The Effects of Lipocalin (LCN2) on Hematopoiesis in Primary Myelofibrosis

Myelofibrosis (MF) is characterized by anemia, thrombocytopenia and marrow fibrosis. Malignant rather than normal hematopoietic progenitor cells (HPC) preferentially proliferate. A form of MF can also occur in patients with a prior history of PV or ET which is referred to as PV or ET related MF (PV-MF and ET-MF). In MF, excessive proliferation of marrow stromal cells occurs in response to cytokines elaborated by the malignant clone. Previously, we found that lipocalin-2 (LCN2) levels were elevated in patient plasmas with MPN, and a greatest degree of elevation was observed in the PMF plasma and PMF conditioned medium. We also found that LCN2 increased PMF HPC proliferation and increased numbers of JAK2V617F homozygous HPCs, but depressed normal BM HPC (Lu. et al, 2013 ASH meeting).

In this study, a series of experiments were performed to further explore the effects of LCN2 on MF pathogenesis. ELISA data showed that plasma LCN2 levels not only were higher in all MF patients, plasma LCN2 levels in PV-MF as well as ET- MF were significantly greater than that in PV or ET plasma (p=0.0002 and p=0.0175, respectively). Importantly, in MPN patients, a non-parametric spearman correlation coefficient analysis suggested an inverse relationship between the patients’ hemoglobin levels and LCN2 levels (coefficient=-0.35, p<0.0001) and between platelet counts and LNC2 levels (coefficient=-0.26, p=0.0044). Immunofluorescence staining showed a greater proportion of PMF MNCs expressed LCN2. And immunohistochemical results revealed that LCN2 expressing cells were present in the PMF bone marrow and spleens samples and the LCN2 expression was restricted to cells belonging to the myeloid lineage rather than erythroid precursors or megakaryocytes. These data strongly confirmed that a higher level of LCN2 was expressed in MF patients. We then found that LCN2 significantly increased the generation of ROS (reactive oxygen species) in normal BM CD34⁺ cells but not PMF CD34⁺ cells (p<0.05). Furthermore, exposure of normal marrow to the LCN2 dramatically increased the CD34⁺/annexin V⁺ cell population in a dose dependent fashion. These apoptotic effects were blocked by addition of N-Acetyl-Cysteine (NAC), a ROS scavenger. By contrast, LCN2 did not increase the degree of apoptosis of PMF CD34⁺ cells when added at similar doses. These data indicate that LCN2 selectively promotes the survival of PMF CD34⁺ cells but leads to the apoptosis of normal BM CD34⁺ cells by increasing ROS. With 6 different individual BM samples, we demonstrated that PMF MNC cells that produce high levels of LCN2 as well as a recombinant form of LCN2 strongly promoted BM marrow adherent cells (MAC) proliferation. Similarly, LCN2 added to semisolid cultures of normal marrow cells resulted in the appearance of increased numbers of fibroblast-like cells. In addition, the ability of LCN2 to promote MAC proliferation was blocked by addition of NAC. These results suggest that LCN2 related generation of ROS induces MAC proliferation. The MAC in both control and LCN2 containing conditions expressed vimentin, Von Willebrand factor (VWF) and α-smooth muscle actin (a-SMA), while in LCN2 cultures expressed higher levels of CD105 (Endoglin) and CD106 (VCAM-1). These MAC therefore expressed features characteristic of mesenchymal stem cells, fibroblasts and endothelial cells. The MAC under control conditions expressed CD90⁺ on their cell surface, while incubation with LCN2 resulted in diminished cell surface CD90 expression, and its localization to the nucleus.

We propose that LCN2 acts in an autocrine fashion to promote malignant PMF hematopoiesis as documented by its ability to preferentially favor the appearance of JAK2V617F homozygous hematopoietic colonies in vitro, but also acts in a paracrine fashion to suppress the proliferation of normal HPC and to affect the marrow microenvironment by promoting stromal cell proliferation. Our findings indicate that LCN2 affects these numerous biological effects by promoting the generation of ROS. We also propose that therapeutic strategies that can impair the actions of LCN2 or inactivate ROS are likely to be beneficial to MPN patients since they improve disease related cytopenias as well as impede the development of progressive marrow fibrosis. Such agents affecting this alternative therapeutic target might potentially be used alone or in combination with other presently available therapeutic modalities.