushing. The phospholipid suspension was stored at 4 C and was usable for at least 2 months. SDS-Polyacrylamide Slab Gel Electrophoresis -It was performed according to Blobel and Dobberstein (22) using 5 to 15 % gradient gels and the buffer system of Maizel (23). Gels were J. Biochem.

3 REGULATION OF ACTIVATED PROTEIN C BY PROTEIN S 701 stained with Coomassie blue and the bands on the gel were scanned by a Helena densitometer at 570 nm. Determination of Activity of Activated Protein C-The activity of activated protein C was determined by measuring both amidolytic and prote olytic activities towards a specific synthetic sub strate, Boc-Leu-Ser-Thr-Arg-MCA, as described in Ref. 24 and Factor Va. The representative assay system for measuring the Factor Va inactivation by activated protein C was as follows using the buffer consisting of 0.05 M Tris-HCl, 0.1 M NaCl, ph 7.5, containing 5 mm CaCI2, and 0.1 % bovine serum albumin. One hundred ƒêl of the buffer, 25 pl of the phospholipid suspension (250 ƒêg/mi), and 25 pl of Factor Va (10 ƒêg/ml) activated by 2 units/ml of thrombin, were incubated at 37 C for 2 min. 50 pl of activated protein C at varying concentrations was then added into the reaction mixture and incubated at 37 C. At intervals, 40 pl aliquots were removed from the incubation mixture and mixed with 10 ƒêl of Factor Xa (3,ƒÊg/ ml) and then 10 ƒêl of prothrombin (65 ƒêg/ml). The mixture was incubated at 37 C for 2.5 min, and 2.5 ml of 100,ƒÊM Boc-Val-Pro-Arg-MCA, a substrate for thrombin, in 0.05 M Tris-HCl, 0.1 M NaCl, 2 mm CaCl2, ph 8.0, was added to measure the activity of the thrombin generated as described (14). The amount of thrombin generated in this system corresponded to the activity of Factor Va in the reaction mixture, as illustrated in Fig. 1. From this calibration curve, the inactivation of Factor Va by activated protein C was determined. Determination of Protein-Phospholipid Binding -The interactions among activated protein C, protein S, nondigested, and digested, and phos pholipid vesicles were determined according to the method of Nelsestuen and Lim (25). Relative 90 light-scattering was measured at 350 nm excitation and emission wavelength, with a Shimadzu spec trofluorophotometer RF-510. Experiments were carried out in 350 pl final volume of 0.02 M Tris- HCl, 0.1 M NaCl, 5 mm CaCl2, and 5 ƒêg/ml of phospholipid, ph 7.5, kept at 25 C. The phos pholipid vesicle suspensions were used within 8-15 h. Protein samples were added into the phos pholipid vesicle suspension and gently mixed, and then 90 light-scattering was measured. The rela tive light-scattering was determined by comparing with the scattering of phospholipid vesicles alone. The Mr of the protein-phospholipid complex was calculated as described in Ref. 25. RESULTS Fig. 1. Relationship between concentration of human Factor Va in prothrombinase complex and activity of thrombin generated from prothrombin. 30ƒÊl of the buffer consisting of 0.05 M Tris-HCl, 0.1 M NaCl, 5 mm CaCl2, and 0.1 % bovine serum albumin, ph 7.5, 5 ƒêl of phospholipid (114 ƒêg/ml), 5 ƒêl of Factor Va at various concentrations and 10 pl of Factor Xa (5.2 ƒêg/ml) were incubated at 37 C for 2 min. 10 pl of prothrombin (65 ƒêg/ml) was then added into the reaction mixture and incubated for 2.5 min. The activity of thrombin generated was measured using Boc-Val-Pro-Arg-MCA as described in Ref. 14, and expressed as the amount of AMC released from the substrate per min. Figure 2 shows the SDS-polyacrylamide slab gel electrophoresis of reduced protein S before (A) and after (B) digestion with thrombin (the mole ratio of substrate to enzyme was 50 to 1). As described (15, 18), human protein S, like its bovine counterpart, in reduced form has a minor component with Mr of 64,000 in addition to the major component with Mr of 70,000. As recently clarified (10, 11), the lower Mr form is a modified protein S lacking the Gla-domain. According to scanning of gels stained with Coomassie blue, the nondigested protein S consisted 81 % of intact form and 19 % of modified form, and the throm bin-digested protein S had 4 % intact form and 96 % modified form. Neither the nondigested protein S nor the thrombin-modified protein S showed any effects Vol. 94, No. 3, 1983

4 702 K. SUZUKI, J. NISHIOKA, and S. HASHIMOTO Fig. 2. SDS-polyacrylamide slab gel electrophoresis of human protein S, nondigested (A) and thrombin digested (B). The samples (10 ƒêg) were reduced with 5% 2-mercaptoethanol before application to the gel. The Mr of each band was determined by comparing with six standard proteins; rabbit muscle phosphorylase b (Mr=94,000), bovine serum albumin (Mr=67,000), ovalbumin (Mr=43,000), bovine erythrocyte carbonic anhydrase (Mr=30,000), soybean trypsin inhibitor (Mr=20,100), and bovine milk a-lactalbumin (Mr= 14,400). on the amidolytic activity of activated protein C towards Boc-Leu-Ser-Thr-Arg-MCA substrate (data not shown); therefore, we developed an assay method using prothrombinase complex to determine the proteolytic activity of activated pro tein C towards Factor Va. According to the method described in " EXPERIMENTAL PRO CEDURES," the activity of activated protein C could be determined as the amount of thrombin generated from prothrombin. Figure 3 shows the relationship between the concentration of activated protein C and the activity of the remaining Factor Va. The amount of active Factor Va, having the ability to accelerate the cleavage of prothrombin by Factor Xa, decreased in proportion to the concentration of activated protein C added to the reaction mixture. Based on this technique, the effect of protein S, in both its nondigested and digested forms, on the inactivation of Factor Va by activated protein C was determined. As shown in Fig. 4, nondigested protein S stimulated the Factor Va inactivation by activated protein C, whereas thrombin-digested protein S suppressed Fig. 3. Relationship between concentration of acti vated protein C (APC) and remaining Factor Va activity. 100 ƒêl of the buffer noted in Fig. 1, 25.ƒÊl of phospholipid (250 ƒêg/ml) and 25 ƒêl of Factor Va (10 ƒêg/ml) were incubated at 37 C for 2 min. 50 ƒêl of activated protein C at varying concentrations was then added and incu bated for 2 min. Thereafter, 10 ƒêl of Factor Xa (5.2 ƒê g/ml) and 10 pl of prothrombin (65 ƒêg/ml) were added and incubated for 2.5 min. The activity of thrombin generated was then measured and expressed as the activity of remaining Factor Va. the activity of activated protein C. Figure 5 shows the effects of nondigested and thrombin digested protein S on the activity of activated protein C. First, without protein S, the activity of Factor Va was decreased to 70 percent of the original value by treating it with activated protein C. Then, in proportion to the increase of nondigested protein S, the activity of the remaining Factor Va decreased and reached a minimum at a nearly equimolar concentration of protein S and the enzyme. The activity of the remaining Factor Va appeared to increase upon the addition of thrombin-digested protein S. These results indi cate that intact protein S stimulates the inactiva tion of Factor Va by activated protein C, whereas protein S modified by thrombin seemingly suppresses the action of activated protein C on Factor Va. According to Walker (9), the stimulating effect of protein S on the Factor Va inactivation can be attributed to the enhancement of the binding of J. Biochem.

5 REGULATION OF ACTIVATED PROTEIN C BY PROTEIN S 703 Fig. 4. Effect of protein S, nondigested and thrombin digested, on Factor Va inactivation by activated protein C. 10 pl of the buffer noted in Fig. 1, 10 pl of protein S (10 jig/ml), nondigested ( ) or digested (O), or buffer (s), 5 Đl of phospholipid (25 Đg/ml), 5 pt of Factor Va (10 jig/ml) and 5 pl of activated protein C (1.3 Đg/ml) were incubated at 37 C for various times. At intervals, 10 pl of Factor Xa (3.0 Đg/ml), 5 pl of phospholipid (250 jig/ml) and 10 Id of prothrombin (65 Đg/ml) were added. After 2.5 min incubation, 5 pl of 0.2 M EDTA was added to stop the reaction. The activity of throm bin generated was then measured and expressed as the remaining Factor Va activity. activated protein C to the phospholipid vesicles. Thus, to confirm his finding and also to clarify the suppressive effect of the thrombin-modified protein S on the action of activated protein C towards Factor Va, we determined the interaction between activated protein C and phospholipid vesicles in the presence of nondigested and throm bin-digested protein S. Figure 6 shows the changes of the Mr of the complexes formed with activated protein C and phospholipid vesicles in the presence of a constant amount of nondigested and digested protein S (4.0 rim). In which, the 90 light-scattering intensity of phospholipid vesi cles alone was assigned the value 1.0 in terms of the Mr, and the Mr appears to increase in proportion to the size of complex according to Nel sestuen and Lim (25), The light-scattering inten sity of the mixture of phospholipid vesicles and the protein S, both nondigested and digested, was almost the same as that of phospholipid vesicles alone, presumably owing to a low affinity of the Fig. 5. Effect of protein S, nondigested or thrombin digested, at varying concentrations on Factor Va inactivation by activated protein C. 10 pl of the buffer noted in Fig. 1, 10 Id of protein S, nondigested ( ) or digested (0), at varying concentrations, 10 Đl of phos pholipid (25 Đg/ml), 5 Đl of Factor Va (10 Đg/ml) and 5 pl of activated protein C (1.1 Đg/ml) were incubated at 37 C for 2 min. Then 5 pl of phospholipid (250 Đ g/ml), 10 pl of Factor Xa (3.0 Đg/m]), and 10 pl of prothrombin (65 jig/ml) were added. After 2.5 min incubation, 5 pl of 0.2 M EDTA was added to stop the reaction. The activity of thrombin generated was measured and then the activity of the remaining Factor Va was expressed as percent of the original activity of Factor Va in the reaction mixture. protein for the phospholipid vesicle used in this experiment. In the presence of nondigested pro tein S, the formation of complex between activated protein C and phospholipid vesicles was evidently enhanced, but the maximum Mr of the complex was the same as that in the absence of nondigested protein S. When the concentration of activated protein C was increased to 10 nm in the presence of nondigested protein S, the Mr of the complex appeared to decrease a little. The reason for this phenomenon is, however, unknown at present. On the other hand, in the presence of the throm bin-digested protein S the complex formation was markedly suppressed. Figure 7 shows the results of a similar ex periment to determine the effects of varying con centrations of protein S, nondigested and digested, human prothrombin and bovine serum albumin Vol. 94, No. 3, 1983

6 704 K. SUZUKI, J. NISHIOKA, and S. HASHIMOTO Fig. 6. Effect of protein S, nondigested and digested, on the binding of activated protein C to phospholipid vesicles. Into 340 ƒêl of phospholipid suspension (5 ƒêg/ml), 4 pl of protein S (4.0 nm final concentration), nondigested (0) or digested (A), or the buffer (0) was added, and then 6 pl of activated protein C at various concentrations was added. The incubation mixture was gently mixed and 90 light-scattering was measured. The Mr was calculated as described by Nelsestuen and Lim (25). on the Mr of the activated protein C-phospholipid complex. Under these conditions, protein S itself, both nondigested and digested, appeared to bind a little onto phospholipid when relatively large amounts of proteins (over 10 rim) were incubated with the phospholipid, and there were no differ ences between the binding abilities of the two forms of protein S. When bovine serum albumin or prothrombin was added into the mixture of activated protein C (4.5 nm) and phospholipid vesicles, neither protein had very much effect on the Mr of the activated protein C-phospholipid complex. Nondigested protein S apparently fur ther increased the Mr of the complex, and forma tion of the complex appeared to reach a maximum at about 6 rim of the protein S. Since this nondigested protein S consisted 81 % of the intact form, the mole ratio of the intact protein S to activated protein C was estimated to be nearly equal (about 1.1). On the other hand, thrombin digested protein S reduced the Mr of the activated protein C-phospholipid complex, suggesting that the modified protein S stimulates the dissociation of the complex. DISCUSSION Fig.{7. Effect of protein S, nondigested and digested, prothrombin and bovine serum albumin on the binding of activated protein C to phospholipid vesicles. Into 340,u] of phospholipid suspension (5 ƒêg/ml), 6ƒÊl of activated protein C (4.5 nm final concentration) was added. Thereafter 4 pl of nondigested (0) or digested (A) protein S, prothrombin (0) or bovine serum albu min ( x ) at various concentrations was added into the incubation mixture, and then 90 light-scattering was measured. As controls, nondigested (0) or digested (p) protein S at varying concentrations was added into 346 ƒêl phospholipid suspension, and the light-scattering was measured. Mr was calculated as described by Nelsestuen and Lim (25). According to Walker (9), 1 mol of protein S combines with 1 mol of activated protein C on the surface of phospholipid vesicles to form a Factor Va inactivation complex. In other words, protein S enhances the binding of activated protein C to phospholipid by decreasing the dissociation constant between the two materials. Thus, any soluble phase interaction between protein S and activated protein C is not thought to affect the rate of Factor Va inactivation by the enzyme in the absence of phospholipid vesicles. The present experiments supported Walker's finding in part and also revealed several new facts. Although the thrombin-modified protein S did not affect the amidolytic activity of activated pro tein C, it appeared to suppress the Factor Va inactivation by the enzyme. The protein-phos pholipid binding experiments indicated that the modified protein S apparently not only suppressed the binding of activated protein C to phospho lipid but also dissociated the complex. Judging from the available information re garding the formation of complex between phos- J. Biochem.

7 REGULATION OF ACTIVATED PROTEIN C BY PROTEIN S 705 pholipid and vitamin K-dependent proteins (25-27), the Gla-residues located at the amino-terminal region of these proteins are essential for the pro teins to bind to phospholipid surface via calcium ions and to function effectively. The stoichio metrical formation of the activated protein C- protein S complex, like prothrombinase complex, on the surface of phospholipid would be effective for the enzyme to digest Factor Va. Based on the present results, we infer that intact protein S enhances the binding of activated protein C to phospholipid to inactivate Factor Va, thus resulting in the decrease in activity of the prothrom binase complex, as previously suggested (9). On the other hand, the modified protein S blocks the binding of activated protein C to phospholipid and furthermore dissociates the enzyme from the phospholipid surface, presumably by decreasing the affinity of the enzyme for phospholipid, thus reducing the inactivation of Factor Va. Thus, we propose that the thrombin-modified protein S may contribute to regulating the action of activated protein C towards Factor Va on phospholipid. We are grateful to Dr. Bjorn Dahlback for providing a preprint of his manuscript. We also wish to thank Miss Yoshimi Yamada for her excellent technical assist ance in determining the binding of proteins to phospho lipid vesicles. REFERENCES 1. Kisiel, W., Canfield, W.M., Ericsson, L.H., & Davie, E.W. (1979) Biochemistry 16, Walker, F.J., Sexton, P.W., & Esmon, C.T. (1979) Biochim. Biophys. Acta 571, Marlar, R.A., Kleiss, A.J., & Griffin, J.H. (1982) Blood 59, Vehar, G.A. & Davie, E.W. (1980) Biochemistry 19, Comp, P.C. & Esmon, C.T. (1981) J. Clin. Invest. 68, Comp, P.C., Jacocks, R.M., Ferrell, G.L., & Esmon, C.T. (1982) J. Clin. Invest. 70, Suzuki, K., Stenflo, J., Dahlback, B., & Teodorsson, B. (1983) J. Biol. Chem. 258, Walker, F.J. (1980) J. Biol. Chem. 255, Walker, F.J. (1981) J. Biol. Chem. 256, Dahlback, B. (1983) Biochem. J. 209, Morita, T., Mizuguchi, J., & Iwanaga, S. (1982) The 5th Congress of The Japanese Society on Thrombosis and Hemostasis (abstract in Japanese) p Jackson, C.M. & Nemerson, Y. (1980) Annu. Rev. Biochem. 49, Iwanaga, S., Morita, T., Kato, H., Harada, T., Adachi, N., Maruyama, I., Takada, K., Kimura, T., & Sakakibara, S. (1979) in KININS-II: Biochemis try, Pathophysiology and Clinical Aspects (Fujii, S., Moriya, H., & Suzuki, T., eds.) pp , Plenum Publishing Corporation, New York 14. Ohno, Y., Kato, H., Morita, T., Iwanaga, S., Takada, K., Sakakibara, S., & Stenflo, J. (1981) J. Biochem. 90, Stenflo, J. & Jonsson, M. (1979) FEBS Lett. 101, Suzuki, K,, Dahlback, B., & Stenflo, J. (1982) J. Biol. Chem. 257, Miletich, J.P., Broze, G.J., Jr., & Majerus, P.W. (1980) Anal. Biochem. 105, Kisiel, W., Hermodson, M.A., & Davie, E.W. (1976) Biochemistry 15, Kisiel, W. (1979) J. Clin. Invest. 64, DiScipio, R.G., Hermodson, M.A., Yates, S.G., & Davie, E.W. (1977) Biochemistry 16, Owen, W.G., Esmon, C.T., & Jackson, C.M. (1974) J. Biol. Chem. 249, Blobel, G.B. & Dobberstein, B. (1975) J. Cell Biol. 17, Maizel, J.V., Jr. (1971) Methods Viol. 5, Suzuki, K., Nishioka, J., & Hashimoto, S. (1983) J. Biol. Chem. 258, Nelsestuen, G.L. & Lim, T.K. (1977) Biochemistry 16, Nelsestuen, G.L., Kisiel, W., & DiScipio, R.G. (1978) Biochemistry 17, Bloom, J.W., Nesheim, M.E., & Mann, K.G. (1979) Biochemistry 18, Vol. 94, No. 3, 1983