Alox5 Blockade Eradicates JAK2V617F-Induced Polycythemia Vera in Mice

The myeloproliferative neoplasms (MPN) are hematologic malignancies with a chronic clinical course and a risk of developing thrombosis and acute leukemia. JAK2 is a member of the Janus family of non–receptor tyrosine kinases and is required for signaling from the erythropoietin (Epo) receptor and other type I cytokine receptors (1). The somatic activating point mutation of JAK2 (JAK2V617F) has been found in Philadelphia chromosome-negative MPNs (2, 3) in approximately 95% of patients with polycythemia vera (PV), 60% of patients with essential thrombocytosis, and 50% of patients with primary myelofibrosis (3, 4). JAK2V617F-positive cells outpace normal hematopoietic cells as a result of its constitutively active kinase activity and activation of growth factor signaling (5–7). This acquired somatic mutation stimulates independent growth of hematopoietic cells (1, 5, 8). Although JAK2 inhibitors have been shown to help to improve symptoms and quality of life in patients (9, 10), it is unlikely that these inhibitors are curative.

Mouse models of JAK2V617F-induced MPNs have been developed in recent years. In the retroviral transduction/transplantation mouse model, JAK2V617F was shown to induce MPN reminiscent of human PV, characterized by erythrocytosis and granulocytosis (11–13). JAK2V617F induces PV-like myeloproliferative disease in mice, although the severity of the disease was variable among different inbred mouse strains. JAK2V617F induces more robust leukocytosis and neutrophilia in BALB/c mice than in C57BL/6 mice (11–13). However, JAK2V617F does not induce thrombocytosis in these mice (11–13). Nevertheless, available mouse models mostly recapitulate myeloproliferative disorder induced by JAK2V617F. In JAK2V617F knock-in mice, hematopoietic stem cells (HSC) were shown to contain PV-initiating cells, but a JAK2 kinase inhibitor failed to eliminate this cell population (7). New therapeutic strategies need to be developed for treating PV more effectively. Because HSCs harbor JAK2V617F in patients with PV (5), PV should be considered as a stem cell–derived disease. Compared with PV, Philadelphia chromosome–positive chronic myeloid leukemia (CML) is also derived from HSCs and has a myeloproliferative phenotype similar to PV in mice (14). Therefore, we reasoned that a gene essential for CML development might also play a critical role in PV development. We have shown that the survival and self-renewal of CML-initiating cells require the arachidonate 5-lipoxygenase gene (Alox5) and that Alox5 is essential for CML development (15). In this study, we investigated the role of Alox5 in pathogenesis of PV in mice. We show that deletion of Alox5 or inhibition of Alox5 function attenuates PV development, suggesting a new strategy for treating PV.

JAK2V617F-expressing Ba/F3 cell line that expresses both JAK2V617F and GFP was kindly provided by Dr. Ross Levine (Human Oncology and Pathogenesis Program, Memorial Sloan-Kettering Cancer Center, New York, NY) in 2014. Ba/F3 cells were transduced with pMSCV-IRES-GFP to only express GFP and used as a control, and this GFP-expressing Ba/F3 cell line was generated about 5 years ago. HEL, a human JAK2V617F-expressing cell line, was purchased from ATCC in 2016. All cell lines were grown in RPMI-1640 medium containing 10% FCS and 50 μmol/L 2-mercaptoethanol without exogenous cytokines, except for Ba/F3 cells that were grown in the presence of IL3. Ba/F3 cell line was obtained in 2001. All these cell lines were authenticated by us for confirming expression of JAK2V617F by Western blot analysis and of GFP by FACS analysis.

C57BL/6J, BALB/c, and homozygous Alox5 knockout (Alox5^−/− in C57BL/6 background) mice were obtained from The Jackson Laboratory.

pMSCV-JAK2V617F-GFP DNA construct was kindly provided by Dr. Richard Van Etten (Tufts-New England Medical Center). To induce PV in mice, the JAK2V617F retrovirus was generated by co-transfection of 293T cells with pMSCV-JAK2V617F-GFP and an ecotropic packaging construct and used to transduce bone marrow cells from donor mice, and after normalizing GFP⁺ donor cells, equal number of the transduced GFP⁺ cells were transplanted into each lethally irradiated recipient mouse as described previously (13).

Prior to FACS analysis, red blood cells (RBC) in bone marrow or peripheral blood cells from PV or normal mice were eliminated using NH₄Cl/NH₄HCO₃ RBC lysis buffer (pH 7.4). For FACS analysis, hematopoietic cells were washed with PBS and stained with fluorescence-labeled antibodies (B220-PE for B cells, Gr-1-APC and Mac-1-PE for neutrophils). For stem cell analysis, bone marrow cells were suspended in staining medium [Hank Balanced Salt Solution (HBSS) with 2% heat-inactivated calf serum] and incubated with biotin-labeled lineage antibodies containing a mixture of antibodies against CD3, CD4, CD8, B220, Gr-1, Mac-1, and Ter119. After washing, the fluorochrome-labeled secondary antibody (APC-Cy7-conjugated streptavidin) for recognizing biotin and phycoerythrin (PE)-conjugated c-Kit and allophycocyanin (APC)-conjugated Sca-1 antibodies were added to the cells. All these antibodies were purchased from eBioscience Biotechnology. Human CD34⁺ cells from patients with PV or healthy individuals were cultured in serum-free medium (SFM) containing SCF, TPO, Flt-3 ligand, and IL3 at 50 ng/mL, and cells were treated with various doses of zileuton for 48 hours. Cells were stained with a CD34 monoclonal antibody (BD Biosciences) initially and then washed with the binding buffer provided by the manufacturer before staining with Annexin-V (BD Biosciences). Data were acquired on a FACS Calibur flow cytometric analyzer (Becton Dickinson).

Antibodies against p-JAK2, JAK2, p-STAT5, STAT5, p-AKT, AKT, p-PI3K, PI3K, p-MAPK, and MAPK were purchased from Cell Signaling Technology. Antibodies against ALOX5 (5-LO) and ACTIN were purchased from Santa Cruz Biotechnology. Protein lysates were prepared by lysing cells in RIPA buffer, and Western blot analysis was carried out as described previously (16).

For treating PV mice, zileuton (Critical Therapeutics) was dissolved in water and given orally in a volume of less than 0.3 mL by gavage (300 mg/kg, once a day) beginning at 2 months after induction of PV. Water was used as a placebo. For in vitro assay, zileuton (Cayman Chemicals) was dissolved in DMSO to make 100 mmol/L stock solution before it was diluted in culture medium for use. DMSO was used as a control. Ruxolitinib (Cayman Chemicals) was dissolved in ethanol to make 10 μmol/L stock solution before it was diluted in culture medium for use.

Plasma was collected by centrifugation of peripheral blood from mice at 1,000 × g for 5 minutes. EPO levels in plasma were determined by ELISA kit (R&D Systems).

Plasma samples were collected from mice and stored at −80°C before use. The levels of leukotriene B4 (LTB4) in plasma were determined using an ELISA kit (Cayman Chemical) following the manufacturer's instruction. Briefly, 50 μL of each standard or plasma sample was added to each well of a 96-well plate and then 50 μL of LTB4 AchE Tracer was added to each well. Next, 50 μL of LTB4 EIA antiserum was added to each well, and the 96-well plates were incubated at 4°C overnight. The plates were read using an ELISA microplate reader.

Peripheral blood was collected from patients with PV at the Icahn School of Medicine at Mount Sinai (ISMMS; Mount Sinai, NY). Written informed consent was obtained from all patients according to guidelines established by the Institutional Review Board of the ISMMS. All patients met the World Health Organization diagnostic criteria for PV. The blood samples were layered onto Ficoll-Hypaque (1.077 g/mL; GE Healthcare) and low-density mononuclear cells were separated after centrifugation. The CD34⁺ cell population was isolated using a human CD34⁺ cell selection kit (StemCell Technologies) according to the manufacturer's instructions. Normal human CD34⁺ cells were purchased from AllCells.

Mononuclear cells isolated from patients with PV or healthy donors were cultured in 12-well plates with IMDM and supplemented with 20% FBS, 100 U/mL penicillin, and 10 μg/mL streptomycin (Invitrogen). Equal cell numbers were plated in both untreated and treated well at day zero. Cells were treated with various doses of zileuton (50–500 μmol/L) for 4 days. Cell proliferation was determined using the PrestoBlue Cell Viability Reagent (Life Technologies) according to the instructions provided by the manufacturer. Briefly, the reagent was added directly to the cell culture wells at 1:10 dilution, and the cells were incubated for 2 hours. The viable cell number was determined by measuring fluorescence intensity. The fluorescence intensity was measured at 560 nm (excitation) and 590 nm (emission) using a Spectra Max M5e Microplate Reader and was corrected for background fluorescence by including control wells containing culture media alone (no cells) for each plate. The data were analyzed using SoftMax Pro software (Molecular Devices). Absolute cell number also was determined using trypan blue solution (Sigma). The proportion of CD34⁺ cells was measured flow cytometrically after staining with an anti-CD34 antibody (BD Biosciences). Data were acquired on a FACS Calibur flow cytometric analyzer (Becton Dickinson).

Bone marrow cells were collected from PV mice and then normalized to seed 1 × 10⁴ GFP⁺ cells and then plated into each well of a 6-well plate and cultured in M3434 medium in the presence and absence of zileuton (100 μmol/L; n = 3 for each treatment group). The colonies were counted at day 7 after drug treatment. For CFU-GM and CFU-E/BFU-E assays, bone marrow cells collected from PV mice were analyzed by FACS for the percentage of GFP⁺ cells and then normalized to seed 1 × 10⁴ GFP⁺ cells in each dish in Stem Cell Methocult with 3 dishes for each sample in the presence and absence of zileuton. Colonies derived from CFU-E were counted after 48 hours of culture, and colonies from BFU-E and CFU-GM were counted after 10 and 7 days, respectively.

In some cases, CD34⁺ cells were assayed for colony formation assay in semisolid media as previously described (17). Briefly, 500 CD34⁺ cells were plated in duplicate in tissue culture dishes (30 mm diameter) containing 1-mL IMDM with 1.1% methylcellulose and 20% FBS, to which stem cell factor (SCF), thrombopoietin (TPO), fms-like tyrosine kinase 3 ligand (Flt-3 ligand), IL3, granulocyte macrophage-colony stimulating factor (GM-CSF) at 50 ng/mL, and 2 U/mL EPO were added. Various doses of zileuton (50–500 μmol/L; Cayman) were added to the medium. Colonies were enumerated after 14 days of incubation, and individual colonies were plucked and genotyped for JAK2V617F as previously described (17).

Genomic DNA was isolated from randomly plucked colonies using the Extract-N-Amp Blood PCR Kits (Sigma). JAK2V617F was detected by using a nested allele-specific PCR. The final PCR products were analyzed on 2.0% agarose gels. The nested PCR product had a size of 453 bp. A 279-bp product indicated allele-specific JAK2V617F-positive, whereas a 229-bp product denoted allele-specific wild-type product. Colonies were classified as homozygous for JAK2V617F, if they contained only the 279-bp band, whereas heterozygous colonies were identified on the basis of the presence of both the 279-bp and 229-bp bands (17).

CD34⁺ cells from patients with PV and normal CD34⁺ human cells were cultured in SFM containing SCF, TPO, Flt-3 ligand, and IL3 at 50 ng/mL, and cells were treated with various doses of zileuton (50–500 μmol/L). Cells were stained with a CD34 monoclonal antibody (BD Biosciences) initially and then washed again with binding buffer before staining with Annexin-V (BD Biosciences). Data were acquired on a FACS Calibur flow cytometric analyzer (Becton Dickinson).

Statistical analysis was performed by using the Student t test (GraphPad Prism v5.01 software for Windows, GraphPad Software).

To study the role of Alox5 in JAK2V617F-induced PV, we decided to use Ba/F3 and JAK2V617F-expressing Ba/F3 cells, as well as JAK2V617F-expressing human HEL cells. We characterized these cells for the activation of JAK/STAT signaling by JAK2V617F using a JAK2 inhibitor ruxolitinib. We found that compared with Ba/F3 cells, phosphorylation of JAK2 and STAT5 was greatly increased in JAK2V617F-expressing Ba/F3 and HEL cells and that inhibition of JAK2 kinase activity by ruxolitinib largely reduced the phosphorylation (Fig. 1A), indicating that JAK/STAT signaling is highly activated by JAK2V617F in these cells. In addition, expression of JAK2V617F also led to activation of β-catenin and AKT (Fig. 1B). Using these cells, we then tested whether Alox5 expression is regulated by JAK2V617F. We found that in JAK2V617F-expressing Ba/F3 cells, expression of 5-LO encoded by Alox5 was significantly increased as compared with Ba/F3 cells (Fig. 1C), suggesting that Alox5 is involved in JAK2V617F signaling. Next, we wondered whether upregulation of 5-LO expression by JAK2V617F is abolished upon inhibition of JAK2V617F signaling. We treated JAK2V617F-expressing Ba/F3 and HEL cells with ruxolitinib using Ba/F3 cells as control and found that inhibition of JAK2 phosphorylation by Ruxolitinib did not downregulate 5-LO expression (Fig. 1C). This result suggests that JAK2V617F regulates 5-LO expression independently of its kinase activity, providing a unique mechanism for the involvement of Alox5 in JAK2V617F signaling. To study the functional relevance of JAK2V617F/Alox5 signaling, we further tested whether inhibition of Alox5 function abrogates JAK2V617F-stimulated cell proliferation by treating JAK2V617F-expressing Ba/F3 and HEL cells with the Alox5 inhibitor zileuton and observed a dose-dependent growth inhibition of the cells (Fig. 1D). In contrast, zileuton had no significant inhibitory effect on parental Ba/F3 cells that do not express JAK2V617F (Fig. 1D), indicating that zileuton inhibits cell growth stimulated by JAK2V617F. Furthermore, we examined the effect of Alox5 inhibition on apoptosis and cell-cycle progression of Ba/F3, JAK2V617F-expressing Ba/F3 and HEL cells. We found that zileuton treatment induced apoptosis of JAK2V617F-expressing Ba/F3 and HEL cells with a minimal effect on Ba/F3 cells (Fig. 1E). In addition, zileuton treatment caused a blockade of cell-cycle progression of JAK2V617F-expressing Ba/F3 cells from G₀ to G₁ phase but did not affect the S + G₂–M phase (Fig. 1F). No effect of zileuton on the S + G₂–M phase was confirmed by staining the cells with propidium iodide (PI) and subsequent FACS analysis (Fig. 1G).

Next, we examined whether Alox5 is required for induction of PV by JAK2V617F in the retrovirus transduction/transplantation mouse model (13). We transduced bone marrow cells from wild-type (WT) or Alox5 homozygous knockout (Alox5^−/−) mice in C57BL/6 background with retrovirus expressing JAK2V617F, followed by transplantation into lethally irradiated recipient mice. We found that the white blood cell counts (WBC) in recipients of JAK2V617F-transduced Alox5^−/− donor bone marrow cells was significantly lower than that in recipients of JAK2V617F-transduced WT donor bone marrow cells (Fig. 2A), indicating that Alox5 is required for JAK2V617F-induced myeloproliferative disorder. In addition, we found that the average percentage of JAK2V617F-expressing Gr-1⁺ WBC count (GFP⁺Gr-1⁺ WBCs) in recipients of JAK2V617F-transduced Alox5^−/− donor bone marrow cells was significantly lower than that in recipients of JAK2V617F-transduced WT donor bone marrow cells (Fig. 2B; P < 0.01), indicating further that Alox5 is required for JAK2V617F-induced myeloproliferative disorder. We also analyzed WBCs that did not express JAK2V617F (GFP⁻) in mice and found that the percentage of GFP⁻ Gr-1⁺ WBCs was not reduced in the recipients of JAK2V617F-transduced Alox5^−/− donor bone marrow cells (Fig. 2B), suggesting that Alox5 deficiency does not significantly affect normal Gr-1⁺ WBCs. The effect of Alox5 deficiency on PV development was further supported by the lower number of JAK2V617F-expressing (GFP⁺) Gr-1⁺ WBCs compared with the WT group (Fig. 2C). When we compared total numbers of cells in these PV mice, we found that RBCs (Fig. 2D), hemoglobin (Fig. 2E), and hematocrit (Fig. 2F) were significantly lower in recipients of JAK2V617F-transduced Alox5^−/− bone marrow cells than in recipients of the transduced WT bone marrow cells, further indicating that Alox5 is required for PV development. In bone marrow, the average percentage of JAK2V617F-expressing Gr-1⁺ WBCs (GFP⁺ Gr-1⁺ WBCs) in recipients of JAK2V617F-transduced Alox5^−/− donor bone marrow cells was also significantly lower than that in recipients of JAK2V617F-transduced WT donor bone marrow cells (Fig. 2G; P < 0.01). In addition, the number of bone marrow erythroid cells (Ter119⁺GFP⁺) in mice receiving JAK2V617F-transduced Alox5^−/− bone marrow cells was significantly lower than that in mice receiving JAK2V617F-transduced WT bone marrow cells (P < 0.05; Fig. 2H) and the same was true in the spleen (Fig. 2I), which was consistent with the smaller spleen size in the absence of Alox5 (Fig. 2J). Because JAK2V617F-transformed HSCs is shown to be a PV-initiating cell population (7), we further tested whether loss of Alox5 affects this cell population in PV mice. We found that the number of HSCs (Lin⁻Sca-1⁺c-Kit⁺) in mice receiving JAK2V617F-transduced Alox5^−/− bone marrow cells was significantly lower than that in mice receiving JAK2V617F-transduced WT bone marrow cells (Fig. 2K).

Because deletion of Alox5 caused inhibition of the development of JAK2V617F-induced PV in mice (Fig. 2), we tested whether Alox5 is a potential target gene for PV by treating PV mice with zileuton (300 mg/kg, once a day) or a placebo. BALB/c mice were used because JAK2V617F induces PV more robustly in BALB/c mice than in C57BL/6 mice (13). Western blot analysis of protein lysates from the spleens of PV mice showed that 5-lipoxyginase (5-LO, a protein product of the Alox5 gene) is upregulated by JAK2V617F, and inhibition of Alox5 function by zileuton restored the 5-LO level back to the control level (Fig. 3A). We believe that the reduced 5-LO levels in zileuton-treated PV mice simply reflect the reduction or disappearance of JAK2V617F-expressing cells in the spleen of the treated mice, as zileuton is thought to inhibit the enzymatic activity of 5-LO with no known effect on 5-LO expression. In these mice, peripheral blood smears and tissue sections of bone marrow, spleen, and liver showed a reduced number of peripheral blood WBCs and less tissue infiltration in zileuton-treated PV mice as compared with placebo-treated PV mice (Fig. 3B). We also compared WBC counts and found that the number of WBCs in placebo-treated mice rose significantly with time, whereas the number of WBCs and percentage of JAK2V617F-expressing (GFP⁺) Gr-1⁺ WBCs in zileuton-treated mice did not (Fig. 3C and D). The inhibition of PV development by zileuton correlated with much smaller infiltrated spleen in zileuton-treated PV mice than in placebo-treated PV mice (Fig. 3E). To further demonstrate the effect of Alox5 inhibition by zileuton on PV development, we compared the levels of GFP⁺ cells in PV mice. FACS analysis showed that the percentage of Gr-1⁺ or Mac⁺ PV cells in peripheral blood was significantly lower in zileuton-treated PV mice than in placebo-treated PV mice (Fig. 3F), although there was no significant difference in the percentage of B220⁺ cells between the 2 treatment groups (Fig. 3F). Similar result was observed in bone marrow of these mice (Fig. 3G). Consistent with the inhibition of PV cells by zileuton in mice, the average LTB4 level in zileuton-treated PV mice was much lower than that in placebo-treated PV mice (Fig. 3H), and survival of zileuton-treated PV mice was significantly improved compared with placebo-treated PV mice (Fig. 3I). Beginning 6 months after induction of PV, about 40% placebo-treated PV mice gradually died, but all zileuton-treated mice survived (Fig. 3I). In addition, quantitative colony-forming assay showed that zileuton inhibits PV cell growth in vitro (Fig. 3J). Furthermore, the colony formation of granuloctye macrophage progenitor (CFU-GM) and erythroid cells (BFU-E/CFU-E) was also inhibited by zileuton (Fig. 3K). Together, these results indicate that inhibition of Alox5 function suppresses the growth of JAK2V617F-expressing cells, providing a novel therapeutic strategy for JAK2V617F-induced PV.

We further evaluated the effect of Alox5 inhibition by zileuton on PV development by examining RBCs and platelets. We found that zileuton treatment did not significantly affect the level of RBCs (Fig. 4A) but had a significant inhibitory effect on mean corpuscular volume (MCV; Fig. 4B), hematocrit (Fig. 4C), mean corpuscular hemoglobin (MCH; Fig. 4D), and hemoglobin (Fig. 4E) as compared with placebo-treated PV mice. Zileuton treatment did not have an inhibitory effect on the level of PLT in PV mice (Fig. 4F). Furthermore, we found that in both bone marrow and spleen, the number of JAK2V617F-expressed erythroid cells (Ter-119⁺GFP⁺) in zileuton-treated PV mice was significantly lower than that in placebo-treated PV mice (P < 0.01 and P < 0.001; Fig. 4G and H). The above-described inhibitory effect of zileuton was consistent with the smaller size of the spleen in zileuton-treated PV mice than in placebo-treated PV mice (Fig. 4I).

Because JAK2V617F affects plasma levels of Epo in patients with PV (12), we also investigated whether zileuton treatment alters the level of Epo regulated by JAK2V617F by using mice receiving empty vector–transduced bone marrow cells as a control. We found that the EPO level in placebo-treated PV mice was significantly lower than that in the control mice and that zileuton treatment largely restored the plasma EPO level after 30-day treatment (Fig. 5A). To investigate whether JAK2V617F regulates the EPO signaling pathway via Alox5, we treated JAK2V617F-expressing BaF/3 and HEL cells with zileuton using Ba/F3 cells as control because the EPO signaling is linked to the JAK2/STAT5 pathway and subsequently the PI3K/AKT pathway in JAK2V617F-expressing cells (2, 3). We found that at higher dose of zileuton with 6-hour treatment, JAK2/STAT5 signaling was inhibited (Fig. 5B). In JAK2V617F-expressing Ba/F3 cells, zileuton also inhibited activation of AKT and β-catenin by JAK2V617F (Fig. 5C) and did not affect activation of PI3K and MAPK (Fig. 5D).

We next evaluated whether inhibition of Alox5 function by zileuton reduces proliferation of human CD34⁺ cells from patients with PV. We cultured peripheral blood cells from patients with PV under the stem cell culture conditions in the absence or presence of zileuton (100 and 250 μmol/L) for 48 hours. The percentages of CD34⁺ cells were analyzed by FACS, and total cell numbers were compared at the end of the culture. We found that zileuton treatment significantly reduced proliferation of CD34⁺ cells (Fig. 6A). Human CD34⁺ cells from patients with PV were also analyzed in a colony-forming assay. The colonies were enumerated after 14 days of incubation (Table 1), and individual colonies were plucked and genotyped for detecting the presence of JAK2V617F. We found that zileuton treatment significantly reduced colony number of CD34⁺ progenitor cells carrying 2 JAK2V617F mutant alleles (Fig. 6B) but had no effect on colony formation of normal CD34⁺ progenitor cells. Zileuton treatment also reduced colony formation of human CFU-GM and BFU-E (Fig. 6C).

In this study, we find that loss of the Alox5 gene or pharmacologic inhibition of Alox5 function attenuates the development of PV induced by JAK2V617F, providing a strong rationale for targeting Alox5 in PV treatment. This idea is supported by our finding that inhibition of Alox5 function reduces proliferation of human CD34⁺ cells from patients with PV. Recently, Alox5 has been shown to play a role in functional regulation of disease-initiating cells for BCR-ABL- or TEL-PDGFR–induced myeloid leukemia and human glioma cells (15, 18, 19). Similar to CML induced by BCR-ABL, JAK2V617F was detected in HSCs of patients with PV (5), and JAK2V617F-expressing long-term HSCs were shown to be responsible for the initiation and maintenance of PV in mice (7). The effect of Alox5 loss on PV-initiating cells could complement the benefit from the use of a JAK2 inhibitor that does not have an inhibitory effect on PV-initiating cells (7).

It is interesting for us to observe that inhibition of JAK2 activity by ruxolitinib inhibits JAK2 phosphorylation but does not cause a reduced 5-LO expression, suggesting that JAK2V617F stimulates 5-LO expression independently of its kinase activity. This result supports a rationale for simultaneously inhibiting both JAK2V617F and 5-LO in treating PV. A number of JAK2 inhibitors (such as SAR302503 and Jakafi) have been developed and tested in clinic (20), and a long-term treatment benefit of these inhibitors still remains to be seen. Thus, an anti-Alox5 agent alone or in combination with a JAK2 inhibitor will provide a new therapeutic strategy for PV.

β-Catenin pathway plays a critical role in self-renewal and differentiation of both HSCs (21) and leukemic stem cells (LSC) in CML (22–24). We show that the molecular mechanism for the role of Alox5 in PV is linked to the β-catenin pathway. Our previous study shows that Alox5 regulates the level of β-catenin expression in LSCs of CML (15), and similarly, we find that JAK2V617F also induces Alox5 expression. These findings suggest that the Alox5 may represent a shared signaling pathway by Philadelphia chromosome–positive and -negative MPNs. Besides β-catenin, we also show that JAK2V617F activates the AKT pathway but does not activate PI3K and MAPK, which is different from BCR-ABL that activates these two pathways in CML cells (25). This signaling difference may reflect pathologic differences between PV and CML and provides clues for better understanding of these 2 types of MPNs. It is likely that Alox5 functions upstream of β-catenin and AKT because inhibition of Alox5 function by zileuton causes downregulation of β-catenin expression and reduced AKT phosphorylation in JAK2V617F-expressing cells. Our future study will focus on explaining how Alox5 signals to β-catenin and AKT in PV cells.

Alox5 is involved in numerous physiologic and pathologic processes, including oxidative stress response, inflammation, and cancer (26–30). Increasing evidence shows that the metabolites of the arachidonic acid cascade play a role in epithelial cell proliferation and tumorigenesis (31, 32). A known function of Alox5 is to produce inflammatory metabolites called leukotrienes. There are several types of leukotrienes (33), among which LTB4 is a major product of the Alox5 pathway. We show that the level of LTB4 is decreased in PV mice treated with zileuton. Although there is no available evidence indicating that LTB4 is a mediator of Alox5 signaling in activation of β-catenin and AKT, this possibility needs to be examined in the future. At least, it is reasonable to think that LTB4 could serve as a biologic indicator for PV pathogenesis.

In summary, we show that Alox5 plays critical role in the development of PV induced by JAK2V617F, which identified Alox5 as a potential target gene for PV. Future studies will test more robustly whether inhibition of Alox5 function is effective in suppressing human PV cells.

No potential conflicts of interest were disclosed.

Conception and design: Y. Chen, A. Liang, S. Li

Development of methodology: Y. Chen, Y. Shan, N. DeSouza, R. Hoffman

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Y. Chen, Y. Shan, N. DeSouza, R. Hoffman, S. Li

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): Y. Chen, Y. Shan, N. DeSouza, R. Hoffman, S. Li

Writing, review, and/or revision of the manuscript: Y. Chen, M. Lu, N. DeSouza, R. Hoffman, A. Liang, S. Li

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Y. Chen, M. Lu, S. Li

Study supervision: Y. Chen, S. Li

Other (technical support for animal experiment): Z. Guo

We thank Dr. Ross Levine (Memorial Sloan-Kettering Cancer Center) for providing JAK2V617F-expressing Ba/F3 cell line.

This work was supported mostly by a grant from the MPN Research Foundation (S. Li). The work was also partially supported by grants from the Leukemia & Lymphoma Society and NIH grants R21-HL113603, R01-CA122142, and R01-CA176179 (S. Li).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.