Aberrant Responsiveness of Erythropoiesis to Iron Deficiency in Polycythemia Vera

Polycythemia vera (PV) presents with iron deficiency at diagnosis, and the mainstay of treatment, i.e. therapeutic phlebotomy, worsens iron deficiency. Iron deficiency is associated with anemia and symptoms of cognitive impairment and fatigue even in the absence of anemia, and patients with low risk PV often suffer from iron deficiency related symptoms How iron deficiency develops in PV patients prior to phlebotomy is not well understood. We previously demonstrated that PV patients exhibit a greater extent of iron deficiency relative to wild type JAK2 patients with other causes of erythrocytosis [Ginzburg Leuk 2018]. We hypothesize that mutated JAK2 leads to aberrant insensitivity of erythropoiesis to iron deficiency. To explore this hypothesis, we first analyzed serum data from iron deficient PV patients (n=14) and blood donors (n=15), normalized for age (51 vs. 42 years, P=0.12) and serum ferritin concentration (22 vs. 22 ng/ml, P=0.95); our data demonstrate that PV patients have significantly lower hepcidin, transferrin saturation, MCV, a higher HCT, and a trend toward higher erythroferrone (ERFE) relative to controls (Fig 1a-1e). Secondly, CD34+ cells were isolated from mononuclear cells and plated with erythropoietin and either 100% or 10% transferrin saturation to mimic iron replete and iron deficient conditions, respectively; differentiation and proliferation were analyzed using flow cytometry. These experiments revealed that although glycophorin A (GPA) and CD36+ cells were decreased in iron deficient relative to iron replete control cells, PV cells continue to proliferate irrespective of iron status (Fig 1f). In addition, only iron deficient control, not PV, cells demonstrated an erythroid lineage specific decrease in proliferation relative to iron replete cells (Fig 1g), demonstrating that iron restriction in PV does not limit erythroid differentiation or proliferation in vitro. Thirdly, we transplanted JAK2 V617F (PV) and wild type (WT) cells into recipient females and placed mice on 35ppm (iron replete (IR)) or 2.5ppm (iron deficient (ID)) diets. IR PV mice exhibited the expected erythrocytosis and decrease in MCV and MCH relative to WT controls (Fig 2a-2d). WT mice on an ID diet exhibited decreased MCV, MCH, and RET-He while PV mice had decreased RBC counts with an increased MCHC (Fig 2a, 2c-2f). These findings demonstrate altered iron regulation in PV erythroblasts with a preference for decreasing RBC count, rather than cellular Hb production, in iron deficiency in vivo. No differences were found in the total number of bone marrow erythroblasts in PV mice relative to WT mice on IR or ID diets. IR and ID PV mice demonstrated splenomegaly relative to WT controls. In addition, IR PV erythroblasts expressed significantly more ERFE relative to WT controls with decreased ERFE expression in ID PV erythroblasts (Fig 2g). Similarly, liver hepcidin expression was lower in IR PV relative to WT controls, but was restored in ID PV mice (Fig 2h), the later likely a response to decreased ERFE expression (Fig 2g); no changes are observed in liver iron concentration in IR relative to ID PV mice. Furthermore, the expected decrease in pSTAT is observed only in ID WT (Fig 2i), not ID PV (Fig 2j) erythroblasts. Lastly, TfR2 protein expression was increased in IR PV relative to WT controls (Fig 2k) and decreased only in ID WT (Fig 2l) but not ID PV (Fig 2m) erythroblasts. Since TfR2 degradation is enhanced during iron deficiency [Khalil JEM 2018] and TfR2 enables iron delivery for mitochondrial heme synthesis [Khalil Blood Adv 2017], persistently increased TfR2 in PV erythroblasts may explain why cellular Hb synthesis (i.e. MCH and RET-He) remains unaffected by iron deficiency in PV mice. Taken together, these findings demonstrate that in vitro iron deficiency fails to limit differentiation and proliferation in PV erythroblasts and enhances STAT5 signaling in ID PV erythroblasts in vivo but results in decreased circulating RBCs in PV mice. In addition, decreased erythroblast ERFE expression in ID PV mice results in increased hepcidin to limit erythroblast iron availability, but persistently increased TfR2 concentration enables mitochondrial iron delivery for Hb synthesis despite cellular iron deficiency. Our studies provide novel mechanistic insights into the dysregulation of iron utilization for erythropoiesis in PV. Disclosures Levine: Prelude Therapeutics: Research Funding; Imago Biosciences: Membership on an entity's Board of Directors or advisory committees; Roche: Consultancy, Research Funding; C4 Therapeutics: Membership on an entity's Board of Directors or advisory committees; Amgen: Honoraria; Novartis: Consultancy; Gilead: Consultancy; Lilly: Honoraria; Loxo: Membership on an entity's Board of Directors or advisory committees; Qiagen: Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Research Funding; Isoplexis: Membership on an entity's Board of Directors or advisory committees. Hoffman:Merus: Research Funding. Ginzburg:La Jolla Pharma: Membership on an entity's Board of Directors or advisory committees.