Wild-type Factor VIII Research Articles

Rapid clearing CT-001 restored hemostasis in mice with coagulopathy induced by activated protein C.

Activated protein C (APC) is one of the mechanisms contributing to coagulopathy, which is associated with high mortality. The counteraction of the APC pathway could help ameliorate bleeding. However, patients also transform frequently from a hemorrhagic state to a prothrombotic state at a later time. Therefore, a prohemostatic therapeutic intervention should take this thrombotic risk into consideration. CT-001 is a novel factor VIIa (FVIIa) with enhanced activity and desialylated N-glycans for rapid clearance. We assessed CT-001 clearance in multiple species and its ability to reverse APC-mediated coagulopathic blood loss. The N-glycans on CT-001 were characterized by liquid chromatography-mass spectrometry. Three species were used to evaluate the pharmacokinetics of the molecule. The potency and efficacy of CT-001 under APC pathway-induced coagulopathic conditions were assessed by coagulation assays and bleeding models. The N-glycosylation sites of CT-001 had high occupancy of desialylated N-glycans. CT-001 exhibited 5 to 16 times higher plasma clearance in human tissue factor knockin mice, rats, and cynomolgus monkeys than wildtype FVIIa. CT-001 corrected the activated partial thromboplastin time and thrombin generation of coagulopathic plasma to normal in in vitro studies. In an APC-mediated saphenous vein bleeding model, 3 mg/kg of CT-001 reduced bleeding time in comparison with wildtype FVIIa. The correction of bleeding by CT-001 was also observed in a coagulopathic tail amputation severe hemorrhage mouse model. The efficacy of CT-001 is independent of the presence of tranexamic acid, and the combination of CT-001 and tranexamic acid does not lead to increased thrombogenicity. CT-001 corrected APC pathway-mediated coagulopathic conditions in preclinical studies and could be a potentially safe and effective procoagulant agent for addressing APC-mediated bleeding.

Antidrug antibodies against the polyethylene glycol moiety inhibit the procoagulant activity of therapeutic polyethylene glycolated factor VIII

BackgroundThe standard therapy for patients with hemophilia A (HA) is the replacement with factor VIII (FVIII) therapeutics. To overcome the limitation of short half-life of wild-type FVIII protein, polyethylene glycol (PEG) can be coupled to therapeutic FVIII to improve pharmacokinetics. ObjectivesWe aimed to characterize antibodies developed against a FVIII therapeutic PEGylated with a 40-kDa PEG (40PEG-BDDFVIII) in 2 patients with mild HA. MethodsAn inhouse bead-based immunoassay was developed to characterize and confirm the specificity of the detected antibodies. The neutralizing nature of the antibodies toward PEGylated therapeutics was determined by a modified Nijmegen-Bethesda assay. ResultsTwo out of 46 patients treated with 40PEG-BDDFVIII developed inhibitory antibodies toward the drug. Switching to a non-PEGylated FVIII successfully increased the FVIII activity in both patients. In patient 1, antibodies were raised against FVIII and PEG. Anti-FVIII antibodies were of the immunoglobulin (Ig)G isotype, whereas anti-PEG antibodies were of IgG, IgM, and IgA isotypes. In patient 2, antibodies of IgG and IgA isotypes were directed only against the PEG moiety. Competitive assays confirmed the specificity of the antibodies against PEG. The applied Nijmegen-Bethesda assay revealed that patients’ anti-PEG antibodies and AGP3, an antibody against the backbone of PEG, can inhibit all currently available PEGylated therapeutics but to different degrees. No inhibitory FVIII antibodies were detected. ConclusionAntibodies against the PEG moiety of 40PEG-BDDFVIII abolished the efficacy of the drug. This is the first report on real-world experiences with the development of neutralizing anti-PEG antibodies after treatment with PEGylated FVIII therapeutics in mild HA.

CT-001 is a rapid clearing factor VIIa with enhanced clearance and hemostatic activity for the treatment of acute bleeding in non-hemophilia settings

IntroductionAcute bleeding leads to significant morbidity and mortality. Recombinant wildtype Factor VIIa (WT FVIIa) had been reported to have some therapeutic effects in some clinical trials, however, its use was associated with thromboembolic events. We sought to develop a novel FVIIa molecule (CT-001) with enhanced activity and lowered thrombogenicity risk. Methods and methodsCT-001 has 4 N-glycans (T106N/N145/V253N/N322) with terminal sialic acid residues removed to promote active clearance via the asialoglycoprotein receptor, and P10Q/K32E substitutions introduced to its gamma-carboxyglutamic acid (Gla) domain for enhanced phospholipid affinity and activity. ResultsIn mice, CT-001 had a half-life of 5 min and a clearance of 467 mL/h/kg at 3 mg/kg, significantly faster than WT FVIIa (t1/2 = 1.8 h, Cl = 39 mL/h/kg). Interestingly, CT-001 was efficacious in reducing blood loss even with its rapid clearance. In a severe hemorrhage mouse model with tail amputated 5 cm from the tip, 1 mg/kg CT-001 provided efficacy comparable to 3 mg/kg WT FVIIa. The fast clearance of CT-001 resulted in significantly reduced thrombogenicity in comparison to WT FVIIa in a FeCl3-induced carotid artery thrombosis mouse model, and further confirmed in a soluble tissue factor-induced thrombosis model. ConclusionsThe data on CT-001 demonstrate that a short duration of highly active FVIIa procoagulant activity has the potential to be an optimal paradigm for the treatment of acute bleeds.

FVIII Proteins with a Modified Immunodominant T-Cell Epitope Exhibit Reduced Immunogenicity and Normal Procoagulant Activity

Factor VIII (FVIII) neutralizing antibodies (inhibitors) are a serious complication in hemophilia A (HA). The peptide FVIII2194-2213 (legacy FVIII numbering, sequence: SYFTNMFATWSPSKARLHLQ) has been shown to contain an immunodominant HLA-DRA*01-DRB1*01:01 (DRB1*01:01)-restricted epitope recognized by CD4+ T-effector cells isolated from HA subjects. The aim of this study was to identify amino acid substitutions to de-immunize this epitope while retaining pro-coagulant function and expression levels comparable to those of wild-type (WT) FVIII proteins. The minimal DRB1*01:01-binding motif was mapped to FVIII2194-2205 by quantitative peptide-MHC Class II (MHCII) binding assays. Further binding assays with sequence-modified FVIII peptides established that the residues important for DRB1*01:01 affinity were F2196, M2199, A2201, and S2204, corresponding to MHCII anchor positions 1, 4, 6 and 9. T-cell proliferation experiments with Ala-substituted FVIII2194-2205 peptides identified F2196A as a substitution that abrogated proliferation and cytokine secretion of five clones specific for FVIII2194-2213. The substitution F2196A was then engineered into a recombinant (r)FVIII-C2 protein. T-cell clones that were strongly stimulated by wild-type rFVIII-C2 did not proliferate when cultured with rFVIII-C2-F2196A, indicating the immunogenic synthetic peptide includes a naturally processed T-cell epitope. The measured and predicted effects of additional amino acid substitutions within this binding motif were further evaluated by peptide-MHCII binding assays, CD4+ T-cell proliferation assays, and MHCII-binding prediction algorithms. Amino acid homologs within this binding motif were evaluated with sequence homologies to FVIII and FV proteins from different species. On the basis of all these results, six rB-domain-deleted (BDD)-FVIII proteins with amino acid substitutions F2196A, F2196L, F2196K, M2199A, M2199W or M2199R, respectively, were designed and produced; all but the F2196A variant had specific activities comparable to that of rWT-BDD-FVIII. The F2196A variant also had the lowest expression level among the rBDD-FVIIII proteins produced. Peptide-MHCII binding assays confirmed that FVIII2194-2205 peptides with substitutions F2196A, F2196K, M2199A and M2199R did not bind to DRB1*01:01 or to an additional 9 HLA-DR proteins tested, with the exception of the M2199A peptide binding to DRB1*09:01. The predicted effects of these FVIII sequence modifications on binding to additional MHCII alleles included in reference sets representing ~99% of the human population HLA diversity were evaluated using ProPred and Immune Epitope Database algorithms. These programs correctly predicted that the substitutions would abrogate binding to DRB1*01:01 and also indicated that affinities for additional MHCII would be reduced by each substitution. The substitutions were predicted to increase peptide-MHCII affinities for a smaller number of MHCII alleles, indicating a possible risk of creating neo-epitopes restricted to these specific HLA alleles. Importantly, proliferation of patient-derived T-cell clones in response to rBDD-FVIII-F2196K and rBDD-FVIII-M2199A proteins was dramatically reduced compared with responses to rWT-BDD-FVIII. We have previously demonstrated that a severe HA inhibitor subject had an oligoclonal T-cell response to only one T-cell epitope within the FVIII A2, C1 and C2 domains and that one epitope was FVIII2194-2213 (Ettinger et al., Blood 2106; 128(16):2043). This observation, together with the results presented here, further suggests that modification of only a small number of immunodominant epitopes may be sufficient to significantly de-immunize FVIII and decrease FVIII inhibitors in HA patients. The present results provide proof-of-principle for the design of less immunogenic FVIII proteins targeted to specific subsets of HA patients. DisclosuresEttinger:Bloodworks NW: Patents & Royalties: Inventor on FVIII patents. James:Bloodworks NW: Patents & Royalties: inventor on FVIII patent(s). Pratt:Bloodworks NW: Patents & Royalties: inventor on FVIII patents.

A Spacer Sequence Inserted between the A1 and A2 Domains of Factor VIII Overcomes the Requirement for Proteolytic Cleavage at Arg 372 in the Generation of Cofactor Activity

Factor VIII (FVIII) with a multi-domain structure (A1-a1-A2-a2-B-a3-A3-C1-C2) is a procofactor and precursor for the anti-hemophilic cofactor protein, FVIIIa. Following the intracellular processing within the B domain, secreted FVIII circulates as a heterodimer with variably sized (90K-200K) heavy chain (A1-a1-A2-a2-B) and an 80K light chain (a3-A3-C1-C2). Proteolytic activation of FVIII by thrombin that yields heterotrimeric FVIIIa (A1-a1/A2-a2/A3-C1-C2), the cofactor for intrinsic tenase, involves cleavage of three peptide bonds between Arg372-Ser373, Arg740-Ser741, and Arg1689-Ser1690. Cleavage at Arg740 removes the B-domain, and cleavage at Arg1689 removes the a3-acidic region and releases FVIII from vWF, its carrier protein, and exposes membrane binding sites within the FVIII light chain. Cleavage at Arg372 separates A1-a1 and A2-a2 domains and is implicated in the cofactor-dependent recognition and enhancement in the rate of factor X (FX) activation by intrinsic tenase. Subsequently, the separated A2-a2 domain dissociates spontaneously from the heterotrimeric FVIIIa resulting in the rapid loss of cofactor activity. We speculated that the requirement for cleavage at Arg372 might be obviated by the insertion of an optimized linker sequence between A1-a1 and A2-a2 domains on an uncleavable Gln372 backbone. To investigate this possibility, we prepared cDNA constructs of B-domain deleted FVIII variants; FVIII wild-type (FVIIIWT), FVIII372Q, and FVIII372Q followed by a rigid (Ala-Pro)5 linker sequence (FVIII372Q-AP5). All three FVIII constructs were stably transfected into BHK cells and high expressing clones were selected by one stage aPTT and western blotting of expression media. Selected stable clones were further expanded to collect 15L of expression media over 5-day period, and recombinant FVIII variants were purified using a three-step chromatographic approach. These FVIII variants were studied using SDS-PAGE, western blotting, aPTT assays, thrombin generation assay (TGA) and purified assays to assess kinetics of FX activation and spontaneous loss of cofactor activity. In contrast to FVIIIWT, FVIII372Q and FVIII372Q-AP5 were completely resistant to cleavage at Gln372 by thrombin, yielding bands corresponding to A1-a1-A2-a2 (90K) and A3-C1-C2 (73K). In one stage aPTT assays, FVIII372Q showed prolonged clotting times with specific activity in the range of 200-400 U/mg, while FVIIIWT and FVIII372Q-AP5 displayed comparable clotting times with specific activities ranging between 8000-10000 U/mg and 4500-5500 U/mg, respectively. In TGA initiated with either 0.1 pM tissue factor or 1 pM factor XIa, both FVIIIWT and FVIII372Q-AP5 displayed similar TGA profiles. In steady state kinetic studies of FX activation using limiting concentrations of factor IXa, saturating concentrations of FVIII variants pretreated with thrombin, membranes and increasing concentrations of FX, the cofactor function of thrombin-cleaved FVIII372Q was severely impaired. However, despite lack of cleavage at Gln372 in FVIII372Q-AP5, catalytic efficiency for FX activation by intrinsic tenase assembled by this variant was comparable to that seen with FVIIIaWT. At the physiological concentration of FX, the initial velocity for Xa formation (v/E) for intrinsic tenase assembled with FVIIIa372Q-AP5 was within a factor of 2 of that observed with FVIIIaWT while the rate observed with FVIIIa372Q was >10-fold lower. Following rapid activation with thrombin, loss of cofactor function was significantly slower for FVIIIa372Q-AP5(t1/2 ~ 10 min) compared to FVIIIaWT (t1/2 ~ 2 min). Our findings indicate that the requirement for cleavage at Arg372 for the development of full FVIIIa cofactor function can be overcome by modulating the A1-A2 connector with an optimized linker sequence. Failure to yield an infinitely stable cofactor in the case of FVIIIa372Q-AP5 suggests that cleavage at Arg372 does not solely explain the spontaneous loss of FVIIIa cofactor function. Disclosures Krishnaswamy: Bayer: Research Funding.

Blood coagulation factor VIII D1241E polymorphism leads to a weak malectin interaction and reduction of factor VIII posttranslational modification and secretion

Blood coagulation factor VIII (FVIII) is a key cofactor in regulation of blood coagulation. This study investigated the mechanism by which FVIII is translated and transported into the endoplasmic reticulum (ER) and processed in the Golgi apparatus before secretion using an in vitro cell model. HEK-293T cells were transfected with vectors carrying wild-type (WT) FVIII or polymorphic FVIII D1241E for coexpression with ER lectins and treatment with tunicamycin (an N-linked glycosylation inhibitor), 1-deoxynojirimycin (an alpha-glucosidase inhibitor), endoglycosidase H, or MG132 (Cbz-Leu-Leu-leucinal; a proteasome inhibitor). The data showed that the minor allele of FVIII D1241E was able to reduce FVIII secretion into the conditioned medium but maintain a normal level of procoagulation ability, although both FVIII WT and the minor allele of FVIII D1241E showed similar levels of transcription and translation capacities. Functionally, the D1241E polymorphism led to a reduced level of FVIII in the Golgi apparatus because of its reduced association with malectin, which interacts with newly synthesized glycoproteins in the ER for FVIII folding and trafficking, leading to degradation of the minor allele of FVIII D1241E in the cytosol. This study demonstrated that malectin is important for regulation of the FVIII posttranslational process and that the minor allele of FVIII D1241E had a reduced association with malectin but an increased capacity for proteasomal FVIII degradation. These data imply the role of the ER quality control in future recombinant FVIII development.

Functional identification of factor VIII B domain regions in hepatocyte cells

Factor VIII (FVIII) functions as a cofactor within the intrinsic pathway of blood coagulation in process of FX activation by FIXa, for which deficiency results in the bleeding disorder hemophilia A. The gene of FVIII contains 26 exons that code for a 19 amino acid signal peptide and a 2332 amino acid polypeptide with a domain structure designated A1-A2-B-A3-C1-C2, of which the A domains are homologous with each other, as are the C domains. It has been well-documented that both the domains are the necessary elements for FVIII activities. The B domain is highly glycosylated and has a variable sequence, even among FVIIIs from different species. The B domain plays versatile roles in FVIII lifespan except for coagulation activity, but the functional characteristics of its specific regions remain still obscure. A series of recombinant FVIIIs (rFVIIIs) with B domain truncated were constructed and transiently expressed in hepatocyte cells. Media and cell lysates were collected after 72 h for the analyses of FVIII biosynthesis, secretion, activity and stability in ex vivo plasma relative to the full length wild-type FVIII. Unexpectedly, various regions in B domain exhibited different contribution to these functionalities. The discovery might facilitate the bioengineered rFVIIIs and gene therapeutics.

Restore the Secretion and Activity of Factor VIII Mutants By Proteostasis Regulator

Quantitative or qualitative deficiency in factor VIII (FVIII) result in hemophilia A. FVIII deficiency is caused by broad spectrum mutations, and missense mutation is the most common type of mutation, which lead to various defects of FVIII from decreased secretion to impaired function. Cellular protein homeostasis is disturbed by FVIII mutants, and thus excessive misfolding and degradation of FVIII mutant result in FVIII deficiency. We hypothesized that folding and/or trafficking of FVIII can be enhanced by proteostasis regulators, leading to increased coagulation factor levels sufficient to restore hemostasis. We performed a cell-based screening using a FDA-approved drug library and a natural compound library looking for compounds that enhance secretion of FVIII with a B domain deletion. We identified SAHA (Suberanilohydroxamic acid) as a compound that increased both wild-type (WT) and a missense mutant FVIII. SAHA is a histone deacetylase inhibitor that was also reported to enhance proteostaisis of mutant CFTR, a-1 antitrypsin, lysosomal glucocerebrosidase and the GABA receptor. Here we show that SAHA can increase the secretion and activity of FVIII mutants by enhancing their folding and trafficking as well transcription. To confirm the ability of SAHA to enhance FVIII mutant secretion, we first used the WT FVIII A2 domain and the M614T mutant A2 domain to detect the proper working time and concentration of SAHA in HEK 293T cells. M614T is a missense mutation that was shown to cause secretion defect of FVIII in our previous study. Western blot showed a dose-dependent increase in levels of both A2-WT and A2-M614T in cell culture media between 0 and 2.5ug/ml of SAHA after 24-hour treatment. Levels of both WT and mutant A2 domain dropped when SAHA concentration exceeded 5 ug/ml due to cellular toxicity. SAHA enhanced the secretion of A2-M614T much more than that of A2-WT. Of note, protein levels of both A2-WT and A2-M614T in cell lysate were also slightly elevated by SAHA. Time course study showed that the maximum effect on secretion was achieved after 24 hours of SAHA treatment. We next studied the effect of SAHA on activity and secretion of several full-length FVIII mutants (A469S, R527W, R531G, I566T, N582D, S584T, R593C, M614T, R1997G, R2150C), which were shown to decrease FVIII secretion due to various defects in intracellular trafficking in our previous studies. Except for N582D, activity levels of all FVIII mutants in cell culture media were elevated 3-14 folds after a 24-hour SAHA treatment (P<0.05), whereas the activity level of FVIII WT only increased by ~2 folds (P<0.05). Antigen levels of FVIII mutants in both cell lysates and cell culture media increased by ~3 folds following SAHA treatment (P<0.05). The increase in activities of several FVIII mutants (I566T, S584T, R593C, R1997G, R2150C) is higher than the increase in antigen levels of the same mutants in cell culture media, which indicates that, in addition to enhancing secretion, SAHA treatment also increased the proportions of functional proteins in secreted FVIII mutants in cell culture media. To investigate the mechanism, we studied the interactions of WT FVIII and several responsive mutant FVIII (I566T, S584T, R593C and R1997C) with endoplasmic reticulum (ER) chaperones BiP and calreticulin by co-Immunoprecipitation. The protein levels of BiP and calreticulin in HEK 293T cells were not affected by SAHA. However, the interactions of BiP with both WT FVIII and FVIII mutants were remarkably enhanced by SAHA, while no significant differences in the interactions of calreticulin with FVIII WT or FVIII mutants were observed. The effect of SAHA on FVIII secretion can be blocked by simultaneously treating cells with a BiP inhibitor SubAB. We next measured mRNA levels of BiP as well as FVIII in cells expressing FVIII WT and FVIII mutants with or without SAHA treatment via real time PCR. Similar to protein levels, BiP mRNA levels were not changed by SAHA. However, FVIII mRNA levels were elevated by 1.2 to 3.1 folds (P<0.05) by SAHA, in keeping with the histone deacetylase inhibitor activity of SAHA. In summary, we have shown that SAHA increases FVIII antigen and activity levels in cell culture media by increasing FVIII mRNA levels and by enhancing its folding and trafficking via specifically promoting its interaction with BiP. SAHA is a potential proteostaisis regulator that can increase the secretion and restore the function of certain FVIII missense mutants. Disclosures No relevant conflicts of interest to declare.

The Endothelial Protein C Receptor (EPCR) Regulates Endogenous Factor VII Levels in Mice

The endothelial protein C receptor (EPCR) has been demonstrated to bind activated FVII (FVIIa) through the Gla domain with equal affinity to Protein C (PC). Mouse studies suggest that EPCR is involved in the extravasation of infused human FVIIa, leading to an extended extravascular tissue persistence, longer than expected based on its circulating half-life. This provides a plausible explanation for the long-term benefits of hemophilic patients on human FVIIa prophylaxis. Collectively, these data suggest that EPCR sequesters administered FVIIa in tissues where it may have a hemostatic effect. However, the role of the endogenous FVII-EPCR interaction in normal conditions is largely unknown. For this, we have developed a mouse model to better understand this interaction in vivo. Endogenous mouse FVII and FVIIa (mFVII/FVIIa) do not bind mouse EPCR. However, our laboratory has demonstrated that L4F, L8M, and T9R substitutions in the Gla domain of mFVIIa enable its interaction with mouse EPCR while retaining full enzymatic activity in vitro. Based on that data, we utilized CRISPR/Cas9 technology to knock-in L4F, L8M, and T9R into the mFVII Gla domain in the mouse F7 locus (F7FMR), thereby developing mice with a chimeric endogenous FVII capable of binding EPCR. Founder animals were generated and capable of producing offspring, indicating that the gain-of-function in mFVII was compatible with life. Animals were subsequently backcrossed to wildtype C57BL/6 mice in order to remove potential off-target effects of the CRISPR/Cas9. Resultant heterozygous animals (F7FMR/WT) from the final cross were bred to generate F7FMR/FMR, F7FMR/WT, and F7WT/WTlittermates. We generated 59 male and 52 female animals and a binomial distribution test demonstrated that sex is equally distributed in the population. Moreover, the genotypes expected from the heterozygous crosses were inherited in a 1:2:1 ratio, further indicating that the gain-of-function in FVII is not lethal during development. As additional metrics of health, we measured weight longitudinally during weeks 1-10 of life and found no differences between the three genotypes for either gender. Complete blood counts (CBCs) revealed no differences between the F7FMR/FMR, F7FMR/WT, and F7WT/WTgenotypes, with the exception of a mild elevation in F7FMR/WTanimals compared to animals with wildtype FVII. Collectively, we found that the gain-of-function in EPCR binding by endogenous FVII is not detrimental to the overall health of the mice. Subsequently, we determined the mFVII levels in the F7FMR/FMR, F7FMR/WT, and F7WT/WTanimals using an in-house ELISA. We observed that plasmatic mFVII levels were dependent on the EPCR-binding capacity of the endogenous mFVII. Specifically, F7WT/WTmice, whose mFVII does not bind EPCR, had a plasmatic mFVII concentration of ~690 ng/ml. In contrast, F7FMR/FMRhomozygote mice had ~350 ng/ml of mouse FVII, approximately half the plasma levels of the F7WT/WT. Heterozygote animals F7FMR/WThad an intermediate plasmatic mFVII level (~550 ng/ml), suggesting that EPCR may regulate plasmatic FVII levels in vivo. Lastly, we determined the hemostatic response to injury in the F7FMR/FMR, F7FMR/WT, and F7WT/WTanimals. We did this in two ways, by measuring blood loss following tail clip assay and by determining time to vessel occlusion following ferric chloride injury of the carotid artery. We observed no differences between the three genotypes in response to either injury model. In conclusion, we have generated and characterized a novel mouse model in which endogenous FVII is capable of binding EPCR. Using this model, we demonstrated that EPCR can modulate plasmatic FVII levels in vivo but does not appear to affect hemostasis. Since this model mimics the FVII-EPCR interaction in humans, it can now be used to further investigate how this interaction participates in other normal or pathologic states that depend on FVII and/or EPCR. Disclosures Margaritis: Bayer Hemophilia Awards: Research Funding; Bristol-Myers Squibb: Other: Salary (spouse); CSL Behring: Other: Salary (spouse); NovoNordisk A/S: Research Funding.

N-Glycan-calnexin interactions in human factor VII secretion and deficiency.

Factor VII (FVII) is a key serine protease in blood coagulation. N-glycosylation in FVII has been shown to be critical for protein secretion. To date, however, the underlying biochemical mechanism remains unclear. Recently, we found that N-glycans in the transmembrane serine protease corin are critical for calnexin-assisted protein folding and extracellular expression. In this study, we tested the hypothesis that N-glycans in the FVII protease domain mediate calnexin-assisted protein folding and that naturally occurring F7 mutations abolishing N-glycosylation impair FVII secretion. We expressed human FVII wild-type (WT) and mutant proteins lacking one or both N-glycosylation sites in HEK293 and HepG2 cells in the presence or absence of a glucosidase inhibitor. FVII expression, secretion and binding to endoplasmic reticulum chaperones were examined by immune staining, co-immunoprecipitation, Western blotting, and ELISA. We found that N-glycosylation at N360 in the protease domain, but not N183 in the pro-peptide domain, of human FVII is required for protein secretion. Elimination of N-glycosylation at N360 impaired calnexin-assisted FVII folding and secretion. Similar results were observed in WT FVII when N-glycan-calnexin interaction was blocked by glucosidase inhibition. Naturally occurring F7 mutations abolishing N-glycosylation at N360 reduced FVII secretion in HEK293 and HepG2 cells. These results indicate that N-glycans in the FVII protease domain mediate calnexin-assisted protein folding and subsequent extracellular expression. Naturally occurring F7 mutations abolishing N-glycosylation in FVII may impair this mechanism, thereby reducing FVII levels in patients.

Human platelets express endothelial protein C receptor, which can be utilized to enhance localization of factor VIIa activity

Background High-dose factor VIIa (FVIIa) is routinely used as an effective bypassing agent to treat hemophilia patients with inhibitory antibodies that compromise factor replacement. However, the mechanism by which FVIIa binds activated platelets to promote hemostasis is not fully understood. FVIIa-DVQ is an analog of FVIIa with enhanced tissue factor (TF)-independent activity and hemostatic efficacy relative to FVIIa. Our previous studies have shown that FVIIa-DVQ exhibits greater platelet binding, thereby suggesting that features in addition to lipid composition contribute to platelet-FVIIa interactions. Objectives Endothelial cell protein C receptor (EPCR) also functions as a receptor for FVIIa on endothelial cells. We therefore hypothesized that an interaction with EPCR might play a role in platelet-FVIIa binding. Methods/results In the present study, we used flow cytometric analyses to show that platelet binding of both FVIIa and FVIIa-DVQ is partially inhibited in the presence of excess protein C or an anti-EPCR antibody. This decreased binding results in a corresponding decrease in the activity of both molecules in FXa and thrombin generation assays. Enhanced binding to EPCR was sufficient to account for the increased platelet binding of FVIIa-DVQ compared with wild-type FVIIa. As EPCR protein expression has not previously been shown in platelets, we confirmed the presence of EPCR in platelets using immunofluorescence, flow cytometry, immunoprecipitation, and mass spectrometry. Conclusions This work represents the first demonstration that human platelets express EPCR and suggests that modulation of EPCR binding could be utilized to enhance the hemostatic efficacy of rationally designed FVIIa analogs.

Molecular mechanisms of missense mutations that generate ectopic N-glycosylation sites in coagulation factor VIII.

N-glycosylation is a common posttranslational modification of secreted and membrane proteins, catalyzed by the two enzymatic isoforms of the oligosaccharyltransferase, STT3A and STT3B. Missense mutations are the most common mutations in inherited diseases; however, missense mutations that generate extra, non-native N-glycosylation sites have not been well characterized. Coagulation factor VIII (FVIII) contains five consensus N-glycosylation sites outside its functionally dispensable B domain. We developed a computer program that identified hemophilia A mutations in FVIII that can potentially create ectopic glycosylation sites. We determined that 18 of these ectopic sites indeed become N-glycosylated. These sites span the domains of FVIII and are primarily associated with a severe disease phenotype. Using STT3A and STT3B knockout cells, we determined that ectopic glycosylation exhibited different degrees of dependence on STT3A and STT3B. By separating the effects of ectopic N-glycosylation from those due to underlying amino acid changes, we showed that ectopic glycans promote the secretion of some mutants, but impair the secretion of others. However, ectopic glycans that enhanced secretion could not functionally replace a native N-glycan in the same domain. Secretion-deficient mutants, but not mutants with elevated secretion levels, show increased association with the endoplasmic reticulum chaperones BiP (immunoglobulin heavy chain-binding protein) and calreticulin. Though secreted to different extents, all studied mutants exhibited lower relative activity than wild-type FVIII. Our results reveal differential impacts of ectopic N-glycosylation on FVIII folding, trafficking and activity, which highlight complex disease-causing mechanisms of FVIII missense mutations. Our findings are relevant to other secreted and membrane proteins with mutations that generate ectopic N-glycans.

Functional and Molecular Characterization of C91S Mutation in the Second Epidermal Growth Factor-Like Domain of Factor VII.

BackgroundCoagulation Factor VII is a vitamin K-dependent serine protease which has a pivotal role in the initiation of the coagulation cascade. The congenital Factor VII deficiency is a recessive hemorrhagic disorder that occurs due to mutations of F7 gene. In the present study C91S (p.C91S) substitution was detected in a patient with FVII deficiency. This mutation has not been characterized by a functional study.ObjectivesIn this study, we aimed to evaluate the impact of C91S substitution on factor VII expression and function.Materials and MethodsThe F7 complete cDNA was isolated from HepG2 cell line and inserted into the pcDNA3.1 mammalian expression vector. The desired mutation was generated by the site-directed mutagenesis and the wild-type and mutated constructs were transfected into CHO-K1 cells. The protein activity and antigen level (antigen concentration) were validated in the culture medium and cell lysate of the transiently transformed cells. An immunocytochemistry procedure was also performed to evaluate the intracellular localization of the mutated and the wild-type FVII, as well.ResultsThe present in vitro study has demonstrated that C91S antigen expression was increased in the transfected CHO-K1 cells compared to the wild-type (WT) protein. Despite an increased protein secretion, the factor VII coagulant activity was diminished following C91S substitution when it was assessed by a standard one-stage analysis. In addition, the immunocytochemistry procedure revealed that there was no difference in the intracellular localization of the C91S mutated FVII compared to the WT protein.ConclusionsOur results present that C91S mutation has an effect on the coagulation activity, secretion, biosynthesis, and probably folding of the FVII leading to the FVII deficiency.

Factor VII deficiency: Unveiling the cellular and molecular mechanisms underlying three model alterations of the enzyme catalytic domain

Activated factor (F) VII is a vitamin K-dependent glycoprotein that initiates blood coagulation upon interaction with tissue factor. FVII deficiency is the most common of the rare congenital bleeding disorders. While the mutational pattern has been extensively characterized, the pathogenic molecular mechanisms of mutations, particularly at the intracellular level, have been poorly defined. Here, we aimed at elucidating the mechanisms underlying altered FVII biosynthesis in the presence of three mutation types in the catalytic domain: a missense change, a microdeletion and a frameshift/elongation, associated with severe or moderate to severe phenotypes. Using CHO-K1 cells transiently transfected with expression vectors containing the wild-type FVII cDNA (FVIIwt) or harboring the p.I289del, p.G420V or p.A354V-p.P464Hfs mutations, we found that the secretion of the FVII mutants was severely decreased compared to FVIIwt. The synthesis rate of the mutants was slower than the FVIIwt and delayed, and no degradation of the FVII mutants by proteasomes, lysosomes or cysteine proteases was observed. Confocal immunofluorescence microscopy studies showed that FVII variants were localized into the endoplasmic reticulum (ER) but were not detectable within the Golgi apparatus. These findings suggested that a common pathogenic mechanism, possibly a defective folding of the mutant proteins, was triggered by the FVII mutations. The misfolded state led to impaired trafficking of these proteins causing ER retention, which would explain the low to very low FVII plasma levels observed in patients carrying these mutations.

Proangiogenic TF-FVIIa-PAR2 Signaling Requires Matriptase-Independent Integrin Interaction

The close link between coagulation activation and cancer progression is supported by clinical and experimental studies. A central molecular pathways by which tumor cells interact with the hemostatic system is through the expression of the cell surface receptor tissue factor (TF) that in complex with coagulation factor VIIa (FVIIa) triggers the extrinsic pathway of blood coagulation, contributes to cancer associated thrombosis, and promotes direct tumor cell signaling through protease-activated receptors (PARs). Genetic and pharmacological evidence shows that epithelial and tumor cell TF-FVIIa signaling induces a diverse set of proangiogenic and immune modulatory cytokines, chemokines and growth factors. However, the mechanistic details of this signaling pathway remain incompletely understood. Furthermore, a recent study has suggested that the TF-FVIIa complex does not directly cleave PAR2, but rather induces PAR signaling through initiating a cell surface protease cascade involving the transmembrane serine protease matriptase. Here, we employ mutants of FVIIa to characterize the roles of cell surface proteolysis and endosomal signaling by TF in keratinocytes and cancer cells.We studied TF-expressing MDA-MB-231 breast cancer cells, A7 melanoma cells transduced with TF and PAR2, and non-tumorigenic HaCaT keratinocytes. Stimulation with 10 nM FVIIa or 5 nM recombinant matriptase promoted productive PAR2 signaling in all three cell types as assessed by quantification of cytokine IL-8 mRNA up-regulation. Addition of the serine protease inhibitor aprotinin abolished IL-8 upregulation by recombinant matriptase, but not by FVIIa. Importantly, matriptase was expressed by HaCaT and A7 melanoma cells, whereas MDA-MB-231 cells did not express matriptase, while responding to FVIIa with efficient proangiogenic signaling. HaCaT and A7 cells showed shedding of the matriptase protease domain following incubation with FVIIa, suggesting activation of the serine protease. However, the signaling-defective FVIIa Q40A mutant also promoted matriptase shedding without inducing IL-8. We had previously shown that FVIIa induces association of TF with integrin β1 hetorodimers. We characterized an integrin-binding deficient FVIIa mutant, E26A, rendered incapable to interact with active integrin β1, resulting in defective TF-FVIIa internalization, TF-FVIIa stimulated cell migration and induction of IL-8. FVIIa E26A promoted matriptase shedding as observed with wild-type FVIIa. We found that inhibition of the small rho GTPase arf6, which controls cellular trafficking of β1 integrins, by pharmacological inhibitors or expression of a dominant negative arf6 mutant prevented IL-8 induction by TF-FVIIa, but not by direct stimulation of PAR2 with agonist peptide. Pharmacological activation of arf6 or expression of arf6 constitutive active mutant on the other hand promoted translocation of TF-FVIIa, which unlike PAR2 agonist peptide induced delayed and prolonged MAPK activation downstream of PAR2. This data show that TF-FVIIa triggers PAR2 signaling dependent on integrin-arf6-regulated trafficking and prolonged activation of endosomal signaling.Our data support the conclusion that the FVIIa induced interaction of TF with integrins and endocytosis of the TF-FVIIa-integrin complex plays a pivotal role in inducing proangiogenic cytokines downstream of proteolytic cleavage of PAR2. Mutants of FVIIa that are defective in supporting this pathway fully promote cell surface matriptase activation, but fail to upregulate tumor cell proangiogenic responses or tumor cell migration. These results argue against a role for matriptase as an intermediary in TF-FVIIa induced PAR2 activation, but do not exclude that matriptase modulates homeostatic epithelial responses through PAR2 signaling in non-malignant cells. These data furthermore suggest that matriptase inhibitors cannot substitute for targeting TF as potential therapeutic strategy to attenuate cancer proangiogenic signaling of the hemostatic system DisclosuresOstergaard:Novo Nordisk A/S: Employment. Ruf:Iconic Therapeutics: Consultancy.

Infusion of Coagulation Factor VIII-Containing Induced Pluripotent Stem Cell (iPSC)-Derived Megakaryocytes (iMks) Shows Potential As Hemophilia a Treatment

Ectopically expressed factor VIII (FVIII) in megakaryocytes (MKs) and platelets (pFVIII) is stored in a-granules and released at sites of vascular injury by activated platelets (Plts), restoring hemostasis in FVIIInull mice, even in the presence of neutralizing inhibitors. These studies support the idea that unlike therapies that correct plasma levels of FVIII, pFVIII may be a useful therapy in patients with hemophilia A who have intractable inhibitors and significant bleeds. Expressing FVIII in Plts, however, has limitations that make pFVIII gene therapy through bone marrow transplantation (BMT) problematic: 1) pFVIII expressed during megakaryopoiesis can injure the Mks, potentially exacerbating post-BMT thrombocytopenia, and 2) pFVIII’s efficacy in joint and intracranial bleeds has yet to be shown, especially in the presence of inhibitors. Due to these limitations we propose an alternative strategy: infusing patient-specific iMks derived from personalized iPSCs and expressing either human B-domain-deleted (BDD) FVIII or variants of FVIII that have greater stability and longer half-lives. Our group has shown that infusing in vitro-grown Mks into mice releases functional Plts in the recipient animals. iPSCs are a renewable source of stem cells that can be pre-screened to select clones that both express high levels of pFVIII and also release high numbers of Plts after differentiation into iMks. As proof-of-principle, iMks were transfected with a self-inactivating lentivirus containing cDNA for 1 of 3 FVIII variants: wildtype BDD FVIII (WT FVIII), a PACE/furin cleavage site FVIII (FVIIIR1645H) variant, and an amino acid 1645 to 1648 deletion FVIII (FVIIIΔ) variant that removes the entire PACE/furin cleavage site. FVIIIR1645H and FVIIIΔ showed greater stability and consequently greater specific activity with no increase in injuring Mks. We previously published that hemophilia A mice expressing pFVIIIR1645H were more hemostatically corrected than comparable mice expressing WT pFVIII. All of the FVIII variant iMks expressed at least a 40-fold higher level of mRNA compared to the non-transduced control (N=6) and integration levels show the same number of viral copies between the groups (N=6). All variants expressed >550 pg FVIII/106 CD42b+ iMKs (N=6). Upon activation with thrombin, transduced Mks released the FVIII into the supernatant. To examine whether this pFVIII injured the developing Mks, baseline PAC-1 binding for Mk activation in culture (N=3), TUNEL staining and Annexin-5 binding for apoptosis (N=4) were analyzed with no differences observed with WT Mks not expressing pFVIII. To test the ability of FVIII-expressing iMks to correct the coagulopathy in hemophilia A, 5x105 iMks were added to FVIIInull murine whole blood (0.11 ml) and evaluated for clot formation using rotational thromboelastometry (ROTEM). Each pFVIII iMk variant showed a decrease in clotting time, clot formation time, and an increase in maximum clot firmness when compared to the non-transduced control (p<0.007 for each, N=4). These FVIII expressing iMks were also tested in vivo in a FeCl3 carotid artery injury murine model. 24 hours prior to infusion, recipient hemophilia A mice were treated with clodronate liposomes to eliminate circulating monocytes and to improve the survival of infused human iMks and their released Plts. Immediately post iMks (5x106)infusion, a 20% FeCl3 solution was applied to the carotid artery for 3 mins and flow rate through the injured vessel was measured for 30 mins. Both WT FVIII and FVIIIR1645H showed a significant decrease in blood flow through the injured vessel from 1.2 ml/min seen in FVIIInull mice receiving control iMks to 0.4 ml/min (p<0.05, N=10). Wild-type mice had a flow rate of 0.13 ml/min. These data indicate that pFVIII within iMKs or their derived Plts expressing FVIII can improve hemostasis in vitro and in vivo. These studies provide the groundwork to examine whether infused iMks pFVIII can improve hemostasis in the setting of inhibitors. DisclosuresArruda:Pfizer: Patents & Royalties, Research Funding. Sabatino:Spark Therapeutics: Research Funding. Camire:Bayer: Consultancy; Spark Therapeutics: Membership on an entity’s Board of Directors or advisory committees, Patents & Royalties; Pfizer: Consultancy, Patents & Royalties, Research Funding; Novo Nordisk: Research Funding.

Differential functional readthrough over homozygous nonsense mutations contributes to the bleeding phenotype in coagulation factor VII deficiency

Background Whereas the rare homozygous nonsense mutations causing factor (F)VII deficiency may predict null conditions that are almost completely incompatible with life, they are associated with appreciable differences in hemorrhagic symptoms. The misrecognition of premature stop codons (readthrough) may account for variable levels of functional full-length proteins. Objectives To experimentally evaluate the basal and drug-induced levels of FVII resulting from the homozygous p.Cys132X and p.Ser112X nonsense mutations that are associated with moderate (132X) or life-threatening (112X) symptoms, and that are predicted to undergo readthrough with (132X) or without (112X) production of wild-type FVII. Methods We transiently expressed recombinant FVII (rFVII) nonsense and missense variants in human embryonic kidney 293 cells, and evaluated secreted FVII protein and functional levels by ELISA, activated FX generation, and coagulation assays. Results The levels of functional FVII produced by p.Cys132X and p.Ser112X mutants (rFVII-132X, 1.1% ± 0.2% of wild-type rFVII; rFVII-112X, 0.5% ± 0.1% of wild-type rFVII) were compatible with the occurrence of spontaneous readthrough, which was magnified by the addition of G418 - up to 12% of the wild-type value for the rFVII-132X nonsense variant. The predicted missense variants arising from readthrough abolished (rFVII-132Trp/Arg) or reduced (rFVII-112Trp/Cys/Arg, 22-45% of wild-type levels) secretion and function. These data suggest that the appreciable rescue of p.Cys132X function was driven by reinsertion of the wild-type residue, whereas the minimal p.Ser112X function was explained by missense changes permitting FVII secretion and function. Conclusions The extent of functional readthrough might explain differences in the bleeding phenotype of patients homozygous for F7 nonsense mutations, and prevent null conditions even for the most readthrough-unfavorable mutations.

728. Tissue-Directed Transgene Engineering for AAV and Lentivector Gene Therapy Approaches

Clinical gene therapy frequently is encumbered by low transgene product biosynthesis at predictably safe vector doses. It has been hypothesized that the presence of rare codons may regulate transgene product expression through depletion of the available cognate tRNA pool. Codon optimization is the prominent strategy utilized to overcome this hypothesized limitation and involves replacing rare, presumably translation rate-limiting, codons with the most frequent ones. Typical algorithms attempt to match the codon usage frequency of the transgene coding sequence to that of target organism's total mRNA pool, which has been shown to approximate the overall available tRNA concentrations. Upon closer examination, it appears that both codon frequency and tRNA content vary between tissue types. Therefore, we hypothesize that codon-optimization can be improved by tailoring to the codon-frequency of the most highly expressed genes present in target cell types. Our tissue-directed codon optimization algorithm utilizes novel codon usage indices generated from target cell gene expression data. As proof of concept, we developed tissue-codon optimized variants of coagulation factor VIII (FVIII) to be utilized in lenti- and adeno-associated viral (LV and AAV, respectively) vectors. These two vectors are the leading platforms for clinical gene therapy of hemophilia A. LV is utilized to target autologous hematopoietic cell types ex vivo while AAV is delivered intravenously to genetically-modify hepatocytes. However to date, development-stage LV and AAV gene therapy products for hemophilia A have been characterized by low-level FVIII transgene product biosynthesis. Initially, we designed human hepatocyte-, monocyte- and standard overall human-optimized FVIII constructs (LCO, MCO and HCO, respectively) to be compared to wild-type FVIII. Upon initial examination, expression of LCO-FVIII was shown to be 3-fold greater than either MCO- or wild-type FVIII from the human hepatoma cell line, HepG2, following transient transfection. In contrast, LCO-FVIII and MCO-FVIII expression was diminished 12 and 4-fold, respectively, compared to wild-type in the ‘neutral’ human embryonic kidney 293T cell line. Furthermore, following hydrodynamic injection of naked plasmid DNA into hemophilia A mice, LCO-FVIII exhibited a sustained 10-fold increase in FVIII expression relative to the HCO-FVIII comparator. In attempt to generate a lead candidate for clinical translation, we utilized several of our most promising vector components to construct a liver-optimized AAV8 vector consisting of LCO-ET3, our previously described high-expression bioengineered FVIII variant, transcriptionally driven by a novel 146bp liver-directed promoter and adjacent MVM intron. Following intravenous delivery into hemophilia A mice, vector doses of 1e12 and 1e11 vector genomes per kg achieved sustained, predictable curative plasma FVIII levels of 200% and 20% of normal human levels, respectively. These initial results support the utility of our novel approach of clinical tissue-directed transgene optimization.

Mild hemophilia A patient with novel Pro1809Leu mutation develops an anti‐C2 antibody inhibiting allogeneic but not autologous factor VIII activity

In mild hemophilia A (MHA) patients, the risk of inhibitor development is generally low, but some factor VIII (FVIII) gene missense mutations are associated with a higher inhibitor incidence. To investigate the mechanism(s) of inhibitor development in MHA. A patient, HA78, with MHA with a novel P1809L missense mutation in the A3 domain, exhibited significant residual FVIII activity ( C ~10IUdL(-1) ), despite the development of an inhibitor (5.6BUmL(-1) ). Purified HA78-IgG significantly depressed C from normal plasma but not from patient's plasma without inhibitor, indicating that this IgG inhibited allogeneic but not autologous FVIII. The HA78-IgG blocked thrombin and FXa-catalyzed FVIII cleavage but had little effect on FVIII binding to von Willebrand factor and phospholipid. The IgG recognized a C2 epitope close or overlapping the previously described anti-C2 ESH8 epitope. Similarly, a recombinant FVIII-P1809L mutant was little inactivated by HA78-IgG. This mutant demonstrated ~3-fold lower binding affinities to von Willebrand factor and phospholipid compared with wild-type, while reactions with thrombin or FXa were not impaired. Reaction of FVIII-P1809L with the alternative anti-C2 ESH4 showed only an ~20% inhibition compared with wild-type FVIII but was similar to wild-type after incubation with ESH8. Asurface plasmon resonance-based assay demonstrated that anti-C2 ESH4 bound to FVIII-P1809L with ~10(2) -fold lower affinity compared with ESH8. These results indicated that the P1809L mutation in A3 induced the conformational change in the FVIII molecule that hampered antigenic determinant(s) located in the C2 domain and might result in the inhibitor development.

Modulating the Antithrombin-Mediated in Vivo Clearance of Coagulation Factor VIIa

The short half-life of coagulation factor VIIa (FVIIa) in circulation is the result of elimination through multiple pathways of which inactivation by the plasma inhibitor antithrombin (AT) accounts for as much as 65% of the total clearance in humans. Remarkably, the rate of inhibition in vivo is about 30 times greater than the uncatalyzed rate of inhibition in vitro suggesting the presence of rate enhancing components in vivo (Agersø et al. (2011) J Thromb Haemost, 9:333-338). Prime candidates include endogenous heparin-like glycosaminoglycans (GAGs) potentiating the reactivity of antithrombin, or tissue factor (TF) which upon binding to FVIIa increases its susceptibility to inhibition.In the present study site-directed mutagenesis of FVIIa was undertaken to identify variants with altered AT reactivity in order to explore the relationship between the reactivity of FVIIa with AT in vitro and in vivo as well as the nature of endogenous rate enhancing components. The pharmacokinetic properties of FVIIa variants were determined in Sprague Dawley rats as this model recapitulates the aspects of AT-mediated FVIIa clearance observed in humans and allows for interaction of human FVIIa with endogenous rat TF. Similar to the human situation, inactivation of wild-type FVIIa in rat is evident as an accumulation of circulating FVIIa-AT complexes and a progressive divergence of the pharmacokinetic profiles representing FVIIa clot activity and total FVIIa antigen.Initially, the ability to modulate the in vivo complex formation with AT was investigated using two FVIIa variants exhibiting enhanced (>200%) or reduced (<10%) in vitro reactivity with AT, respectively, regardless of the type of cofactor present. Reflecting the in vitro reactivity, clot activity and antigen PK profiles in rats were found to coincide for the AT resistant variant along with essentially no detectable AT complex formation, whereas exacerbated AT complex formation and clot activity:antigen discrepancy was observed for the variant exhibiting enhanced in vitro reactivity.Interestingly, among the generated FVIIa variants with altered AT reactivity, two subsets were identified that displayed differential in vitro reactivity with AT depending on the type cofactor present. Accordingly, one group exhibited a greater susceptibility to inhibition relative to wild-type FVIIa in the presence of heparin but not in the presence of TF, while the other group demonstrated the opposite behavior. Endowed with the ability to report on the cofactor identity from the rate of inhibition relative to wild-type FVIIa, variants from each group were tested for their tendency to accumulate as complexes with AT following intravenous administration to rats. Supporting a contribution from endogenous GAGs to the in vivo inactivation of FVIIa, the measured in vivo peak levels of accumulated FVIIa-AT complexes were found to directly correlate with the in vitro rate constants determined for the variants in the presence of heparin, but not when the cofactor was TF or the combination of TF and heparin.Altogether, these results 1) demonstrate a direct relationship between the in vitro reactivity of FVIIa with AT in the presence of heparin and the clearance of FVIIa through this pathway in vivo, and 2) identify heparin-like GAGs as the likely rate enhancing component of FVIIa inhibition in vivo. DisclosuresØstergaard:Novo Nordisk A/S: Employment. Hansen:Novo Nordisk A/S: Employment. Pelzer:Novo Nordisk A/S: Employment. Agersø:Novo Nordisk A/S: Employment. Pedersen:Novo Nordisk A/S: Employment. Glue:Novo Nordisk A/S: Employment. Johnsen:Novo Nordisk A/S: Employment. Andersen:Novo Nordisk A/S: Employment. Bjelke:Novo Nordisk A/S: Employment. Breinholt:Novo Nordisk A/S: Employment. Stennicke:Novo Nordisk A/S: Employment. Gandhi:Novo Nordisk A/S: Employment. Olsen:Novo Nordisk A/S: Employment. Hermit:Novo Nordisk A/S: Employment.