Phenotypic and Transcriptomic Profiling of MPN Endothelium

Introduction

The classical myeloproliferative neoplasms (MPNs) - essential thrombocythemia (ET), polycythemia (PV) and primary myelofibrosis (PMF) – are myeloid disorders characterized by excess clonal hematopoiesis. This results in expansion of mature blood cells and increased risk of transformation to acute myeloid leukemia, but most significantly, an elevated risk of thromboembolic events. Thromboembolic events, both arterial and venous, are the most common complications associated with MPNs adding significantly to the morbidity and mortality associated with this disease. While the association of MPN with thrombosis remains well-recognized, the pathophysiology remains incompletely understood. Multiple factors, including platelet dysfunction, monocytic activation, and procoagulant extracellular vesicles have been reported as potential contributors of thrombotic risk in MPNs. Endothelial cell activation is known to occur in MPN and may play a direct role in pathophysiology of MPN-associated thromboinflammation. Blood outgrowth endothelial cells (BOECs), which are phenotypically and transcriptomically endothelium, are progeny of circulating, transplantable, precursor cells. BOECs can be harvested from peripheral blood and have been utilized to evaluate many pathophysiologic states, including bleeding, thrombotic and inflammatory disorders. We studied BOECs from individuals with JAK2V617F-mutated MPN and describe associated phenotypic and transcriptomic abnormalities.

Methods

Our studies were approved by the institutional review board. Blood outgrowth endothelial cells were harvested as previously described. CD45, VE-cadherin and P-selectin surface expression were determined by immune-flow cytometry. von Willebrand factor (vWF) expression was determined by ELISA and confocal immunofluorescence microscopy (IF). DNA and RNA were extracted using commercial kits. JAK2 mutant allele frequency was estimated using real-time PCR. Bulk RNAseq was carried out using Illumina Novoseq S4. T test was carried out when appropriate. P value <0.05 was considered statistically significant.

Results

8 subjects with JAK2V617F mutated MPN as well as 13 healthy controls were enrolled. 60% of the individuals with MPN had a history of thromboembolism versus none of the controls. Peripheral blood collected, mononuclear fraction isolated and BOEC cultures started on all study subjects. As previously described, CD45-negative, VE-Cadherin- as well as vWF-positive colonies were classified as BOECs or otherwise excluded. BOECs were successfully harvested in 7 (88%) MPN subjects vs only 6 (34%) controls. The mean number of BOEC colonies were also significantly higher in MPN, 7 vs 1.3 (p=0.016). Time to appearance of the first colony (13.7 vs 14 days; p=0.4) and cell proliferation rate (MTT assay at 24 hours, OD 0.5 vs 0.65; p=0.15) was similar in controls and MPN respectively. MPN BOECs had significant higher vWF expression (5260 vs 1160 arbitrary units per mg of lysate; p=0.004), as well as P-selectin surface expression (mean AFI 358 vs. 190; p=0.008), with significant correlation between the two (R²=0.74). None of the MPN colonies were positive for JAK2V617F mutation as tested by a sensitive real-time PCR assay. 10 MPN colonies and 5 control colonies were sent for bulk RNAseq analysis. Differential gene expression was performed using DESeq2. Although there was heterogeneity within the cohorts, there was clear hierarchical clustering between the two clusters. Pathway enrichment analysis was performed using the Enrichr suit (Ma’ayan lab). The top Gene Ontology Biologic Process pathways altered included heparan sulfate proteoglycan biosynthetic process, regulation of blood coagulation, negative regulation of platelet activation, regulation of plasminogen activation, positive regulation of secretion, among others. Human Phenotype Ontology also identified many pathways associated with vascular diseases, such as vasculitis and aneurysm, in addition to ion transport and complement pathway abnormalities.

Conclusions

We have characterized primary blood outgrowth endothelial cells from individuals with MPNs and identified a phenotypic and transcriptomic signature of MPN-associated endothelial dysfunction.