Abstract

Steady-state and dynamic regulation of erythrocyte production occurs by altering the balance of cell-survival versus apoptosis signaling in maturing erythroblasts. Previously, the pro-apoptotic factor Nix was identified as a critical death signal in normal erythropoietic homeostasis, acting in opposition to erythroblast-survival signaling by erythropoietin and Bcl-xl. However, the role of Nix in stress-erythropoiesis is not known. Here, by comparing the consequences of erythropoietin administration, acute phenylhydrazine-induced anemia, and aging in wild-type and Nix-deficient mice, we show that complete absence of Nix, or its genetic ablation specifically in hematopoietic cells, mimics the effects of erythropoietin (Epo). Both Nix ablation and Epo treatment increase early erythroblasts in spleen and bone marrow and increase the number of circulating reticulocytes, while maintaining a pool of mature erythroblasts as an “erythropoietic reserve”. As compared with WT, Nix null mice develop polycythemia more rapidly after Epo treatment, consistent with enhanced sensitivity to erythropoietin observed in vitro. After phenylhydrazine administration, anemia in Nix-deficient mice is less severe and recovers more rapidly than in WT mice, despite lower endogenous Epo levels. Anemic stress depletes mature erythroblasts in both WT and Nix null mice, but Nix null mice with basal erythroblastosis are resistant to anemic stress. These findings show that Nix null mice have greatly expanded erythroblast reserve and respond normally to Epo- and anemia-stimulated induction of erythropoiesis. However, the hematocrits of young adult Nix null mice are not elevated, and these mice paradoxically develop anemia as they age with decreased hemoglobin content (10g/dl) and hematocrit (36%; at 80±3 weeks of age) compared to WT mice (13g/dl and 46%; 82±5 weeks of age), inspite of persistent erythoblastosis observed in the bone marrow and spleen. Nix null erythrocytes, which are macrocytic and exhibit membrane abnormalities typically seen in immature cells or with accelerated erythropoiesis, demonstrate shorter life span with a half life of 5.2±0.6 days in the peripheral circulation by in vivo biotin labeling (as compared with a half life of 11.7±0.9 days in WT), and increased osmotic fragility as compared with normal erythrocytes. This suggests that production and release of large numbers of reticulocytes in Nix null mice can decrease erythrocyte survival. To rule out a non-hematopoietic consequence of Nix ablation that contributes to or causes increased erythrocyte fragility and in vivo consumption, such as primary hypersplenism, we undertook Tie2-Cre mediated conditional Nix gene ablation. Nixfl/fl + Tie2-Cre mice (hematopoietic-cell specific Nix null) develop erythroblastosis with splenomegaly, reticulocytosis, absence of polycythemia and increased erythrocyte fragility; suggesting that erythroblastosis and accelerated erythrocyte turnover are a primary consequence of Nix ablation in hematopoietic cells. Hence, dis-inhibition of erythropoietin-mediated erythroblast survival pathways by Nix ablation enhances steady-state and stress-mediated erythropoiesis.

Author notes

Disclosure: No relevant conflicts of interest to declare.