Ruxolitinib Reduces Endothelial Pro-Adhesive Interactions: Implications for JAK2^V617+ MPN Thrombosis

Introduction

The JAK2^V617F+ mutation occurs in up to 95% of patients with polycythemia vera (PV) and increases the risk of thrombosis 6-fold. Recent studies demonstrate that JAK2^V617F+ endothelial cells express pro-adhesive proteins, suggesting that the endothelium may contribute to increased thrombosis.^1-3 The targeted JAK1/JAK2 inhibitor ruxolitinib is an approved second-line therapy for PV patients and is effective in alleviating constitutional symptoms, lowering hematocrit, and reducing cell number. However, there is limited data regarding efficacy of ruxolitinib in reducing thrombosis. Although recent work has demonstrated ruxolitinib reduces neutrophil extracellular trap formation,⁴ the vascular effects of ruxolitinib are unknown. Therefore, we hypothesize that ruxolitinib reduces endothelial cell pro-adhesive activation leading to decreased rolling and adhesion of JAK2^V617F+ samples.

Methods

To mimic JAK2^V617F activation, primary human umbilical vein endothelial cells (HUVEC, passage 1-5) were treated with TNF-α (10 ng/mL) +/- ruxolitinib (400 nM-4 µM, Selleckchem) and characterized 4h later. For confocal microscopy, cells were fixed with paraformaldehyde, permeabilized and stained for either VWF, VCAM-1, or P-selectin along with 2-(4-amidinophenyl)-1H-indole-carboxamide (DAPI, Thermo Fisher). Images were captured on Olympus FluoView FV1000 IX2 Inverted Confocal microscope. Secretion of VWF, VCAM-1 and P-selectin into conditioned media was assessed with ELISA (Molecular Innovations; BioLegend). Citrated normal and JAK2^V617F+whole blood was obtained per IRB and labeled with calcein AM (ThermoFisher). To quantify leukocyte and platelet velocity and adhesion, an endothelialized poly-di-methyl-siloxane (PDMS) microchannel was prepared and treated with TNF-α (10 ng/mL) +/- ruxolitinib (400 nM-4 µM). Samples were perfused through the microfluidic at a shear stress of 0.35 dynes/cm², and 10 s images were captured using a fluorescence microscope (Olympus). Cell velocity and adhesion were quantified using FIJI (NIH).

Results

Compared to TNF-α treated HUVEC alone, the combination of ruxolitinib + TNF-α reduced expression of VWF, VCAM-1 and P-selectin using both immunofluorescence and ELISA. In endothelialized microfluidic devices treated with TNF-α alone, both normal (n=3) and JAK2^V617F+ (n=9) leukocyte and platelet velocity was significantly decreased and cell adhesion increased. This result was independent of hematocrit and platelet levels. In normal controls (n=2), TNF-α+ruxolitinib lead to a trend toward normal leukocyte and platelet velocity with decreased cell adhesion. Lastly, in a small subset of JAK2^V617F+ patients (n=4), addition of ruxolitinib increased cell velocity.

Conclusions

In conclusion, in TNF-α-activated endothelial cells, treatment with ruxolitinib decreases pro-adhesive VWF, VCAM-1, and P-selectin expression. Using normal controls and JAK2^V617F+ MPN blood samples, our TNF-α-activated endothelialized microfluidics model demonstrates significant reduction in leukocyte and platelet velocity and increased cell adhesion. In normal and JAK2^V617F+ MPN whole blood, treatment of TNF-α-stimulated endothelium with ruxolitinib improves cell velocity. Further evaluation using JAK2^V617F+ endothelial cells is planned. Collectively, these results suggest that ruxolitinib may reduce JAK2^V617F+ thrombotic risk through reduction of pro-adhesive endothelial activation.