Original Paper Haemostasis 1992;22:165-172
Blood Systems Research Foundation Laboratory, Department of Pathology, University of New Mexico School of Medicine, Albuquerque, N. Mex., USA
Key Words
Activated protein C Tumor cells Factor Va Protein S
Inactivation of Factor Va by Activated Protein C on Selected Human Tumor Cell Lines Abstract
Previous studies have demonstrated that platelets or aortic endothelial cells provide an appropriate surface that augments the proteolytic inactivation of factor Va by activated protein C (APC). We have examined the ability of three human tumor cell lines (HepG2, CAPAN-2 and J82) to support the inactiva tion of human factor Va by human APC in the presence and absence of human protein S. APC-mediated factor Va inacti vation on these tumor cell lines was assessed by measuring the ability of residual cell-bound factor Va to augment the proteo lytic activation of prothrombin by factor Xa. Each of the tumor cell lines studied supported factor Va inactivation by APC in the presence of calcium ions. HepG2 cell monolayers supported this reaction most effectively, with CAPAN-2 and J82 cell monolayers exhibiting moderate and weak effective ness, respectively. Although not essential for this reaction, protein S moderately enhanced the rate of factor Va inactiva tion by APC on these tumor cell lines. In addition, pretreat ment of each tumor cell line with rabbit antihuman protein S IgG had little, if any, effect on its ability to support factor Va inactivation by APC. Our data suggest that these, and perhaps other, tumor cells can provide an appropriate phospholipid surface for promoting factor Va binding and rapid inactiva tion by APC.
Introduction
Protein C is a vitamin-K-dependent glyco protein that plays an important role in the reg ulation of coagulation. Protein C is synthe
Received: March 15,1991 Accepted in revised formbyProf.Hcmkcr: December 19, 1991
sized in the liver and secreted into the blood where it circulates as a precursor to a serine protease, activated protein C (APC) [1, 2]. Human protein C is converted to APC by pro teolytic cleavage of a dodecapeptide from the
Walter Kisiel, PhD Department of Pathology University of New Mexico School of Medicine Albuquerque, NM 87131 (USA)
© 1992 S. Karger AG, Basel 0301-0147/92/ 0224-0165S2.75/0
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
Tomohiro Nakagaki Tatsuya Sueyoshi Walter Kisiel
Methods Materials H-D-Phc-Pip-Arg-pNA (S-2238) was obtained from Helena Laboratories, Beaumont, Tex., USA. Tis sue culture flasks and plates (24 wells) were purchased from Corning. Minimum essential medium (Eagle), RPMI medium 1640, McCoy’s 5a medium, and nonessential amino acids were purchased from Mediatech. Minimum essential medium alpha, trypsin (1 X solu tion; tissue culture grade), penicillin-streptomycin (104 U penicillin and 10 mg streptomycin/ml) and bovine
166
scrum abumin were obtained from Sigma. Fetal bo vine serum was purchased from Hyclone Laboratories. Human brain thromboplastin was prepared according to Quick [20]. All other reagents were the highest qual ity available from commercial sources. Proteins All proteins were purified from human plasma. Protein C was purified to homogeneity by a combina tion of barium citrate adsorption and elution, DEAESepharose CL-6B chromatography, dextran sulfate agarose chromatography and preparative electropho resis essentially as described [2], Protein C was acti vated by incubation with the purified protein C activa tor (ACC-C) from Agkistrodon contorlrix contorlrix venom as described [21 ]. The final product was pure as assessed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis [22], Factors X and Xa were prepared as previously described [23], Human prothrombin was prepared according to Kisiel and Hanahan [24]. Hu man factor V was purified by a combination of proce dures involving barium citrate adsorption, 8-12% po lyethylene glycol 6000 precipitation. DEAE-Sepharose CL-6B column chromatography and Ultrogel AcA-22 column chromatography essentially as described by Suzuki ct al. [8], Conversion of factor V to factor Va was carried out by incubation of factor V with the puri fied factor V activator from Russell’s viper venom (RVV-V) at an enzyme-substrate weight ratio of 1:100 [25]. Protein S was purified to homogeneity as de scribed [26], Monospecific antibodies against human protein S were produced in rabbits and IgG purified from antisera by protein A-Sepharose (Pharmacia) af finity chromatography. Cell Culture Each cell line was cultured as described by Sakai et al. [19], Factor Va inactivation studies were performed on cell monolayers within 24 h after reaching confluency. Inactivation o f Factor Va by APC The medium from 2.0-cm2 wells with confluent tumor cells was aspirated, and the cells were washed with 0.5 ml ofHepes saline buffer [HSB; 10 mA/Hepes (pH 7.45)/137 mM NaCl/4 mM KC1, 11 mM glucose] supplemented with 10 mA/ EDTA followed by three washes of HSB (0.5 ml/wash). Incubation buffer [0.5 ml of HSB supplemented with 0.5 % bovine serum albumin and 5 mM CaCL (HSB+)] containing various concentrations of APC (0-3 |iM ) in the absence or presence of protein S (160 n M) was added to each well. Following a 60-min incubation at room temperature,
Nakagaki/Sueyoshi/Kisiel
Factor Va Inactivation on Tumor Cells
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
amino terminus of the heavy chain of the mol ecule by a complex of thrombin and thrombo modulin [3]. APC, in turn, regulates thrombin formation through the selective proteolytic degradation of factors Va and Villa [4-6]. Factor Va is a heterodimeric protein [7, 8] that serves as an essential nonenzymatic com ponent of the prothrombinase complex. As a cofactor, factor Va accelerates prothrombin activation by factor Xa on a membrane surface approximately 10,000-fold [9-13]. Thus, inac tivation of membrane-bound factor Va mark edly reduces the rate of thrombin generation. Earlier studies using either platelets [ 14-16] or bovine aortic endothelial cells [ 17] have shown that these cells effectively support factor Va inactivation by APC. Protein S enhances the rate of this reaction, although differences have been noted among studies with respect to the effectiveness of protein S in augmenting the reaction on cell surfaces [14-17], While a considerable amount of data has accumulated in recent years on the ability of tumor cells to support the assembly and func tional expression of the prothrombinase com plex [18, 19], little is known as to whether or not these cells support the protein C anticoag ulant pathway. In this paper, we report on the ability of three human tumor cell lines, of diverse histologic origin, to support the inacti vation of factor Va by APC in the presence or absence of protein S.
(AA4o5/min)* - (AA4o5/min)§
(AA405/min)* where (AA4os/min)* is the absorbance change at 405 nm in the absence of APC, and (AA4os/min)§ is the absorbance change at 405 nm in the presence of each concentration of offered APC. The ability of cell-bound APC to inactivate fluidphase factor Va in the presence and absence of protein S was also determined. In this system, cells were ini tially preincubated at room temperature with HSB+ containing varying concentrations of APC (0-3 \xM) in the absence and presence of protein S (160nA/) for 60 min followed by six washes with HSB\ Factor Va (2 pg/ml), dissolved in 0.5 ml of HSB\ was then of fered to the cells and incubated at room temperature with gentle mixing for 30 min. At this time, aliquots (10 pi) of the supernatant were removed and imme diately assayed in a one-stage factor V clotting assay [27] by mixing 100 pi of diluted sample, 100 pi of human thromboplastin, 100 pi of factor-V-deficient human plasma [28] and 100 pi of 25 mM CaCk Clot ting times were determined from the point of calcium addition, and factor Va activity interpolated from linear log-log standard curve relating factor Va activity and clotting time. In experiments designed to assess whether tumor cells contain functional cell surface protein S, tumor cells were first incubated for 60 min at 37 °C with either nonimmune rabbit IgG (500 pg/ml) or rabbit antihuman protein S IgG (500 pg/ml) followed by six washes prior to the addi tion of APC. Control experiments indicated that this concentration of antihuman protein S IgG was suffi cient to completely abrogate the rate-enhancing effect observed with exogenous protein S (160 nM).
Results
In initial experiments, we compared the inactivation of cell-bound and fluid-phase factor Va by HepG2 cell-bound APC. In those experiments designed to assess cell-bound factor Va inactivation, HepG2 monolayers were offered varying concentrations of APC and incubated in the presence of 5 xnM cal cium for 60 min at room temperature. Fol lowing several washes to remove unbound APC, factors Va and Xa were offered to the cells and incubated for an additional 30 min to allow for factor Va-Xa binding to the HepG2 cell surface. Following this 30-min incubation, the remaining cell-bound factor Va cofactor activity was quantitated through its ability to support the factor-Xa-dependent conversion of prothrombin to thrombin. The concentration of factor Va selected for these studies was 2pg/ml, which from previous studies produced maximal prothrombinase activity on several tumor cell lines including those cell lines employed in the present study [19]. The factor Xa concentration employed, 3 nM, was twice the Kp observed for factor Xa binding to J82 and HepG2 cells reported in an earlier study from this laboratory [29]. In those experiments designed to measure fluidphase factor Va inactivation by cell-bound APC, factor Xa was omitted and only factor Va (2 pg/ml) was offered to the APC-treated cells. Fluid-phase factor Va activation was then assessed in aliquots of the supernatants above the cells in a factor V clotting assay. As shown in figure 1, the inactivation of cellbound factor Va required a considerably lower concentration of APC than that re quired for the relatively slow inactivation of fluid-phase factor Va. In both systems, pro tein S exhibited a small, but measurable, en hancement of the APC-mediated activation of factor Va on and above the HepG2 monolayers. Similar results with respect to cell-
167
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
cells were washed six times with HSB+. Incubation buffer (0.5 ml of HSB+) containing factor Va (2 gg/ml) and factor Xa (3 nA/) was then added to the cell monolayers and incubated at room temperature with con stant gentle mixing for 30 min. At this time, cells were washed six times with HSB+, and residual cell-bound factor Va cofactor activity was assessed in an amidolytic assay as follows. HSB+ (0.5 ml) containing pro thrombin (2 \iM) was added to the washed monolayers and incubated at room temperature with gentle mixing for 10 min. Aliquots (10 pi) of the incubation mixture were removed and added to 790 pi of 50 m/V/Tris-HCl (pH 8.3) containing 150 mM NaCl, 5 mM EDTA and 0.1 mM S-2238. The absorbance at 405 nm was con tinuously recorded using a Beckman DU 65 spectro photometer. The relative rate of factor Va inactivation was expressed by the following empirical relationship:
Fig. 2. Calcium dependence of APC binding to HepG2 cells. Confluent FlepG2 monolayers were incu bated with either 80 nM APC (•) or 80 nM APC with 160 nM protein S (o) for 60 min at room temperature in the presence of 0-20 m.V/ CaCG. The cells were washed six times with HSB containing the respective calcium concentrations, and a mixture of factor Va (2 pg/ml) and factor Xa (3 n.V/). in IISB". was added and incubated for an additional 30 min. Residual fac tor Va cofactor activity was determined using the S2238 amidolvtic assay.
bound versus fluid-phase factor Va inactiva tion were observed on monolayers of J82 and CAPAN-2 tumor cell lines (data not shown). In addition, essentially identical results were obtained if factor Va was first incubated with the cells for 15 min prior to the addition of factor Xa and an additional 15-min incuba tion before the washing steps (data not shown). This latter finding suggests that on these tumor cells, factor Xa docs not appear to confer a significant protective effect with respect to factor Va inactivation by APC as noted earlier for human platelets [ 16].
The results of the above studies, carried out in the presence of 5 mM calcium, sug gested that the binding of the APC to the tumor cell occurred in a calcium-dependent manner. To confirm this calcium depen dence, HepG2 monolayers were incubated with a constant concentration of APC in the presence and absence of protein S for 60 min in the presence of 0-20 mM CaCfi. After washing the cells with Hepes saline buffer containing the calcium concentration used in the incubation, cells were offered factors VaXa in the presence of 5 mM calcium. Follow ing a 30-min incubation, cell-bound factor Va
168
Nakagaki/Sueyoshi/Kisiel
Factor Va Inactivation on Tumor Cells
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
Fig. 1. Inactivation of factor Va by APC on HepG2 cells. HepG2 monolayers were incubated with varying concentrations of APC (0-350 nM) in the presence (o) or absence (•) of protein S (160 mW) for 60 min at room temperature. Following six washes with HSB+, the cells were either offered a mixture of factor Va (2 pg/ml) and factor Xa (3 nM), or factor Va alone (2 ,ug/ml). After a 30-min incubation, residual factor Va activity in the supernatant was determined by a dotting assay (-----), or residual cell-bound factor Va (----- ) was determined with a S-2238 amidolvtic assay as described in Methods.
cr
Fig. 3. Inactivation of factor Va by APC bound to J82 cells (a), CAPAN-2 cells (b), and HepG2 cells (c). Varying concentrations of APC (0-3 \iM) were incu bated 60 min at room temperature in the absence (•) and presence (o) of 160 nM protein S followed by six washes of HSEP. A mixture of factor Va (2 gg/ml) and factor Xa (3 nM), in 0.5 ml HSB+, was then added to each monolayer, and incubated for an additional 30 min at room temperature. After six washes with HSB+, prothrombin (2 \\.M in HSB+) was added, and residual factor Va cofactor activity was measured us ing the S-2238 amidolytic assay.
169
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
cofactor activity remaining was then quanti tated according to its relative ability to sup port the factor-Xa-dependcnt activation of prothrombin. As shown in figure 2, the bind ing of APC, as reflected in its ability to inacti vate cell-bound factor Va, was dependent on calcium, and approximately 1 mM calcium was required for maximal binding and inacti vation of factor Va. Higher concentrations of calcium (2-20 mM) had no measurable effect on APC binding to the HcpG2 cells as mani fest in its ability to inactivate cell-bound fac tor Va (fig. 2). In the next series of experiments, we com pared the ability of three tumor cell lines of diverse histological origin to support APC binding and the APC-mcdiated inactivation of cell-bound Va. Each of the tumor cell line monolayers tested augmented the APC-mediated factor Va inactivation. Pretreatment of each tumor cell line with antihuman protein S IgG had no significant effect on the rate of APC-mediated inactivation of cell-bound fac tor Va, suggesting that these tumor cells do not contain functional cell surface protein S of tumor origin (data not shown). The order of effectiveness was HepG2 > CAPAN-2 > J82 in the presence and absence of protein S (fig. 3). Protein S exhibited a modest rateenhancing effect on the rate of factor Va inac tivation on all tumor cell lines (fig. 3). The concentrations of APC required for half-max imal inactivation of cell-bound factor Va were obtained from double reciprocal plots of the data presented in figure 3. In the absence of protein S, the concentrations of APC required for half-maximal inactivation of factor Va on HepG2, CAPAN-2 and J82 cell monolayers were 8.3, 30 and 136 nM, respectively. In the presence of 160 nM protein S, these values were reduced to 2.0, 17 and 66 nM, respec tively. It is perhaps appropriate to mention at this point that the above concentrations of APC may, in fact, overestimate the concen-
Fig. 4 . Inactivation of cellbound factor Va by APC. A mix ture of factor Va (2 |ig/ml) and fac tor Xa (3 nM) were incubated with HepG2 monolayers for 60 min at room temperature, and washed six times with HSB+. Varying concen tration of APC (0-25 nM) were offered to the cells with (o) or without (•) 160 nM protein S and incubated an additional 30 min at room temperature. After six washes with HSB+, prothrombin (2 \xM in HSB+) was added and residual factor Va cofactor activity measured using the S-2238 amidolytic assay. Inset: double reciprocal plot of the data.
170
~ 400 pM in the absence of protein S (fig. 4), and are to be compared to APC concentration of 2 and 8.3 nM, respectively, obtained on HepG2 monolayers in the preceding experi ments where the order of addition was APC followed by factor Va. A similar reduction of APC concentrations required for half-maxi mal inactivation of factor Va using this order of component addition was also observed on J82 and CAPAN-2 monolayers (data not shown).
Discussion
Previous studies have shown that several human tumor cell lines can readily assemble a functional cell surface prothrombinase com plex [18,19]. Factor Va is a key component of the prothrombinase complex and augments the factor-Xa-catalyzed activation of pro thrombin three orders of magnitude in the presence of calcium ions and an appropriate phospholipid membrane [14, 17], In the
Nakagaki/Sueyoshi/Kisiel
Factor Va Inactivation on Tumor Cells
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
trations of APC necessary to achieve halfmaximal inactivation of cell-bound factor Va on each cell line due to the potential mechani cal removal of APC by the cell-washing steps following the incubation period. To further characterize the APC-mediated inactivation of cell-bound factor Va, the order of offering APC and factor Va to the cells was reversed. In the preceding series of studies factor Va was offered to the cells previously treated with APC. In these experiments, fac tor Va and Xa were first incubated with tumor cells for 60 min, followed by six washes and the subsequent incubation with varying concentrations of APC in the presence and absence of protien S. The results of these stud ies, shown in figure 4, reveal that the concen trations of APC required for half-maximal inactivation of HepG2 cell-bound factor Va was dramatically reduced relative to the op posite order of component addition. The con centration of APC required for half-maximal inactivation of factor Va in these studies was ~ 50 pM in the presence of protein S and
ability to accelerate the formation of APC by offered thrombin in relation to the absence of cells [Nakagaki T., unpubl. data]. Whether or not other tumor cell lines constitutively syn thesize thrombomodulin, or whether pro teases in the culture medium degrade cell sur face thrombomodulin will require several additional studies. Conceivably, mechanisms which utilize the factor-Va-dependent acti vation of protein C by thrombin [30-32] are important in the formation of APC in the tumor microenvironment. Alternatively, plasma-derived APC may suffice to catalyze tumor-cell-bound factor Va inactivation and thereby modulate fibrin formation and depo sition. Immunochemical studies have shown that circulating blood contains less than 500 pM APC [33]. Based on our studies, which indicate that half-maximal rates of fac tor Va inactivation on HepG2 cells are ob served at 50 pM APC in the presence of pro tein S, it is not unlikely that these concentra tions of APC are present in the tumor mi croenvironment given the hyperpermeable nature of the tumor microvasculature [34].
Acknowledgments We wish to thank Nancy Basore for excellent tech nical assistance throughout these studies. This work was supported by research grants from the National Institutes of Health (HL35246) and Blood Systems. Inc. T.S. is a recipient of a grant from Toyobo Biotech nology Foundation.
171
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
present study, we demonstrate that three hu man tumor cell lines, HepG2, CAPAN-2 and J82, each support, to varying degrees, the inactivation of cell-bound factor Va by APC in the presence of factor Xa. The order of effectiveness in this regard was HcpG2 > CAPAN-2 > J82. In contrast to the results of studies using bovine platelets and bovine en dothelial cells [14, 15, 17], human protein S was not essential for the APC-mediated inac tivation of cell-bound factor Va. In our stud ies, protein S moderately increased the rate of factor Va inactivation by APC. Inasmuch as kinetic parameters for the inactivation of fac tor Va by APC were not obtained in this study, it is uncertain whether protein S in creased the kcat or decreased the Km of this reaction. The results of our studies demonstrate that tumor cells bind APC in a calcium-dependent manner that results in the expression of func tional cell surface proteolytic activity in the absence of protein S. Our experimental find ings using human APC and human tumor cell lines corroborate earlier studies by Solymoss et al. [16], who were the first investigators to report the protein-S-independent inactivation of factor Va by APC on phospholipid vesicles and human platelets. Whether or not human tumor cells phenotypically express a unique cell surface APC-binding protein is unknown and will require further studies. The physiological relevance of our findings with respect to APC-mediated factor Va inac tivation on tumor cells and its impact on fibrin investment of solid tumors is uncertain. In addition, whether or not tumor cells syn thesize and express cell surface thrombomo dulin that participates in the formation of APC in situ is unknown. The results of pre liminary studies carried out in our laboratory indicate that the three tumor cell lines exam ined in this study do not express cell surface thrombomodulin on the basis of the cells’
1 Stenflo JA: A new vitamin K-dependent protein: Purification from bo vine plasma and preliminary char acterization. J Biol Chem 1976:251: 355-363. 2 Kisiel W: Human plasma protein C: Isolation, characterization, and mechanism of activation by athrombin. J Clin Invest 1979:64: 761-769. 3 Esmon CT. Owen WG: Identifica tion of an endothelial cell cofactor for thrombin-catalyzed activation of protein C. Proc Natl Acad Sci USA 1981;78:2249-2252. 4 Kisiel W. Canfield WM. Ericsson LH, Davie EW: Anticoagulant prop erties of bovine plasma protein C following activation by thrombin. Biochemistry 1977;16:5824-5831. 5 Suzuki K, Stenflo J. Dahlback B. Teodorsson B: Inactivation of hu man coagulation factor V by acti vated protein C. J Biol Chcm 1983; 258:1914-1920. 6 Vehar GA, Davie EW: Preparation and properties of bovine factor VIII (antihemophilic factor). Biochemis try 1980:19:401-410. 7 Nesheim ME, Mann KG: Throm bin-catalyzed activation of single chain bovine factor V. J Biol Chem 1979;254:1326-1334. 8 Suzuki K, Dahlback B. Stenflo J: Thrombin-catalyzed activation of human coagulation factor V. J Biol Chem 1982;257:6556-6564. 9 Jobin F, Esnouf MP: Studies on the formation of the prothrombin-converting complex. Biochem J 1967; 102:666-674. 10 Esmon CT, Owen WG, Jackson CM: A plausible mechanism for pro thrombin activation by factor Xa, factor Va, phospholipid, and cal cium ions. J Biol Chem 1974;249: 8045-8047. 11 Nesheim ME, Taswell JB, Mann KG: The contribution of bovine fac tor V and factor Va to the activity of prothrombinase. J Biol Chem 1979; 254:10952-10962. 12 Rosing J, Tans G, Grovers-Riemslag JWP. Zwaal RFA, Hemker HC: The role of phospholipid and factor Va in the prothrombinase complex. J Biol Chem 1980;255:274-283.
13 Kane WH, Majerus PW: The inter action of human coagulation factor Va with platelets. J Biol Chem 1982; 257:3963-3969. 14 Harris KW, Esmon CT: Protein S is required for bovine platelets to sup port activated protein C binding and activity. J Biol Chem 1985;260: 2007-2010. 15 Suzuki K, Nishioka J, Matsuda M, Murayama H, Hashimoto S: Protein S is essential for the activated pro tein C-catalyzed inactivation of platelet-associated factor Va. J Bio chem (Tokyo) 1984:96:455-460. 16 Solymoss S. Tucker MM, Tracy PB: Kinetics of inactivation of mem brane-bound Va by activated pro tein C. J Biol Chem 1988;263: 14884-14890. 17 Stern DM, Nawroth PP. Harris K, Esmon CT: Cultured bovine aortic endothelial cells promote activated protein C - protein S - mediated inactivation of factor Va. J Biol Chem 1986:261:713-718. 18 Van DeWater L. Tracy PB, Aronson D. Mann KG, Dvorak HF: Tumor cell generation of thrombin via func tional prothrombinase assembly. Cancer Res 1985;45:5521-5525. 19 Sakai T, Noguchi M. Kisiel W: Tu mor cells augment factor Xa-catalyzed conversion of prothrombin to thrombin. Haemostasis 1990;20: 125-135. 20 Quick AJ: Hemorrhagic Disease and Thrombosis. Philadelphia, Lea & Febiger, 1966. 21 Kisiel W. Kondo S, Smith KJ, McMullen BA, Smith LF: Charac terization of a protein C activator from Agkistrodon contortrix contortrix venom. J Biol Chem 1987;262: 12607-12613. 22 Laemmli U: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;277:680-685. 23 Kondo S, Kisiel W: Evidence that plasma lipoproteins inhibit the fac tor VII tissue factor complex by a different mechanism than extrinsic pathway inhibitor. Blood 1987;70: 1947-1954.
24 Kisiel W, Hanahan DJ: Purification and characterization of human fac tor II. Biochem Biophys Acta 1973; 304:103-110. 25 Kisiel W, Canfield WM: Snake ve nom proteases that activate blood coagulation factor V. Methods Enzymol 1981;80:275-285. 26 Dahlback B: Purification of human vitamin K-dependent protein S and its proteolysis by thrombin. Bio chem J 1983:209:2007-2010. 27 Nesheim ME, Katzmann JA, Tracy PB, Mann KG: Factor V. Methods Enzymol 1981:80:249-274. 28 Bloom JW, Nesheim ME, Mann KG: A rapid technique for the prep aration of factor V-deficient plasma. Thromb Res 1979:15:595. 29 Sakai T. Kisiel W: Binding of hu man factors X and Xa to HepG2 and J82 human tumor cell lines: Evidence that factor Xa binds to tumor cells independent of factor Va. J Biol Chem 1990;265:91059113. 30 Salem HH, Broze GJ. Miletich JP. Majerus PW: Human coagulation factor VA is a cofactor for the acti vation of protein C. Proc Natl Acad Sci USA 1983:80:1584-1588. 31 Salem HH, Broze GJ. Miletich JP. Majerus PW: The light chain factor Va contains the activity of factor Va that accelerates protein C activation by thrombin. J Biol Chem 1983;258: 8531-8534. 32 Maruyama I, Salem HH, Majerus PW: Coagulation factor Va binds to human umbilical vein endothelial cells and accelerates protein C acti vation. J Clin Invest 1984;74:224— 230. 33 Bauer KA, Kass BL, Beeler DL. Ro senberg RD: Detection of human protein C activation in humans. J Clin Invest 1984;74:2033-2041. 34 Dvorak HF, Nagy JA, Dvorak JT, Dvorak AM: Identification and characterization of the blood vessels of solid tumors that are leaky to cir culating macromolecules. Am J Pa thol 1988;133:95-109.
172
Nakagaki/Sueyoshi/Kisiel
Factor Va Inactivation on Tumor Cells
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
References
Blood Systems Research Foundation Laboratory, Department of Pathology, University of New Mexico School of Medicine, Albuquerque, N. Mex., USA
Key Words
Activated protein C Tumor cells Factor Va Protein S
Inactivation of Factor Va by Activated Protein C on Selected Human Tumor Cell Lines Abstract
Previous studies have demonstrated that platelets or aortic endothelial cells provide an appropriate surface that augments the proteolytic inactivation of factor Va by activated protein C (APC). We have examined the ability of three human tumor cell lines (HepG2, CAPAN-2 and J82) to support the inactiva tion of human factor Va by human APC in the presence and absence of human protein S. APC-mediated factor Va inacti vation on these tumor cell lines was assessed by measuring the ability of residual cell-bound factor Va to augment the proteo lytic activation of prothrombin by factor Xa. Each of the tumor cell lines studied supported factor Va inactivation by APC in the presence of calcium ions. HepG2 cell monolayers supported this reaction most effectively, with CAPAN-2 and J82 cell monolayers exhibiting moderate and weak effective ness, respectively. Although not essential for this reaction, protein S moderately enhanced the rate of factor Va inactiva tion by APC on these tumor cell lines. In addition, pretreat ment of each tumor cell line with rabbit antihuman protein S IgG had little, if any, effect on its ability to support factor Va inactivation by APC. Our data suggest that these, and perhaps other, tumor cells can provide an appropriate phospholipid surface for promoting factor Va binding and rapid inactiva tion by APC.
Introduction
Protein C is a vitamin-K-dependent glyco protein that plays an important role in the reg ulation of coagulation. Protein C is synthe
Received: March 15,1991 Accepted in revised formbyProf.Hcmkcr: December 19, 1991
sized in the liver and secreted into the blood where it circulates as a precursor to a serine protease, activated protein C (APC) [1, 2]. Human protein C is converted to APC by pro teolytic cleavage of a dodecapeptide from the
Walter Kisiel, PhD Department of Pathology University of New Mexico School of Medicine Albuquerque, NM 87131 (USA)
© 1992 S. Karger AG, Basel 0301-0147/92/ 0224-0165S2.75/0
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
Tomohiro Nakagaki Tatsuya Sueyoshi Walter Kisiel
Methods Materials H-D-Phc-Pip-Arg-pNA (S-2238) was obtained from Helena Laboratories, Beaumont, Tex., USA. Tis sue culture flasks and plates (24 wells) were purchased from Corning. Minimum essential medium (Eagle), RPMI medium 1640, McCoy’s 5a medium, and nonessential amino acids were purchased from Mediatech. Minimum essential medium alpha, trypsin (1 X solu tion; tissue culture grade), penicillin-streptomycin (104 U penicillin and 10 mg streptomycin/ml) and bovine
166
scrum abumin were obtained from Sigma. Fetal bo vine serum was purchased from Hyclone Laboratories. Human brain thromboplastin was prepared according to Quick [20]. All other reagents were the highest qual ity available from commercial sources. Proteins All proteins were purified from human plasma. Protein C was purified to homogeneity by a combina tion of barium citrate adsorption and elution, DEAESepharose CL-6B chromatography, dextran sulfate agarose chromatography and preparative electropho resis essentially as described [2], Protein C was acti vated by incubation with the purified protein C activa tor (ACC-C) from Agkistrodon contorlrix contorlrix venom as described [21 ]. The final product was pure as assessed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis [22], Factors X and Xa were prepared as previously described [23], Human prothrombin was prepared according to Kisiel and Hanahan [24]. Hu man factor V was purified by a combination of proce dures involving barium citrate adsorption, 8-12% po lyethylene glycol 6000 precipitation. DEAE-Sepharose CL-6B column chromatography and Ultrogel AcA-22 column chromatography essentially as described by Suzuki ct al. [8], Conversion of factor V to factor Va was carried out by incubation of factor V with the puri fied factor V activator from Russell’s viper venom (RVV-V) at an enzyme-substrate weight ratio of 1:100 [25]. Protein S was purified to homogeneity as de scribed [26], Monospecific antibodies against human protein S were produced in rabbits and IgG purified from antisera by protein A-Sepharose (Pharmacia) af finity chromatography. Cell Culture Each cell line was cultured as described by Sakai et al. [19], Factor Va inactivation studies were performed on cell monolayers within 24 h after reaching confluency. Inactivation o f Factor Va by APC The medium from 2.0-cm2 wells with confluent tumor cells was aspirated, and the cells were washed with 0.5 ml ofHepes saline buffer [HSB; 10 mA/Hepes (pH 7.45)/137 mM NaCl/4 mM KC1, 11 mM glucose] supplemented with 10 mA/ EDTA followed by three washes of HSB (0.5 ml/wash). Incubation buffer [0.5 ml of HSB supplemented with 0.5 % bovine serum albumin and 5 mM CaCL (HSB+)] containing various concentrations of APC (0-3 |iM ) in the absence or presence of protein S (160 n M) was added to each well. Following a 60-min incubation at room temperature,
Nakagaki/Sueyoshi/Kisiel
Factor Va Inactivation on Tumor Cells
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
amino terminus of the heavy chain of the mol ecule by a complex of thrombin and thrombo modulin [3]. APC, in turn, regulates thrombin formation through the selective proteolytic degradation of factors Va and Villa [4-6]. Factor Va is a heterodimeric protein [7, 8] that serves as an essential nonenzymatic com ponent of the prothrombinase complex. As a cofactor, factor Va accelerates prothrombin activation by factor Xa on a membrane surface approximately 10,000-fold [9-13]. Thus, inac tivation of membrane-bound factor Va mark edly reduces the rate of thrombin generation. Earlier studies using either platelets [ 14-16] or bovine aortic endothelial cells [ 17] have shown that these cells effectively support factor Va inactivation by APC. Protein S enhances the rate of this reaction, although differences have been noted among studies with respect to the effectiveness of protein S in augmenting the reaction on cell surfaces [14-17], While a considerable amount of data has accumulated in recent years on the ability of tumor cells to support the assembly and func tional expression of the prothrombinase com plex [18, 19], little is known as to whether or not these cells support the protein C anticoag ulant pathway. In this paper, we report on the ability of three human tumor cell lines, of diverse histologic origin, to support the inacti vation of factor Va by APC in the presence or absence of protein S.
(AA4o5/min)* - (AA4o5/min)§
(AA405/min)* where (AA4os/min)* is the absorbance change at 405 nm in the absence of APC, and (AA4os/min)§ is the absorbance change at 405 nm in the presence of each concentration of offered APC. The ability of cell-bound APC to inactivate fluidphase factor Va in the presence and absence of protein S was also determined. In this system, cells were ini tially preincubated at room temperature with HSB+ containing varying concentrations of APC (0-3 \xM) in the absence and presence of protein S (160nA/) for 60 min followed by six washes with HSB\ Factor Va (2 pg/ml), dissolved in 0.5 ml of HSB\ was then of fered to the cells and incubated at room temperature with gentle mixing for 30 min. At this time, aliquots (10 pi) of the supernatant were removed and imme diately assayed in a one-stage factor V clotting assay [27] by mixing 100 pi of diluted sample, 100 pi of human thromboplastin, 100 pi of factor-V-deficient human plasma [28] and 100 pi of 25 mM CaCk Clot ting times were determined from the point of calcium addition, and factor Va activity interpolated from linear log-log standard curve relating factor Va activity and clotting time. In experiments designed to assess whether tumor cells contain functional cell surface protein S, tumor cells were first incubated for 60 min at 37 °C with either nonimmune rabbit IgG (500 pg/ml) or rabbit antihuman protein S IgG (500 pg/ml) followed by six washes prior to the addi tion of APC. Control experiments indicated that this concentration of antihuman protein S IgG was suffi cient to completely abrogate the rate-enhancing effect observed with exogenous protein S (160 nM).
Results
In initial experiments, we compared the inactivation of cell-bound and fluid-phase factor Va by HepG2 cell-bound APC. In those experiments designed to assess cell-bound factor Va inactivation, HepG2 monolayers were offered varying concentrations of APC and incubated in the presence of 5 xnM cal cium for 60 min at room temperature. Fol lowing several washes to remove unbound APC, factors Va and Xa were offered to the cells and incubated for an additional 30 min to allow for factor Va-Xa binding to the HepG2 cell surface. Following this 30-min incubation, the remaining cell-bound factor Va cofactor activity was quantitated through its ability to support the factor-Xa-dependent conversion of prothrombin to thrombin. The concentration of factor Va selected for these studies was 2pg/ml, which from previous studies produced maximal prothrombinase activity on several tumor cell lines including those cell lines employed in the present study [19]. The factor Xa concentration employed, 3 nM, was twice the Kp observed for factor Xa binding to J82 and HepG2 cells reported in an earlier study from this laboratory [29]. In those experiments designed to measure fluidphase factor Va inactivation by cell-bound APC, factor Xa was omitted and only factor Va (2 pg/ml) was offered to the APC-treated cells. Fluid-phase factor Va activation was then assessed in aliquots of the supernatants above the cells in a factor V clotting assay. As shown in figure 1, the inactivation of cellbound factor Va required a considerably lower concentration of APC than that re quired for the relatively slow inactivation of fluid-phase factor Va. In both systems, pro tein S exhibited a small, but measurable, en hancement of the APC-mediated activation of factor Va on and above the HepG2 monolayers. Similar results with respect to cell-
167
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
cells were washed six times with HSB+. Incubation buffer (0.5 ml of HSB+) containing factor Va (2 gg/ml) and factor Xa (3 nA/) was then added to the cell monolayers and incubated at room temperature with con stant gentle mixing for 30 min. At this time, cells were washed six times with HSB+, and residual cell-bound factor Va cofactor activity was assessed in an amidolytic assay as follows. HSB+ (0.5 ml) containing pro thrombin (2 \iM) was added to the washed monolayers and incubated at room temperature with gentle mixing for 10 min. Aliquots (10 pi) of the incubation mixture were removed and added to 790 pi of 50 m/V/Tris-HCl (pH 8.3) containing 150 mM NaCl, 5 mM EDTA and 0.1 mM S-2238. The absorbance at 405 nm was con tinuously recorded using a Beckman DU 65 spectro photometer. The relative rate of factor Va inactivation was expressed by the following empirical relationship:
Fig. 2. Calcium dependence of APC binding to HepG2 cells. Confluent FlepG2 monolayers were incu bated with either 80 nM APC (•) or 80 nM APC with 160 nM protein S (o) for 60 min at room temperature in the presence of 0-20 m.V/ CaCG. The cells were washed six times with HSB containing the respective calcium concentrations, and a mixture of factor Va (2 pg/ml) and factor Xa (3 n.V/). in IISB". was added and incubated for an additional 30 min. Residual fac tor Va cofactor activity was determined using the S2238 amidolvtic assay.
bound versus fluid-phase factor Va inactiva tion were observed on monolayers of J82 and CAPAN-2 tumor cell lines (data not shown). In addition, essentially identical results were obtained if factor Va was first incubated with the cells for 15 min prior to the addition of factor Xa and an additional 15-min incuba tion before the washing steps (data not shown). This latter finding suggests that on these tumor cells, factor Xa docs not appear to confer a significant protective effect with respect to factor Va inactivation by APC as noted earlier for human platelets [ 16].
The results of the above studies, carried out in the presence of 5 mM calcium, sug gested that the binding of the APC to the tumor cell occurred in a calcium-dependent manner. To confirm this calcium depen dence, HepG2 monolayers were incubated with a constant concentration of APC in the presence and absence of protein S for 60 min in the presence of 0-20 mM CaCfi. After washing the cells with Hepes saline buffer containing the calcium concentration used in the incubation, cells were offered factors VaXa in the presence of 5 mM calcium. Follow ing a 30-min incubation, cell-bound factor Va
168
Nakagaki/Sueyoshi/Kisiel
Factor Va Inactivation on Tumor Cells
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
Fig. 1. Inactivation of factor Va by APC on HepG2 cells. HepG2 monolayers were incubated with varying concentrations of APC (0-350 nM) in the presence (o) or absence (•) of protein S (160 mW) for 60 min at room temperature. Following six washes with HSB+, the cells were either offered a mixture of factor Va (2 pg/ml) and factor Xa (3 nM), or factor Va alone (2 ,ug/ml). After a 30-min incubation, residual factor Va activity in the supernatant was determined by a dotting assay (-----), or residual cell-bound factor Va (----- ) was determined with a S-2238 amidolvtic assay as described in Methods.
cr
Fig. 3. Inactivation of factor Va by APC bound to J82 cells (a), CAPAN-2 cells (b), and HepG2 cells (c). Varying concentrations of APC (0-3 \iM) were incu bated 60 min at room temperature in the absence (•) and presence (o) of 160 nM protein S followed by six washes of HSEP. A mixture of factor Va (2 gg/ml) and factor Xa (3 nM), in 0.5 ml HSB+, was then added to each monolayer, and incubated for an additional 30 min at room temperature. After six washes with HSB+, prothrombin (2 \\.M in HSB+) was added, and residual factor Va cofactor activity was measured us ing the S-2238 amidolytic assay.
169
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
cofactor activity remaining was then quanti tated according to its relative ability to sup port the factor-Xa-dependcnt activation of prothrombin. As shown in figure 2, the bind ing of APC, as reflected in its ability to inacti vate cell-bound factor Va, was dependent on calcium, and approximately 1 mM calcium was required for maximal binding and inacti vation of factor Va. Higher concentrations of calcium (2-20 mM) had no measurable effect on APC binding to the HcpG2 cells as mani fest in its ability to inactivate cell-bound fac tor Va (fig. 2). In the next series of experiments, we com pared the ability of three tumor cell lines of diverse histological origin to support APC binding and the APC-mcdiated inactivation of cell-bound Va. Each of the tumor cell line monolayers tested augmented the APC-mediated factor Va inactivation. Pretreatment of each tumor cell line with antihuman protein S IgG had no significant effect on the rate of APC-mediated inactivation of cell-bound fac tor Va, suggesting that these tumor cells do not contain functional cell surface protein S of tumor origin (data not shown). The order of effectiveness was HepG2 > CAPAN-2 > J82 in the presence and absence of protein S (fig. 3). Protein S exhibited a modest rateenhancing effect on the rate of factor Va inac tivation on all tumor cell lines (fig. 3). The concentrations of APC required for half-max imal inactivation of cell-bound factor Va were obtained from double reciprocal plots of the data presented in figure 3. In the absence of protein S, the concentrations of APC required for half-maximal inactivation of factor Va on HepG2, CAPAN-2 and J82 cell monolayers were 8.3, 30 and 136 nM, respectively. In the presence of 160 nM protein S, these values were reduced to 2.0, 17 and 66 nM, respec tively. It is perhaps appropriate to mention at this point that the above concentrations of APC may, in fact, overestimate the concen-
Fig. 4 . Inactivation of cellbound factor Va by APC. A mix ture of factor Va (2 |ig/ml) and fac tor Xa (3 nM) were incubated with HepG2 monolayers for 60 min at room temperature, and washed six times with HSB+. Varying concen tration of APC (0-25 nM) were offered to the cells with (o) or without (•) 160 nM protein S and incubated an additional 30 min at room temperature. After six washes with HSB+, prothrombin (2 \xM in HSB+) was added and residual factor Va cofactor activity measured using the S-2238 amidolytic assay. Inset: double reciprocal plot of the data.
170
~ 400 pM in the absence of protein S (fig. 4), and are to be compared to APC concentration of 2 and 8.3 nM, respectively, obtained on HepG2 monolayers in the preceding experi ments where the order of addition was APC followed by factor Va. A similar reduction of APC concentrations required for half-maxi mal inactivation of factor Va using this order of component addition was also observed on J82 and CAPAN-2 monolayers (data not shown).
Discussion
Previous studies have shown that several human tumor cell lines can readily assemble a functional cell surface prothrombinase com plex [18,19]. Factor Va is a key component of the prothrombinase complex and augments the factor-Xa-catalyzed activation of pro thrombin three orders of magnitude in the presence of calcium ions and an appropriate phospholipid membrane [14, 17], In the
Nakagaki/Sueyoshi/Kisiel
Factor Va Inactivation on Tumor Cells
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
trations of APC necessary to achieve halfmaximal inactivation of cell-bound factor Va on each cell line due to the potential mechani cal removal of APC by the cell-washing steps following the incubation period. To further characterize the APC-mediated inactivation of cell-bound factor Va, the order of offering APC and factor Va to the cells was reversed. In the preceding series of studies factor Va was offered to the cells previously treated with APC. In these experiments, fac tor Va and Xa were first incubated with tumor cells for 60 min, followed by six washes and the subsequent incubation with varying concentrations of APC in the presence and absence of protien S. The results of these stud ies, shown in figure 4, reveal that the concen trations of APC required for half-maximal inactivation of HepG2 cell-bound factor Va was dramatically reduced relative to the op posite order of component addition. The con centration of APC required for half-maximal inactivation of factor Va in these studies was ~ 50 pM in the presence of protein S and
ability to accelerate the formation of APC by offered thrombin in relation to the absence of cells [Nakagaki T., unpubl. data]. Whether or not other tumor cell lines constitutively syn thesize thrombomodulin, or whether pro teases in the culture medium degrade cell sur face thrombomodulin will require several additional studies. Conceivably, mechanisms which utilize the factor-Va-dependent acti vation of protein C by thrombin [30-32] are important in the formation of APC in the tumor microenvironment. Alternatively, plasma-derived APC may suffice to catalyze tumor-cell-bound factor Va inactivation and thereby modulate fibrin formation and depo sition. Immunochemical studies have shown that circulating blood contains less than 500 pM APC [33]. Based on our studies, which indicate that half-maximal rates of fac tor Va inactivation on HepG2 cells are ob served at 50 pM APC in the presence of pro tein S, it is not unlikely that these concentra tions of APC are present in the tumor mi croenvironment given the hyperpermeable nature of the tumor microvasculature [34].
Acknowledgments We wish to thank Nancy Basore for excellent tech nical assistance throughout these studies. This work was supported by research grants from the National Institutes of Health (HL35246) and Blood Systems. Inc. T.S. is a recipient of a grant from Toyobo Biotech nology Foundation.
171
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
present study, we demonstrate that three hu man tumor cell lines, HepG2, CAPAN-2 and J82, each support, to varying degrees, the inactivation of cell-bound factor Va by APC in the presence of factor Xa. The order of effectiveness in this regard was HcpG2 > CAPAN-2 > J82. In contrast to the results of studies using bovine platelets and bovine en dothelial cells [14, 15, 17], human protein S was not essential for the APC-mediated inac tivation of cell-bound factor Va. In our stud ies, protein S moderately increased the rate of factor Va inactivation by APC. Inasmuch as kinetic parameters for the inactivation of fac tor Va by APC were not obtained in this study, it is uncertain whether protein S in creased the kcat or decreased the Km of this reaction. The results of our studies demonstrate that tumor cells bind APC in a calcium-dependent manner that results in the expression of func tional cell surface proteolytic activity in the absence of protein S. Our experimental find ings using human APC and human tumor cell lines corroborate earlier studies by Solymoss et al. [16], who were the first investigators to report the protein-S-independent inactivation of factor Va by APC on phospholipid vesicles and human platelets. Whether or not human tumor cells phenotypically express a unique cell surface APC-binding protein is unknown and will require further studies. The physiological relevance of our findings with respect to APC-mediated factor Va inac tivation on tumor cells and its impact on fibrin investment of solid tumors is uncertain. In addition, whether or not tumor cells syn thesize and express cell surface thrombomo dulin that participates in the formation of APC in situ is unknown. The results of pre liminary studies carried out in our laboratory indicate that the three tumor cell lines exam ined in this study do not express cell surface thrombomodulin on the basis of the cells’
1 Stenflo JA: A new vitamin K-dependent protein: Purification from bo vine plasma and preliminary char acterization. J Biol Chem 1976:251: 355-363. 2 Kisiel W: Human plasma protein C: Isolation, characterization, and mechanism of activation by athrombin. J Clin Invest 1979:64: 761-769. 3 Esmon CT. Owen WG: Identifica tion of an endothelial cell cofactor for thrombin-catalyzed activation of protein C. Proc Natl Acad Sci USA 1981;78:2249-2252. 4 Kisiel W. Canfield WM. Ericsson LH, Davie EW: Anticoagulant prop erties of bovine plasma protein C following activation by thrombin. Biochemistry 1977;16:5824-5831. 5 Suzuki K, Stenflo J. Dahlback B. Teodorsson B: Inactivation of hu man coagulation factor V by acti vated protein C. J Biol Chcm 1983; 258:1914-1920. 6 Vehar GA, Davie EW: Preparation and properties of bovine factor VIII (antihemophilic factor). Biochemis try 1980:19:401-410. 7 Nesheim ME, Mann KG: Throm bin-catalyzed activation of single chain bovine factor V. J Biol Chem 1979;254:1326-1334. 8 Suzuki K, Dahlback B. Stenflo J: Thrombin-catalyzed activation of human coagulation factor V. J Biol Chem 1982;257:6556-6564. 9 Jobin F, Esnouf MP: Studies on the formation of the prothrombin-converting complex. Biochem J 1967; 102:666-674. 10 Esmon CT, Owen WG, Jackson CM: A plausible mechanism for pro thrombin activation by factor Xa, factor Va, phospholipid, and cal cium ions. J Biol Chem 1974;249: 8045-8047. 11 Nesheim ME, Taswell JB, Mann KG: The contribution of bovine fac tor V and factor Va to the activity of prothrombinase. J Biol Chem 1979; 254:10952-10962. 12 Rosing J, Tans G, Grovers-Riemslag JWP. Zwaal RFA, Hemker HC: The role of phospholipid and factor Va in the prothrombinase complex. J Biol Chem 1980;255:274-283.
13 Kane WH, Majerus PW: The inter action of human coagulation factor Va with platelets. J Biol Chem 1982; 257:3963-3969. 14 Harris KW, Esmon CT: Protein S is required for bovine platelets to sup port activated protein C binding and activity. J Biol Chem 1985;260: 2007-2010. 15 Suzuki K, Nishioka J, Matsuda M, Murayama H, Hashimoto S: Protein S is essential for the activated pro tein C-catalyzed inactivation of platelet-associated factor Va. J Bio chem (Tokyo) 1984:96:455-460. 16 Solymoss S. Tucker MM, Tracy PB: Kinetics of inactivation of mem brane-bound Va by activated pro tein C. J Biol Chem 1988;263: 14884-14890. 17 Stern DM, Nawroth PP. Harris K, Esmon CT: Cultured bovine aortic endothelial cells promote activated protein C - protein S - mediated inactivation of factor Va. J Biol Chem 1986:261:713-718. 18 Van DeWater L. Tracy PB, Aronson D. Mann KG, Dvorak HF: Tumor cell generation of thrombin via func tional prothrombinase assembly. Cancer Res 1985;45:5521-5525. 19 Sakai T, Noguchi M. Kisiel W: Tu mor cells augment factor Xa-catalyzed conversion of prothrombin to thrombin. Haemostasis 1990;20: 125-135. 20 Quick AJ: Hemorrhagic Disease and Thrombosis. Philadelphia, Lea & Febiger, 1966. 21 Kisiel W. Kondo S, Smith KJ, McMullen BA, Smith LF: Charac terization of a protein C activator from Agkistrodon contortrix contortrix venom. J Biol Chem 1987;262: 12607-12613. 22 Laemmli U: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;277:680-685. 23 Kondo S, Kisiel W: Evidence that plasma lipoproteins inhibit the fac tor VII tissue factor complex by a different mechanism than extrinsic pathway inhibitor. Blood 1987;70: 1947-1954.
24 Kisiel W, Hanahan DJ: Purification and characterization of human fac tor II. Biochem Biophys Acta 1973; 304:103-110. 25 Kisiel W, Canfield WM: Snake ve nom proteases that activate blood coagulation factor V. Methods Enzymol 1981;80:275-285. 26 Dahlback B: Purification of human vitamin K-dependent protein S and its proteolysis by thrombin. Bio chem J 1983:209:2007-2010. 27 Nesheim ME, Katzmann JA, Tracy PB, Mann KG: Factor V. Methods Enzymol 1981:80:249-274. 28 Bloom JW, Nesheim ME, Mann KG: A rapid technique for the prep aration of factor V-deficient plasma. Thromb Res 1979:15:595. 29 Sakai T. Kisiel W: Binding of hu man factors X and Xa to HepG2 and J82 human tumor cell lines: Evidence that factor Xa binds to tumor cells independent of factor Va. J Biol Chem 1990;265:91059113. 30 Salem HH, Broze GJ. Miletich JP. Majerus PW: Human coagulation factor VA is a cofactor for the acti vation of protein C. Proc Natl Acad Sci USA 1983:80:1584-1588. 31 Salem HH, Broze GJ. Miletich JP. Majerus PW: The light chain factor Va contains the activity of factor Va that accelerates protein C activation by thrombin. J Biol Chem 1983;258: 8531-8534. 32 Maruyama I, Salem HH, Majerus PW: Coagulation factor Va binds to human umbilical vein endothelial cells and accelerates protein C acti vation. J Clin Invest 1984;74:224— 230. 33 Bauer KA, Kass BL, Beeler DL. Ro senberg RD: Detection of human protein C activation in humans. J Clin Invest 1984;74:2033-2041. 34 Dvorak HF, Nagy JA, Dvorak JT, Dvorak AM: Identification and characterization of the blood vessels of solid tumors that are leaky to cir culating macromolecules. Am J Pa thol 1988;133:95-109.
172
Nakagaki/Sueyoshi/Kisiel
Factor Va Inactivation on Tumor Cells
Downloaded by: Nagoya University 133.6.82.173 - 1/15/2019 12:49:06 AM
References
