PHARMACEUTICAL COMPOSITIONS FOR THE TREATMENT OF THROMBOSIS IN PATIENTS SUFFERING FROM A MYELOPROLIFERATIVE NEOPLASM

Abstract

Thrombosis is the main cause of morbidity and mortality in patients with JAK2V617F positive myeloproliferative neo-plasms (MPN). Recent works reported the presence of JAK2V617F in endothelial cells in some MPN patients. Here, the inventors show that JAK2V617F endothelial cells promote thrombosis through induction of endothelial P-selectin expression and thus demonstrate that P-selectin blockade was sufficient to reduce the increased propensity of thrombosis. Accordingly the present invention relates to a method of treating thrombosis in a patient suffering from a myeloproliferative neoplasm comprising administering to the patient a therapeutically effective amount of a P-selectin antagonist.

Claims

1. A method of treating or prophylactically treating thrombosis in a patient suffering from a myeloproliferative neoplasm comprising administering to the patient a therapeutically effective amount of a P-selectin antagonist.

2. The method of claim 1 wherein the patient suffers from polycythemia vera (PV), essential thrombocythemia (ET) or primary myelofibrosis (PMF).

3. The method of claim 1 wherein the patient harbours one mutation in JAK2.

4. The method of claim 3 wherein the one mutation is the JAK2V617F mutation.

5. The method of claim 1 wherein the P-selectin antagonist is administered to the patient for prophylactically treating thrombosis.

6. The method of claim 1 wherein the P-selectin antagonist is an antibody against P-selectin.

7. The method of claim 6 wherein the antibody is Crizanlizumab.

8. The method of claim 1 wherein the P-selectin antagonist is hydroxycarbamide.

9. The method of claim 1 wherein the P-selectin antagonist is an inhibitor of P-selectin expression.

10. The method of claim 10, wherein the inhibitor of P-selectin expression is a siRNA or an antisense oligonucleotide.

Description

FIGURES

[0030] FIG. 1 Increased endothelial P-selectin expression is responsible for the pro-adhesive phenotype of JAK2V617F endothelial cells. In static conditions, increased adhesion of normal mononuclear cells (a) and neutrophils (b) on JAK2V617F HUVECs is reversed in the presence of a P-selectin blocking antibody. In Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice, increased rolling (c) and adhesion (d) of leukocytes is abolished in the presence of a P-selectin blocking antibody. (e) Increased thrombus formation in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice is abrogated in the presence of a P-Selectin blocking antibody. Results are mean value+/s.e.m. Statistical significance by 2-way ANOVA analysis of variance and Sidak post-hoc test. **P<0.05; **P<0.01; ****P<0.0001.

EXAMPLE

[0031] Materials. For Western-Blotting in human endothelial cells, the following antibodies were used: JAK2 (rabbit, Santa Cruz Biotechnology, Dallas, Tex.), JAK2 Tyr1007/1008 (rabbit, Cell Signaling, Danvers, Mass.) STAT-3 (mouse, Cell Signaling), STAT3 Tyr705 (rabbit, Cell Signaling), pan-Akt (9272, rabbit, Cell Signaling), Ser473 p-Akt (rabbit, Cell Signaling), -Tubuline (rabbit, Cell Signaling). For P-Selectin immunostaining and blocking in mice, anti-mouse P-Selectin monoclonal antibody was used (RB40.34 clone, BD Biosciences, Franklin Lakes, N.J.). Blocking effect of this antibody was previously described.sup.39,40. For vWF immunostaining in human endothelial cells, primary antibody rabbit anti-human vWF (EMD Millipore) was used. For VE-Cadherin staining in human endothelial cells, monoclonal antibody anti-VE-Cadherin (SantaCruz Biotechnology) was used.

[0032] Cell culture/Lentivirus transduction. Human umbilical venous endothelial cells (HUVECs, Lonza, Basel, Swiss) were cultured in EGM-2 media (Lonza, CC-3156) supplemented with EGM-2 Single Quots (Lonza, CC-4176). Cells were transduced with GFP lentivirus encoding human JAK2V617F or JAK2WT by adding lentiviral supernatant to the medium and incubating overnight at 37 C. An empty lentivirus encoding only for GFP was used as a negative control. Lentivirus were transduced at a multiplicity of infection of 20, allowing more than 95% of GFP positive cells. Transgene stability was verified after each passage and remaining stable throughout the passages. HUVECs were used for experiments between passage 5 and passage 7.

[0033] Western-Blot in HUVECs. HUVECs were platted in 6 wells plats to reach confluence. Confluent cells were then starved for 4 hours in EBM2 medium (0% SVF, 0% BSA). Cells were lysed in Laemmli buffer (Tris 10 mM, Saccharose 7%, SDS 2%, (-mercapto-ethanol 3.92%, Blue of bromophenol 0.04 g/L) after congelation at 20 c. Cell lysates were resolved by SDS-PAGE and probed with following antibodies: JAK2 (rabbit, Santa Cruz Biotechnology), JAK2 Tyr1007/1008 (rabbit, Cell Signaling), STAT-3 (mouse, Cell Signaling), STAT3 Tyr705 (rabbit, Cell Signaling), pan-Akt (rabbit, Cell Signaling), Ser473 p-Akt (rabbit, Cell Signalling), -Tubuline (mouse, Cell Signaling). Binding of antibodies to the blots was detected using Odyssey Infrared Imaging System (LI-COR Biosciences, Lincoln, Nebr.).

[0034] Thrombin generation. Thrombin generation was measured in freshly prepared platelet free plasma by means of the Calibrated Automated Thrombogram (CAT) method (Thrombinoscope BV, Maastricht, Netherlands). Thrombin generation was performed in 96-well plates containing confluent HUVECs washed with HEPES buffer (20 mM Hepes, 140 mM NaCl, 5 mg/ml BSA, pH 7.35). Polystyrene was considered as the control condition since it is the reference material to study thrombin generation.sup.41. Thrombin generation was triggered by tissue factor (1 or 2.5 pM final concentration). The velocity index (nM/min) was calculated as the ratio Peak/(time to Peak-LagTime)]; LagTime (LT) is related to the initiation phase of coagulation, time to Peak (ttPeak) and Peak are the reflection of the amplification phase of coagulation and the Endogenous Thrombin Potential (ETP) reflects the global quantity of thrombin produced during the experiment.

[0035] Protein C activation. Confluent monolayer HUVECs in 96-well plates were washed twice with warm sterile PBS and then incubated with thrombin (2 nM thrombin) and CaCl2 (2.5 mM) for 10 min at 37 C. Protein C was then added to each well at a final concentration of 0.2 M. The plate was further incubated for different times at 37 C. Aliquots were then collected and transferred to a clean 96-well plate. Hirudin (100 U/ml) was added to block thrombin. Activated protein C activity was monitored during 30 minutes at 405 nm on a FLUOStar Optima plate reader (BMG Labtech GMBH, Ortenberg, Germany) using a specific substrate (PNAPEP1566, 100 L at 0.4 mM). Results were plotted as the rate of substrate hydrolysis as the function of time of PC activation.

[0036] In vitro static adhesion on endothelial cells. Blood samples were performed from well healthy voluntary witnesses. Blood was diluted with PBS (1/2 dilution) and deposed on lymphocyte separation medium (Pancoll, Dominique Dutscher, Brumath, France) before centrifugation at 450 g during 20 minutes to obtain mononuclear cells (MNC). Monocytes were isolated from MNC by magnetic immunoseparation and selection of CD14 positive cells (EasySep Monocyte Extraction kit, R&D Systems, Minneapolis, Minn.). Neutrophils were obtained from blood samples deposed on neutrophils separation medium (Polymorphprep, Fresenius Kabi, Oslo, Norway) before centrifugation at 450 g during 45 minutes. After isolation, MNCs, monocytes and neutrophils were marked with membrane dye (CellTracker Orange, ThermoFisher Scientific, Waltham, Mass.). Transduced JAK2V617F HUVECs, wild type HUVECs or negative control HUVECs were platted in 24 wells plats to reach confluence. Mononuclear cells, monocytes, and neutrophils were added on top of transduced HUVECS for 1 hour at 37 C., using 500 000 cells by well. After 1 hour, three washes with EGM2 medium were done. We visualized adherent cells using a fluorescent microscope (AxioObserver, Zeiss, Oberkochen, Germany) and analysed images by ZEN imaging software (Zeiss). Well surface was quantified in order to obtain the following ratio: number of adherent cells/mm2. For P-Selectin inhibition experiments, we used P-Selectin blocking antibody (AK4 clone, BioLegend, San Diego, Calif.) during 30 minutes before deposit of cells on HUVECs.

[0037] In vitro Neutrophils adhesion on HUVEC cells in flux conditions. Canals in a cell chamber (Vena8 Endothelial+, Celix No: 1510-02) were coated with Human fibronectin (100 ng/ml) (Promocell, Heidelberg, Germany, No: C-43060), before seeding of pre-cultivated HUVECs at the concentration of 3.10.sup.6 Cells/ml. After two hours at 37 C., cells were cultivated with flux in a closed circuit where a pump was linked to medium and linked to cells in canals (KIMA-IPOD TOUCH MICROFLUIDIC PUMP, Cellix). A flux with two periods was imposed to the cells: a perfusing period (3 minutes with 600 L/min) and a rest period (20 minutes with 0 L/min). Cells were incubated with this flux for 48 hours at 37 C. 3.10.sup.6 Neutrophils/ml were isolated from a healthy donor with a kit (MACSxpress, Neutrophil cocktail human-lyophilized, Miltenyi Biotec, Bergisch Gladbach, Germany, No: 130-104-434) before introduction in canals with a venous flux of 10 L/min and observed at 20 in phase contrast with a microscope ZEISS during a film of 5 minutes. After five minutes canals were washed with medium and pictures of every canal were performed. Neutrophils quantification was performed blindly on canal's picture and Neutrophils Velocity is studied with FIJI (Image J).

[0038] Generation and characterization of Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice. The conditional flexed JAK2 (JAK2V617F/WT) mice were generously provided by J. L Villeval and have been previously described.sup.19. The double-heterozygous Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice were generated by crossing JAK2.sup.V617F/WT mice with Pdgfb-iCreERT2 mice allowing tamoxifen-inducible adult expression of JAK2.sup.V617F in endothelium. Littermate Pdgfb-iCreERT2-negative;JAK2.sup.V617F/WT mice were used as controls. To induce Cre activity in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice, we used oral gavage of a single dose of 8 mg of tamoxifen. Tamoxifen induction was performed in 5 weeks-old animals. Mutant mice were analysed 2-3 weeks after tamoxifen administration. Ears of adult mice were genotyped by PCR. Haematocrit, hemoglobin level, platelet, and white cells count were determined using an automated counter (scil Vet abc Plus+) on blood collected from the sublingual vein in EDTA containing tubes. For blood flow cytometry analysis in Pdgfb-iCreERT2;mT/mG;JAK2.sup.V617F/WT mice, cells were stained with TER-119 APC (BD Biosciences, Ter-119 clone), CD42d-APC (BioLegend, 1C2 clone) and Ly6-APC (BD Biosciences, RB68C5 clone). FACS analysis was performed using an Accuri C6 flow cytometer (BD Biosciences). Data were interpreted using BD Accuri C6 Analysis Software.

[0039] Isolation of endothelial cells from mice. Mice were euthanized followed by exposure of the thoracic and abdominal cavity. In order to isolate ECs from lungs, after the right atrium was cut, physiologic sera was injected in the left ventricle to completely flush blood cells from the lungs. Kidneys and lungs were removed and minced into small pieces, following by incubation for 60 minutes at 37 C. with 5 ml 0.1% type 4 collagenase. The digested tissue suspension was aspirated into to a 10-ml syringe with a 14-gauge cannula, and clumps were triturated into a single-cell suspension. The single-cell suspension was filtered through a 70m strainer. The filtered cell suspension was centrifuged for 10 minutes at 300 g, and the cell pellet was washed with 0.5% BSA, 2 mM EDTA, and PBS containing CaCl2 and MgCl2 (Gibco, ThermoFisher Scientific). The cell pellet was suspended with 190 L 0.5% BSA, 2 mM EDTA, PBS, following by the addition of 5 L anti-CD31 antibody (BD Biosciences, 553370) and incubation at 37 C. during 30 minutes. After washing in PBS-EDTA-BSA, 200 L of anti-rat beads were added to the cell suspension, and cells were incubated 15 minutes at 4 C. Cell suspension was next washed in PBS-EDTA-BSA and endothelial cells were isolated using magnetic microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany, 120-000-291).

[0040] DNA purification and quantitative allele-specific PCR. Genomic DNAs were purified using NucleoSpin Tissue kit (Macherey-Nagel, Duren, Germany). For quantification of wild type (WT) and mutated JAK2 DNA, quantitative allele specific PCR from gDNA was conducted to identify amplified fragments from the mutated or WT JAK2 DNA, respectively. It was performed on a 7500 Real Time PCR System AB (Applied Biosystems, Foster City, USA) and analyzed with associated software.

[0041] Intravital microscopy of mesenteric venules. Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice and Pdgfb-iCreERT2negative;JAK2.sup.V617F/WT mice were used 20 days post tamoxifen injection. Intra-peritoneal injection of TNF-alpha (R&D Systems) at the dose of 250 ng/mice was performed. 4 hours after TNF-alpha administration, mice were anesthetized with intra-peritoneal injection of ketamine/xylazine. Rhodamin 6G (Sigma-Aldrich, Saint-Louis, Mo., ref: 4127) injection was performed to stain leukocytes five minutes before incision. An incision was made through the abdominal wall to expose the mesentery, and mesenteric venules of 150- to 250-m diameter were studied. 5 venules by mice were sequentially observed for 1 minute and 30 seconds during the 30 minutes after surgical procedure, using a fluorescent microscope (AXIO Zoom.V16, Zeiss). For P-Selectin inhibition, 25 g/mouse of P-selectin blocking antibody (RB40 clone, BD Biosciences) or isotype control antibody (A110-1 clone, BD Biosciences) was injected 5 minutes before starting the analysis.

[0042] Video analysis. Rolling leukocytes were quantitated by counting the number of rhodamin-marked cells passing a given plane perpendicular to the vessel axis in 30 seconds. Adherent leukocytes were quantitated by counting the number of rhodamin-marked cells motionless during 30 seconds. Vessel surface was quantitated to perform the following ratio: adherent cells/vessel surface (number of cells/cm2). Videos analyses were performed using ZEN imaging software (Zeiss). Rolling and adhesion quantification were performed by two independent observers, blindly.

[0043] Mouse model of thrombus formation. 20 days post-tamoxifen induction, PDGFb-iCreERT2;JAK2V617F.sup.V617F/WT mice and PDGFb-iCreERT2-negative;JAK2WT.sup.V617F/WT mice were utilized for experiments. We studied spontaneous thrombus formation, and two thrombosis-induced models. To induce platelet and coagulation activation, intra-peritoneal injection of collagen (Nycomed Pharma, Zurich, Swiss, 7806141/450) at the dose of 75 g/kg and epinephrine (Helena Laboratories, Beaumont, Tex., 5367) at the dose of 30 g/kg was performed 3 minutes before euthanasia. To induce low inflammation, TNF-alpha injection (RD Systems, 210-TA-020) at the dose of 250 ng/mice was injected 4 hours before euthanasia. The dose of 500 ng/mice TNF-alpha is commonly used to trigger inflammation.sup.43,44 and we chose a lower dose to reveal a potential hypersensitivity. After injection of collagen-epinephrine or TNF-alpha, mice were anesthetized with isoflurane and blood was obtained by sub-lingual sampling in polypropylene Eppendorf tubes containing 5 L of EDTA in order to perform blood count. After euthanasia, an incision was performed in thoracic wall to expose mice heart and lungs were washed with intra-cardiac perfusion of PBS without CaCl2 and MgCl2 (Gibco ThermoFisher Scientific) during 3 minutes. Lungs were fixed with secondary three-minutes injection of 10% formalin and collected before formalin fixation and paraffin embedding. For P-selectin inhibition, we used P-selectin blocking antibody (RB40 clone, BD Biosciences) at the dose of 25 g per mouse, 4 hours before euthanasia.

[0044] Histology. To quantify thrombus formation in mice, Carstair's staining was performed. Slides were hydrated in xylol and ethanol to distilled water, followed by incubation in 5% ferric ammonium sulfate for 5 min, washing, and staining by Mayer hematoxylin for 5 min, washing, and Picric Acid-orange G solution for 1 hour, and washing, 1% phosphotungstic acid for 10 min. After washing, slides were stained by Ponceau Fuchsin solution for 7 min, washing, 1% phosphotungstic acid for 10 minutes, Anilin blue solution for 30 min, and rinsed in distilled water. Slides were dehydrated covered with a coverslip using mounting medium. Thrombus quantification was performed using an optical microscope and pulmonary area was quantitated to perform the following ratio: thrombi number/pulmonary area (number of thrombi/cm2). Three slides by mice were analysed by two independent observers, blindly.

[0045] Immunostaining. Immunofluorescence analyses were realized on HUVECs, with or without activation of cells by TNF-alpha (10 ng/ml overnight). Cells were fixed with 2% paraformaldehyde (PFA) for 10 minutes. After saturation in 5% bovine serum albumin for 1 hour, cells were incubated with primary antibody rabbit anti-human vWF (Merck Millipore), vWF primary antibody was resolved with Alexa Fluor 588 conjugated secondary antibody (Invitrogen, Carlsbad, Calif.). Cells were mounted in Vectashield mounting medium containing 4,6-diamidino-2-phenylindole (DAPI, Vector Laboratories, Burlingame, Calif.), imaged with a confocal microscope (Zeiss LSM 700) and analysed by Imaris software (Bitplane). In mice, P-Selectin immunostaining was performed in carotid arteries of PDGFb-iCreERT2;JAK2V617F.sup.V617F/WT and control mice. Briefly, mice were euthanized followed by exposure of the thoracic cavity. After the right atrium was cut, PBS was injected in the left ventricle to completely flush blood cells from the carotid arteries. Fixation of endothelial cells was performed by injection of 10% formalin. After exposure, carotid arteries were removed from the mice and fixed in 10% formalin during 10 minutes, following by washing in PBS and saturation in 10% donkey serum-PBS. Arteries were incubated with primary anti P-Selectin antibody (RB40.34, BD Biosciences) and primary anti VE-Cadherin antibody (SantaCruz, sc-6458,) overnight at 4 C. After three washing with PBS, arteries were incubated with Alexa Fluor 568 conjugated secondary antibody (VE-Cadherin) or and Alexa Fluor 488 conjugated secondary antibody (P-Selectin) during 2 hours at room temperature. Arteries were then mounted in Vectashield mounting medium containing 4,6-diamidino-2-phenylindole (DAPI, Vector Laboratories), imaged with a confocal microscope (Zeiss LSM 700) and analyzed with Image J.

[0046] Quantification of VWF in HUVEC supernatant and in HUVECs. HUVEC cell were seeded in 12 wells plate (Costar). When they were confluent, they were washed with 500 L of PBS (Gibco ThermoFisher Scientific) before adding 500 L of medium not deprived. Experiments were performed in absence or in presence of TNF-alpha (10 ng/ml during 24 hours) (Merck Millipore). After 24 hours, supernatant was removed, centrifuged at 12000 g during 5 minutes and stored at 80 C. To quantify vWF in HUVECs, cells were washed with PBS before adding in wells 500 L of PBS. Cells were destroyed in wells with scrapper. Liquid was removed and 5 freezing and thawing successions were performed. Finally, the intracellular lysates were centrifuged at 12000 g during 5 minutes and stored at 80 C. As previously published.sup.45, vWF was quantified following this method: 98 wells plate (Greiner, Flat Bottom) were coated with an anti-vWF antibody (DAKO, Les Ulis, France, A0082) diluted at 1/660 in a coating buffer overnight at 4 C. After washing and blocking, sample of HUVEC supernatant and intracellular lysate were put in wells. Standards were realized with a platelet poor plasma from a healthy donor. After 2 hours and a second wash, antibody anti-vWF coupled to HRP (DAKO, P0226) diluted at 1/6000 in wash buffer was deposit in wells. After 2 hours and a third wash, HRP was revealed with an OPD and H2O2 solution. After 2 minutes of coloration, the reaction was blocked with a H2SO4 3M solution and reading was performed at 492 nm with OPTIMA plate reader.

[0047] Study of hydroxyurea effect on endothelial cells. HUVECs were treated with HU (Sigma-Aldrich, H8627) during 24 hours at the concentration of 100 M before washing with EGM2 medium. Neutrophils static adhesion was performed as described. To study the effect of hydroxyurea in mice, Pdgfb-iCreERT2;JAK2.sup.V617F/WT and control mice were treated with HU at the dose of 200 mg/kg/day (oral gavage) during 10 days before experiments.

[0048] ELISA for soluble P-Selectin. Mice were anesthetized, and blood was obtained by retro-orbital venous plexus sampling in polypropylene Eppendorf tubes containing 100 L of ethylenediaminetetraacetic acid (EDTA). Plasma was prepared by centrifugation of the blood within 30 minutes at 1000 g for 10 min at +4 C. then 10 000 g for 15 min at +4 C. Enzyme-linked immunoabsorbent assay (ELISA) was performed according to manufacturer's instructions (R&D Systems).

[0049] Von Willebrand Factor quantification in mice. Mice were anesthetized, and blood was obtained by retro-orbital venous plexus sampling in polypropylene Eppendorf tubes containing 0.138 M sodium citrate (1/10 volume). Plasma was prepared by centrifugation of the blood 20 minutes 1500 g. Plasma vWF concentration was measured by ELISA using a polyclonal antibody against vWF (Dako France, Les Ulis, France, ref A0082) and a horseradish peroxidase-conjugated secondary antibody anti VWF (Dako, ref P0226). Pooled plasma from 40 C56BL/6 WT mice was used as reference and set at 100%. Results were expressed as a percentage of the normal murine vWF level.

[0050] Statistics. Results were expressed as meanSEM. Statistical significance was calculated using the Student t test or Mann-Whitney statistical test to compare differences between 2 groups. To compare difference among multiple groups, 1-way-ANOVA followed by Tukey post-hoc test or 2-way-ANOVA followed by Sidak post-hoc test were used. GraphPad Prism 6 was used. P value of less than 0.05 was considered significant.

[0051] Study approval. All mice used in this study were bred and maintained at the institute. This study was conducted in accordance with both Bordeaux University institutional committee guidelines (committee CEEA50) and those in force in the European community for experimental animal use (L358-86/609/EEC).

[0052] Data availability. The authors declare that the data supporting the findings of this study are available within the article and from the authors on reasonable request.

Results

The Expression of JAK2V617F by Endothelial Cells Leads to Increased Thrombus Formation

[0053] To analyze the specific role of JAK2V617F endothelial cells in thrombus formation, we crossed Pdgfb-iCreERT2 mice with conditional flexed JAK2 (JAK2V617F/WT) mice.sup.19, to allow heterozygous expression of JAK2V617F in endothelial cells but not in hematopoietic cells after tamoxifen administration. We then investigated whether Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice displayed a higher propensity for thrombosis. We looked at pulmonary thrombus formation using experimental conditions that allowed assessment of endothelial involvement, i.e. without exposition of the subendothelium. We thus used three conditions to strongly validate our observations: (i) spontaneous thrombosis, (ii) a mild thrombosis model consisting of the administration of low doses of collagen plus epinephrine, as we reasoned we would induce specifically a weak activation of platelets and vasoconstriction to better evidence an intrinsic prothrombotic phenotype of JAK2V617F endothelial cells, (iii) a weak inflammatory trigger of thrombosis with injection of low doses of TNF-alpha to reveal a potential hypersensitivity to inflammation. Small spontaneously formed thrombi were observed in the lungs of Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice, but not in littermate JAK2.sup.V617F/WT control mice. With the 2 models of mild induction of thrombosis, we observed significantly increased thrombus formation in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice compared with controls. Altogether, these results demonstrated that JAK2V617F endothelial cells have a prothrombotic phenotype.

JAK2V617F Endothelial Cell Have Normal Anticoagulant Activity

[0054] We then wanted to decipher the mechanisms by which endothelial JAK2V617F expression leads to a prothrombotic phenotype. Healthy endothelial cells are capable of inhibiting coagulation and thus thrombin generation in several ways: secretion of Tissue Factor Pathway Inhibitor, expression of Thrombomodulin and Endothelial Protein C receptor. Indeed, in the presence of human umbilical vein endothelial cells (HUVEC), Tissue Factor (TF)-triggered thrombin generation in platelet poor plasma is considerably impaired.sup.20. To assess whether the expression of JAK2V617F by endothelial cells decreased their anticoagulant properties or even triggered procoagulant properties, we measured thrombin generation at the surface of HUVECs transduced either with lentivirus encoding human JAK2V617F or JAK2 wild-type (JAK2WT) or empty lentivirus as controls. Western blot analysis revealed induced protein expression of JAK2 and an increase in the phosphorylation level of JAK2, STAT3 and AKT in JAK2V617F HUVECs, in agreement with an hyperactivation of the JAK/STAT pathway. Under resting conditions, measurement of thrombin generation at the surface of control, JAK2WT- or JAK2V617F-lentivirus infected endothelial cells did not reveal any significant difference in the kinetic and extent of thrombin generation. These results rule-out a significant gain of procoagulant activity in response to JAK2WT or JAK2V617F-induced endothelial expression. We reasoned that JAK2V617F endothelial cells could acquire a procoagulant phenotype due to the exposure to circulating inflammatory stimuli. We thus repeated the same experiments after overnight activation with TNF-alpha, but did not observe any difference. Additionally, we measured the rate of thrombin-triggered protein C activation and did not observe any difference between the cells demonstrating that JAK2V617F endothelial cells have normal anticoagulant properties.

JAK2V617F Endothelial Cells Have a Pro-Adhesive Phenotype

[0055] Exposition of endothelial cells to inflammatory stimuli leads to endothelial expression of adhesion molecules that which allows the rolling and adhesion of leukocytes. Teofili et al. reported increased adhesion of normal human mononuclear cells (MNC) on patients' endothelial cells derived from JAK2V617F ECFC, in static conditions. We first assessed whether our model of JAK2V617F transduced HUVECs reproduced such a proadhesive phenotype. We indeed observed increased adhesion of healthy MNCs and polymorphonuclear neutrophils (PMN). We then addressed the adhesive properties of JAK2V617F HUVECs in flow conditions and reported that more normal PMNs rolled and stably adhered to JAK2V617F-HUVECs as compared to JAK2WT-HUVECs. To assess whether this proadhesive phenotype was also observed in vivo, we visualized leukocyte interactions with mesenteric venules from Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice. We found that both leukocyte rolling and adhesion were significantly increased in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice only when they were previously administered with low dose TNF-alpha.

P-Selectin Expression is Increased in JAK2V617F Endothelial Cells

[0056] The adhesion of leukocytes to endothelial cells is mediated by Cell Adhesion Molecules (CAM) and selectins.sup.21. Flow cytometry analysis showed that JAK2V617F HUVECs expressed inter-CAM (ICAM), vascular-CAM (VCAM) and E-selectin at the same levels than JAK2WT HUVECs, whether or not they were previously activated with TNF-alpha. Immunostaining of non permeabilized carotid arteries from Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice showed an increased exposure of P-selectin at the endothelial cell surface in vivo, whether or not they were previously administered with TNF-alpha. Cell-surface P-selectin is susceptible to proteolytic cleavage that results in the shedding of its extracellular domain in the circulation. We thus reasoned that soluble P-selectin might be increased in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice. Using ELISA quantification, we observed increased levels of soluble P-selectin in the plasma of Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice, even after normalization with platelet count. Lastly we excluded increased platelet activation in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice by quantifying soluble Platelet Factor 4. Altogether, these results are in favor of increased membrane-attached and plasmatic soluble P-selectin from endothelial origin without increase in endothelial ICAM, VCAM or E-selectin. Interestingly, the levels of soluble P-selectin and the number of rolling leukocytes correlated in both groups (Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice and control) (Pearson correlation test, r=0.8244), indicating that increased P-selectin expression is associated with increased leukocyte rolling.

Endothelial Expression and Release of Von Willebrand Factor is Increased by JAK2V617F

[0057] Within endothelial cells, P-selectin is stored in exocytotic organelles called Weibel-Palade bodies together with von Willebrand factor (vWF). Exocytosis of Weibel Palade bodies leads to cell surface expression of vWF and P-selectin. In vitro, using immunostaining on non-permeabilized HUVECs, we observed increased vWF expression at the surface of JAK2V617F HUVECs, spontaneously and after overnight activation with 10 ng/ml TNF-alpha. Furthermore, quantification of vWF in the supernatant of HUVECs revealed higher amounts of vWF released by JAK2V617F HUVECs. Intra-cellular concentration of vWF was also increased in JAK2V617F HUVECs. Treating cells with TNF-alpha increased the secretion of vWF by JAK2WT HUVECs and to even greater levels by JAK2V617F HUVECs. Intracellular vWF levels were strongly reduced in TNF-treated cells and differences between JAK2V617F and JAK2WT HUVECs was abrogated after TNF-alpha treatment as the majority of Weibel-Palade bodies had likely already been released. These results were confirmed in vivo with higher levels of vWF antigen in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice compared with control mice. All together, our data showed that endothelial JAK2V617F increased vWF protein level and soluble vWF release, in line with increased P-Selectin expression at cell surface as a consequence of increased degranulation of Weibel-Palade bodies.

Increased P-Selectin Exposure is Involved in the Pro-Adhesive and Pro-Thrombotic Phenotype of JAK2V617F Endothelial Cells

[0058] To investigate a potential causal link between increased P-Selectin exposure and the pro-adhesive phenotype of JAK2V617F endothelial cells, we reproduced the same experiments as previously described, but in the presence of a P-selectin blocking antibody. Our in vitro approach showed a complete reversion of the hyperadhesive properties of JAK2V617F HUVECs after exposition with the P-selectin blocking antibody (FIGS. 1a and b). Quantification of leukocytes in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice treated with TNF-alpha and the P-selectin blocking antibody revealed a complete inhibition of leukocyte rolling and adhesion (FIG. 1c and d). To examine whether increased P-selectin was also responsible for thrombus formation, we used the model of low dose TNF-alpha induced lung thrombus formation. We observed that pre-treatment of the mice with the P-selectin blocking antibody completely abrogated thrombus formation in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice, but had no effect in control mice (FIG. 1e). Altogether, these results show that the pro-thrombotic phenotype of JAK2V617F endothelial cells is mainly the consequence of increased adhesive properties, due to overexpression of membrane P-selectin, secondary to degranulation of Weibel-Palade bodies.

Treatment with Hydroxyurea Abrogates TNF-Alpha Induced Thrombosis in Pdgfb-iCreERT2;JAK2.SUP.V617F/WT .Mice Through Decreased P-Selectin Expression

[0059] Hydroxyurea is an antimetabolite frequently used in MPN to reduce the occurrence of thrombosis. Its anti-thrombotic effect is reported to be via the reduction of blood cell counts. But hydroxyurea is also used in sickle cell disease to reduce vasoocclusive crisis, and its beneficial effect is in part mediated by a direct effect on endothelial cells, with a reduction in leukocyte adhesion.sup.22. We wondered whether hydroxyurea was capable of reducing the prothrombotic effect of JAK2V617F endothelial cells. We treated Pdgfb-iCreERT2;JAK2.sup.V617F/WT and control mice during 10 days with hydroxyurea, injected low dose TNF-alpha and quantified thrombus formation in lungs. We observed a significant inhibition of thrombus formation in Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice together with a reduction in leukocyte rolling and adhesion. In vitro, we observed a significant reduction of neutrophil adhesion on JAK2V617F HUVECs that had been pretreated 24 hours with hydroxyurea, demonstrating that the anti-adhesive effect of hydroxyurea observed in vivo is a direct effect on endothelial cells. We then examined whether the protective effect of hydroxyurea acted through reduction of endothelial P-selectin. We observed that P-selectin/platelets ratio was significantly lower in hydroxyurea-treated than in non-treated Pdgfb-iCreERT2;JAK2.sup.V617F/WT mice. In these same mice, we observed a significant decrease of endothelial membrane P selectin expression, confirming a direct effect of hydroxyurea on JAK2V617F endothelial cells. Lastly, we measured vWF in the supernatant of JAK2V617F HUVECs that had been treated during 24 hours with 100 M hydroxyurea. We observed decreased vWF concentrations, in favor of a reduced release of Weibel Palade bodies. Altogether, our in vitro and in vivo results show that hydroxyurea has a direct effect on JAK2V617F endothelial cells, decreasing endothelial P-selectin release and surface expression, thus decreasing endothelial cell pro-thrombotic phenotype.

Discussion

[0060] Despite significant advances in deciphering the molecular mechanisms responsible for MPN occurrence and transformation, the mechanisms that lead to thrombosis, the first cause of morbidity and mortality, remain largely elusive. Recent identification of the JAK2V617F mutation in endothelial cells of MPN patients.sup.15-17 opens new perspectives in the pathogenesis of thrombosis in MPNs. Here we demonstrate using a transgenic mouse model, that JAK2V617F-positive endothelial cells promote spontaneous thrombosis in basal conditions and have increased thrombotic response to weak inflammatory stimuli. We demonstrate that the mechanism that leads to thrombosis involves endothelial P selectin release and cell surface exposure and subsequent leukocyte rolling and adhesion. We also demonstrate that treatment with hydroxyurea decreases P-selectin endothelial expression and thrombus formation in mice expressing JAK2V617F only in endothelial cells. The link between P-selectin expression and thrombosis has already been described, and in most cases P-selectin is from platelet origin.sup.23-25. The mechanism of P-selectin mediated thrombosis involves neutrophil activation, either through tissue factor expression and activation of the extrinsic coagulation pathway, or through priming for neutrophil extracellular trap formation.sup.26, a process that is now well recognized as participating in thrombus formation.sup.27. Increase of endothelial P-selectin and subsequent thrombosis was reported in response to venous flow reduction and local hypoxia.sup.24. Here, we show for the first time that endothelial cells can have a constitutive increased expression of P-selectin, even without any hypoxic or inflammatory stimuli. Further studies are now required to decipher the specific molecular mechanism that is responsible for increased endothelial P-selectin expression in JAK2V617F expressing endothelial cells.

[0061] Given that MPN are acquired hematological malignancies, the description of prothrombotic JAK2V617F endothelial cells raises the question of their origin. JAK2V617F endothelial cells have been found using two approaches: culture of endothelial progenitors.sup.17,28-30 and microdissection.sup.15,16. Endothelial progenitors comprise (i) Colony Forming Unit-Endothelial Cell (CFU-EC) which are of hematopoietic origin, give rise to endothelial cells unable to proliferate nor form vessels in transplantation experiments, and (ii) Endothelial Colony Forming Cells (ECFC) which generate a progeny of phenotypically and functionally competent endothelial cells. Three groups have looked for the presence of JAK2V617F in ECFC and CFU-EC in MPN patients. Two out of three found JAK2V617F ECFC.sup.17,30, suggesting that the JAK2V617F mutation would have occurred in a progenitor cell of the hematopoietic and endothelial lineages. Such a cell certainly exists in the embryo but its existence in adults is a matter of debate. On the contrary, all groups found JAK2V617F CFU-EC in all MPN patients, a result that is not surprising given that CFU-EC are of hematopoietic origin. In the case of microdissection experiments.sup.15,16, it is possible that the JAK2V617F endothelial cells that have been microdissected are of hematopoietic origin, as monocytes are known to integrate into the vessel wall after an injury, acquiring the phenotype of mature endothelial cells.sup.31. Altogether, in MPN, the presence of JAK2V617F endothelial cells of real endothelial origin is probably rare but the presence of JAK2V617F endothelial cells of hematopoietic origin is common. Such cells probably integrate into the vessel wall after a vascular lesion, thus giving rise to a bedding of JAK2V617F endothelial cells. Demonstrating that JAK2V617F endothelial cells have a prothrombotic phenotype is thus particularly relevant in our understanding of the pathogenesis of thrombosis in MPN.

[0062] Our study has important therapeutic implications. We demonstrate that treatment with hydroxyurea inhibits the pathological hyperadhesive phenotype of JAK2V617F endothelial cells. The results presented here challenge current thinking, according to which the antithrombotic effect of hydroxyurea in MPN is only mediated by lowering blood cell count. However, a direct effect of hydroxyurea on endothelial cells was already reported with increased NO and cGMP production.sup.32, probably via inhibition of eNOS degradation by proteasome.sup.33. Moreover, in patients with sickle cell disease, treatment with hydroxyurea efficiently decreases vaso-occlusive crisis frequency.sup.34. This antithrombotic effect was shown to be a direct effect on endothelial cells via stimulation of the NO-cGMP pathway, and reduction of leukocyte rolling in the microvasculature of TNF-alpha treated sickle cell mice.sup.22. Our result suggests that hydroxyurea should be considered in priority in patients with MPN and a history of thrombosis. This is often the case as hydroxyurea is the first line therapy in these high-risk patients. But some patients, due to young age or intolerance to high doses of hydroxyurea, take second line therapies such as interferon or anagrelide. We question whether they would still benefit from hydroxyurea treatment in association with other drugs to maintain thrombosis protection.

[0063] Another therapeutic implication of our work comes from our demonstration that increased endothelial P-selectin favors thrombosis in MPN. In sickle cell disease, various factors such as thrombin, platelet-activating factor, TNF-alpha, and sickle cells themselves are responsible for endothelial activation. Increased endothelial P-selectin expression was shown to participate in the occurrence of vasoocclusive crisis.sup.35 and very recently, a clinical trial demonstrated that treatment with an anti-P-selectin antibody, crizanlizumab, efficiently prevented pain crises in sickle cell disease.sup.36. It is thus tempting to speculate that such treatment could have therapeutical benefits in MPN patients with high thrombotic risk, who need to receive anticoagulant treatment for a MPN-related thrombosis (such as splanchnic thrombosis).

[0064] We believe our work raises a new concept as we show for the first time that endothelial cells bearing a genetic mutation acquire a prothrombotic phenotype. It is well known that endothelial cells, when stimulated by extrinsic factors such as inflammatory cytokines, hypoxia or antiphospholipid antibodies, become activated and promote thrombosis. There are some examples of constitutional mutations affecting endothelial cell functions and particularly angiogenic properties.sup.37,38 but, to our knowledge, no constitutional or acquired mutation affecting endothelial cell prothrombotic profile had been described to date. Using a mouse model with a specific expression of JAK2V617F in endothelial cells mimicking the presence of JAK2V617F in endothelial cells found in patients.sup.15-17, we provide insights into the pathogenesis of thrombosis in MPN, describe the pathological phenotype of intrinsically mutated endothelial cells, and provide evidences for the use of hydroxyurea in preventing thrombosis.

[0065] Altogether, we believe that our work is clinically relevant for three reasons: (i) it suggests that specific biomarkers of endothelial cell activation should be intensively looked for to determine which MPN patients are at high thrombotic risk. (ii) it opens the route to new therapeutic options in MPN, such as hydroxyurea in patients with high thrombotic risk or high endothelial activation markers, and anti-P-selectin antibody instead or in addition to standard care to prevent thrombosis in high risk MPN patients. (iii) it provides the proof of concept that an acquired genetic mutation can alter the phenotype of endothelial cells as shown for the acquired prothrombotic phenotype acquired upon expression of the JAK2 mutation. This suggests that other activating mutations in endothelial cells could be causal in thrombotic disorders of unknown cause, which account for 50% of recurrent venous thrombosis.

REFERENCES

[0066] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.

[0067] 1. Arber, D. A., et al. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood 127, 2391-2405 (2016).

[0068] 2. Baxter, E. J., et al. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet 365, 1054-1061 (2005).

[0069] 3. James, C., et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature 434, 1144-1148 (2005).

[0070] 4. Kralovics, R., et al. A gain-of-function mutation of JAK2 in myeloproliferative disorders. The New England journal of medicine 352, 1779-1790 (2005).

[0071] 5. Levine, R. L., et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer cell 7, 387-397 (2005).

[0072] 6. Elliott, M. A. & Tefferi, A. Pathogenesis and management of bleeding in essential thrombocythemia and polycythemia vera. Current hematology reports 3, 344-351 (2004).

[0073] 7. Barbui, T., Finazzi, G. & Falanga, A. Myeloproliferative neoplasms and thrombosis. Blood 122, 2176-2184 (2013).

[0074] 8. Falanga, A., Marchetti, M., Vignoli, A., Balducci, D. & Barbui, T. Leukocyte-platelet interaction in patients with essential thrombocythemia and polycythemia vera. Experimental hematology 33, 523-530 (2005).

[0075] 9. Barbui, T., Carobbio, A., Rambaldi, A. & Finazzi, G. Perspectives on thrombosis in essential thrombocythemia and polycythemia vera: is leukocytosis a causative factor? Blood 114, 759-763 (2009).

[0076] 10. Landolfi, R., et al. Leukocytosis as a major thrombotic risk factor in patients with polycythemia vera. Blood 109, 2446-2452 (2007).

[0077] 11. De Grandis, M., et al. JAK2V617F activates Lu/BCAM-mediated red cell adhesion in polycythemia vera through an EpoR-independent Rap1/Akt pathway. Blood 121, 658-665 (2013).

[0078] 12. Alonci, A., et al. Evaluation of circulating endothelial cells, VEGF and VEGFR2 serum levels in patients with chronic myeloproliferative diseases. Hematological oncology 26, 235-239 (2008).

[0079] 13. Belotti, A., et al. Circulating endothelial cells and endothelial activation in essential thrombocythemia: results from CD146+ immunomagnetic enrichment-flow cytometry and soluble E-selectin detection. American journal of hematology 87, 319-320 (2012).

[0080] 14. Cella, G., et al. Nitric oxide derivatives and soluble plasma selectins in patients with myeloproliferative neoplasms. Thrombosis and haemostasis 104, 151-156 (2010).

[0081] 15. Sozer, S., et al. The presence of JAK2V617F mutation in the liver endothelial cells of patients with Budd-Chiari syndrome. Blood 113, 5246-5249 (2009).

[0082] 16. Rosti, V., et al. Spleen endothelial cells from patients with myelofibrosis harbor the JAK2V617F mutation. Blood 121, 360-368 (2012).

[0083] 17. Teofili, L., et al. Endothelial progenitor cells are clonal and exhibit the JAK2(V617F) mutation in a subset of thrombotic patients with Ph-negative myeloproliferative neoplasms. Blood 117, 2700-2707 (2011).

[0084] 18. Wakefield, T. W., Myers, D. D. & Henke, P. K. Mechanisms of venous thrombosis and resolution. Arteriosclerosis, thrombosis, and vascular biology 28, 387-391 (2008).

[0085] 19. Hasan, S., et al. JAK2V617F expression in mice amplifies early hematopoietic cells and gives them a competitive advantage that is hampered by IFNalpha. Blood 122, 1464-1477 (2013).

[0086] 20. Ollivier, V., et al. Bioreactivity of stent material: Activation of platelets, coagulation, leukocytes and endothelial cell dysfunction in vitro. Platelets, 1-11 (2016).

[0087] 21. Ley, K., Laudanna, C., Cybulsky, M. I. & Nourshargh, S. Getting to the site of inflammation: the leukocyte adhesion cascade updated. Nature reviews 7, 678-689 (2007).

[0088] 22. Almeida, C. B., et al. Hydroxyurea and a cGMP-amplifying agent have immediate benefits on acute vaso-occlusive events in sickle cell disease mice. Blood 120, 2879-2888 (2012).

[0089] 23. Andre, P., Hartwell, D., Hrachovinova, I., Saffaripour, S. & Wagner, D. D. Pro-coagulant state resulting from high levels of soluble P-selectin in blood. Proceedings of the National Academy of Sciences of the United States of America 97, 13835-13840 (2000).

[0090] 24. von Bruhl, M. L., et al. Monocytes, neutrophils, and platelets cooperate to initiate and propagate venous thrombosis in mice in vivo. The Journal of experimental medicine 209, 819-835 (2012).

[0091] 25. Zetterberg, E., et al. Abnormal P-selectin localization during megakaryocyte development determines thrombosis in the gata1low model of myelofibrosis. Platelets 25, 539-547 (2014).

[0092] 26. Etulain, J., et al. P-selectin promotes neutrophil extracellular trap formation in mice. Blood 126, 242-246 (2015).

[0093] 27. Martinod, K. & Wagner, D. D. Thrombosis: tangled up in NETs. Blood 123, 2768-2776 (2014).

[0094] 28. Helman, R., et al. Granulocyte whole exome sequencing and endothelial JAK2V617F in patients with JAK2V617F positive Budd-Chiari Syndrome without myeloproliferative neoplasm. British journal of haematology (2016).

[0095] 29. Piaggio, G., et al. Endothelial colony-forming cells from patients with chronic myeloproliferative disorders lack the disease-specific molecular clonality marker. Blood 114, 3127-3130 (2009).

[0096] 30. Yoder, M. C., et al. Redefining endothelial progenitor cells via clonal analysis and hematopoietic stem/progenitor cell principals. Blood 109, 1801-1809 (2007).

[0097] 31. Murdoch, C., Muthana, M., Coffelt, S. B. & Lewis, C. E. The role of myeloid cells in the promotion of tumour angiogenesis. Nature reviews. Cancer 8, 618-631 (2008).

[0098] 32. Cokic, V. P., et al. Hydroxyurea induces the eNOS-cGMP pathway in endothelial cells. Blood 108, 184-191 (2006).

[0099] 33. Cokic, V. P., Beleslin-Cokic, B. B., Noguchi, C. T. & Schechter, A. N. Hydroxyurea increases eNOS protein levels through inhibition of proteasome activity. Nitric oxide: biology and chemistry 16, 371-378 (2007).

[0100] 34. Charache, S., et al. Effect of hydroxyurea on the frequency of painful crises in sickle cell anemia. Investigators of the Multicenter Study of Hydroxyurea in Sickle Cell Anemia. The New England journal of medicine 332, 1317-1322 (1995).

[0101] 35. Matsui, N. M., et al. P-selectin mediates the adhesion of sickle erythrocytes to the endothelium. Blood 98, 1955-1962 (2001).

[0102] 36. Ataga, K. I., et al. Crizanlizumab for the Prevention of Pain Crises in Sickle Cell Disease. The New England journal of medicine 376, 429-439 (2017).

[0103] 37. Labauge, P., Denier, C., Bergametti, F. & Tournier-Lasserve, E. Genetics of cavernous angiomas. The Lancet. Neurology 6, 237-244 (2007).

[0104] 38. Larrivee, B., et al. ALK1 signaling inhibits angiogenesis by cooperating with the Notch pathway. Developmental cell 22, 489-500 (2012).

[0105] 39. Falati, S., et al. Accumulation of tissue factor into developing thrombi in vivo is dependent upon microparticle P-selectin glycoprotein ligand 1 and platelet P-selectin. The Journal of experimental medicine 197, 1585-1598 (2003).

[0106] 40. Su, Y., Lei, X., Wu, L. & Liu, L. The role of endothelial cell adhesion molecules P-selectin, E-selectin and intercellular adhesion molecule-1 in leucocyte recruitment induced by exogenous methylglyoxal. Immunology 137, 65-79 (2012).

[0107] 41. Dargaud, Y., et al. Evaluation of a standardized protocol for thrombin generation measurement using the calibrated automated thrombogram: an international multicentre study. Thrombosis research 130, 929-934 (2012).

[0108] 42. Claxton, S., et al. Efficient, inducible Cre-recombinase activation in vascular endothelium. Genesis 46, 74-80 (2008).

[0109] 43. Chauhan, A. K., et al. ADAMTS13: a new link between thrombosis and inflammation. The Journal of experimental medicine 205, 2065-2074 (2008).

[0110] 44. Wegmann, F., et al. ESAM supports neutrophil extravasation, activation of Rho, and VEGF-induced vascular permeability. The Journal of experimental medicine 203, 1671-1677 (2006).

[0111] 45. Rayes, J., et al. Mutation and ADAMTS13-dependent modulation of disease severity in a mouse model for von Willebrand disease type 2B. Blood 115, 4870-4877 (2010).