Lipocalin-2 secreted from JAK2V617F⁺ hematopoietic cells is associated with paracrine DNA damage. (A) Thorough investigation of genes with elevated expression in murine and human JAK2V617F⁺ hematopoietic cells compared with their normal counterparts in the published gene expression data (GSE21842 and GSE9827). Genes coding for soluble factors are underlined. (B) Quantitative real-time polymerase chain reaction (PCR) analysis of the candidate genes in JAK2V617F-32D cells relative to EV-32D cells (n = 3). (C) Schematic representation of the following experiments. JAK2V617F-32D cells were transduced with the shRNAs against candidate genes, and parental 32D cells were incubated with each CM. (D) Relative cell numbers of 32D cells 36 hours after incubation with each CM (n = 3 each). (E) Lipocalin-2 secretory ability in the respective 32D cells assessed by enzyme-linked immunosorbent assay (ELISA) in cultured media (n = 4 each). (F) Lipocalin-2 ELISA in the plasma of mice transplanted with EV, wild-type JAK2, or JAK2V61F-transduced bone marrow cells (n = 4 each). (G) Serum levels of lipocalin-2 in patients with MPNs. The concentration of lipocalin-2 for patients in complete remission for hematologic malignancies was measured as control. (H) Lipocalin-2 concentration of media conditioned by bone marrow mononuclear cells derived from patients with MPNs and controls. (I) Immunofluorescence staining of γH2AX foci formation (Alexa Fluor 488: green) in 32D cells incubated with CM from JAK2V617F-32D with shRNA against Lcn2 or control shRNA. Nuclei were stained with TO-PRO3. (J) Intracellular levels of ROS were measured by H2DCFDA assay following incubation of parental 32D cells with each CM for 24 hours. Average FITC fluorescence intensity is shown (n = 3 each). Scale bars: 10 μm. Error bars indicate SD.