Regular Article

PLATELETS AND THROMBOPOIESIS Hemostatic disorders in a JAK2V617F-driven mouse model of myeloproliferative neoplasm

Lamia Lamrani,1,2 Catherine Lacout,3-5 V´eronique Ollivier,1,2 C´ecile V. Denis,4,6 Elizabeth Gardiner,7 Benoit Ho Tin Noe,1,2 William Vainchenker,3-5 Jean-Luc Villeval,3-5 and Martine Jandrot-Perrus1,2

1INSERM, U1148, Paris, France; 2Universit´e Paris Denis Diderot 7, Paris, France; 3INSERM, U1009, Villejuif, France; 4Universit´e Paris-Sud 11, Villejuif, France; 5Institut Gustave Roussy, Villejuif, France; 6INSERM, U770, Le Kremlin-Bicˆetre, France; and 7Monash University, Clayton, VIC, Australia

Key Points Thrombosis is common in patients suffering from myeloproliferative neoplasm (MPN), whereas bleeding is less frequent. JAK2V617F, the main mutation involved in MPN, is • Mice constitutively developing considered as a risk factor for thrombosis, although the direct link between the mutation a JAK2V617F-induced PV and hemostatic disorders is not strictly established. We investigated this question using exhibit a bleeding tendency conditional JAK2V617F knock-in mice with constitutive and inducible expression of V617F combined with the accelerated JAK2 in hematopoietic cells, which develop a polycythemia vera (PV)-like disorder evolving into myelofibrosis. In vitro, thrombosis was markedly impaired with an 80% formation of unstable clots. V617F • Hemostatic defects are not decrease in platelet-covered surface, when JAK2 blood was perfused at arterial shear over collagen. JAK2V617F platelets presented only a moderate glycoprotein (GP) VI deficiency concomitant with JAK2V617F not responsible for the defective platelet accumulation. In contrast, a decreased proportion expression, suggesting they of high-molecular-weight von Willebrand factor multimers could reduce platelet adhesion. are not directly caused by the Accordingly, the tail bleeding time was prolonged. In the FeCl3-induced thrombosis model, mutation. platelet aggregates formed rapidly but were highly unstable. Interestingly, vessels were considerably dilated. Thus, mice developing PV secondary to constitutive JAK2V617F ex- pression exhibit a bleeding tendency combined with the accelerated formation of unstable clots, reminiscent of observations made in patients. Hemostatic defects were not concomitant with the induction of JAK2V617F expression, suggesting they were not directly caused by the mutation but were rather the consequence of perturbations in blood and vessel homeostasis. (Blood. 2014;124(7):1136-1145) Introduction

Myeloproliferative neoplasms (MPNs) are clonal diseases arising often conflicting or of uncertain clinical or pathogenic relevance. from hematopoietic stem cells harboring molecular defects that However, thrombosis appears to be more frequent in patients with the promote an excessive and cytokine-deregulated production of mature JAK2V617F mutation. This feature has been critically reevaluated myeloid blood cells. According to the classification of the World Health in 3 independent large meta-analyses that reported an increased Organization, classical BCR-ABL-negative MPN includes polycythe- risk of venous and arterial thrombosis (odds ratio 2.09 and 1.68, mia vera (PV), essential thrombocythemia (ET), and primary myelo- 1.96 and 2.5, respectively),8,9 of thrombosis at presentation (odds fibrosis (PMF). The unique acquired somatic JAK2V617F mutation is ratio 1.88),8 and an increased rate of thrombotic events (32% vs present in .95% of PV and 50% of ET and PMF patients.1-4 20%)10 in patients with the JAK2V617F mutation compared with The V617F mutation in the pseudokinase domain constitutively their wild-type (WT) counterparts. Furthermore, rare ET patients activates cytokine signaling via the formation of active complexes harboring a homozygous JAK2V617F mutation had a higher risk of with type I receptors, such as the erythropoietin receptor or the thrombosis than WT and heterozygous subjects and presented an thrombopoietin (TPO) receptor.1 increased risk of thrombosis recurrence.11,12 Platelets are assumed Disease-related hemostatic complications impair life expectancy of to impact the risk of thrombosis, not only because they are in- patients with PV and ET. Arterial and venous thrombosis are the main creased in number but also because of their hyperreactivity, as sug- cause of death and morbidity and are often the initial event leading gested by the increased P-selectin exposure on their surface.13-15 to diagnosis.5 Their incidence ranges from 12% to 39% in PV and 11% Enhanced urinary excretion of thromboxane metabolites in untreated to 25% in ET.6 Hemorrhages are reported with a low ;3% incidence ET patients is also in favor of in vivo platelet activation.16 Therefore, in ET and PV but more frequently (;10%) in PMF patients.7 an association between the JAK2V617F mutation and platelet Mechanisms underlying the MPN thrombotic diathesis are still activation is suggested, but the direct assessment of this mutation’s largely elusive, and biological studies have provided data that were effect on platelet function in patients with MPNs is skewed by the

Submitted October 7, 2013; accepted June 2, 2014. Prepublished online as The publication costs of this article were defrayed in part by page charge Blood First Edition paper, June 20, 2014; DOI 10.1182/blood-2013-10- payment. Therefore, and solely to indicate this fact, this article is hereby 530832. marked “advertisement” in accordance with 18 USC section 1734.

The online version of this article contains a data supplement. There is an Inside Blood Commentary on this article in this issue. © 2014 by The American Society of Hematology

1136 BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7 BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7 HEMOSTATIC DISORDERS IN A JAK2V617F PV MOUSE MODEL 1137 variable amplification of the JAK2V617F mutant clone, additional The tail bleeding time (TBT) from VavCre/JAK2V617F KI mice molecular defects, and the frequently associated prophylactic treat- was assessed 2 months posttransplantation. It was characterized ment of thrombosis. by frequent, short transient arrests followed by immediate intense The recently described knock-in (KI) mouse MPN models, in and prolonged bleeding that required stopping by the experimenter which JAK2V617F expression is inducible and tissue specific, offer an after 15 minutes (Figure 1Ai). Accordingly, the volume of blood attractive alternative for the precise analysis of the impact of the lost was significantly increased (Figure 1Aii). In vehicle-treated mutation on hemostatic functions.17-20 These mice faithfully model SCLCreERt/JAK2V617F KI mice or tamoxifen-treated WT mice human PV evolving into myelofibrosis (MF), and reports of cardiac (not shown), neither the TBT nor the blood loss significantly changed thrombosis and gangrenous bowel are consistent with clinical events whatever the time after induction. In contrast, the TBT and the observed in MPN patients.17,19,20 Our objective was to use these KI volume of blood lost progressively increased from 2 weeks after mice to assess the role of JAK2V617F in the deregulation of hemostasis, treatment by tamoxifen of SCLCreERt/JAK2V617F KI mice, and with a special interest in platelets. 2 months posttreatment, bleeding needed to be stopped as in VavCre/ JAK2V617F KI mice (Figure 1Bi-ii). The frequency of transient bleeding interruptions increased from treatment to day 30 and de- Materials and methods creased at day 60 when bleeding became continuous (Figure 1C).

V617F A detailed methods section can be found in the supplemental Methods Thrombosis is impaired ex vivo in JAK2 mice available on the Blood Web site. 2 When whole blood was perfused at 1500 s 1 over collagen, WT Animal models platelets rapidly adhered to the surface and formed aggregates. In contrast, the adhesion of VavCre/JAK2V617F platelets was delayed All animal experiments were carried out at the animal facilities of the Institut and the formation of stable aggregates was markedly impaired Gustave Roussy and of Inserm U1148 and approved by the local review (Figure 2A). After 5 minutes of flow, surface coverage was decreased boards (protocol no. 2012-061), under the conditions established by the by 80% for VavCre/JAK2V617F KI compared with WT platelets European Community (Directive 86/609/CCE). (Figure 2B). The capacity of platelets from SCLCreERt/JAK2V617F JAK2V617F KI mice. Conditional JAK2V617F KI mice were crossed with VavCre or SCLCreERt transgenic mice for fetal or tamoxifen-inducible KI mice to adhere to collagen dramatically dropped 2 weeks after adult expression of JAK2V617F in hematopoietic tissues, respectively. Bone induction of the mutation and was comparable to that of platelets V617F marrow (BM) cells from VavCre/JAK2V617F KI or SCLCreERt /JAK2V617F of VavCre/JAK2 KI (Figure 2C-D). KI mice were grafted into lethally irradiated WT recipients leading to con- Thrombus formation was further analyzed in vitro by thromboe- stitutive or inducible heterozygous expression of the mutation in hemato- lastometry on whole blood from VavCre/JAK2V617F KI mice and poietic cells only. Recipients from both donor mice develop a PV-like their matched controls (supplemental Figure 2). The clotting time disease evolving into MF.18,20 Studies were performed 9 weeks after and clot formation time of JAK2V617F blood were both prolonged as V617F transplantation of VavCre/JAK2 KI BM cells or 6, 13, 30, and 60 compared with controls, whereas the maximum clot firmness was t V617F days after tamoxifen administration to SCLCreER /JAK2 KI recipient decreased, and maximum lysis increased. mice. Control mice were WT mice grafted with WT BM cells or vehicle- treated SCLCreERt/JAK2V617F KI recipient mice. JAK2V617F KI platelets show significantly reduced TPOhigh model. BM cells transduced with a retrovirus expressing the activation response murine TPO gene were grafted into irradiated WT recipient mice resulting in 21 an ET-like disease rapidly evolving into MF, as previously described. These The platelet surface expression levels of the main glycoproteins (GPs), mice were used 12 weeks posttransplantation. GPIba,GPV,andintegrinsa2andaIIbb3, were similar to controls Secondary polycythemia model. WT mice were treated by intraperitoneal in VavCre/JAK2V617F, EPO-treated, TPOhigh,andSCLCreERt/ injection of human recombinant EPO (epoetine alfa 5000 UI/kg) every 2 days V617F during 3 weeks. JAK2 KI mice, whatever the time of analysis (supplemental Table 2). In contrast, surface-expressed GPVI was systematically de- creased in VavCre/JAK2V617F KI mice compared with control Results mice as indicated by a 40% mean reduction of the GPVI labeling, Hematologic parameters in JAK2V617F KI mice although variations between mice were observed (Figure 3A). Accordingly, on immunoblots, the intensity of the GPVI band was As previously reported, a PV-like disorder was observed 2 months deceased to 45 6 12% of the mean value obtained for WT platelets after transplantation in VavCre/JAK2V617F BM recipients20 but also (Figure 3Biii). Furthermore, the FcRg chain that is constitutively in SCLCreERt/JAK2V617F BM recipient mice treated with tamoxifen associated with GPVI was also decreased by 36 6 10% in VavCre/ (supplemental Table 1A-B). The size of the spleen was doubled in JAK2V617F KI platelets (Figure 3Bi). The treatment of control platelets VavCre/JAK2V617F mice as compared with WT (supplemental by N-ethylmaleimide (NEM) resulted in the appearance of a band Figure 1Ai). corresponding to the intracellular GPVI remnant that reflects GPVI Transplantation of TPO-transduced BM cells (TPOhigh model)21 shedding. This band was not observed in the extracts of JAK2V617F resulted, 12 weeks after transplantation, in a reduction of the initial platelets unless they were treated by NEM, ruling out the hypothesis thrombocytosis to normal platelet levels (1029 6 492 3 106/mL), that GPVI was cleaved in vivo by metalloproteases (Figure 3Bii). In variable leukocytosis (14 6 12 3 106/mL), severe anemia (hematocrit SCLCreERt/JAK2V617F KI mice, the expression of GPVI remained 19 6 6.6%), and a 1.7-fold enlargement of the spleen. EPO-treated normal until 2 months after tamoxifen administration and then mice exhibited isolated erythrocytosis (red blood cell [RBC] con- dropped to the same level as observed on VavCre/JAK2V617F KI centration: 1150 6 66 3 109/mL, hematocrit 60 6 5.5%). Only slight platelets (Figure 3C). Of note, GPVI was also decreased by 20% on splenomegaly was observed in EPO-treated mice (supplemental platelets from TPOhigh mice but was normal on platelets from EPO- Figure 1Aiii). treated mice (Figure 3D). 1138 LAMRANI et al BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7

Figure 1. TBT. The TBT was measured 2 months post- BM grafting in WT and VavCre/JAK2V617F KI mice (A, n 5 8) and at different times after administration of tamoxifen (TM, n 5 8) or vehicle (VH, n 5 22) in SCLCreERt/JAK2V617F KI mice (B). The arrow at 15 minutes indicates that bleeding was arrested by the experimenter. Each point represents an individual mouse. The volume of blood lost in 15 minutes was measured in VavCre/JAK2V617F KI (Aii) and in SCLCreERt/JAK2V617F KI mice (Bii). (C) The frequency of transient bleeding arrests observed in SCLCreERt/ JAK2V617F KI mice at different times after administra- tion of tamoxifen was quantified. *P , .05; **P , .004; ***P , .0001.

In nonstimulated whole blood, the levels of P-selectinand activated of medium- and low-sized multimers (Figure 5B). In SCLCreERt/ integrin aIIbb3 were low both on JAK2V167F and control platelets JAK2V617F KI mice, the plasma concentration of VWF tended to (Figure 4A). After stimulation by convulxin, rhodocytin, or thrombin increase with time after treatment by tamoxifen (Figure 5Aii). The receptor-activating peptide 4 (TRAP4), integrin aIIbb3 became proportion of low-molecular-weight multimers increased from day activated but to a lesser extent on VavCre/JAK2V617F platelets than 13 posttreatment (Figure 5C). These data indicate an enhanced in controls, and similar data were obtained for P-selectin exposure proteolysis of VWF in JAK2V617F-expressing mice. (data not shown). In SCLCreERt/JAK2V617F KI mice, alterations in P-selectin exposure in response to convulxin (Figure 4Bi) or Vascular abnormalities in JAK2V617F mice rhodocytin (Figure 4Bii) were only observed 2 weeks after induction by tamoxifen. Similar data were obtained for the activation of When mesenteric vessels were exposed to study in vivo thrombosis, integrin aIIbb3, and dose-response curves showed that increasing we observed that the diameter of the vessels was increased in convulxin doses did not restore a full response (Figure 4C). VavCre/JAK2V617F KI mice as compared with controls (Figure 6A). In addition, a similar increased diameter of mesenteric vessels was high VWF proteolysis in JAK2V617F mice observed in EPO-treated mice but not in TPO mice, suggesting an adaptive process to the high hematocrit (Figure 6Ai). In SCLCreERt/ Because neither the moderate GPVI deficiency nor the platelet JAK2V617F mice, the diameters of mesenteric vessels were normal hyporesponsiveness was sufficient to explain the defect in collagen- 1 month after tamoxifen administration but were increased as in induced thrombosis in flow conditions, we analyzed plasma von VavCre/JAK2V617F KI mice 2 months after tamoxifen administra- Willebrand factor (VWF) in VavCre/JAK2V617F and SCLCreERt/ tion (Figure 6B). JAK2V617F KI mice (Figure 5). The plasma concentration of VWF Increased vessel diameter was also observed in other vascular was not different in VavCre/JAK2V617F compared with WT mice, territories, such as the aorta and ear vessels. Histologic analysis of but electrophoretic analysis revealed a decreased intensity of the aortic sections from VavCre/JAK2V617F mice showed an increased bands of high molecular weight, together with an increased intensity luminal surface, and the vessel wall presented signs of distension: in BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7 HEMOSTATIC DISORDERS IN A JAK2V617F PV MOUSE MODEL 1139

Figure 2. Collagen-induced in vitro thrombosis. Whole blood was perfused for 5 minutes over an immo- bilized collagen surface at 1500 s21. The adhesion of

DiOC6-labeled platelets was recorded. (A) Views of platelet adhesion at different times (30 seconds, 1 minute, 2 minutes, and 5 minutes) of 1 out of 6 significant experiments from VavCre/JAK2V617F (KI) and WT mice are shown. Original magnification 320 for all panels. Quantification of the covered surface obtained with VavCre/JAK2V617F platelets (KI, n 5 6) and of SCLCreERt/JAK2V617F platelets (TM, n 5 4 per group) and their matched controls is shown in panels B and C, respectively. **P , .007. (D) Mosaic views of SCLCreERt/JAK2V617F (TM) platelets and their con- trols (VH) are shown at day 6 (D6) and day 60 (D60).

the media, the numbers of smooth muscle cells and elastic laminae It is noteworthy that platelet accumulation at the site of vascular were unchanged compared with control mice, but they appeared injury occurred at a normal rate in TPOhigh mice. However, complete distended (Figure 6C). We suspected that the increased vascular occlusion was hardly ever reached, whereas EPO- and vehicle- diameter represented a process of adaptation to the increased blood treated mice were indistinguishable with regard to occlusion time viscosity (approximately twofold) because of the high hematocrit. and stability of thrombi, strongly suggesting that erythrocytosis Indeed, VavCre/JAK2V617F KI mice had a normal heart rate was not involved in the thrombus defects observed in JAK2V617F (492 6 15.28inWTmicevs5096 85.58 bpm in KI mice) and KI mice (supplemental Figure 3). electrocardiogram. Furthermore, calculation of the shear indicated that, because of this dilatation, it was increased by only 35% in the aortas of VavCre/JAK2V617F KI mice compared with controls. Discussion Thrombosis is accelerated in vivo, but thrombi are unstable in JAK2V617F mice Arterial and venous thrombosis is the main cause of death and morbidity in PV and ET, whereas hemorrhages are more often An FeCl3-induced thrombosis model was then used to evaluate the observed in PMF. However, because of the heterogeneity of influence of JAK2V617F on localized thrombus formation. Vessels of patients and associated prophylactic treatment, these important similar diameter were selected in WT and KI mice. Platelet accumu- health issues are difficult to investigate in human patients. lation started very rapidly, only a few minutes after the injury in JAK2V617F KI mice were generated, and several studies have VavCre/JAK2V617F KI mice, and time to occlusion was shorter than shown that they faithfully model human PV, evolving into MF. in control mice (19.1 6 5.0 minutes, n 5 7 vs 39.2 6 5.8 minutes, Furthermore, lethality in these models is strongly suspected to be of n 5 6) (Figure 7A). In SCLCreERt/JAK2V617F KI mice, the rate of hemostatic/thrombotic origin.17,19 Therefore, investigation of the thrombus formation was identical to controls up to 1 month after role of JAK2V617F expression in hemostasis and thrombosis using tamoxifen administration; however, 1 month later, it was increased this mouse model is justified. Both constitutive (VavCre) and as in VavCre/JAK2V617F KI mice. Once the thrombus was formed, it inducible (SCLCreERt) JAK2V617F KI models were studied to was stable in control animals. However, in VavCre/JAK2V617F KI determine the impact of JAK2V617F on these parameters with mice and SCLCreERt/JAK2V617F KI mice 2 months postinduction, a special focus on platelet function. VavCre or SCLCreERt mice we observed images of thrombolysis (Figure 7B) and also of express the recombinase Cre in hematopoietic cells but also in embolization (Figure 7C). some endothelial cells. Transplantations of BM cells taken from 1140 LAMRANI et al BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7

Figure 3. JAK2V617F platelets present a GPVI deficiency. (A) GPVI expression was quantified by flow cytometry (mean fluorescence intensity, MFI) on the platelets of WT (n 5 11) and VavCre/JAK2V617F (n 5 24) mice 2 months after BM graft. ***P , .0003. (Bi) Representative (of 3) immunoblots of platelet extracts (10 mg) from 4 different WT mice and 4 different VavCre/JAK2V617F (KI) mice are shown. The same membrane was stripped and reblotted with antibodies against GPVI (JAQ1), the FcRg chain, and b-actin. (Bii) The shedding of GPVI was analyzed. One representative immunoblot is shown (n 5 3). Platelets from WT mice and VavCre/JAK2 V617F (KI) mice were treated ornotwithNEM.Plateletextracts(10mg) were immunoblotted using a polyclonal antibody to the intracellular domain of GPVI. GPVI-IC indicates the GPVI remnant cytoplasmic domain that remained associated with platelets after shedding of the extracellular domain. (Biii) Densitometric analysis of the GPVI band from 4 different mice per group. *P , .03. (C) Surface expression of GPVI was measured on SCLCreERt /JAK2 V617F (TM) at different times after administration of tamoxifen (n 5 7 for each group). **P , .002. (D) Platelet surface expression of GPVI in TPOhigh (n 5 9) and EPO-treated mice (n 5 3) is shown.

VavCre/JAK2V617F KI or SCLCreERt/JAK2V617F KI mice were the morphology of vessels in patients, most studies have been performed in order to express the JAK2V617F mutation exclusively in performed in the BM. Zetterberg et al observed vessels of larger hematopoietic cells. Furthermore, SCLCreERt/JAK2V617F KI, in perimeter in the BM of patients with MF.22 Also, interestingly, contrast to VavCre/JAK2V617F KI, allows expression of the mutation a reduction in flow-mediated vasodilatation was observed in PV in adult mice as in sporadic MPN with dissociation of the direct patients consistent with constitutive dilatation.23 The observation effect of JAK2 activation from the indirect effect of the resulting of similarly dilated vessels in EPO-treated mice, but not in TPOhigh long-term disease. Finally, we evaluated the relative roles of mice, is consistent with the proposal that vasodilatation represents thrombocytosis/MF or erythrocytosis in the thrombotic defects by a process of adaptation to the increased blood viscosity. using 2 non-JAK2V617F models: the TPOhigh retroviral (RV) model or Vasodilatation reduces vascular resistance and shear forces. Indeed, EPO-treated mice, respectively. recording the electrocardiogram showed no change in heart rate or We report an increased diameter of the vessels in JAK2V617F KI signs of overload in JAK2V617F KI mice. The observation that the mice in different vascular territories including mesentery, ear, and vessels of EPO-treated mice are similarly dilated to those of aorta. Histology performed on transverse sections indicated that JAK2V617F KI mice, whereas EPO-treated mice do not present the increased diameter was attributable to distension of the vascular hemostatic alterations, rules out the possibility that the changes in wall, which was not modified in terms of numbers of cells and the vessel diameter are sufficient to impact hemostasis and elastic laminae, but both had a stretched appearance. Regarding thrombosis. BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7 HEMOSTATIC DISORDERS IN A JAK2V617F PV MOUSE MODEL 1141

Figure 4. P-selectin exposure or JON/A binding on resting and activated platelets. Whole blood was incubated with buffer, convulxin (Cvx, 2.5 mg/mL), rhodocytin (rhodo, 50 nM), or TRAP4 (1000 mM) for 15 minutes at room temperature before measurement of P-selectin or JON/A (GPIIbIIa active form) exposure at the platelet surface. (A) Experiments performed in VavCre/JAK2V617F mice (KI) and their controls (WT). ***P , .0001. (B) Experiments were performed in SCLCreERt/JAK2V617F mice at different times after administration of tamoxifen (white squares) or vehicle (black circles), incubated with convulxin (i) or rhodocytin (ii). (C) Surface expression of the active form of integrin aIIbb3 (JON/A) was quantified on SCLCreERt/JAK2V617F platelets (TM) activated in whole blood by 3 doses of convulxin (0.5, 1, and 2.5 mg/mL) at different times after administration of tamoxifen.

Hemostasis and thrombosis were analyzed in vitro and in vivo. unlikely to affect adhesion and activation in thrombus formation Globally, VavCre/JAK2V617F KI mice were characterized by (1) a according to a previous report indicating that a 20% expression of marked deficiency in thrombus formation in vitro in collagen-coated GPVI is sufficient to permit platelet activation by collagen.27 Because flow chambers, (2) a mild GPVI deficiency, (3) hyporesponsive the GPVI deficiency occurred late in the course of the disease when platelets, (4) a decreased plasma content in large VWF multimers, (5) signs of MF were observed, we hypothesized that the GPVI deficiency a prolonged bleeding time, and (6) the rapid formation of occlusive could be related to fibrosis. The observation that platelets from high but unstable thrombi in FeCl3-injured mesenteric vessels. In TPO mice also exhibited significant GPVI deficiency supports this SCLCreERt/JAK2V617F KI mice, impaired in vitro thrombosis, low hypothesis. Cleavage of the GPVI extracellular domain was undetect- platelet reactivity, VWF proteolysis, and prolonged bleeding time able in JAK2V617F platelets, arguing against GPVI shedding as a cause occurred 2 weeks after induction of the mutation by tamoxifen. A of platelet GPVI deficiency. GPVI appears late in megakaryocytic 2-month delay was necessary to observe the GPVI deficiency, maturation, and its deficiency may result from impaired/dysmorphic vasodilation, and the rapid in vivo formation of unstable thrombi. megakaryopoiesis, as reported in MF. Alternatively, GPVI may be These observations call for several comments: (1) Only a few internalized and degraded, as described for antibody-induced GPVI JAK2V617F platelets attached to collagen, and they were unable to deficiency, but requires platelet activation.28 form thrombi when using whole blood at arterial shear rates; because The defective in vitro thrombosis and low responsiveness of the appearance of the moderate GPVI deficiency and of the JAK2V617F platelets occurred simultaneously in SCLCreERt/JAK2V617F vasodilatation is markedly delayed with respect to the defect in KI mice. In PV and ET patients, platelets have been proposed to be less thrombus formation (day 60 vs day 13), their responsibility in the responsive in vitro as a consequence of their in vivo activation,29,30 inability of platelets to form thrombi on collagen can be ruled out. with elevated markers of platelet activation reported in some, but not (2) Controversial data have been reported regarding GP expression all, studies on PV,13,31 ET,16,32,33 and PMF13 patients. In contrast to at the platelet surface in PV, ET,13,24-26 and PMF13 patients; however, these observations, no increased P-selectin expression was observed we found only a mild GPVI deficiency in JAK2V617F KI mice. Such on platelets from VavCre/JAK2V617F and SCLCreERt/JAK2V617F KI a defect has been previously reported in PV and ET patients27 and is mice. Accordingly, integrin aIIbb3 was not activated on freshly drawn 1142 LAMRANI et al BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7

Figure 5. Von Willebrand analysis in JAK2V617F mice. (A) Plasma concentrations of VWF in VavCre/JAK2V617F (Ai, KI, n 5 21), SCLCreERt/JAK2V617F (Aii, TM, n 5 4 per group), and their controls. Mouse plasma samples were analyzed 2-3 months after BM graft. *P , .03. (Bi) Representative electrophoreses (n 5 3) of the multimeric structure of VWF from 2 different WT mice and 4 different VavCre/JAK2V617F (KI). The dotted line square represents the decreased intensity of the high-molecular-weight bands (HMW). Quantification of HMW (Bii) and low-molecular-weight bands (LMW) (Biii) of VWF from 9 VavCre/JAK2V617F (KI) and WT. **P , .002. (Ci) Representative electrophoreses of the multimeric structure of VWF from SCLCreERt/JAK2V617F mice at D6 and D60 compared with controls (VH). Quantification of HMW (Cii) and LMW bands (Ciii) of VWF from 4 SCLCreERt/JAK2V617F per group at different times after treatment by tamoxifen.

JAK2V617F platelets. The impaired responses of JAK2V617F platelets to adhesion in the case of JAK2V617F blood, as the approximately several agonists are thus not the consequence of a desensitization process. twofold increased blood viscosity produces higher shear rates. In VavCre/JAK2V617F, the content in hemostatically active VWF In contrast to the defective thrombus formation in vitro, the multimers was severely reduced. In SCLCreERt/JAK2V617F KI formation of occlusive platelet aggregates was considerably acceler- V617F mice, the proportion of highly reactive multimers started to decrease ated in response to vascular injury by FeCl3 in JAK2 KI mice. 2 weeks after treatment by tamoxifen. The electrophoretic pattern of However, once formed, thrombi were less stable in JAK2V617F KI VWF indicated that it had been submitted to increased proteolysis than in WT mice. Frequent signs of embolization were observed as in JAK2V617F mice. It is likely that VWF multimer deficiency is well as clot lysis, a process never observed previously.42 The observation involved in the impaired adhesion of JAK2V617F platelets in flow of transient arrests of tail bleeding before the reoccurrence of persistent conditions. This is reminiscent of the acquired von Willebrand bleeding is consistent with clot instability in JAK2V617F KI mice. syndrome (AVWS) described in MPN patients with extremely high The differences between in vitro (defective) and in vivo (accelerated) platelet counts causing an increased clearance of platelet-bound thrombus formation can have many causes, including, in particular, 34,35 large VWF multimers. To determine whether VWF proteolysis the initiation mechanisms. (1) In the FeCl3 model used here, the is caused by the disintegrin and metalloprotease with thrombo- event triggering the formation of the thrombus is complex and still spondin type I repeats-13 (ADAMTS-13), or by other proteases, in a subject of debate, particularly regarding the earliest stages.43,44 increased shear conditions requires additional studies. RBCs interacting with endothelial cells appear to have the ability The prolonged TBT of JAK2V617F KI mice, also observed in the to participate in these early stages by recruiting platelets. On one JAK2V617F RV model mice,36,37 is consistent with in vitro defective hand, we have not observed accelerated thrombosis in EPO- thrombus formation. In MPN patients, bleeding has been reported to treated mice, suggesting that an increased RBC count is not suffi- be the consequence of an AVWS.7,38-40 Indeed, thrombus formation cient to accelerate platelet adhesion. On the other hand, the recent in vitro was measured using flow adjusted to provide a shear rate of observation that RBC adhesion to endothelial laminin is increased, 1500 s21 with WT blood such as found in medium-sized arteries, because of the phosphorylation of the Lutheran/basal cell adhesion conditions in which VWF is assumed to trigger the first contact between molecule (Lu/BCAM) by the JAK2V617F/Rap1/Akt pathway,45 platelets and collagen.41 VWF must be even more important for platelet suggests that RBCs could contribute to the rapid initiation of BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7 HEMOSTATIC DISORDERS IN A JAK2V617F PV MOUSE MODEL 1143

Figure 6. Vasodilation in JAK2V617F mice. (Ai) Rep- resentative images of the macroscopic aspect of mesenteric vessels from VavCre/JAK2V617F (KI), EPO- treated, and TPOhigh mice. Bars represent 5 mm. (Aii) Diameters of mesenteric VavCre/JAK2V617F vessels (KI) and matched controls (CTR) (n 5 12). ***P , .0001. (B) Representative images showing that the increase in diameter of mesenteric vessel of SCLCreERt/JAK2V617F mice appeared 60 days (D60) after treatment by tamoxifen. (C) Representative transverse sections (n 5 3) of the aorta from 1 WT and 1 VavCre/JAK2V617F (hematoxylin and eosin staining). Bars represent 500 mm.

V617F V617F thrombosis by FeCl3 in JAK2 KI mice. (2) Although thrombin Importantly, the effect of JAK2 on hemostasis was delayed formation is blocked when thrombosis is analyzed in anticoagulated with respect to the expression of JAK2V617F induced by the ad- t V617F blood in vitro, it is known to be important in FeCl3-induced ministration of tamoxifen to SCLCreER /JAK2 KI mice: thrombosis.42 Unfortunately, thrombin generation could not be tested a 2-week delay was required to observe the defective in vitro because not enough plasma could be collected from high-hematocrit thrombosis, prolonged bleeding time, low platelet reactivity, and JAK2V617F KI blood, but thromboelastometry on whole anticoagulated VWF proteolysis; and a 2-month delay was necessary to observe blood showed that, in vitro, tissue factor–induced coagulation was the GPVI deficiency, vasodilation, and rapid in vivo formation of delayed. However, the possibility that the oxidative stress attributable to unstable thrombi. Together, these observations rule out a direct FeCl3 would injure leukocytes in vivo, leading to the formation effect of the JAK2 mutation on hemostasis but are in favor of its role of neutrophil extracellular traps and activation of coagulation, should in changes associated with the progression of MPN. The observa- be considered.46 Indeed, activation of leukocytes in MPN patients has tions that (1) EPO-driven erythrocytosis resulted in vasodilatation been reported, and leukocytosis is a powerful independent risk and (2) thrombocytosis from TPOhigh RV mice was associated with factor for thrombosis.47-50 aGPVIdeficiency are also consistent with indirect JAK2V617F effects. An interesting point is also that, once formed, clots are unstable In contrast to our results, Hobbs et al using a different JAK2V617F in JAK2V617F KI mice. Decreased firmness and accelerated clot KI mouse model showed increased platelet reactivity to some lysis was indeed observed in whole blood by thromboelastometry. agonists (collagen and thrombin) and decreased tail bleeding Together with the increased proteolysis of VWF, these observations volume.51 Differences may be attributable to the ET phenotype of evoke the activation of proteases, possibly released by leukocytes, this model instead of the PV/PMF phenotype of our model. On the capable of destabilizing platelet aggregates and of activating other hand, Etheridge et al,34 using Tie2Cre/JAK2V617F TG mice fibrinolysis in JAK2V617F KI mice. Further studies are necessary to (expressing human JAK2V617F in endothelial and hematopoietic elucidate the role of leukocytes in clot formation and instability. cells) displaying an ET/PMF phenotype, showed attenuated arterial 1144 LAMRANI et al BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7

Figure 7. In vivo thrombosis. Representative images of n 5 3 (for each mouse) of in vivo thrombosis experiments. FeCl3 injury of mesenteric vessels was performed in VavCre/JAK2V617F mice (A) or SCLCreERt/ JAK2V617F mice 2 months after treatment by tamoxifen (B and C). The accumulation of rhodamin 6G-stained platelets (in red) was recorded. The white arrow in panel B indicates signs of accelerated thrombus lysis, and the white circles in panel C indicate thrombus embolization. Note that the time to occlusion in JAK2V617F KI mice is reduced as compared with their controls. Original magnification 35 for all panels. thrombosis after injury and an AVWS. The fact that RBCs were not Le Kremlin Bicetre,ˆ France) for VWF analysis, Dr Katsue Suzuki-Inoue increased in these mice argues in favor of a contribution of RBCs to (University of Yamanashi, Japan) for the kind gift of rhodocytin; the initiation of thrombus formation in FeCl3-induced thrombosis in and Mary Osborne-Pellegrin for help in editing the manuscript. our PV model. Furthermore, transplantation experiments showed This work was supported by INSERM and Universite´ Paris that delayed occlusion after injury required JAK2V617F expression Diderot, the Cancerop´ oleˆ Ile-de-France, and the Ligue Nationale both in hematopoietic and endothelial cells arguing, in agreement Contre le Cancer (labeled team 2009), as well as the Ministeredela` with our data, against the responsibility of an isolated JAK2V617F Recherche et de l’Enseignement Superieur´ (L.L.) and the French platelet defect. Interestingly, decreased clot stability was observed in Group on Thrombosis and Haemostasis (GEHT) (L.L.). Tie2Cre/JAK2V617F as in VavCre and SCLCreERt/JAK2V617F mice (this study), all models displaying a PMF-like phenotype.34 These results (1) emphasize the role of MPN disease phenotype and not JAK2V617F in the hemostasis disorders observed in different Authorship mouse models of MPN, (2) are consistent with the variability of the Contribution: L.L. designed and performed experiments, analyzed disorders observed in MPN patients, and (3) offer the opportunity to data, and wrote the manuscript; C.L. generated KI mice; C.V.D., V.O., develop biomarkers for the preclinical study of thrombosis and to and B.H.T.N. performed experiments; E.G. provided crucial reagents determine the etiology of disturbed hemostasis in MPN. and analyzed data; W.V. provided funding and scientific advice; and J.-L.V. and M.J.-P. conceived the study, designed experiments, analyzed data, and wrote the manuscript. Conflict-of-interest disclosure: The authors declare no competing Acknowledgments financial interests. Correspondence: Martine Jandrot-Perrus, Inserm U1148, The authors thank Dr Joachim Goethert (University of Duisburg, Hopitalˆ Bichat, 46, rue Henri Huchard, 75877 Paris cedex 18, Essen, Germany) for providing the HSC-SCLCreERt mice; Dr Warren France; e-mail: [email protected]; and Jean-Luc Alexander (Walter Elisa Hall Institute, Melbourne, Australia) for the Villeval, Inserm U1009, Institut Gustave Roussy, PR1, 114 rue VavCre mice; the staff of the animal facilities of the Institut Gustave Edouard Vaillant, 94805 Villejuif, France; e-mail: jean-luc.villeval@ Roussy, directed by Patrick Gonin; Paulette Legendre (Inserm U770 gustaveroussy.fr. BLOOD, 14 AUGUST 2014 x VOLUME 124, NUMBER 7 HEMOSTATIC DISORDERS IN A JAK2V617F PV MOUSE MODEL 1145

References

1. James C, Ugo V, Le Cou´edic JP, et al. A unique promoter induces a polycythemia vera-like 34. Tatewaki W, Takahashi H, Shibata A. Multimeric clonal JAK2 mutation leading to constitutive disease. Blood. 2010;115(17):3589-3597. composition of plasma von Willebrand factor in signalling causes polycythaemia vera. Nature. 18. Marty C, Lacout C, Martin A, et al. chronic myeloproliferative disorders. Clin Lab 2005;434(7037):1144-1148. Myeloproliferative neoplasm induced by Haematol. 1988;10(4):417-425. 2. Baxter EJ, Scott LM, Campbell PJ, et al; Cancer constitutive expression of JAK2V617F in 35. Budde U, Scharf RE, Franke P, Hartmann-Budde Genome Project. Acquired mutation of the knock-in mice. Blood. 2010;116(5):783-787. K, Dent J, Ruggeri ZM. Elevated platelet count as tyrosine kinase JAK2 in human myeloproliferative 19. Mullally A, Lane SW, Ball B, et al. Physiological a cause of abnormal von Willebrand factor disorders. Lancet. 2005;365(9464):1054-1061. Jak2V617F expression causes a lethal multimer distribution in plasma. Blood. 1993; 82(6):1749-1757. 3. Kralovics R, Passamonti F, Buser AS, et al. myeloproliferative neoplasm with differential A gain-of-function mutation of JAK2 in effects on hematopoietic stem and progenitor 36. Etheridge SL, Roh ME, Cosgrove ME, et al. myeloproliferative disorders. N Engl J Med. 2005; cells. Cancer Cell. 2010;17(6):584-596. JAK2V617F-positive endothelial cells contribute 352(17):1779-1790. 20. Hasan S, Lacout C, Marty C, et al. JAK2V617F to clotting abnormalities in myeloproliferative neoplasms. Proc Natl Acad Sci USA. 2014; 4. Levine RL, Wadleigh M, Cools J, et al. Activating expression in mice amplifies early hematopoietic 111(6):2295-2300. mutation in the tyrosine kinase JAK2 in cells and gives them a competitive advantage that polycythemia vera, essential thrombocythemia, is hampered by IFNa. Blood. 2013;122(8): 37. Zaleskas VM, Krause DS, Lazarides K, et al. and myeloid metaplasia with myelofibrosis. 1464-1477. Molecular pathogenesis and therapy of Cancer Cell. 2005;7(4):387-397. 21. Villeval JL, Cohen-Solal K, Tulliez M, et al. High polycythemia induced in mice by JAK2 V617F. PLoS ONE. 2006;1(1):e18. 5. Elliott MA, Tefferi A. Thrombosis and haemorrhage in thrombopoietin production by hematopoietic cells polycythaemia vera and essential thrombocythaemia. induces a fatal myeloproliferative syndrome in 38. Landolfi R, Rocca B, Patrono C. Bleeding and Br J Haematol. 2005;128(3):275-290. mice. Blood. 1997;90(11):4369-4383. thrombosis in myeloproliferative disorders: mechanisms and treatment. Crit Rev Oncol 6. Tefferi A. Polycythemia vera and essential 22. Zetterberg E, Vannucchi AM, Migliaccio AR, et al. Hematol. 1995;20(3):203-222. thrombocythemia: 2012 update on diagnosis, risk Pericyte coverage of abnormal blood vessels in stratification, and management. Am J Hematol. myelofibrotic bone marrows. Haematologica. 39. Castaman G, Lattuada A, Ruggeri M, Tosetto A, 2012;87(3):284-293. 2007;92(5):597-604. Mannucci PM, Rodeghiero F. Platelet von Willebrand factor abnormalities in myeloproliferative 7. Kessler CM. Propensity for hemorrhage and 23. Neunteufl T, Heher S, Stefenelli T, Pabinger I, syndromes. Am J Hematol. 1995;49(4):289-293. thrombosis in chronic myeloproliferative disorders. Gisslinger H. Endothelial dysfunction in patients Semin Hematol. 2004;41(2 suppl 3):10-14. with polycythaemia vera. Br J Haematol. 2001; 40. Budde U, Schaefer G, Mueller N, et al. Acquired 115(2):354-359. von Willebrand’s disease in the myeloproliferative 8. Ziakas PD. Effect of JAK2 V617F on thrombotic 24. Thibert V, Bellucci S, Cristofari M, Gluckman E, syndrome. Blood. 1984;64(5):981-985. risk in patients with essential thrombocythemia: Legrand C. Increased platelet CD36 constitutes measuring the uncertain. Haematologica. 2008; 41. Konstantinides S, Ware J, Marchese P, a common marker in myeloproliferative disorders. 93(9):1412-1414. Almus-Jacobs F, Loskutoff DJ, Ruggeri ZM. Br J Haematol. 1995;91(3):618-624. Distinct antithrombotic consequences of platelet 9. Dahabreh IJ, Zoi K, Giannouli S, Zoi C, 25. Clezardin P, McGregor JL, Dechavanne M, glycoprotein Ibalpha and VI deficiency in a mouse Loukopoulos D, Voulgarelis M. Is JAK2 V617F Clemetson KJ. Platelet membrane glycoprotein model of arterial thrombosis. J Thromb Haemost. mutation more than a diagnostic index? A meta- abnormalities in patients with myeloproliferative 2006;4(9):2014-2021. analysis of clinical outcomes in essential disorders and secondary thrombocytosis. Br J thrombocythemia. Leuk Res. 2009;33(1):67-73. 42. Boulaftali Y, Lamrani L, Rouzaud MC, et al. The Haematol. 1985;60(2):331-344. mouse dorsal skinfold chamber as a model for the 10. Lussana F, Caberlon S, Pagani C, Kamphuisen 26. Bolin RB, Okumura T, Jamieson GA. Changes in study of thrombolysis by intravital microscopy. PW, B¨uller HR, Cattaneo M. Association of V617F distribution of platelet membrane glycoproteins in Thromb Haemost. 2012;107(5):962-971. Jak2 mutation with the risk of thrombosis among patients with myeloproliferative disorders. Am J patients with essential thrombocythaemia or 43. Eckly A, Hechler B, Freund M, et al. Mechanisms Hematol. 1977;3(1):63-71. idiopathic myelofibrosis: a systematic review. underlying FeCl3-induced arterial thrombosis. Thromb Res. 2009;124(4):409-417. 27. Best D, Senis YA, Jarvis GE, et al. GPVI levels in J Thromb Haemost. 2011;9(4):779-789. platelets: relationship to platelet function at high 44. Barr JD, Chauhan AK, Schaeffer GV, Hansen JK, 11. Vannucchi AM, Antonioli E, Guglielmelli P, et al. shear. Blood. 2003;102(8):2811-2818. Clinical profile of homozygous JAK2 617V.F Motto DG. Red blood cells mediate the onset of mutation in patients with polycythemia vera or 28. Rabie T, Varga-Szabo D, Bender M, et al. thrombosis in the ferric chloride murine model. essential thrombocythemia. Blood. 2007;110(3): Diverging signaling events control the pathway Blood. 2013;121(18):3733-3741. 840-846. of GPVI down-regulation in vivo. Blood. 2007; 45. De Grandis M, Cambot M, Wautier MP, et al. 110(2):529-535. 12. De Stefano V, Za T, Rossi E, et al; GIMEMA JAK2V617F activates Lu/BCAM-mediated red cell CMD-Working Party. Recurrent thrombosis in 29. Michiels JJ, Berneman Z, Schroyens W, Finazzi adhesion in polycythemia vera through an EpoR- patients with polycythemia vera and essential G, Budde U, van Vliet HH. The paradox of platelet independent Rap1/Akt pathway. Blood. 2013; thrombocythemia: incidence, risk factors, and effect activation and impaired function: platelet-von 121(4):658-665. Willebrand factor interactions, and the etiology of of treatments. Haematologica. 2008;93(3):372-380. 46. von Br¨uhl ML, Stark K, Steinhart A, et al. thrombotic and hemorrhagic manifestations in Monocytes, neutrophils, and platelets cooperate 13. Jensen MK, de Nully Brown P, Lund BV, Nielsen essential thrombocythemia and polycythemia vera. to initiate and propagate venous thrombosis in OJ, Hasselbalch HC. Increased platelet activation Semin Thromb Hemost. 2006;32(6):589-604. and abnormal membrane glycoprotein content mice in vivo. J Exp Med. 2012;209(4):819-835. 30. Castaldi PA, Berndt MC, Booth W, Gregory C, and redistribution in myeloproliferative disorders. 47. Carobbio A, Finazzi G, Antonioli E, et al. Bull H, Greaves M. Evidence for a platelet Br J Haematol. 2000;110(1):116-124. Thrombocytosis and leukocytosis interaction membrane defect in the myeloproliferative in vascular complications of essential 14. Falanga A, Marchetti M, Vignoli A, Balducci D, syndromes. Thromb Res. 1982;27(5):601-609. Barbui T. Leukocyte-platelet interaction in patients thrombocythemia. Blood. 2008;112(8):3135-3137. 31. Santilli F, Romano M, Recchiuti A, et al. with essential thrombocythemia and polycythemia 48. Vannucchi AM, Antonioli E, Guglielmelli P, et al; Circulating endothelial progenitor cells and vera. Exp Hematol. 2005;33(5):523-530. MPD Research Consortium. Prospective residual in vivo thromboxane biosynthesis in low- identification of high-risk polycythemia vera 15. Arellano-Rodrigo E, Alvarez-Larr´an A, Reverter dose aspirin-treated polycythemia vera patients. patients based on JAK2(V617F) allele burden. JC, Villamor N, Colomer D, Cervantes F. Blood. 2008;112(4):1085-1090. Increased platelet and leukocyte activation as Leukemia. 2007;21(9):1952-1959. 32. Arellano-Rodrigo E, Alvarez-Larr´an A, Reverter contributing mechanisms for thrombosis in 49. Passamonti F. Prognostic factors and models in JC, et al. Platelet turnover, coagulation factors, essential thrombocythemia and correlation with polycythemia vera, essential thrombocythemia, and soluble markers of platelet and endothelial the JAK2 mutational status. Haematologica. 2006; and primary myelofibrosis. Clin Lymphoma activation in essential thrombocythemia: relationship 91(2):169-175. Myeloma Leuk. 2011;11(suppl 1):S25-S27. with thrombosis occurrence and JAK2 V617F allele 16. Rocca B, Ciabattoni G, Tartaglione R, et al. burden. Am J Hematol. 2009;84(2):102-108. 50. Falanga A, Marchetti M. Thrombotic disease in the Increased thromboxane biosynthesis in essential myeloproliferative neoplasms. Hematology Am Soc 33. Griesshammer M, Beneke H, Nussbaumer B, thrombocythemia. Thromb Haemost. 1995;74(5): Hematol Educ Program. 2012;2012(1):571-581. Gr¨unewald M, Bangerter M, Bergmann L. 1225-1230. Increased platelet surface expression of 51. Hobbs CM, Manning H, Bennett C, et al. JAK2V617F 17. Akada H, Yan D, Zou H, Fiering S, Hutchison RE, P-selectin and thrombospondin as markers of leads to intrinsic changes in platelet formation and Mohi MG. Conditional expression of heterozygous platelet activation in essential thrombocythaemia. reactivity in a knock-in mouse model of essential or homozygous Jak2V617F from its endogenous Thromb Res. 1999;96(3):191-196. thrombocythemia. Blood. 2013;122(23):3787-3797.