Search

The inside of all blood vessels is lined with a tight network of specialized endothelial cells that form a semipermeable barrier between blood and underlying tissues. By virtue of their strategic position at the interface between blood and tissue, endothelial cells are constantly exposed to blood cells, proteins and other blood components. To sense their environment, endothelial cells are equipped with a wide range of transmembrane receptors that convert information from the extracellular environment into intracellular signaling pathways. To better understand the swift adaptive endothelial response it is crucial to map these communication networks. To cover the rapid and reversible post-translational modifications a system-wide analysis of cellular signaling networks is required. Therefore in the present thesis, we investigated endothelial signaling pathways, with particular emphasis on coagulation, using an unbiased quantitative mass spectrometry approach. Thrombin is the key serine protease of the coagulation cascade and its predominant receptor is protease-activated receptor 1 (PAR1). Our data demonstrated that thrombin dynamically regulates a plethora of phosphorylation events, including actin cytoskeleton and cell junctions signaling. Next, we showed that thrombin-induced phosphoregulation is mediated exclusively through PAR1. In addition, we showed that PAR1 inhibitors vorapaxar and parmodulin-2 demonstrated distinct antagonistic properties and that only canonical PAR1 cleavage by thrombin, and not by other proteases, generates a tethered ligand that potently induces early signaling. These data provided mechanistic inside into the activation and inhibition of PAR1-mediated signaling in endothelial cells. To dissect diverging and converging signaling pathways, we analyzed the communication networks initiated by histamine and other members of the PAR family (PAR1,2,3,4). Our data showed that PAR1 and PAR2 induced similar phosphoregulation of proteins involved in Gαq- and the MAPK pathway. In contrast, they differentially regulated proteins related to clathrin-mediated endocytosis. Thrombin and histamine induced highly similar phosphoregulation, while they differentially phosphorylated proteins related to the Gα12/13 pathway. Furthermore, we assessed phosphoproteomic regulation in endothelial cells induced by platelet proteins and compared it to vascular endothelial growth factor (VEGF)-mediated phosphoregulation. These data showed that platelet releasate and lysate induced highly similar signal transduction and that their effect was not primarily driven by VEGF. In parallel, we used a proteomics approach to compare cord blood platelets with adult peripheral blood platelets and showed that their protein abundance was highly similar, while limited differences were related to proteins involved in mitochondrial energy metabolism. Finally, we performed an integrated analysis of all phosphoproteomic analyses to reveal unique signaling signatures and core activation pathways. Taken together, this thesis provides a unique overview of communication networks in endothelial cells, which may contribute to a better understanding of the interplay of coagulation factors, platelets and the vessel wall and may prompt future investigations to identify key targets for therapeutic intervention.

The inside of all blood vessels is lined with a tight network of specialized endothelial cells that form a semipermeable barrier between blood and underlying tissues. By virtue of their strategic position at the interface between blood and tissue, endothelial cells are constantly exposed to blood cells, proteins and other blood components. To sense their environment, endothelial cells are equipped with a wide range of transmembrane receptors that convert information from the extracellular environment into intracellular signaling pathways. To better understand the swift adaptive endothelial response it is crucial to map these communication networks. To cover the rapid and reversible post-translational modifications a system-wide analysis of cellular signaling networks is required. Therefore in the present thesis, we investigated endothelial signaling pathways, with particular emphasis on coagulation, using an unbiased quantitative mass spectrometry approach. Thrombin is the key serine protease of the coagulation cascade and its predominant receptor is protease-activated receptor 1 (PAR1). Our data demonstrated that thrombin dynamically regulates a plethora of phosphorylation events, including actin cytoskeleton and cell junctions signaling. Next, we showed that thrombin-induced phosphoregulation is mediated exclusively through PAR1. In addition, we showed that PAR1 inhibitors vorapaxar and parmodulin-2 demonstrated distinct antagonistic properties and that only canonical PAR1 cleavage by thrombin, and not by other proteases, generates a tethered ligand that potently induces early signaling. These data provided mechanistic inside into the activation and inhibition of PAR1-mediated signaling in endothelial cells. To dissect diverging and converging signaling pathways, we analyzed the communication networks initiated by histamine and other members of the PAR family (PAR1,2,3,4). Our data showed that PAR1 and PAR2 induced similar phosphoregulation of proteins involved in Gαq- and the MAPK pathway. In contrast, they differentially regulated proteins related to clathrin-mediated endocytosis. Thrombin and histamine induced highly similar phosphoregulation, while they differentially phosphorylated proteins related to the Gα12/13 pathway. Furthermore, we assessed phosphoproteomic regulation in endothelial cells induced by platelet proteins and compared it to vascular endothelial growth factor (VEGF)-mediated phosphoregulation. These data showed that platelet releasate and lysate induced highly similar signal transduction and that their effect was not primarily driven by VEGF. In parallel, we used a proteomics approach to compare cord blood platelets with adult peripheral blood platelets and showed that their protein abundance was highly similar, while limited differences were related to proteins involved in mitochondrial energy metabolism. Finally, we performed an integrated analysis of all phosphoproteomic analyses to reveal unique signaling signatures and core activation pathways. Taken together, this thesis provides a unique overview of communication networks in endothelial cells, which may contribute to a better understanding of the interplay of coagulation factors, platelets and the vessel wall and may prompt future investigations to identify key targets for therapeutic intervention.

Blood coagulation (hemostasis) is a process that is driven by plasma proteins which form a complex network of molecular interactions. An essential part of the coagulation system is formed by a cascade of proenzymes (i.e. coagulation factors) that are sequentially converted into active enzymes. Activation of these coagulation enzymes is regulated by cofactors present in the blood plasma. However, the way regulation is achieved on a molecular level remains an open question for most of the coagulation factors. In the present thesis we have addressed these issues by use of footprinting techniques in combination with mass spectrometry. Footprinting makes use of chemical modification of amino acids. The extent of chemical modification can be seen as a measure of accessibility of the modified amino acids, either within a single protein or in a complex of proteins. The footprinting approach can, therefore, be used to probe changes within a single protein or to explore binding sites within a protein complex. As proof-of-principle, footprinting with lysine-reactive tags was applied in a previously well-characterized protein complex between the plasma proteins Receptor-Associated Protein and Low-density-lipoprotein Receptor-related Protein. The same footprinting technique was used to characterize chimeras of the cofactor, factor VIII which contained protein domains of the homologous coagulation factor V. Subsequent studies were aimed to elucidate the activation process of plasma proenzymes like Hyaluronan-binding-protein 2 and coagulation factors IX, X and prothrombin. The active protease domains of these serine proteases share essentially the same protein fold. By use of footprinting, subtle changes could be probed in the interplay between different parts of these protease domains. This way, activation of factor IX appeared different compared to those of factor X and prothrombin. The activity of activated IX was furthermore studied by alanine substitutions made in the factor IX protease domain.

Blood coagulation (hemostasis) is a process that is driven by plasma proteins which form a complex network of molecular interactions. An essential part of the coagulation system is formed by a cascade of proenzymes (i.e. coagulation factors) that are sequentially converted into active enzymes. Activation of these coagulation enzymes is regulated by cofactors present in the blood plasma. However, the way regulation is achieved on a molecular level remains an open question for most of the coagulation factors. In the present thesis we have addressed these issues by use of footprinting techniques in combination with mass spectrometry. Footprinting makes use of chemical modification of amino acids. The extent of chemical modification can be seen as a measure of accessibility of the modified amino acids, either within a single protein or in a complex of proteins. The footprinting approach can, therefore, be used to probe changes within a single protein or to explore binding sites within a protein complex. As proof-of-principle, footprinting with lysine-reactive tags was applied in a previously well-characterized protein complex between the plasma proteins Receptor-Associated Protein and Low-density-lipoprotein Receptor-related Protein. The same footprinting technique was used to characterize chimeras of the cofactor, factor VIII which contained protein domains of the homologous coagulation factor V. Subsequent studies were aimed to elucidate the activation process of plasma proenzymes like Hyaluronan-binding-protein 2 and coagulation factors IX, X and prothrombin. The active protease domains of these serine proteases share essentially the same protein fold. By use of footprinting, subtle changes could be probed in the interplay between different parts of these protease domains. This way, activation of factor IX appeared different compared to those of factor X and prothrombin. The activity of activated IX was furthermore studied by alanine substitutions made in the factor IX protease domain.

Although platelets are small anucleate cell fragments, they are highly organized cells rich in different types of organelles. Platelets contain three secretory granules, dense granules, α-granules and lysosomes which are released upon platelet activation at sites of injury. Most of these α-granule proteins are synthesized by the platelet precursor cells, the megakaryocytes (MKs). Platelets and MKs are, however, unique in that they also take up plasma proteins. The mechanisms behind the endocytosis of plasma proteins as well as the exocytosis machinery are, however, still poorly understood. In our work, we employ protein-protein interaction studies and cell-based internalization studies to get more insight into the mechanisms underlying the endocytosis of plasma proteins in MKs. In addition, we utilize mass spectrometry-based approaches to assess the role of the platelet integrins in α-granule cargo enrichment as well as the mechanism of the release thereof. α-granule coagulation factor V (FV) derives entirely from receptor-mediated endocytosis by megakaryocytes. In our work we provided evidences of a novel role for the lectin galectin8 (Gal8) as mediator of FV uptake. Surface plasmon resonance (SPR) analysis showed a reversible and dose-dependent binding of Gal8 and FV. By inhibiting Gal8 functions and/or targeting its expression in the megakaryocyte-like cells (DAMI), we demonstrated that Gal8-FV complex formation is a prerequisite for FV uptake. Further we explored the potential role of other interaction partners of Gal8, i.e. GPIIbIIIa and GPIb-GPIX-GPV, in FV endocytosis. Mass spectrometry (MS) analysis of platelet lysates from leukocyte adhesion deficiency type III (LAD-III) and Glanzmann's thrombasthenia (GT) patients show that the level of GPV and not of GPIIbIIIa correlates with the amount of FV. Biochemical approaches show that Gal8 can act as a cross-linker between FV and GPV, suggesting a potential role for GPV as Gal8 counter-receptor. Fibrinogen, the most abundant α-granule protein, can be taken up from plasma by both circulating platelets and megakaryocytes. Here, by mass spectrometry analysis of the platelet content from GT and LAD-III patients we explored the role of integrins on platelets cargo enrichment.The results show that GPIIbIIIa only contributes to the uptake of fibrinogen, and that activation of this integrin is not required to facilitate fibrinogen endocytosis. Novel differentially expressed proteins that may be associated with the investigated diseases are identified as well. Moreover MS analysis further revealed the genetic defect of a previously uncharacterized GT patient. In a different line of research, we explored the platelet secretory efficiency of α-granulesin patients with Familial Hemophagocytic Lymphohistiocytosis-5 (FHL-5). Disease-causing mutations have been mapped onto the gene encoding syntaxin-binding-protein2 (Munc18-2). We use MS, flow cytometry and electron microscopy to identify the protein product of a Munc18-2 splice-site variant (1247-1G>C) and the effect thereof on platelet degranulation. This Munc18-2 variant has been established to cause impaired degranulation of neutrophils. In spite of previous suggestions, flow cytometry, electron microscopy and MS studies reveal that the splice-site variant does not significantly affect platelet. Moreover MS analysis revealed the still unknown protein product resulting from the 1247-1G>C mutation.

Although platelets are small anucleate cell fragments, they are highly organized cells rich in different types of organelles. Platelets contain three secretory granules, dense granules, α-granules and lysosomes which are released upon platelet activation at sites of injury. Most of these α-granule proteins are synthesized by the platelet precursor cells, the megakaryocytes (MKs). Platelets and MKs are, however, unique in that they also take up plasma proteins. The mechanisms behind the endocytosis of plasma proteins as well as the exocytosis machinery are, however, still poorly understood. In our work, we employ protein-protein interaction studies and cell-based internalization studies to get more insight into the mechanisms underlying the endocytosis of plasma proteins in MKs. In addition, we utilize mass spectrometry-based approaches to assess the role of the platelet integrins in α-granule cargo enrichment as well as the mechanism of the release thereof. α-granule coagulation factor V (FV) derives entirely from receptor-mediated endocytosis by megakaryocytes. In our work we provided evidences of a novel role for the lectin galectin8 (Gal8) as mediator of FV uptake. Surface plasmon resonance (SPR) analysis showed a reversible and dose-dependent binding of Gal8 and FV. By inhibiting Gal8 functions and/or targeting its expression in the megakaryocyte-like cells (DAMI), we demonstrated that Gal8-FV complex formation is a prerequisite for FV uptake. Further we explored the potential role of other interaction partners of Gal8, i.e. GPIIbIIIa and GPIb-GPIX-GPV, in FV endocytosis. Mass spectrometry (MS) analysis of platelet lysates from leukocyte adhesion deficiency type III (LAD-III) and Glanzmann's thrombasthenia (GT) patients show that the level of GPV and not of GPIIbIIIa correlates with the amount of FV. Biochemical approaches show that Gal8 can act as a cross-linker between FV and GPV, suggesting a potential role for GPV as Gal8 counter-receptor. Fibrinogen, the most abundant α-granule protein, can be taken up from plasma by both circulating platelets and megakaryocytes. Here, by mass spectrometry analysis of the platelet content from GT and LAD-III patients we explored the role of integrins on platelets cargo enrichment.The results show that GPIIbIIIa only contributes to the uptake of fibrinogen, and that activation of this integrin is not required to facilitate fibrinogen endocytosis. Novel differentially expressed proteins that may be associated with the investigated diseases are identified as well. Moreover MS analysis further revealed the genetic defect of a previously uncharacterized GT patient. In a different line of research, we explored the platelet secretory efficiency of α-granulesin patients with Familial Hemophagocytic Lymphohistiocytosis-5 (FHL-5). Disease-causing mutations have been mapped onto the gene encoding syntaxin-binding-protein2 (Munc18-2). We use MS, flow cytometry and electron microscopy to identify the protein product of a Munc18-2 splice-site variant (1247-1G>C) and the effect thereof on platelet degranulation. This Munc18-2 variant has been established to cause impaired degranulation of neutrophils. In spite of previous suggestions, flow cytometry, electron microscopy and MS studies reveal that the splice-site variant does not significantly affect platelet. Moreover MS analysis revealed the still unknown protein product resulting from the 1247-1G>C mutation.

Hemophilia A is an X chromosome-linked bleeding disorder caused by a reduction or com­plete absence of coagulation factor VIII (FVIII). The bleeding tendency in patients suffering from hemophilia A can be treated by regular intravenous administrations of FVIII. A severe complication that occurs in approximately 30% of hemophilia A patients is the recognition of administered FVIII as “non-self” by the recipients’ immune system, leading to the formation of FVIII-neutralizing antibodies. The initial steps in the formation of these antibodies are recognition and uptake of the administered FVIII by antigen-presenting cells, which is followed by presentation of FVIII fragments to the immune system on MHC class II complexes. In this thesis the mechanism of FVIII endocytosis and presentation on MHC class II by antigen-presenting cells was investigated. The thesis describes that the uptake of FVIII by antigen-presenting cells can be inhibited by shielding or modifying certain amino acids in the C1 domain of FVIII. A FVIII variant was constructed that is less immunogenic. Administration of this FVIII variant to hemophilic mice led to significantly less antibody formation than administration of normal FVIII. Additionally, it was shown that recognition of FVIII by antigen-presenting cells leads to donor-dependent differences in the presentation of FVIII fragments on MHC class II. The results described in this thesis contribute to the development of alternative treatments for hemophilia A patients resulting in less formation of antibodies against FVIII.

A defect in coagulation factor VIII (FVIII) results in the inherited bleeding disorder hemophilia A. Current treatment of hemophilia A is hampered by the need of frequent administration of costly FVIII products. Therefore gene therapy is an attractive alternative for protein replacement to treat hemophilia A patients. Recent insights have suggested that it may be beneficial to express FVIII in cells that also secrete its natural carrier protein von Willebrand factor (VWF), i.e. platelets and endothelial cells. In this thesis our aim was to explore the potential of endothelial cells as a cellular release-on-demand device for co-delivery of FVIII and VWF. We demonstrate that blood outgrowth endothelial cells (BOECs) transduced with a lentiviral vector encoding FVIII are capable of long-term production of high FVIII levels. A substantial part of FVIII synthesized by BOECs is stored with VWF in storages organelles called Weibel-Palade bodies (WPBs), and can be released upon stimulation of WPB exocytosis. Using confocal microscopy we observed that the appearance of FVIII-containing WPBs was round as opposed to the typical elongated shape of normal WPBs. We found that FVIII expression prohibits the formation of long VWF tubules that normally determine the elongated shape of WPBs. In addition, our data suggest that an intracellular interaction of FVIII with VWF impairs recruitment of platelets by VWF released from WPBs. The effect of FVIII expression on VWF stored in WPBs implies that FVIII and VWF associate with each other inside WPBs. Therefore we studied whether or not WPB-derived FVIII and VWF remain associated after release from WPBs. We found that FVIII co-stored with VWF in WPBs binds to secreted VWF strings without a need of the extracellular high-affinity VWF binding site Y1680. Our data further suggest that intracellular and extracellular FVIII-VWF complex assembly differ with regard to the contribution of the Y1680 residue in VWF binding. We also determined the role of VWF for storage of FVIII in WPBs by studying the sorting ability of FVIII variants displaying reduced VWF interaction. We found that FVIII sorting to WPBs is not affected by a reduced extracellular FVIII-VWF interaction. Moreover, using domain deletion mutants and FVIII-FV chimeras, we demonstrate that FVIII sorting to WPBs is driven by FVIII-specific structural elements in both C domains. In conclusion, BOECs comprise a promising cell type as these cells synthesize and store considerable quantities of FVIII with VWF in WPBs. Furthermore, FVIII is probably released from WPBs in complex with VWF as exocytosis of WPB content under flow results in strong association of FVIII with VWF strings.

A defect in coagulation factor VIII (FVIII) results in the inherited bleeding disorder hemophilia A. Current treatment of hemophilia A is hampered by the need of frequent administration of costly FVIII products. Therefore gene therapy is an attractive alternative for protein replacement to treat hemophilia A patients. Recent insights have suggested that it may be beneficial to express FVIII in cells that also secrete its natural carrier protein von Willebrand factor (VWF), i.e. platelets and endothelial cells. In this thesis our aim was to explore the potential of endothelial cells as a cellular release-on-demand device for co-delivery of FVIII and VWF. We demonstrate that blood outgrowth endothelial cells (BOECs) transduced with a lentiviral vector encoding FVIII are capable of long-term production of high FVIII levels. A substantial part of FVIII synthesized by BOECs is stored with VWF in storages organelles called Weibel-Palade bodies (WPBs), and can be released upon stimulation of WPB exocytosis. Using confocal microscopy we observed that the appearance of FVIII-containing WPBs was round as opposed to the typical elongated shape of normal WPBs. We found that FVIII expression prohibits the formation of long VWF tubules that normally determine the elongated shape of WPBs. In addition, our data suggest that an intracellular interaction of FVIII with VWF impairs recruitment of platelets by VWF released from WPBs. The effect of FVIII expression on VWF stored in WPBs implies that FVIII and VWF associate with each other inside WPBs. Therefore we studied whether or not WPB-derived FVIII and VWF remain associated after release from WPBs. We found that FVIII co-stored with VWF in WPBs binds to secreted VWF strings without a need of the extracellular high-affinity VWF binding site Y1680. Our data further suggest that intracellular and extracellular FVIII-VWF complex assembly differ with regard to the contribution of the Y1680 residue in VWF binding. We also determined the role of VWF for storage of FVIII in WPBs by studying the sorting ability of FVIII variants displaying reduced VWF interaction. We found that FVIII sorting to WPBs is not affected by a reduced extracellular FVIII-VWF interaction. Moreover, using domain deletion mutants and FVIII-FV chimeras, we demonstrate that FVIII sorting to WPBs is driven by FVIII-specific structural elements in both C domains. In conclusion, BOECs comprise a promising cell type as these cells synthesize and store considerable quantities of FVIII with VWF in WPBs. Furthermore, FVIII is probably released from WPBs in complex with VWF as exocytosis of WPB content under flow results in strong association of FVIII with VWF strings.

Coagulation factor VIII (FVIII) comprises multiple domains that in a concerted manner mediate the role of FVIII as a cofactor for activated factor IX (FIXa) in the coagulation cascade. These domains of FVIII are grouped in a heavy chain (domains A1-A2-B) and a light chain (domains A3-C1-C2). At sites of vascular injury, FVIII is activated and assembles with FIXa on procoagulant membranes comprising phosphatidylserine (PS) in the outer leaflet of the bilayer. Defects therein are associated with the bleeding disorder hemophilia A, which can be treated with intravenous infusion with purified FVIII. Effective treatment, however, is severely hampered by the fact that FVIII is rapidly removed from the circulation. One relevant biological modifier of the FVIII half-life is von Willebrand factor (VWF), which forms a tight complex with FVIII in the circulation. In this complex, FVIII is protected from even faster clearance from plasma. The role of the A domains and the C2 domain in the interaction with phospholipid membranes, FIXa, VWF and the FVIII clearance receptors has been extensively studied. The potential function of the C1 domain in these molecular mechanisms has been hardly investigated and is therefore poorly understood. Yet, anti-C1 domain antibodies have been identified that effectively inhibit multiple FVIII functions. We investigated the role of the C1 domain in the biological processes that involve FVIII by using single chain variable domain antibody fragment (scFV) KM33 that requires residues 2092-2093 in the C1 domain for binding FVIII. Furthermore, we utilised FVIII YFP constructs in which Lys-2092 and/or Phe-2093 were replaced by alanines. The results show that this scFV is particularly effective in blocking the interaction of FVIII with phospholipid membranes and with activated platelets. We furthermore report that C1 domain residues 2092 and 2093 contribute to binding phospholipid membranes comprising a low percentage of PS. To assess the role of these residues in the interaction with physiological relevant procoagulant surfaces, we performed functional studies employing the FVIII YFP derivatives and activated platelets. The results show that residues 2092-2093 are also of major importance for the interaction of FVIII with the activated platelets. Employing LRP-expressing U87MG cells, we furthermore demonstrated that this C1 domain region is also critical for the uptake of FVIII. An intriguing finding is that there is a cell surface binding step for FVIII that is independent of LRP. This binding step is of critical importance in the mechanism of LRP-dependent uptake of FVIII. We show that the C1 domain residues 2092-2093 contribute to both cell surface binding as well as receptor binding and found that the cell binding step of FVIII is not mediated by PS. Furthermore, we determined the role of VWF in endocytosis of FVIII and investigated the inhibitory mechanism of KM33 and variants thereof on the biological processes where FVIII is involved. The results described in the present thesis contribute to the knowledge of the function of coagulation FVIII.

Coagulation factor VIII (FVIII) comprises multiple domains that in a concerted manner mediate the role of FVIII as a cofactor for activated factor IX (FIXa) in the coagulation cascade. These domains of FVIII are grouped in a heavy chain (domains A1-A2-B) and a light chain (domains A3-C1-C2). At sites of vascular injury, FVIII is activated and assembles with FIXa on procoagulant membranes comprising phosphatidylserine (PS) in the outer leaflet of the bilayer. Defects therein are associated with the bleeding disorder hemophilia A, which can be treated with intravenous infusion with purified FVIII. Effective treatment, however, is severely hampered by the fact that FVIII is rapidly removed from the circulation. One relevant biological modifier of the FVIII half-life is von Willebrand factor (VWF), which forms a tight complex with FVIII in the circulation. In this complex, FVIII is protected from even faster clearance from plasma. The role of the A domains and the C2 domain in the interaction with phospholipid membranes, FIXa, VWF and the FVIII clearance receptors has been extensively studied. The potential function of the C1 domain in these molecular mechanisms has been hardly investigated and is therefore poorly understood. Yet, anti-C1 domain antibodies have been identified that effectively inhibit multiple FVIII functions. We investigated the role of the C1 domain in the biological processes that involve FVIII by using single chain variable domain antibody fragment (scFV) KM33 that requires residues 2092-2093 in the C1 domain for binding FVIII. Furthermore, we utilised FVIII YFP constructs in which Lys-2092 and/or Phe-2093 were replaced by alanines. The results show that this scFV is particularly effective in blocking the interaction of FVIII with phospholipid membranes and with activated platelets. We furthermore report that C1 domain residues 2092 and 2093 contribute to binding phospholipid membranes comprising a low percentage of PS. To assess the role of these residues in the interaction with physiological relevant procoagulant surfaces, we performed functional studies employing the FVIII YFP derivatives and activated platelets. The results show that residues 2092-2093 are also of major importance for the interaction of FVIII with the activated platelets. Employing LRP-expressing U87MG cells, we furthermore demonstrated that this C1 domain region is also critical for the uptake of FVIII. An intriguing finding is that there is a cell surface binding step for FVIII that is independent of LRP. This binding step is of critical importance in the mechanism of LRP-dependent uptake of FVIII. We show that the C1 domain residues 2092-2093 contribute to both cell surface binding as well as receptor binding and found that the cell binding step of FVIII is not mediated by PS. Furthermore, we determined the role of VWF in endocytosis of FVIII and investigated the inhibitory mechanism of KM33 and variants thereof on the biological processes where FVIII is involved. The results described in the present thesis contribute to the knowledge of the function of coagulation FVIII.

Summary Factor (F)VIII is an essential cofactor in the enzymatic coagulation cascade. A defect in the gene encoding for FVIII results in the X-linked inherited bleeding disorder hemophilia A. In the circulation, FVIII is protected from proteolytic degradation and premature clearance by forming a complex with its carrier protein von Willebrand factor (VWF). While the complex of FVIII and VWF involves one of the strongest protein-protein interactions in the human body, it has remained unclear when and where FVIII and VWF first interact. The most widely accepted view is that FVIII and VWF are secreted from different cells and form a complex in the circulation, merely by virtue of their high-affinity interaction. During the last decade however, increasing evidence has been presented that suggests that the human body should also comprise a releasable storage pool of both proteins. In this thesis, we have studied the relationship between high-affinity interaction in the circulation and intracellular co-storage of FVIII and VWF. We have taken a new approach towards this issue by creating fluorescently tagged VWF and FVIII proteins that are established as being deficient in high-affinity interaction. We have demonstrated that FVIII is stored in VWF-containing storage organelles in absence of high-affinity interaction and that the extent of FVIII storage is independent of its affinity for VWF. In addition, we studied the therapeutic benefit of intracellular co-expression of FVIII and VWF in terms of potential treatment options for hemophilia A. We have shown that blood outgrowth endothelial cells can produce high levels of functional FVIII over a prolonged period of time. In addition, we have demonstrated that a substantial amount of FVIII can be stored in VWF-containing storage organelles whereas the majority of FVIII is released in a constitutive manner. From our data, we conclude that the structural requirements that determine high-affinity interaction between FVIII and VWF in the circulation differ from those that determine intracellular co-trafficking. Our data provide a molecular explanation for the fact that hemophilia A and von Willebrand’s disease patients that suffer from impaired complex assembly of FVIII and VWF in the circulation show concomitant increase of FVIII and VWF plasma levels following treatment with desmopressin. The mechanism that determines FVIII storage in VWF-containing granules as well as the therapeutic potential of FVIII expression in VWF producing cells now awaits further exploration.