Abstract
Polycythemia vera (PV) is a clonal stem cell disorder characterized by the unregulated production of red cells, white cells and platelets. This phenotype was explained by the discovery of an activating point mutation (V617F) in the gene for JAK2, a tyrosine kinase utilized for signal transduction by the erythropoietin, granulocyte colony-stimulating factor and thrombopoietin receptors. The constitutively active kinase enhances the terminal differentiation of erythroid, myeloid and megakaryocytic progenitor cells. However, gene expression profiling of terminally differentiated JAK2 V617F-positive PV neutrophils reflected only their activated state but not differential gene expression specific to PV. Furthermore, PV is a stem cell disorder and JAK2 is not essential for normal stem cell function. Therefore, to define the molecular abnormalities in PV stem cells, and to determine the influence of JAK2 V617F on these, we used DNA microarray technology to examine gene expression in the circulating CD34+ cells of 19 JAK2 V617F-positive PV patients (11 women and 8 men) in comparison to mobilized CD34+ cells from six normal individuals (3 women and 3 men). Differential changes in gene expression were validated by quantitative PCR (QPCR). Both PV CD34+ cells and mobilized CD34+ cells were greater than 97 % CD38+ and 90 % of both CD34+ cell populations were in a G0/G1 state. Principle component analysis revealed that gene expression in the normal controls segregated according to gender and gender also confounded disease effects in the PV patients as well. When differential gene expression in the PV patients was analyzed separately according to gender, women differentially up or down regulated 249 genes and men, 484 genes. This allowed us to identify a core set of 90 genes that were differentially regulated (59 up regulated and 31 down regulated) in both men and women, and thus, specific for PV and likely to be involved in its pathogenesis. When CD34+ cell gene expression was stratified according to both JAK2 V617F status and gender, the gender effect persisted but there was virtually no influence of the JAK2 V617F allele burden on core PV gene expression. Amongst the 59 up regulated genes were the chemokine genes, CCL3, CCL5 and PF4; the hemoglobin genes, epsilon, gamma, alpha and beta; the extracellular matrix genes, collagen types 1, 3, 4 and 6, TIMP1, heparanase, osteonectin, periostin, fibulin 3, ficolin 1 and thrombospondin; the procoagulant genes, PAI-1and fibroleukin; the transcription factor genes NF1B and HES1; the cell cycle gene, CDC20; the DNA synthesis gene, ribonucleotide reductase; and the folate and leptin receptor genes. The 31 down regulated genes included the transcription factor genes, HOXA9, HOP, KLF6 and hepatic leukemia factor; the FLT3 receptor gene and the PDGF-D and the CD133 genes. Overall, there was down regulation of tumor suppressor genes and up regulation of genes involved in Notch, Wnt and TGF-β signaling in the PV patients as compared to the controls. Using the 90 core gene set, unsupervised hierarchical clustering separated the 19 PV patients in two groups that did not differ with respect to age (mean, 63 vs. 65 years), disease duration (median, 10.5 vs. 7 years), gender or JAK2 V617F neutrophil burden (92 % vs. 85 %), but did differ with respect to the frequency of thromboembolic complications (6/6 vs. 1/13, p= 0.01), splenectomy (4/6 vs. 1/13, p=0.001), and a requirement for chemotherapy (5/6 vs. 3/13, p=0.04). Interestingly, based on their gene expression, the 6 PV patients with the more aggressive clinical course also segregated closely with the normal controls. These observations indicate that PV is not a monolithic disorder, that clinical differences between men and women PV patients have in part a genetic basis, that JAK2 V671F does not have a significant impact on gene expression at the CD34+ cell level, and that it may be possible to determine which PV patients would benefit from early therapeutic intervention by examining gene expression in circulating CD34+ cells.
Disclosures: No relevant conflicts of interest to declare.
Author notes
Corresponding author