Binding of coagulation factors IX and X to the endothelial cell surface

Emicizumab promotes factor Xa generation on endothelial cells

BACKGROUND

Until recently, the treatment of hemophilia A relied on factor (F)VIII replacement. However, up to one-third of patients with severe hemophilia A develop neutralizing alloantibodies that render replacement therapies ineffective. The development of emicizumab, a bispecific antibody that partially mimics FVIIIa, has revolutionized the treatment of these patients. However, the use of an activated prothrombin complex concentrate [FEIBA (Takeda)] to treat breakthrough bleeding in patients on emicizumab has been associated with thrombotic complications including a unique microangiopathy.

OBJECTIVES

We hypothesized that the thrombotic complications observed with the combination of emicizumab and FEIBA might be due to excessive expression of procoagulant activity on the surface of endothelial cells.

METHODS

We examined the ability of emicizumab to promote FX activation on endothelial cells using 2 cell culture models.

RESULTS

We found that endothelial cells readily support emicizumab-mediated activation of FX by FIXa. The level of FXa generation depends on the concentration of available FIXa. The addition of FEIBA to emicizumab increased FXa generation in a dose-dependent manner on endothelial cells in both models. The rate of FXa generation was further enhanced by endothelial cell activation. However, unlike emicizumab, we found limited FXa generation in the presence of FVIII(a), which followed a significant lag time and was not dependent on FIXa concentration under these conditions.

CONCLUSION

Emicizumab promotes FXa generation on the surface of endothelial cells, which is markedly enhanced in the presence of FEIBA. These findings demonstrate a potential mechanism for the thrombotic complications seen with the combined use of emicizumab and FEIBA.

Unresolved hemostasis issues in haemophilia

Despite rapid technological advancement in factor and nonfactor products in the prevention and treatment of bleeding in haemophilia patients, it is imperative that we acknowledge gaps in our understanding of how hemostasis is achieved. The authors will briefly review three unresolved issues in persons with haemophilia (PwH) focusing on the forgotten function that red blood cells play in hemostasis, the critical role of extravascular (outside circulation) FIX in hemostasis in the context of unmodified and extended half-life FIX products and finally on the role that skeletal muscle myosin plays in prothrombinase assembly and subsequent thrombin generation that could mitigate breakthrough muscle hematomas.

Biodistribution of recombinant factor IX, extended half-life recombinant factor IX Fc fusion protein, and glycoPEGylated recombinant factor IX in hemophilia B mice

Extended half-life recombinant FIX (rFIX) molecules have been generated to reduce the dosing burden and increase the protection of patients with hemophilia B. Clinical pharmacology studies with recombinant factor IX Fc fusion protein (rFIXFc) report a similar initial peak plasma recovery to that of rFIX, but with a larger volume of distribution. Although the pegylation of N9-GP results in a larger plasma recovery, there is a smaller volume of distribution, suggesting less extravasation of the latter drug. In this study, we set out to compare the biodistribution and tissue localization of rFIX, rFIXFc, and glycoPEGylated rFIX in a hemophilia B mouse model. Radiolabeled rFIX, rFIXFc, and rFIX-GP were employed in in vivo single-photon emission computed tomography imaging (SPECT/CT), microautoradiography (MARG), and histology to assess the distribution of FIX reagents over time. Immediately following injection, vascularized tissues demonstrated intense signal irrespective of FIX reagent. rFIX and rFIXFc were retained in joint and muscle areas through 5 half-lives, unlike rFIX-GP (assessed by SPECT). MARG and immunohistochemistry showed FIX agents localized at blood vessels among tissues, including liver, spleen, and kidney. Microautoradiographs, as well as fluorescent-labeled images of knee joint areas, demonstrated retention over time of FIX signal at the trabecular area of bone. Data indicate that rFIXFc is similar to rFIX in that it distributes outside the plasma compartment and is retained in certain tissues over time, while also retained at higher plasma levels. Overall, data suggest that Fc fusion does not impede the extravascular distribution of FIX.

Modulation of Extravascular Binding of Recombinant Factor IX Impacts the Duration of Efficacy in Mouse Models

BACKGROUND

 There is an emerging concept that in addition to circulating coagulation factor IX (FIX), extravascular FIX contributes to hemostasis.

OBJECTIVE

 Our objective was to evaluate the efficacy of extravascular FIX using animal models of tail clip bleeding and ferric chloride-induced thrombosis.

METHODS

 Mutant rFIX proteins with described enhanced (rFIXK5R) or reduced (rFIXK5A) binding to extracellular matrix were generated and characterized using in vitro aPTT, one-stage clotting, and modified FX assays. Using hemophilia B mice, pharmacokinetic (PK) parameters and in vivo efficacy of these proteins were compared against rFIX wild-type protein (rFIXWT) in a tail clip bleeding and FeCl3-induced thrombosis model. Respective tissue disposition of FIX was evaluated using immunofluorescence.

RESULTS

 In vitro characterization demonstrated comparable clotting activity of rFIX proteins. The PK profile showed that rFIXK5A displayed the highest plasma exposure compared to rFIXWT and rFIXK5R. Immunofluorescence evaluation of liver tissue showed that rFIXK5R was detectable up to 24 hours, whereas rFIXWT and rFIXK5A were detectable only up to 15 minutes. In the tail clip bleeding model, rFIXK5R displayed significant hemostatic protection against bleeding incidence for up to 72 hours postintravenous administration of 50 IU/kg, whereas the efficacy of rFIXK5A was already reduced at 24 hours. Similarly, in the mesenteric artery thrombus model, rFIXK5R and rFIXWT demonstrated prolonged efficacy compared to rFIXK5A.

CONCLUSION

 Using two different in vivo models of hemostasis and thrombosis, we demonstrate that mutated rFIX protein with enhanced binding (rFIXK5R) to extravascular space confers prolonged hemostatic efficacy in vivo despite lower plasma exposure, whereas rFIXK5A rapidly lost its efficacy despite higher plasma exposure.

The Function of extravascular coagulation factor IX in haemostasis

INTRODUCTION

The majority of clotting factor IX (FIX) resides extravascularly, in the subendothelial basement membrane, where it is important for haemostasis.

AIM

We summarize preclinical studies demonstrating extravascular FIX and its role in haemostasis and discuss clinical observations supporting this. We compare the in vivo binding of BeneFIX^® and the extended half-life FIX, Alprolix^® , to extravascular type IV collagen (Col4).

METHODS

Three mouse models of haemophilia were used: the FIX knockout as the CRM^- model and two knock-in mice, representing a CRM⁺ model of a commonly occurring patient mutation (FIX_R333Q ) or a mutation that binds poorly to Col4 (FIX_K5A ). The murine saphenous vein bleeding model was used to assess haemostatic competency. Clinical publications were reviewed for relevance to extravascular FIX.

RESULTS

CRM status affects recovery and prophylactic efficacy. Prophylactic protection decreases ~5X faster in CRM⁺ animals. Extravascular haemostasis can explain unexpected breakthrough bleeding in patients treated with some EHL-FIX therapeutics. In mice, both Alprolix^® and BeneFIX^® bind Col4 with similar affinities (Kd~20-40 nM) and show dose-dependent recoveries. As expected, the concentration of binding sites in the mouse calculated for Alprolix^® (574 nM) was greater than for BeneFIX^® (405 nM), due to Alprolix^® binding to both Col4 and the endothelial cell neonatal Fc receptor.

CONCLUSION

Preclinical and clinical results support the interpretation that FIX plays a role in haemostasis from its extravascular location. We believe that knowing the CRM status of haemophilia B patients is important for optimizing prophylactic dosing with less trial and error, thereby decreasing clinical morbidity.

Setting the stage for individualized therapy in hemophilia: What role can pharmacokinetics play?

Replacement therapy with clotting factor concentrates (CFC) is the mainstay of treatment in hemophilia. Its widespread application has led to a dramatic decrease in morbidity and mortality in patients, with concomitant improvement of quality of life. However, dosing is challenging and costs are high. This review discusses benefits and limitations of pharmacokinetic (PK)-guided dosing of replacement therapy as an alternative for current dosing regimens. Dosing of CFC is now primarily based on body weight and based on its in vivo recovery (IVR). Benefits of PK-guided dosing include individualization of treatment with better targeting, more flexible blood sampling, increased insight into association of coagulation factor levels and bleeding, and potential overall lowering of overall costs. Limitations include a slight burden for the patient, and availability of closely collaborating, experienced clinical pharmacologists.

Emerging drugs for the treatment of hemophilia A and B

INTRODUCTION

Replacement therapy with clotting factor concentrates is the most appropriate and effective way to treat bleedings of Hemophilia A&B to prevent chronic arthropathy. Unfortunately, the short half-life (HL) of FVIII/IX concentrates obliges the patients to receive frequent infusions, a big concern for children. The development of inhibitors in about 30-45% of hemophilia A and in 3-5% of hemophilia B patient is the major adverse event of replacement therapy.

AREAS COVERED

In the last few years, new rFIX have been developed with HL. New rFVIII concentrates are displaying small increase of PK characteristics. The new bio-engineering methods allowed the production of molecules fused with Fc fragment of IgG or Albumin or linked to PEG. A new approach to improve hemostasis is represented by Mab against TFPI and small RNA interfering with Antithrombin synthesis. Another innovative drug seems to be the new bi-specific antibody which mimics FVIII function in linking FXa and FX to tenase production.

EXPERT OPINION

The emerging drugs for hemophilia treatment seem to be very promising. The extended half-life will improve the adherence of patients to therapy. Accurate post-marketing surveillance studies will be necessary to check the efficacy, safety and immunogenicity of these new molecules.

Protein replacement therapy and gene transfer in canine models of hemophilia A, hemophilia B, von willebrand disease, and factor VII deficiency

Dogs with hemophilia A, hemophilia B, von Willebrand disease (VWD), and factor VII deficiency faithfully recapitulate the severe bleeding phenotype that occurs in humans with these disorders. The first rational approach to diagnosing these bleeding disorders became possible with the development of reliable assays in the 1940s through research that used these dogs. For the next 60 years, treatment consisted of replacement of the associated missing or dysfunctional protein, first with plasma-derived products and subsequently with recombinant products. Research has consistently shown that replacement products that are safe and efficacious in these dogs prove to be safe and efficacious in humans. But these highly effective products require repeated administration and are limited in supply and expensive; in addition, plasma-derived products have transmitted bloodborne pathogens. Recombinant proteins have all but eliminated inadvertent transmission of bloodborne pathogens, but the other limitations persist. Thus, gene therapy is an attractive alternative strategy in these monogenic disorders and has been actively pursued since the early 1990s. To date, several modalities of gene transfer in canine hemophilia have proven to be safe, produced easily detectable levels of transgene products in plasma that have persisted for years in association with reduced bleeding, and correctly predicted the vector dose required in a human hemophilia B liver-based trial. Very recently, however, researchers have identified an immune response to adeno-associated viral gene transfer vector capsid proteins in a human liver-based trial that was not present in preclinical testing in rodents, dogs, or nonhuman primates. This article provides a review of the strengths and limitations of canine hemophilia, VWD, and factor VII deficiency models and of their historical and current role in the development of improved therapy for humans with these inherited bleeding disorders.

Persistent factor VIII-dependent factor X activation on endothelial cells is independent of von Willebrand factor

Endothelial cells are able to support the activation of coagulation factor X by activated factor IX in the presence of its cofactor, factor VIII. We have previously reported that this reaction is persistent on endothelial cells, but transient on activated platelets and phospholipid vesicles when activated factor X (Xa) is used as activator of factor VIII. Aim of the present study was to explore the influence of von Willebrand factor and that of the factor VIII activator, either factor Xa or thrombin, on the decay of factor X activation on the endothelial cell surface. Kinetics of factor X activation on human umbilical vein endothelial cells was compared with that on phospholipid vesicles employing purified coagulation factors from plasma as well as recombinant factor VIII variants. Employing factor Xa as factor VIII activator, rate constants for decay of membrane-bound factor X activation were consistently low on endothelial cells (0.02 min) as compared with phospholipid vesicles (0.2 min). Activation of factor VIII by thrombin resulted in two-fold increased decay rates. In the presence of excess of von Willebrand factor over factor VIII, decay rates were not significantly changed. Factor VIII variants with and without a Tyr to Phe substitution, which abolishes high-affinity binding to von Willebrand factor, displayed the same factor X activation decay kinetics. Although previous studies have shown that von Willebrand factor modulates factor VIII activation and stabilisation, this apparently does not affect the progression of factor X activation at the endothelium.

Pharmacokinetic Study of Recombinant Human Factor IX in Previously Treated Patients with Hemophilia B in Taiwan

BACKGROUND/PURPOSE

The pharmacokinetic (PK) study of recombinant human factor IX (rFIX) has been done in patients with hemophilia B in areas other than Taiwan. However, wide patient-related variability in recovery is noted and there is no PK study of rFIX in Taiwanese patients with hemophilia B. The purpose of this study is to evaluate the PKs of the rFIX in Taiwanese patients with hemophilia B.

METHODS

A PK study of rFIX was performed in 10 previously treated Taiwanese patients with hemophilia B. Nine of them had severe hemophilia B and the other one had moderately severe hemophilia B. The mean age of our patients was 24.7 +/- 8.6 years (mean +/- SD) with a range of 15.5-47.0 years.

RESULTS

The infusion of 75 IU/kg of rFIX in our patients resulted in a mean FIX activity increase of 1.08 +/- 0.27 IU/dL per IU/kg with a range of 0.3-1.49 IU/dL per IU/kg, and a mean in vivo recovery of 44.0 +/- 9.5% with a range of 31.4-60.4%. The mean elimination half-life was 24.4 +/- 6.4 hours with a range of 14.1-35.3 hours. The mean values of area under the curve, total body clearance, volume of distribution at steady state were 1363 +/- 166 IU x hour/dL, 4.84 +/- 1.03 mL/hour/kg, and 144.3 +/- 41.8 mL/kg, respectively. In addition, rFIX was well tolerated in our patients. Only two patients experienced mild treatment-related adverse events, including fever and feeling sleepy, respectively. Both of them resolved spontaneously without any sequences, and there was no thrombosis formation or inhibitor development in the following 6 months after rFIX infusion.

CONCLUSION

The results of PK study of rFIX in our previously treated Taiwanese patients with hemophilia B are comparable to the reported results from other related studies conducted in different geographical areas.

Reduced bleeding events with subcutaneous administration of recombinant human factor IX in immune-tolerant hemophilia B dogs

Intravenous administration of recombinant human factor IX (rhFIX) acutely corrects the coagulopathy in hemophilia B dogs. To date, 20 of 20 dogs developed inhibitory antibodies to the xenoprotein, making it impossible to determine if new human FIX products, formulations, or methods of chronic administration can reduce bleeding frequency. Our goal was to determine whether hemophilia B dogs rendered tolerant to rhFIX would have reduced bleeding episodes while on sustained prophylactic rhFIX administered subcutaneously. Reproducible methods were developed for inducing tolerance to rhFIX in this strain of hemophilia B dogs, resulting in a significant reduction in the development of inhibitors relative to historical controls (5 of 12 versus 20 or 20, P <.001). The 7 of 12 tolerized hemophilia B dogs exhibited shortened whole blood clotting times (WBCTs), sustained detectable FIX antigen, undetectable Bethesda inhibitors, transient or no detectable antihuman FIX antibody titers by enzyme-linked immunosorbent assay (ELISA), and normal clearance of infused rhFIX. Tolerized hemophilia B dogs had 69% reduction in bleeding frequency in year 1 compared with nontolerized hemophilia B dogs (P =.0007). If proven safe in human clinical trials, subcutaneous rhFIX may provide an alternate approach to prophylactic therapy in selected patients with hemophilia B.

Circulating and binding characteristics of wild-type factor IX and certain Gla domain mutants in vivo

Residue K5 in factor IX gamma-carboxyglutamic acid (Gla) domain participates in binding endothelial cells/collagen IV. We injected recombinant factor IX containing mutations at residue 5 (K5A, K5R) into factor IX-deficient mice and compared their behavior with that of wild-type factor IX. The plasma concentration of factor IX that binds to endothelial cells/collagen IV (recombinant wild type and K5R) was consistently lower than that of the one that does not bind (K5A). Mice treated with wild type or K5R had 79% of the injected factor IX in the liver after 2 minutes, whereas 17% remained in circulation. In mice injected with K5A, 59% of the injected factor IX was found in liver and 31% was found in plasma. When we blocked the liver circulation before factor IX injection, 74% of K5A and 64% of K5R remained in the blood. When we treated the mouse with EDTA after injecting exogenous factor IX, the blood levels of factor IX that bind to endothelial cells/collagen IV increased, presumably because of release from endothelial cell/collagen IV binding sites. In contrast, the levels of the mutants that do not bind were unaffected by EDTA. In immunohistochemical studies, factor IX appears on the endothelial surfaces of mouse arteries after factor IX injection and of human arteries from surgical specimens. Thus, we have demonstrated that factor IX binds in vivo to endothelial cell-collagen IV surfaces. Our results suggest that factor IX Gla-domain mediated binding to endothelial cells/collagen IV plays a role in controlling factor IX concentration in the blood.

Pharmacokinetics of coagulation factors: clinical relevance for patients with haemophilia

Haemophilia is a recessively inherited coagulation disorder, in which an X-chromosome mutation causes a deficiency of either coagulation factor VIII (FVIII) in haemophilia A, or factor IX (FIX) in haemophilia B. Intravenous administration of FVIII or FIX can be used to control a bleeding episode, to provide haemostasis during surgery or for long term prophylaxis of bleeding. In special cases, activated factor VII (FVIIa) may be used instead of FVIII or FIX. The aim of this work is to review the pharmacokinetics of FVIII, FIX and FVIIa and to give an outline of the use of pharmacokinetics to optimise the treatment of patients with haemophilia. The pharmacokinetics of FVIII are well characterised. The systemic clearance (CL) of FVIII is largely determined by the plasma level of von Willebrand factor (vWF), which protects FVIII from degradation. Typical average CL in patients with normal vWF levels is 3 ml/h/kg, with an apparent volume of distribution at steady state (Vss) that slightly exceeds the plasma volume of the patient, and the average elimination half-life (t1/2) is around 14 hours. There are still some discrepancies in the literature on the pharmacokinetics of FIX. The average CL of plasma-derived FIX seems to be 4 ml/h/kg, the Vss is 3 to 4 times the plasma volume and the elimination t1/2 often exceeds 30 hours. FVIIa has a much higher CL (average of 33 ml/h/kg), and a short terminal t1/2 (at 2 to 3 hours). The Vss is 2 to 3 times the plasma volume. Since the therapeutic levels of coagulation factors are well defined in most clinical situations, applied pharmacokinetics is an excellent tool to optimise therapy. Individual tailoring of administration in prophylaxis has been shown to considerably increase the cost effectiveness of the treatment. Dosage regimens for the treatment of bleeding episodes or for haemostasis during surgery are also designed using pharmacokinetic data, and the advantages of using a constant infusion instead of repeated bolus doses have been explored. The influence of antibodies (inhibitors) on the pharmacokinetics of FVIII and FIX is in part understood, and the doses of coagulation factor needed to treat a patient can tentatively be calculated from the antibody titre. In conclusion, therapeutic monitoring of coagulation factor levels and the use of clinical pharmacokinetics to aid therapy are well established in the treatment of patients with haemophilia.

Pharmacokinetic analysis of plasma-derived and recombinant F IX concentrates in previously treated patients with moderate or severe hemophilia B

BACKGROUND

Hemophilia B is an X-linked bleeding disorder that affects approximately 1 in 25,000 males. Therapy for acute bleeding episodes consists of transfusions of plasma-derived (pd-F IX) or recombinant (r-F IX) concentrates.

STUDY DESIGN AND METHODS

A double-blind, two-period crossover study was initiated to assess the pharmacokinetics of pd-F IX and r-F IX and to address patient-specific variables that might influence in vivo recovery. Study product was administered by a single bolus infusion (50 IU/kg) to 43 previously treated patients in the nonbleeding state, and F IX:C levels were measured over a period of 48 hours after infusion.

RESULTS

The mean in vivo recovery in the pd-F IX group was 1.71 +/- 0.73 IU per dL per IU per kg compared with 0.86 +/- 0.31 IU per dL per IU per kg with r-F IX (p <or= 0.0001). There was a significant positive correlation (Pearsons r = 0.62, p <or= 0.0001, 95% CI, 0.37-0.78) between the recoveries of the two products and a weak correlation between the recovery of pd-F IX and baseline F IX:Ag levels. There was no significant difference in the terminal half-lives of the two products.

CONCLUSIONS

The study found wide product- and patient-related variability in recovery. Inherent differences among patients, including baseline F IX, may account for some of the interpatient variability. These differences should be taken into account in optimizing treatment regimens for individual patients with hemophilia B.

Location of the platelet binding site in zymogen coagulation factor IX

The assembly of the tenase complex on the surface of the platelet is an essential step in maintaining normal hemostasis as evidenced by the serious hemorrhagic diathesis associated with either factor IX (FIX) or factor VIII deficiencies. Understanding the regions and or residues of FIX crucial for proper binding to platelets has important clinical implications. The ability of FIX to bind activated platelets in the presence of 4 mmol/l CaCl2 was examined using electrophoretic light-scattering experiments. Wild-type FIX binds to activated platelets with dissociation constant Kd = 7.9 nmol/l. Activated FIX binds to activated platelets with Kd = 2 nmol/l. Activated factor VII does not bind activated platelets at physiological concentrations. The Gla domain of FIX is important for the binding of FIX to activated platelets since a chimera with a factor VII (FVII) template and FIX Gla [FVII(FIXGla)] has Kd = 9.6 nmol/l, and a chimera with a FVII template and FIX Gla, A and the first epidermal growth factor domain (EGF1) [FVII(FIXGla,A,EGF1)] has Kd = 9.7 nmol/l, but a chimera with a FIX template and a FVII Gla [FIX(FVIIGla)] does not bind activated platelets. Altering the fifth residue of FIX from a lysine to an alanine (Lys5<--Ala) abolishes the mutant from binding to collagen but does not affect FIX binding to the activated platelet (Kd = 9.8 nmol/l). Point mutations involved with residues 4 and 5 (Gly4<--Phe and Lys5<--no residue), residue 9 (Phe9<--Ala), residue 10 (Val10<--Lys) and residues 9-11 (Phe9<--Met, Val10<--Lys, Glu11<--Lys) do not bind to activated platelets.

Activation of factor IX zymogen results in exposure of a binding site for low-density lipoprotein receptor–related protein

The interaction between the endocytic receptor low density lipoprotein receptor–related protein (LRP) and either coagulation factor IX or its active derivative factor IXa was studied. Purified factor IX was unable to associate with LRP when analyzed by surface plasmon resonance. By contrast, factor XIa–mediated conversion of factor IX into factor IXa resulted in reversible dose- and calcium-dependent binding to LRP. Active-site blocking of factor IXa did not affect binding to LRP, whereas LRP binding was efficiently inhibited in the presence of heparin or antibodies against factor IX or LRP. The factor IXa–LRP interaction could be described by a 2-site binding model with equilibrium dissociation constants of 27 nmol/L and 69 nmol/L. Consistent with this model, it was observed that factor IXa binds to 2 different recombinant receptor fragments of LRP (denoted cluster II and cluster IV) with equilibrium dissociation constants of 227 nmol/L and 53 nmol/L, respectively. The amount of factor IXa degraded by LRP-deficient cells was 35% lower than by LRP-expressing cells, demonstrating that LRP contributes to the transport of factor IXa to the intracellular degradation pathway. Because ligand binding to LRP is often preceded by binding to proteoglycans, the contribution of proteoglycans to the catabolism of factor IXa was addressed by employing proteoglycan-deficient cells. Degradation of factor IXa by proteoglycan-deficient cells proceeded at a 83% lower rate than wild-type cells. In conclusion, the data presented here indicate that both LRP and proteoglycans have the potential to contribute to the catabolism of factor IXa.

Activation of factor IX zymogen results in exposure of a binding site for low-density lipoprotein receptor–related protein

The interaction between the endocytic receptor low density lipoprotein receptor–related protein (LRP) and either coagulation factor IX or its active derivative factor IXa was studied. Purified factor IX was unable to associate with LRP when analyzed by surface plasmon resonance. By contrast, factor XIa–mediated conversion of factor IX into factor IXa resulted in reversible dose- and calcium-dependent binding to LRP. Active-site blocking of factor IXa did not affect binding to LRP, whereas LRP binding was efficiently inhibited in the presence of heparin or antibodies against factor IX or LRP. The factor IXa–LRP interaction could be described by a 2-site binding model with equilibrium dissociation constants of 27 nmol/L and 69 nmol/L. Consistent with this model, it was observed that factor IXa binds to 2 different recombinant receptor fragments of LRP (denoted cluster II and cluster IV) with equilibrium dissociation constants of 227 nmol/L and 53 nmol/L, respectively. The amount of factor IXa degraded by LRP-deficient cells was 35% lower than by LRP-expressing cells, demonstrating that LRP contributes to the transport of factor IXa to the intracellular degradation pathway. Because ligand binding to LRP is often preceded by binding to proteoglycans, the contribution of proteoglycans to the catabolism of factor IXa was addressed by employing proteoglycan-deficient cells. Degradation of factor IXa by proteoglycan-deficient cells proceeded at a 83% lower rate than wild-type cells. In conclusion, the data presented here indicate that both LRP and proteoglycans have the potential to contribute to the catabolism of factor IXa.

Biospecific interactions of vitamin K-dependent factors with phospholipid-like polystyrene derivatives. Part II: factor IX

We previously demonstrated that phosphorylated polystyrene derivatives exhibit phospholipid-like behaviour and therefore are able to interact with factor II, one of the vitamin K-dependent coagulation factors. Under the same conditions as for factor II, we examined the interactions of factor IX with phosphorylated resins of various compositions in phosphate groups: these studies were carried out with or without albumin precoating of the polymers and either in the presence or absence of calcium ions. Adsorption experiments show that, in the absence of calcium ions, only one class of adsorption sites of factor IX can be evidenced with the interactions taking place through the formation of binary complexes, whereas in the presence of calcium ions, the affinity of factor IX for phosphorylated resins becomes very high and two types of adsorption sites have been evidenced with biospecific ternary complexes being formed. The domains of predominance of these complexes were determined. Moreover, the only functional groups borne by the phosphorylated polystyrene resins involved in factor IX-polymer interactions are phosphodiester groups. Comparison between factor II and factor IX adsorption onto the same polymers leads to the conclusion that the observed differences probably reflect the differences in the Gla domains of the vitamin K-dependent factors. Finally, this study demonstrates that phosphorylated polystyrene derivatives can be used as stationary phases for purification of factor IX by highly specific liquid biochromatography.

Human factor IX binds to specific sites on the collagenous domain of collagen IV

The primary region of factor IX that mediates binding to bovine aortic endothelial cells resides in residues 3-11 of the N-terminal region known as the Gla domain. Recently, it was proposed that the observed binding to endothelial cells is actually a measure of the interaction between factor IX and collagen IV (Cheung, W. F., van den Born, J., Kuhn, K., Kjellen, L., Hudson, B. G., and Stafford, D. W. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 11068-11073). To confirm that factor IX binds to collagen IV and to examine the specificity of this interaction, we used scanning force microscopy to examine factor IX binding to collagen IV. We imaged collagen IV in the presence and the absence of factor IX and observed specific interactions between factor IX and collagen IV. Our results demonstrate that factor IX binds to collagen IV at specific sites in the collagenous domain approximately 98 and approximately 50 nm from the C-terminal pepsin-cleaved end.

Dissimilar interaction of factor VIII with endothelial cells and lipid vesicles during factor X activation

A localized and regulated cascade of proteolytic events is a prerequisite for normal haemostasis. The activation of factor X by activated factor IX (factor IXa) in the presence of activated factor VIII (factor VIIIa) is essential for the formation of a fibrin clot at sites of vascular injury. We observed sustained activation of factor X on the surface of vascular endothelial cells, whereas, in agreement with others, on synthetic negatively charged phospholipid vesicles and activated blood platelets factor X activation is transient and starts to decline a few minutes after the onset of the reaction. We examined the mechanism responsible for these differences in factor X activation. Procoagulant membrane and solution were analysed separately for the occurrence of factor VIII and its activation fragments. On negatively charged phospholipid vesicles, on dissociation of factor VIIIa, the 67 kDa light-chain fragment remains associated with the lipid membrane. As a result, factor VIII-binding sites remain occupied, and dampening of factor X activation occurs. In contrast, on monolayers of endothelial cells, no residual factor VIIIa fragments associated with the cell membrane were observed. During endothelial-cell-mediated activation of factor X, accumulation of factor VIIIa fragments was observed in the solution phase only. This finding suggests that, on endothelial cells, factor VIII-binding sites remain accessible for further factor VIII binding, guaranteeing sustained activation of factor X. These data demonstrate that the nature of the procoagulant membrane contributes to the regulation of the cofactor activity of factor VIII and thereby affects the progress of factor X activation.

Identification of the endothelial cell binding site for factor IX

We previously demonstrated that the primary region of factor IX and IXa responsible for saturable specific binding to bovine aortic endothelial cells resides in residues 3-11 at the amino terminus of factor IX. We also demonstrated that mutations of lysine to alanine at residue 5, factor IX K5A, or valine to lysine at residue 10, factor IX V10K, resulted in a molecule unable to bind to endothelial cells. Moreover, a mutation with lysine to arginine at residue 5, factor IX K5R, resulted in a factor IX molecule with increased affinity for the endothelial cell binding site. In this paper we report that collagen IV is a strong candidate for the factor IX binding site on endothelial cells. Factor IX and factor IX K5R compete with 125I-labeled factor IX for binding to tetrameric collagen IV immobilized on microtiter plates, while factor X, factor VII, and factor IX K5A or V10K fail to compete. The Kd for wild-type factor IX binding to collagen IV in the presence of heparin was 6.8 +/- 2 nM, and the Kd for factor IX K5R was 1.1 +/- 0.2 nM, which agrees well with our previously published Kd values of 7.4 and 2.4 nM for binding of the same proteins to endothelial cells. Our working assumption is that we have identified the endothelial cell binding site and that it is collagen IV. Its physiological relevance remains to be determined.

The entire gamma-carboxyglutamic acid- and helical stack-domains of human coagulation factor IX are required for optimal binding to its endothelial cell receptor

The minimal region of the gamma-carboxyglutamic acid (Gla) domain of human factor (f) IX that interacted with its putative bovine aortic endothelial cell (BAEC) receptor was examined by chemical synthesis of peptides with sequence counterparts in this region of the protein, and assessment of their relative abilities to compete with fIX for receptor binding. We found that IC50 values (total peptide concentrations needed to achieve 50% inhibition of binding of [125I]-fIX to BAEC) were ca. 18 nM for unlabeled fIX and 23 nM for the peptide consisting of the entire Gla domain/helical stack (HS) region (residues 1-47) of fIX. The peptide containing only the Gla domain of fIX (residues 1-38) displayed an IC50 value of > 500 nM for this same competitive binding, whereas peptides containing sequences present in positions 1-14 and 1-24 of the Gla domain of human fIX did not significantly compete with [125I]-fIX for BAEC binding. We conclude that whereas a specific receptor recognition element is present within residues 1-14 of fIX, as has previously been concluded by others and by us, full expression of this epitope requires its presence within the entire Gla domain and HS for proper folding. All determinants for proper folding of fIX that lead to BAEC receptor binding appear to be present within these two domains.

Localization of a calcium-dependent epitope to the amino terminal region of the Gla domain of human factor IX

We have used site-directed mutagenesis to define the epitope of calcium-dependent monoclonal antibodies to human factor IX. We demonstrate that the calcium-specific epitope includes residues 1-11 of factor IX, with apparent contributions from other regions of the protein. Antibodies JK.IX-1, -3, and -4 had critical portions of their epitopes in the first eleven amino acids of the Gla domain. These three antibodies could bind to a chimera containing the factor VII Gla domain, in which surface residues 3, 4, 5, 9, 10, and 11 were modified to those of factor IX. In contrast, the epitope of JK.IX-2 was unaffected by mutations in residues 3-11 of factor IX, but was dependent on the amino terminal tyrosine residue. Furthermore, the calcium-dependent monoclonal antibodies, JK.IX-1, -3, and -4, whose epitope include residues 3 through 11. inhibit factor IX's binding to endothelial cells, for which the binding site on factor IX has been localized to this region. Our results, together with previous studies, confirm the existence of discrete calcium and metal-dependent epitopes within the Gla domain of factor IX and show that the calcium-specific epitope lies near the amino terminus region of this domain.

Active site-blocked factors VIIa and IXa differentially inhibit fibrin formation in a human ex vivo thrombosis model

The role of tissue factor/factor VIIa (FVIIa) and factor VIIIa/factor IXa (FVIIIa/FIXa) complexes in thrombus formation was examined in a human ex vivo blood flow system by use of active site-blocked FVIIa (FVIIai) and FIXa (FIXai) as selective inhibitors. Blood was drawn directly from the veins of volunteers into a mixing device where FVIIai and FIXai were mixed with flowing blood. The blood then entered parallel-plate chambers containing coverslips coated with human fibrillar collagen or tissue factor-expressing cell layers of tumor necrosis factor-alpha-stimulated human endothelial cells, human smooth muscle cells, and J82 cells. Exposure of stimulated endothelial cells to blood flowing at a venous shear rate of 65/s led to fibrin deposition, which was inhibited by infusion of FVIIai (IC50, 3 nmol/L), as quantified by micro-densitometry of fibrin-stained coverslips. Whereas FIXai (600 nmol/L) was only a weak inhibitor, FVIIai (60 nmol/L) reduced fibrinopeptide A (FPA) plasma levels from 504 +/- 79 to 171 +/- 27 ng/mL and concomitantly inhibited platelet thrombus deposition. Similarly, experiments with smooth muscle cells and J82 cells showed that FVIIai but not FIXai efficiently reduced FPA levels. Conversely, with tissue factor-free collagen, ,hich induces platelet-dependent fibrin formation, infusion of FIXai but not of FVIIai inhibited fibrin deposition (IC50, 8 nmol/L) and reduced FPA levels from 55 +/- 8 to 9 +/- 5 ng/mL. However, FIXai did not affect the number of platelet thrombi deposited on collagen.(ABSTRACT TRUNCATED AT 250 WORDS)

Calreticulin, an antithrombotic agent which binds to vitamin K-dependent coagulation factors, stimulates endothelial nitric oxide production, and limits thrombosis in canine coronary arteries

Coagulation Factor IX/IXa has been shown to bind to cellular surfaces, and Factor IXa expresses its procoagulant activity by assembling into the intrinsic Factor X activating complex (Factors IXa/VIIIa/X), which also forms on membrane surfaces. This led us to identify cellular proteins which bind Factor IX/IXa; an approximately 55-kDa polypeptide was purified to homogeneity from bovine lung extracts based on its capacity to bind 125I-Factor IX in a dose-dependent and saturable manner. From protein sequence data of the amino terminus and internal peptides, the approximately 55-kDa polypeptide was identified as calreticulin, a previously identified intracellular calcium-binding protein. Recombinant calreticulin bound vitamin K-dependent coagulation factors, 125I-Factor IX, 125I-Factor X, and 125I-prothrombin (Kd values of approximately 2.7, 3.2, and 8.3 nM, respectively), via interaction with its C-domain, although it did not affect the coagulant properties of these proteins. 125I-Calreticulin also bound to endothelial cells in vitro (Kd approximately 7.4 nM), and mouse infusion studies showed an initial rapid phase of clearance in which calreticulin could be localized on the vascular endothelium. Exposure of endothelial cells to calreticulin led to dose-dependent, immediate, and sustained increase in the production of nitric oxide, as measured using a porphyrinic microsensor. In a canine electrically induced thrombosis model, intracoronary infusion of calreticulin (n = 7) prevented occlusion of the left circumflex coronary artery in a dose-dependent manner compared with vehicle-treated controls (n = 5). These results indicate that calreticulin interacts with the endothelium to stimulate release of nitric oxide and inhibit clot formation.

Endothelial cells stimulated with tumor necrosis factor-alpha express varying amounts of tissue factor resulting in inhomogenous fibrin deposition in a native blood flow system. Effects of thrombin inhibitors

TNF-alpha induces changes in endothelial cell functions, such as upregulation of tissue factor, resulting in endothelial procoagulant activity which may play a role in disseminated intravascular coagulation. The procoagulant activity of TNF-alpha-stimulated endothelial cell monolayers was studied in a human ex vivo native (nonanticoagulated) blood flow system using the three thrombin inhibitors recombinant hirudin, Ro 46-6240, and heparin. Under venous blood flow conditions (shear rate 65 s-1) recombinant hirudin, Ro 46-6240, and heparin inhibited fibrin deposition on the endothelial cells by 50% at concentrations of 14, 28, and 412 ng/ml, respectively. The highest tested concentrations of the thrombin inhibitors reduced the postchamber fibrinopeptide A levels from 713 +/- 69 to < 70 ng/ml. Surprisingly, even at relatively high inhibitor concentrations, some local fibrin deposits were found on TNF-alpha-stimulated cells, suggesting that some endothelial cells possess higher procoagulant activity than others. Therefore, the surface expression pattern of tissue factor, the primary initiator of coagulation in this system, was examined by immunogold-silver staining. The results showed that the tissue factor density on the cell surface varied strongly among TNF-alpha-stimulated endothelial cells. Using TNF receptor-selective agonistic mutants of TNF-alpha, it was demonstrated further that the heterogenous surface expression of tissue factor was mediated entirely by the 55-kD TNF receptor and did not involve the 75-kD TNF receptor. We conclude that in this system TNF-alpha induces heterogenous tissue factor expression which may lead to a high local thrombin concentration, such that even in the presence of thrombin inhibitors focal fibrin deposition occurs.

The control of production and release of haemostatic factors in the endothelial cell

Endothelial cell products contribute to many aspects of the regulation of haemostasis. They include potent inhibitors of platelet aggregation (prostacyclin and nitric oxide) rapidly released in response to agonists such as thrombin. Similar agonists also induce the formation of platelet-activating factor by endothelium. Endothelial cell surface ectonucleotidase enzymes control the catabolism of platelet-active adenine nucleotides. The main source of the circulating coagulant cofactor von Willebrand factor is the endothelium, where it is stored in granules for agonist-triggered exocytosis and also secreted constitutively. Surface anticoagulant activities are due to the presence of antithrombin and thrombomodulin. Endothelial cells also secrete plasminogen activator and its inhibitor. Many of these reactions are significantly modulated by exposure of endothelium to cytokines or bacterial endotoxin, the most striking example being the new synthesis and surface expression of the procoagulant tissue factor (thromboplastin).

The endothelial cell binding determinant of human factor IX resides in the gamma-carboxyglutamic acid domain

The blood coagulation factor IX(a) binds specifically to a site on endothelial cells with a Kd of 2.0-3.0 nM. A number of previous studies have attempted to define the region(s) of factor IX(a) that mediate this interaction. These studies suggested that there are two regions of factor IX(a), the gamma-carboxyglutamic acid (Gla) domain and the epidermal growth factor like (EGF-like) domains, that mediate high-affinity binding to endothelial cells. Recently, however, the participation of the EGF1 domain has been excluded from the interaction. This indicated that if there was an EGF component of factor IX contributing to the binding affinity, then it must be in the second EGF-like domain. In order to further evaluate this relationship, we performed competitive binding experiments between 125I plasma factor IX and a set of six chimeric proteins composed of portions of factor VII and factor IX. Our data suggest that the high-affinity interaction between factor IX and the endothelial cell binding site is mediated by the factor IX Gla domain and that the factor IX EGF domains are not involved in binding specificity.

Prothrombotic and fibrinolytic function of normal and perturbed endothelium

Endothelium mediates thrombosis and fibrinolysis in part through cell-based production and binding of coagulation factors and anticoagulant moieties. A variety of biologically active agents (homocysteine, interleukin-1, and endotoxin) and environmental factors (mechanical injury and hypoxia) can upset the homeostasis of the coagulation system leading in vitro and in vivo to abnormal propensity for thrombin formation. None of the above agents or any similar agents have been demonstrated to inhibit clotting factors or thrombosis formation. As one would expect, injury to endothelium even on a sublethal basis promotes a procoagulant response. This procoagulant response appears to be multifactorial and multiphasic. Gross injury leads to factor release by dead or dying cells with rapid promotion of thrombosis by neighboring infact cells. A more subtle prothrombotic state promulgated by viable cells appears after sublethal perturbation and is characterized by protein synthesis-dependent production of endothelial-produced clotting factors in the early stages of injury with subsequent return to normal levels of these factors by 24 hr. A second phase of abnormal levels of fibrinolysis inhibitors and possible reduction in fibrinolysis promoters appears to follow up to 48 hr after the initial injury. Thus the cell can be viewed as producing means of promoting thrombus formation followed by preventing thrombus dissolution. The data presented in this article are primarily cell culture data and are thus limited both in physiologic applicability and in ability to define over long periods the relevance of those findings to other cells in the hemostatic system. Nonetheless, in vivo data of endothelial perturbation correlate with the in vitro findings.(ABSTRACT TRUNCATED AT 250 WORDS)

Release of diamine oxidase into plasma by glycosaminoglycans in rats

Plasma diamine oxidase (DAO) values are enhanced by intravenous injection of heparin which releases the enzyme, synthesized in small bowel enterocytes, from binding sites located on endothelial cells of the intestinal microvasculature. Intestinal DAO, in analogy with lipoprotein lipase (another heparin-released enzyme), is believed to be electrostatically linked to endothelial binding sites composed of a glycosaminoglycan (GAG) which is presumably heparan sulphate, but the complete mechanism of enzyme release is not known. In this study we assayed in rats the DAO-releasing capability of heparan sulphate, dermatan sulphate, chondroitin sulphate A and hyaluronic acid, all heparin related compounds. Heparan sulphate, a compound with the same hexosamine as heparin but with a lower concentration of sulphated iduronic acid, induced a very high release of DAO (3-fold less than heparin), while the other tested GAGs, composed of higher proportions of non sulphated uronic acid and with galactosamine instead of glucosamine, induced a significantly lower release. In rats treated with 60 mg heparan sulphate the significant decrease in ileal mucosal DAO activity indicates that, in analogy with heparin, the high plasma enzymatic activity induced is of enterocytic origin. It is suggested that the high charge density of the compounds tested, due to the degree of sulphatation, is the decisive factor in promoting the release of intestinal DAO.

A 32 kDa lipocortin from human mononuclear cells appears to be identical with the placental inhibitor of blood coagulation

A 32 kDa protein isolated from human mononuclear cells is a member of the lipocortin family, a new group of Ca2+-dependent lipid-binding proteins thought to be involved in the regulation of phospholipase A2, in exocytosis and in membrane-cytoskeleton interactions. Purification of this protein was based on its ability to associate with membrane phospholipids in a Ca2+-dependent manner and its capacity to inhibit purified phospholipase A2 from pig pancreas. Using immunological detection, we show that it is present in various cells involved in the inflammatory and coagulation processes. We present extensive amino acid data that strongly suggest that this protein is identical with a recently described inhibitor of blood coagulation, with endonexin II and with lipocortin V. Sequence alignment with other known proteins show a significant degree of homology with lipocortins I and II, the substrates of the epidermal-growth-factor receptor tyrosine kinase and the oncogene pp60src tyrosine kinase respectively, and with protein II. The possible physiological role of this 32 kDa lipocortin is discussed.

The interactions of factors X and IX with phospholipid

The interactions of human factors X and IX with phospholipid were studied with an ELISA system. In the presence of calcium ions, factor X bound to phosphatidylserine with an affinity of 1.56 x 10(10) M-1 and to phosphatidylcholine with an affinity of 5.6 x 10(9) M-1. In the presence of calcium ions, factor IX bound to phosphatidylserine with an affinity of 8.4 x 10(8) M-1 and no binding to phosphatidylcholine was observed. No competition of factors X and IX for phosphatidylserine binding sites was observed.

Human placental anticoagulant protein: isolation and characterization

An anticoagulant protein was purified from the soluble fraction of human placenta by ammonium sulfate precipitation and column chromatography on DEAE-Sepharose, Sephadex G-75, and Mono S (Pharmacia). The yield of the purified protein was approximately 20 mg from one placenta. The purified protein gave a single band by sodium dodecyl sulfate-polyacrylamide gel electrophoresis with a molecular weight of 36,500. This protein prolonged the clotting time of normal plasma when clotting was induced either by brain thromboplastin or by kaolin in the presence of cephalin and Ca2+. It also prolonged the factor Xa induced clotting time of platelet-rich plasma but did not affect thrombin-induced conversion of fibrinogen to fibrin. The purified placental protein completely inhibited the prothrombin activation by reconstituted prothrombinase, a complex of factor Xa-factor Va-phospholipid-Ca2+. The placenta inhibitor had no effect on prothrombin activation when phospholipid was omitted from the above reaction. Also, it neither inhibited the amidolytic activity of factor Xa, nor did it bind to factor Xa. The placenta inhibitor, however, did bind specifically to phospholipid vesicles (20% phosphatidylserine and 80% phosphatidylcholine) in the presence of calcium ions. These results indicate that the placental anticoagulant protein (PAP) inhibits coagulation by binding to phospholipid vesicles. The amino acid sequences of three cyanogen bromide fragments of PAP aligned with those of two distinct regions of lipocortin I and II with a high degree of homology, showing that PAP is a member of the lipocortin family.