The Thrombopoietin Receptor, MPL, Is Critical for the Development of JAK2V617F-Induced Cardiovascular Disease in a Murine Model of Myeloproliferative Neoplasm

Introduction Vascular endothelial cells (ECs) carrying the JAK2V617F mutation can be detected in many patients with myeloproliferative neoplasms (MPNs). Previously we reported that a JAK2V617F-positive murine model of MPN, in which the mutation is expressed in both blood cells and ECs, developed spontaneous heart failure with a thrombosis and vasculopathy phenotype, and mutant ECs were required to develop this phenotype. In this study, we tested the hypothesis that the JAK2V617F mutation induces endothelial dysfunction to promote cardiovascular diseases (CVDs), and the TPO/MPL signaling plays a critical role in JAK2V617F-induced CVDs.

Methods JAK2V617F Flip-Flop (FF1) mice (from Radek Skoda, Switzerland) and Cdh5-CreERT2 mice were crossed to generate Cdh5-CreERT2^+/-; FF1^+/- mice, in which human JAK2V617F is expressed in vascular ECs after tamoxifen induction (EC^JAK2V617F). The EC^JAK2V617F mice were further crossed with MPL^fl/fl mice (from Warren Alexander, Australia) to knock down the MPL protein in JAK2V617F-bearing ECs (Cdh5-CreERT2^+/-; FF1^+/-; Mpl ^fl/fl, or EC^{JAK2V617F;MPL-/-}).

Results As determined by RT-PCR, human JAK2V617F is expressed in vascular ECs (CD45^-CD31⁺) from the EC^JAK2V617Fmice after tamoxifen induction; no JAK2V617F expression is detected in blood cells from these mice. At 16wk of age, mild thrombocytosis (platelet 689 vs. 478 x 10⁹/L, p=0.0003) and erythrocytosis (red blood cell 11 vs. 10 x 10⁹/L, p=0.017) were observed in EC^JAK2V617F mice compared to age-matched wild-type control mice. Serial transthoracic echocardiography did not reveal any cardiac dysfunction in the EC^JAK2V617F mice during a 9-month follow-up, suggesting that mutant EC alone is not sufficient to develop the CVD.

To test whether JAK2V617F mutant ECs promote the development of CVD in response to stressors, both EC^JAK2V617F mice and control mice (with JAK2 wild-type ECs) were fed a high-fat/high-cholesterol diet. After 6-12wks of diet treatment, there were significant decreases in left ventricular ejection fraction and increases in left ventricular end-diastolic and end-systolic volumes in the EC^JAK2V617F mice compared to control mice, indicative of cardiac dysfunction. Pathological evaluation revealed (1) increased heart weight; (2) presence of thrombosis in segment pulmonary arteries and scattered coronary capillaries; (3) evidence of intimal thickening, smooth muscle cell thickening, and perivascular fibrosis of scattered coronary arterioles; and (4) enlarged cardiomyocytes and increased reticulin fibers in the EC^JAK2V617F mice after high-fat diet challenge. No atherosclerotic lesions were observed in the EC^JAK2V617F mice.

The TPO receptor MPL is expressed on several types of vascular ECs and TPO/MPL signaling can affect cardiovascular function. We found that MPL expression was upregulated in JAK2V617F mutant ECs compared to wild-type ECs, and this upregulation was completely blunted by Ruxolitinib, a clinically available JAK2 inhibitor. These observations prompted us to test whether deregulated TPO/MPL signaling contributes to the increased risks of CVDs in JAK2V617F-positive MPNs. First, we crossed the Cdh5-CreERT2 mice with the MPL^fl mice to knock down the MPL protein in the vascular ECs (EC^MPL-/-). The EC^MPL-/- mice maintained normal blood cell counts and normal cardiac function (on serial echocardiography measurements) during a 6-mo follow-up. These results suggest that endothelial MPL receptor is not required to maintain normal blood count or cardiac function. Next, we crossed the EC^JAK2V617Fmice with the MPL^fl mice to knock down the MPL protein in JAK2V617F-bearing vascular ECs (EC^{JAK2V617F;MPL-/-}). Serial echocardiography revealed normal cardiac function in the EC^{JAK2V617F;MPL-/-} mice even after a high-fat diet challenge. Histology examination did not reveal any thrombosis or vasculopathy in these mice. Last, we found that JAK2V617F mutant ECs had increased senescence and decreased VEGFR2 and VE-cadherin expression compared to wild-type ECs after the high-fat diet challenge, which was rescued by MPL knockdown in the mutant ECs.

Conclusions Our findings indicate that the JAK2V617F mutation induces endothelial dysfunction to promote CVDs, especially when challenged with additional stressors (e.g., high-fat diet). We also showed that inhibiting endothelial TPO/MPL signaling can prevent the development of CVDs in JAK2V617F-positive MPNs.

No relevant conflicts of interest to declare.