Portal de Periódicos da CAPES

Does activated protein C-resistant factor V contribute to thrombin generation in hemophilic plasma?

2005; Elsevier BV; Volume: 3; Issue: 3 Linguagem: Inglês

10.1111/j.1538-7836.2005.01181.x

1538-7933

Mettine H.A. Bos, Daniel W.E. Meijerman, Carmen van der Zwaan, Koen Mertens,

Hemostasis and retained surgical items

Journal of Thrombosis and HaemostasisVolume 3, Issue 3 p. 522-530 Free Access Does activated protein C-resistant factor V contribute to thrombin generation in hemophilic plasma? M. H. A. BOS, M. H. A. BOS The Department of Plasma Proteins, Sanquin Research at CLB, Amsterdam, the Netherlands; and The Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, the NetherlandsSearch for more papers by this authorD. W. E. MEIJERMAN, D. W. E. MEIJERMAN The Department of Plasma Proteins, Sanquin Research at CLB, Amsterdam, the Netherlands; and The Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, the NetherlandsSearch for more papers by this authorC. VAN DER ZWAAN, C. VAN DER ZWAAN The Department of Plasma Proteins, Sanquin Research at CLB, Amsterdam, the Netherlands; and The Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, the NetherlandsSearch for more papers by this authorK. MERTENS, K. MERTENS The Department of Plasma Proteins, Sanquin Research at CLB, Amsterdam, the Netherlands; and The Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, the NetherlandsSearch for more papers by this author M. H. A. BOS, M. H. A. BOS The Department of Plasma Proteins, Sanquin Research at CLB, Amsterdam, the Netherlands; and The Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, the NetherlandsSearch for more papers by this authorD. W. E. MEIJERMAN, D. W. E. MEIJERMAN The Department of Plasma Proteins, Sanquin Research at CLB, Amsterdam, the Netherlands; and The Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, the NetherlandsSearch for more papers by this authorC. VAN DER ZWAAN, C. VAN DER ZWAAN The Department of Plasma Proteins, Sanquin Research at CLB, Amsterdam, the Netherlands; and The Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, the NetherlandsSearch for more papers by this authorK. MERTENS, K. MERTENS The Department of Plasma Proteins, Sanquin Research at CLB, Amsterdam, the Netherlands; and The Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Utrecht University, Utrecht, the NetherlandsSearch for more papers by this author First published: 04 March 2005 https://doi.org/10.1111/j.1538-7836.2005.01181.xCitations: 20 Koen Mertens, Department of Plasma Proteins, Sanquin Research at CLB, Plesmanlaan 125, 1066 CX, Amsterdam, the Netherlands. Tel.: +31 20 5123151; fax: +31 20 5123680; e-mail: K.Mertens@sanquin.nl AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract Summary. In this study we assessed the role of factor V (FV) inactivation in hemophilic plasma with particular reference to the activated protein C (APC)-resistant variants FV-R506Q (FV Leiden) and FV-R306T (FV Cambridge). Purified recombinant full-length FV carrying these single substitutions and FV-R306T/R506Q were used in thrombin generation experiments. Plasma was first immunodepleted of FV, and subsequently of factors VIII, IX, or combinations thereof. Thrombin generation was initiated by low concentrations of recombinant tissue factor. Recombinant soluble thrombomodulin (TM) was used to trigger the APC system. Surprisingly, TM concentrations that reduced thrombin generation in normal plasma by no more than 50% virtually abolished thrombin formation in plasma deficient in the factor VIII/IX complex. This was already apparent at TM levels as low as 0.1 nmol L−1. By varying the concentrations of purified (activated) protein C to plasma that was additionally depleted of protein C, we confirmed that impaired thrombin generation indeed was the result of the action of APC. In contrast, this did not occur when FV-depleted plasma had been reconstituted with FV-R306T/R506Q. Addition of FV-R306T or FV-R506Q partially reduced prothrombin activation, demonstrating the involvement of both APC cleavage sites. FV inactivation also occurred on the surface of human microvascular endothelial cells. Apparently, these cells express sufficient TM to down-regulate thrombin production via the APC pathway. We further conclude that in hemophilic plasma this pathway can induce a secondary defect because of premature FV inactivation. It therefore seems conceivable that APC-resistant FV has the potential of alleviating hemophilic bleeding. Introduction Blood coagulation is regulated by the concerted action of several enzyme–cofactor complexes, which assemble on the membrane of a variety of cell types at the interface between the blood stream and the vasculature. While some of these complexes serve a procoagulant function, others play an anticoagulant role and as such constitute a delicately balanced regulatory system in which thrombin is the key regulatory enzyme (for review see Mann et al. [1]). Thrombin formation is initiated by the factor VIIa (FVIIa) and tissue factor (TF) complex (i.e. initiation phase) and propagated by the action of two enzyme–cofactor complexes (i.e. propagation phase). First, the complex of factor VIIIa (FVIIIa) and factor IXa (FIXa) activates factor X, followed by the prothrombinase complex comprising factor Xa (FXa) and factor Va (FVa) that converts prothrombin into thrombin. Apart from its multiple procoagulant functions, thrombin becomes anticoagulant once assembled with thrombomodulin (TM) on the endothelial cell surface. By activating protein C the thrombin-TM complex triggers the activated protein C (APC) system, which inactivates cofactors FVIIIa and FVa. Thus, the APC systems works in concert with protease inhibitors such as antithrombin in inactivating enzyme–cofactor complexes operating upstream in the thrombin generation cascade (i.e. termination phase). In maintaining the balance between pro- and anticoagulant pathways, FV plays a particular role in that it combines both pro- and anticoagulant properties in a single protein (for reviews see Nicolaes et al. [2] and Mann et al. [3]). Besides its cofactor role in prothrombinase conversion, FV also serves as APC cofactor in FVIIIa inactivation. Both functions of FV are regulated by limited proteolysis. Thrombin-dependent cleavage at Arg709, Arg1018 and Arg1545 activates FV, but abolishes APC cofactor activity at the same time. Conversely, APC-dependent FV cleavage at Arg506 promotes APC cofactor activity, but reduces FXa cofactor activity. Finally, FVa is rapidly cleaved by APC at Arg306, Arg506 and Arg679, thereby eliminating FVa activity. The importance of FVa inactivation in down-regulating thrombin formation is apparent from the notion that mutations in the APC cleavage sites Arg306 or Arg506 lead to APC resistance, which is the most common risk factor for venous thrombosis. Theoretically, excessive thrombin formation should have the potential of counterbalancing impaired thrombin formation in for instance FVIII or FIX deficiency (hemophilia A or B). This has been addressed by Nichols and co-workers, who obtained evidence that some patients have a less severe hemophilia A phenotype because of the simultaneous presence of the FV-R506Q mutation [4]. A similar observation was made by Lee et al. who found that patients carrying the FV-R506Q mutation had lower factor concentrate utilization and fewer bleeding episodes [5]. This is supported by the finding that the APC-resistant FV-R506Q variant is more potent than wild-type FV (wt-FV) in generating thrombin in reconstituted systems lacking FVIII [6]. Other data, however, are inconclusive in this respect. On the one hand, a sporadic case of severe hemophilia B has been described in which heterozygosity for FV-R506Q is associated with an unusually mild bleeding diathesis [7]. On the other hand, extensive studies were unable to link APC resistance with the hemophilic phenotype in vivo[8, 9]. The present study was performed to reappraise the hypothesis that APC resistance effectively counterbalances impaired thrombin formation. Recombinant FV mutants with substitutions of Arg306, Arg506, and the combination thereof were used in thrombin generation experiments in plasmas with various coagulation factor deficiencies. The protein C pathway was either triggered by the addition of recombinant soluble TM or by endogenous TM exposed on human microvascular endothelial cells. Our data suggest that hemophilic plasma displays a dual defect in thrombin generation: a primary deficiency in FVIII or FIX, and a secondary deficiency as a result of premature FV inactivation. Materials and methods Antibodies Monoclonal anti-FIX antibody CLB-FIX 14 has been described [10]. Monoclonal anti-FV antibody CLB-FV 2 was obtained employing purified human FV (Diagnostica Stago, Asnières, France) for immunization and hybridoma screening. Clones producing high-affinity anti-FV antibodies were selected and clone CLB-FV 2 was used for production in 850 cm2 roller bottles (Corning, NY, USA). The antibody was purified from medium employing Protein A–Sepharose (Amersham Biosciences, Roosendaal, the Netherlands) and coupled to CNBr–Sepharose 4B (Amersham Biosciences). Polyclonal anti-FV-immunoglobulin G (IgG) and its peroxidase-labeled equivalent were from Affinity Biologicals (Ancaster, Canada). Construction of recombinant FV From a human fetal liver total RNA library (Clontech Laboratories, Palo Alto, CA, USA), cDNA was obtained using SuperScript II reverse transcriptase (Invitrogen, Breda, the Netherlands). This FV cDNA was then amplified in nine overlapping fragments (FV1–FV9, Table 1). The ATG initiation site was modified according to Kozak [11]. At the 5′-end of each fragment a HindIII restriction site was introduced and a NotI restriction site at each 3′-end. All amplified FV fragments were purified, digested with HindIII/NotI, and ligated separately into the HindIII/NotI-opened pBluescript SK(+) vector (Stratagene, Amsterdam, the Netherlands). The pBluescript-FV constructs displayed the same sequence as reported by Jenny et al. [12] except in the codons for amino acids 1242, 1288 and 2185, which were identical to those described by Kane et al. [13]. The reported polymorphisms at codons 830, 837 and 897 were not present in our FV fragments, which contained the amino acids as described in the Swiss-Prot entry P12259 instead (K, H and K, respectively) (http://us.expasy.org/cgi-bin/niceprot.pl?P12259) and a silent mutation in codon 51 (A to G). The FV fragments were ligated in five successive steps. First, pBluescript-FV1 was digested by HindIII/Bsp1407I, and pBluescript-FV2 by Bsp1407I/PstI, and the two FV fragments were ligated into pBluescript resulting in pBluescript-FV1+2. Subsequently, pBluescript-FV1+2 (HindIII/PstI) was ligated to pBluescript-FV3 (PstI/KpnI) and pBluescript-FV4 (KpnI/BamHI) generating pBluescript-FV1+2+3+4. pBluescript-FV5+6 resulted from ligation of pBluescript-FV5 (BamHI/AflII) and pBluescript-FV6 (AflII/Bsp1407I). pBluescript-FV7 (Bsp1407I/VspI) was linked to pBluescript-FV8 (VspI/XhoI) and pBluescript-FV9 (XhoI/NotI), thus generating pBluescript-FV7+8+9. Finally, full-length FV cDNA was obtained by ligation of pBluescript-FV1+2+3+4 (HindIII/BamHI) to pBluescript-FV5+6 (BamHI/Bsp1407I) and pBluescript-FV7+8+9 (Bsp1407I/NotI). The full-length FV cDNA was ligated into a neomycin-resistant pcDNA3.1(+) expression vector (Invitrogen) after HindIII/NotI digestion. Primers used to introduce the Arg306→Thr substitution in fragment FV2 and Arg506→Gln in FV3 are shown in Table 1. A full-length construct encoding for FV-R306T or FV-R506Q was obtained employing pBluescript-FV2-R306T or pBluescript-FV3-R506Q for digestion and ligation, whereas the FV-R306T/R506Q construct was prepared by linkage of both modified FV2 and FV3 fragments. Table 1. Cloning strategy for the construction of FV variants FV fragment Nucleotide position * Primers † FV1 91–838 sense 5′-TTAAAGCTTACCACCATGTTCCCAGGCTGCCCACGCCTC-3′ antisense 5′-ATTGCGGCCGCGGGCACAAACTGTTATATCTGGCATTG-3′ FV2 758–1202 sense 5′-TTAAAGCTTGCTGGAGCCAGTCATCATCCCTA-3′ antisense 5′-ATTGCGGCCGCATATTCGCTGGTATTACAGGTGCATA-3′ FV3 1131–1816 sense 5′-TTAAAGCTTGGCGGCACATGAAGAGGTGGGAATACT-3′ antisense 5′-ATTGCGGCCGCCACGTTTCACCTCATCAGGATTTTCACA-3′ FV4 1717–2631 sense 5′-TTAAAGCTTGAACAGCAGGCTGTGTTTGCTGTG-3′ antisense 5′-ATTGCGGCCGCTATCCCTGTGACATCTGGCTGTAGA-3′ FV5 2587–3310 sense 5′-TTAAAGCTTGAAGACCCTATAGAGGATCCTCTACAGCCA-3′ antisense 5′-ATTGCGGCCGCTGGATTTATGAAGCACCAACGAATGCT-3′ FV6 3237–4894 sense 5′-TTAAAGCTTCCCTCTAAGAAGTGAAGCCTACAACACA-3′ antisense 5′-ATTGCGGCCGCCTGTTTCCCTTTGTACAAATTCTGAA-3′ FV7 4859–5559 sense 5′-TTAAAGCTTCCTGGGATTATTCAGAATTTGTACAAAGGG-3′ antisense 5′-ATTGCGGCCGCTTGCTCATACATTTTCAGGCCAGGCAA-3′ FV8 5451–6255 sense 5′-TTAAAGCTTGTCCCGAAGTTCTTGGAGACTCAC-3′ antisense 5′-ATTGCGGCCGCCGAAGGGTAGGTCTGTTATAGGCTCG-3′ FV9 6184–6884 sense 5′-TTAAAGCTTTTTGACCCACCTATTGTGGCTAGATA-3′ antisense 5′-ATTGCGGCCGCGAAAAGAAAGAGAAATAGTGGAAAACTG-3′ FV2-R306T 1074–1101 sense 5′-CTGCCCAAAGAAAACCACGAATCTTAAG-3′ antisense 5′-CTTAAGATTCGTGGTTTTCTTTGGGCAG-3′ FV3-R506Q 1681–1705 sense 5′-CTGGACAGGCAAGGAATACAGAGGG-3′ antisense 5′-CCCTCTGTATTCCTTGCCTGTCCAG-3′ * Nucleotide numbering according to Jenny et al. [12]. † Primers used to amplify the FV fragments and to introduce codon substitutions. Mismatched bases are underlined. Production of recombinant FV CHO-k1 cells (ATCC, Manassas, VA, USA) were grown in Dulbecco's modified Eagle's medium-F12 supplemented with 10% heat-inactivated fetal calf serum, 100 U mL−1 penicillin and 100 μg mL−1 streptomycin (all from BioWhittaker, Alkmaar, the Netherlands) at 37 °C in 5% CO2. Cells were transfected employing the calcium phosphate coprecipitation method with the pcDNA3.1(+) plasmid encoding wt-FV and FV variants. Following the selection of transfected cells with neomycin-containing medium (400 μg mL−1) (Invitrogen), neomycin-resistant clones were isolated and propagated using limited dilution. Production of FV was monitored by measuring FV antigen by enzyme-linked immunosorbent assay (ELISA) as described below (see 'Protein concentrations'). Stable cell lines producing appropriate amounts of antigen were selected and used for production in roller bottles. Medium supplemented with 0.4% fungizone (Invitrogen) and 2 μmol L−1 H-D-Phe-Pro-Arg-chloromethylketone (Bachem, Bubendorf, Switzerland) was harvested and 10 mmol L−1 benzamidine (Sigma, St Louis, MO, USA) was added. Medium was concentrated 10-fold employing a Hemoflow F5 hollow fiber cartridge (Fresenius, Bad Homburg, Germany) and stored at −20 °C. Purification of FV Medium containing approximately 50 Units FV was thawed at 37 °C and incubated overnight at 4 °C under stirring in a spinner-flask with 10 mg anti-FV antibody CLB-FV 2 immobilized to 4 mL Sepharose. The Sepharose was collected in a column (11 cm diameter) and washed with 50 mmol L−1 Tris–HCl, 100 mmol L−1 NaCl, 10 mmol L−1 benzamidine, 2 μmol L−1 H-D-Phe-Pro-Arg-chloromethylketone, 10% ethyleneglycol, pH 7.4. After washing steps using the same buffer containing 1 mol L−1 NaCl followed by 100 mmol L−1 NaCl, Sepharose was transferred to a 2.5-cm diameter column. FV was eluted with 25 mmol L−1 lysine, 150 mmol L−1 NaCl, 5 mmol L−1 CaCl2, 10 mmol L−1 benzamidine, 2 μmol L−1 H-D-Phe-Pro-Arg-chloromethylketone, and 50% ethyleneglycol, pH 10. Fractions were neutralized with 1.5 mol L−1 imidazole pH 7, and FV-containing fractions were pooled and concentrated employing Q-Sepharose FF (Amersham Biosciences) in 20 mmol L−1 Tris–HCl, 100 mmol L−1 NaCl, and 10% glycerol, pH 7.4, using 0.5 mol L−1 NaCl in the same buffer for elution. Pooled fractions were dialyzed against 50 mmol L−1 Tris–HCl, 100 mmol L−1 NaCl, 5 mmol L−1 CaCl2, 50% glycerol, pH 7.4, and stored at −20 °C. Plasma FV was obtained by the same procedure. Normal human plasma (Sanquin Plasma Products, Amsterdam, the Netherlands) containing approximately 2000 Units FV was supplemented with 10 mmol L−1 benzamidine, 0.1 mg mL−1 soya bean trypsin inhibitor (Sigma), 5.6 U mL−1 aprotinin (Trasylol®, Bayer, Leverkusen, Germany), 2.3 U mL−1 heparin (Leo pharmaceutical products B.V., Weesp, the Netherlands), 2 μmol L−1 H-D-Phe-Pro-Arg-chloromethylketone, and centrifuged for 30 min (4220 × g at 4 °C) prior to the addition of anti-FV Sepharose. Purified FV (1 μg) was subjected to sodium dodecyl sulphate–polyacrylamide gel electrophoresis on a 5% gel and visualized by silver staining. All FV species displayed a predominant band at 330 kDa, and a secondary band at 220 kDa. The same pattern was observed after immunoblotting using polyclonal anti-FV-IgG. As such, these FV preparations are similar to those described previously [14]. Recombinant wt-FV was similar to plasma-derived FV with respect to thrombin formation both in the absence and presence of APC (Enzyme Research Laboratories, South Bend, IN, USA) as assessed by measuring thrombin generation (see 'Thrombin generation in plasma'). Protein concentrations Protein was quantified by the Bradford method [15], using human serum albumin (Sanquin Plasma Products) as standard. FV antigen was quantified by ELISA. Polyclonal anti-FV-IgG was immobilized to microtiter plates (2.5 μg mL−1) overnight at 4 °C in 50 mmol L−1 NaHCO3, pH 9.8. Wells were washed with 50 mmol L−1 Tris–HCl, 150 mmol L−1 NaCl, 0.1% Tween-20, pH 7.4. Samples were diluted in 50 mmol L−1 Tris–HCl, 150 mm NaCl, 0.1% human serum albumin, 0.1% Tween-20, pH 7.4, and incubated for 2 h at 37 °C. Peroxidase-labeled anti-FV-IgG (2.5 μg/well) was added, and, after washing, peroxidase was detected using 3-3′-5-5′-tetramethylbenzidine. Concentrations were calculated from plots of log absorbance vs. log concentration. FV activity was assessed by measuring thrombin generation in FV-deficient plasma (Dade Behring, Liederbach, Germany) reconstituted with purified FV employing normal human plasma as reference (see 'Thrombin generation in plasma'). One unit of FV activity or antigen corresponds to the amount of FV in 1 mL of normal plasma (∼8 μg mL−1). Preparation of human plasmas depleted of various coagulation factors Plasmas were immunodepleted by chromatography employing appropriate antibodies coupled to CNBr–Sepharose 4B. FIX-depleted plasma, FV-depleted plasma, and FIX/FV-depleted plasma were prepared from normal human plasma applying anti-FIX monoclonal antibody CLB-FIX 14, anti-FV monoclonal antibody CLB-FV 2, or a combination thereof. Protein C/FIX/FV-depleted plasma was prepared from protein C-depleted plasma (Trinity Biotech, Co Wicklow, Ireland). FVIII-depleted plasma was from Trinity Biotech. Depletion of FIX, FV, protein C, or FVIII was verified employing FIX antigen, FV antigen, protein C activity, and FVIII activity measurements. FIX antigen was measured as described previously [10]. Protein C and FVIII activity were assayed using Coamatic® protein C and Coatest® FVIII (Chromogenix, Mölndal, Sweden). Residual levels of depleted coagulation factors were < 0.01 U mL−1. Thrombin generation in plasma Plasma (40 μL) was mixed with 120 μL of 50 mmol L−1 Tris–HCl, 100 mmol L−1 NaCl, 0.1% bovine serum albumin (Sigma), pH 7.3 containing FV and when applicable recombinant human soluble TM (American Diagnostics, Greenwich, CT, USA). Then 160 μL of phospholipid reagent (cephalin, Roche, Almere, the Netherlands) was added and after incubation for 5 min at 37 °C thrombin generation was initiated by adding a similar buffer volume containing recombinant TF (Innovin®, Dade Behring) and CaCl2. Final concentrations were 0.08 U mL−1 FV, 0.1 nmol L−1 TM (unless otherwise stated), 0.10 pmol L−1 TF (1 : 60 000 diluted Innovin®[16]), and 8 mmol L−1 CaCl2. Samples of 25 μL were added to microtiter wells containing 3 μL of 250 mmol L−1 EDTA and the volume was adjusted to 140 μL with 50 mmol L−1 Tris–HCl, 100 mmol L−1 NaCl, 0.1% bovine serum albumin, pH 7.3. Subsequently, 25 μL of 2 mmol L−1 S-2238 (Chromogenix) was added and absorbance was monitored at 405 nm. Thrombin generation curves were calibrated employing active-site titrated thrombin [17]. Thrombin generation curves are described by several parameters [18, 19]. Lag phase and peak time illustrate the time necessary for thrombin generation to start and reach its maximum level, respectively, and IIamax is the concentration of thrombin generated at peak time. The area under the curve represents the total amount of thrombin formed without correcting for end levels of alpha-2-macroglobulin–thrombin complexes; therefore, this parameter was calculated from the initial 20 min of thrombin generation, unless otherwise stated. IIamax and area under the curve are presented as means ± SD of three individual experiments in the text and as means ± SEM of three individual experiments in figures and tables. Thrombin generation on endothelial cells Human microvascular endothelial cells (HMEC-1) [20] were seeded on 24-well dishes coated with fibronectin (Sanquin Plasma Products) in medium 199 (Invitrogen) supplemented with 10% heat-inactivated fetal calf serum, 10% human serum (pool of 24 healthy donors, Sanquin Plasma Products), 5 U mL−1 heparin, 100 U mL−1 penicillin, 100 μg mL−1 streptomycin, 0.03% glutamin (Merck, Darmstadt, Germany), 0.004% fibroblast growth factor (Amersham Biosciences), and 1% Ultroser-G (Invitrogen) at 37 °C in 5% CO2. At 100% confluence cells were washed with 50 mmol L−1 Tris–HCl, 100 mmol L−1 NaCl, 0.1% bovine serum albumin, pH 7.3 and thrombin generation was assessed in a final incubation volume of 480 μL per well at 37 °C as described above. Results Thrombin generation in plasma depleted of various coagulation factors Thrombin generation was assessed in diluted plasma to allow for sampling both before and after clotting had occurred. In preliminary experiments we varied the TF concentration from 20 pmol L−1 to 0.02 pmol L−1 to compare thrombin formation in normal and hemophilic plasma. Thrombin was plotted against time and the area under the thrombin generation curve was calculated from the initial 15 min of thrombin generation. At high TF concentrations we observed little, if any, difference in area under the curve for normal and hemophilic plasma (0.32 μmol L−1 min−1 and 0.37 μmol L−1 min−1, respectively, employing 4 pmol L−1 TF). At lower TF concentrations thrombin formation depends, at least partially, on FVIII and FIX [1, 21]. At a 40-fold lower TF concentration (0.10 pmol L−1) FIX-deficiency indeed became apparent as a reduction in the amount of thrombin formed. The area under the curve decreased from 0.24 µmol L−1 min in normal plasma to 0.15 µmol L−1 min in FIX-deficient plasma. Therefore, thrombin generation was initiated with 0.10 pmol L−1 TF in further experiments addressing thrombin generation in various deficient plasmas. Figure 1 shows that FVIII- or FIX-deficiency prolonged both lag phase and peak time. For normal plasma, peak time was observed at 7.5 min of incubation, whereas for both FVIII- and FIX-deficiency the peak time was prolonged to 15 min. Furthermore, we observed a reduction in area under the curve from 0.43 ± 0.04 μmol L−1 min in normal plasma to 0.30 ± 0.01 μmol L−1 min in FVIII-deficient plasma, and 0.28 ± 0.01 μmol L−1 min in FIX-deficient plasma. In plasma deficient in FV or both FV and FIX the lag phase was prolonged to 15 min, however, a substantial amount of thrombin was still generated in FV-deficient plasma, as indicated by a IIamax of 19 nmol L−1 at 30 min of thrombin formation. The largely extended lag phase in FV-deficient plasma reflects the key role of FV as cofactor in the prothrombinase complex, while the somewhat prolonged lag phase in FVIII- and FIX-deficiency illustrates that under these conditions thrombin formation depends only partially on the action of the FVIII/FIX complex. In conclusion, at the TF concentration used here, deficiencies in FVIII or FIX indeed displayed defective thrombin generation, whereas thrombin generation was lacking in additional FV deficiency. Figure 1Open in figure viewerPowerPoint Thrombin generation in plasma depleted of various coagulation factors. Thrombin generation was assessed in normal human plasma (○), FVIII-depleted plasma (•), FIX-depleted plasma (▵), FV-depleted plasma (▴), or FIX/FV-depleted plasma (□) as described in the 'Materials and methods'. Mean values ± SEM of three independent experiments are presented. Effect of the protein C pathway on thrombin generation To address the anticoagulant effect of the protein C pathway on thrombin generation, recombinant soluble TM was included in various deficient plasma systems. Preliminary experiments showed that TM concentrations as low as 10 pmol L−1 were sufficient to affect thrombin generation. As expected, TM had no effect on the lag phase of thrombin generation in either normal or FIX-depleted plasma (Fig. 2). In contrast, TM did decrease IIamax and thereby the area under the thrombin generation curves in both plasmas. In normal plasma the area under the curve was reduced from 0.37 ± 0.05 μmol L−1 min to 0.18 ± 0.01 μmol L−1 min upon TM addition, resulting in a 50% reduction of thrombin formation. In FIX-depleted plasma, however, a similar amount of TM reduced thrombin generation by 80%, from 0.31 ± 0.02 μmol L−1 min to 0.06 ± 0.01 μmol L−1 min, thereby abolishing thrombin formation almost completely. The extent of this effect in FIX-deficient plasma suggests almost complete inactivation of FV, which is the only relevant APC substrate in this system. Therefore these data imply that under these conditions hemophilic plasma displays an acquired, APC-induced FV deficiency as well. Figure 2Open in figure viewerPowerPoint Effect of the protein C pathway on thrombin generation. Normal human plasma (○, •) or FIX-depleted plasma (▵, ▴) was incubated in the absence (closed symbols) or presence (open symbols) of 0.1 nmol L−1 TM. Thrombin generation was initiated and monitored as described in the 'Materials and methods'. Data represent mean values ± SEM of three independent experiments. Effect of APC-resistant FV on thrombin generation We further monitored thrombin generation in the presence of normal and APC-resistant FV. For this purpose, a FV variant was constructed with substitutions at the two major APC cleavage sites Arg306 and Arg506. Purified FV-R306T/R506Q or wt-FV was added to FIX-deficient plasma that had also been depleted of endogenous FV. Subsequent addition of TM did not effect the lag phase or peak time of thrombin formation. However, IIamax and the area under the curve (Fig. 3A) were reduced in a TM-dependent manner with wt-FV added, from 37 ± 2 nmol L−1 and 0.32 ± 0.05 μmol L−1 min in the absence of TM to 6 ± 2 μmol L−1 and 0.07 ± 0.01 μmol L−1 min at 0.25 nmol L−1 TM, respectively. In contrast, FV-R306T/R506Q did not display any TM-induced reduction of prothrombin activation, indicated by a IIamax of 35 ± 2 nmol L−1 and area under the curve of 0.33 ± 0.06 μmol L−1 min in the absence of TM vs. 34 ± 1 nmol L−1 and 0.29 ± 0.02 μmol L−1 min at 0.25 nmol L−1 TM. In control experiments we verified if the reduction in thrombin was indeed the result of protein C action. Therefore, we additionally depleted the plasma of protein C and added back purified protein C or APC in increasing concentrations. Varying protein C within its physiological range resulted in a dose-dependent down-regulation of thrombin generation, as is shown in Fig. 3(B) (from 0.33 ± 0.08 μmol L−1 min in the absence of protein C to 0.06 ± 0.02 μmol L−1 min with 3.30 nmol L−1 protein C added). Furthermore, the addition of APC instead of protein C also reduced thrombin formation, from 0.21 μmol L−1 min in the absence of APC to 0.06 μmol L−1 min at 33 pmol L−1 APC (data not shown). Thus, protein C down-regulates thrombin generation at physiological concentrations. Moreover, reduced prothrombin activation directly results from FV inactivation. These data support our previous conclusion (Fig. 2) that the lack of thrombin generation in FIX-deficient plasma also reflects the inactivation of FV. Figure 3Open in figure viewerPowerPoint Effect of TM and protein C on thrombin generation in FIX-depleted plasma. (A) FIX/FV-depleted plasma reconstituted with wt-FV (open symbols) or FV-R306T/R506Q (closed symbols) was incubated with TM (0–0.25 nmol L−1). (B) Protein C/FIX/FV-depleted plasma reconstituted with wt-FV was incubated with 0.1 nm TM and 0 nmol L−1 protein C (•), 0.30 nmol L−1 protein C (▪), 0.80 nmol L−1 protein C (▴), or 3.30 nmol L−1 protein C (◆). Thrombin generation was assessed as described in the 'Materials and methods'. Mean values ± SEM of three individual experiments are presented. Effect of individual FV substitutions at Arg306 and Arg506 on thrombin generation It has generally been established that FV variants carrying the Arg306→Thr or Arg506→Gln substitution are, at least partially, APC-resistant [2]. To determine how FV inactivation at a single site influences thrombin generation we employed purified recombinant FV-R306T a