Tissue factor activity is increased in neutrophils from <scp><i>JAK2 V617F</i></scp>‐mutated essential thrombocythemia and polycythemia vera patients

Abstract

Thrombosis is the leading cause of morbidity and mortality in polycythemia vera (PV) and essential thrombocythemia (ET). Twenty percent of events occur at or before the PV/ET diagnosis and the risk remains elevated even after diagnosis.1 A higher prevalence of thrombotic events is seen in PV compared to ET.1 Age > 60 years, cardiovascular risk factors, prior thrombosis, and allelic burden of JAK2 V617F mutation associate with higher thrombotic risk.1 Modest leukocytosis is also implicated.1 Neutrophils are the predominant circulating leukocyte. PV and ET neutrophils are activated as evidenced by the presence of CD11b, upregulation of β1 and β2 integrin, and they demonstrate enhanced inducible ability to extrude extracellular traps.2 Venous thrombosis formation in JAK2 V617F-mutated mouse models is significantly attenuated by blocking neutrophil activation,2 pointing to a dominant role of neutrophils in MPN-associated venous thrombosis. Furthermore, we previously demonstrated that tissue factor (TF), F3, transcripts are increased in PV and ET neutrophils compared to healthy controls.3 TF is the principal initiator of coagulation in vivo. It is an integral membrane protein primarily found in cells that are sequestered from blood, but can also be expressed by circulating leukocytes in various pathologic conditions. Monocytes can robustly and reliably be induced to express functional TF, whereas there remains debate about the ability of neutrophils to synthesize functional TF.4 Since it is known that increased TF mRNA is present in PV and ET neutrophils, we evaluated whether functional TF procoagulant activity was also present. Patients were recruited from the MPN Clinics at the Huntsman Cancer Institute in Salt Lake City, UT and the University of North Carolina Cancer Hospital in Chapel Hill, NC. WHO 2016 guidelines were used to define ET and PV. Peripheral blood was collected into 3.2% sodium citrate, and the first 2 mL was discarded. Neutrophils were isolated from whole blood by negative selection using magnetic beads (Miltyeni Biotec) resulting in 98% purity. For monocyte isolation, peripheral blood was first Ficoll fractionated and then CD14+ cells (monocytes) were selected by FACS, with 99% purity. F3 transcript level was determined by QT-RT PCR using a TaqManExpression Assay (probes F3: Hs01076029_m1 and RPL13A: Hs04194366_g1, Applied Biosystems, CA). Expression levels were normalized against RPL13A (calculated by ΔΔCt method) and were expressed as fold change. To determine neutrophil TF activity, purified neutrophils or monocytes were snap frozen and thawed to lyse. 1 × 106 cells were used per assay. TF activity was measured by a two-step FXa generation assay with and without an inhibitory TF antibody (HTF-1, BD Biosciences) to determine the contribution of TF to total FXa generation. For confocal microscopy, 1 × 105 neutrophils were plated onto Nunc Lab-Tek II chamber slides and fixed and permeabilized with Cytofix/Cytoperm (BD Biosciences). Anti-CD66b PE antibody (BD 561650) was used to identify neutrophils. Biotinylated HTF1 monoclonal antibody (BD 550252; EZ-Link NHS-PEG4, ThermoFisher) and streptavidin BB515 (BD 564453) was used to identify TF. Imaging was performed using Zeiss LSM780 confocal microscope. We found that TF-dependent FXa generation was increased in unstimulated neutrophils from patients with JAK2 V617F-mutated ET and PV compared to healthy controls (Figure 1A). F3 transcripts in neutrophils mirrored TF activity with a statistically significant increase in transcript levels in JAK2 V617F-muated ET and PV versus controls (Figure 1B). Taken together, these findings suggest that PV and ET neutrophils synthesize functional TF and are not merely acquiring TF from other cellular sources, as has previously been described in healthy neutrophils.4 Somatically acquired mutation of JAK2 V617F accounts for 96% of PV and 55% of ET, and leads to constitutive activation of the JAK2-STAT pathway in mutant cells. Increased JAK–STAT signaling results not only in quantitative hematopoietic cell increases, but also qualitative cellular changes. Multiple studies have demonstrated an increased thrombotic tendency in JAK2-mutated versus CALR-mutated ET.1 The mechanism of JAK2 V617F-mediated thrombosis has not been elucidated, although leukocytosis has been implicated as a risk factor independent of hematocrit.5 Leukocytes are increasingly recognized as important mediators of thrombogenesis and clot resolution.6 In our TF activity assay, one million neutrophils per patient were used to determine the TF-dependent FXa generation. To estimate the impact of the neutrophil count on total circulating cell-associated TF procoagulant activity, we calculated the rate of FXa generation per neutrophil and multiplied by the number of neutrophils per mL of whole blood per patient (Figure 1C). We performed the same calculation for monocyte-associated TF activity per mL of blood. Strikingly, unstimulated neutrophils from JAK2 V617F-mutated PV or ET contribute fourfold more TF procoagulant activity per mL blood than unstimulated monocytes from the same patient. By confocal microscopy, TF was present in neutrophils isolated from JAK2-mutated patients, but not in healthy controls (Figure 1D). TF was primarily located near the neutrophil surface, as opposed to the intracellular location reported in sepsis.7 β1 and β2 integrin upregulation in JAK2 V617F-mutated neutrophils, which has been linked to thrombotic risk,2 facilitates increased interaction with vascular endothelium. TF expression on the surface of docked PV/ET neutrophils may thus initiate thrombosis. PV/ET neutrophils also have enhanced ability to extrude extracellular traps2 which may thus be more highly decorated with TF thereby further promoting the thrombotic risk. In PV, erythrocytosis increases whole blood viscosity, resulting in platelet margination that may promote increased platelet interaction with a forming thrombus. While we did not observe any relationship between thrombosis and neutrophil TF activity nor F3 transcripts in our current PV and ET cohorts, we previously found a positive correlation with F3 transcripts in PV granulocytes and thrombosis risk.3 Although our sample size was not large enough for multivariate analysis, we did examine the individual effects of MPN treatment, patient age, and gender on TF activity and F3 transcripts and found no correlation (Figure 1A,B and Table S1). Increasing JAK2 V617F variant allele frequency (VAF) has been associated with thrombotic risk in PV/ET. In our cohort, there was a nonsignificant trend toward increased neutrophil TF activity with higher JAK2 V617F VAF (Table S1), but no correlation was noted between neutrophil JAK2 V617F VAF and neutrophil F3 transcripts. This may again be related to our heterogenous sample population, as we did previously report a positive correlation between JAK2 V617F VAF and F3 transcripts.3 Increased plasma TF antigen has also been reported in JAK2 V617F-mutated ET and correlated with VAF, further suggesting that activation of the JAK–STAT pathway leads to TF upregulation.8 Importantly, hydroxyurea decreases TF expression in monocytes, perhaps contributing to its anti-thrombotic effect in MPN.9 Overall, our results support further evaluation of neutrophils as a potential anti-thrombotic target in JAK2 V617F-mutated PV and ET. Our data further indicate that in PV and ET, neutrophils synthesize and express functional TF and re-ignite an old debate about whether neutrophils can synthesize TF or simply acquire it from other sources. Brandi Reeves received honoraria from Incyte and Bristol Myers Squibb. Brandi N. Reeves: designed research, analyzed data, wrote the manuscript. Jihyun Song, Kathryn J. Wilson, Michael W. Henderson: Performed research. Soo Jin Kim: Performed research, analyzed data. Rafal Pawlinski and Nigel S. Key: Designed research and wrote the manuscript. Josef T. Prchal: Designed research and wrote the manuscript. The data that support the findings of this study are available from the corresponding author upon reasonable request. Table S1 Demographics of study subjects and controls Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts