Neuronal nitric oxide synthase is required for erythropoietin stimulated erythropoiesis in mice

Abstract

Introduction: Erythropoietin (EPO), produced in the kidney in a hypoxia responsive manner, is required for red blood cell production. In non-erythroid tissue, EPO increases endothelial cell production of nitric oxide (NO) and endothelial nitric oxide synthase (eNOS) that regulates vascular tone to improve oxygen delivery. This contributes to EPO cardioprotective activity in mouse models. Nitric oxide treatment in mice shifts hematopoiesis toward the erythroid lineage, increases red blood cell production and total hemoglobin. In erythroid cells, nitric oxide can also be generated by hydroxyurea metabolism that may contribute to hydroxyurea induction of fetal hemoglobin. We find that during erythroid differentiation, EPO induces neuronal nitric oxide synthase (nNOS) and that neuronal nitric oxide synthase is required for normal erythropoietic response.Methods: Wild type (WT) mice and mice with targeted deletion of nNOS (nNOS−/−) and eNOS (eNOS−/−) were assessed for EPO stimulated erythropoietic response. Bone marrow erythropoietic activity was assessed in culture by EPO dependent erythroid colony assay and in vivo by bone marrow transplantation into recipient WT mice. Contribution of nNOS to EPO stimulated cell proliferation was assessed in EPO dependent erythroid cells and in primary human erythroid progenitor cell cultures.Results: EPO treatment increased hematocrit similarly in WT and eNOS−/− mice and showed a lower increase in hematocrit nNOS−/− mice. Erythroid colony assays from bone marrow cells were comparable in number from wild type, eNOS−/− and nNOS−/− mice at low EPO concentration. Colony number increased at high EPO concentration is seen only in cultures from bone marrow cells of wild type and eNOS−/− mice but not from nNOS−/− mice. Colony size with high EPO treatment also exhibited a marked increase in erythroid cultures from wild type and eNOS−/− mice but not from nNOS−/− mice. Bone marrow transplant from nNOS−/− mice into immunodeficient mice showed engraftment at comparable levels to WT bone marrow transplant. With EPO treatment, the increase in hematocrit was blunted in recipient mice that received with nNOS−/− donor marrow compared with recipient mice that received WT donor marrow. In erythroid cell cultures, addition of nNOS inhibitor resulted in decreased EPO dependent proliferation mediated in part by decreased EPO receptor expression, and decreased proliferation of hemin induced differentiating erythroid cells.Discussion: EPO treatment in mice and in corresponding cultures of bone marrow erythropoiesis suggest an intrinsic defect in erythropoietic response of nNOS−/− mice to high EPO stimulation. Transplantation of bone marrow from donor WT or nNOS−/− mice into recipient WT mice showed that EPO treatment post-transplant recapitulated the response of donor mice. Culture studies suggest nNOS regulation of EPO dependent erythroid cell proliferation, expression of EPO receptor and cell cycle associated genes, and AKT activation. These data provide evidence that nitric oxide modulates EPO dose dependent erythropoietic response.