Bcl-x Prevents the Apoptosis of Late-Stage, Hemoglobin-Synthesizing Erythroblasts, but It Does Not Mediate the Anti-Apoptotic Effect of Erythropoietin on Early-Stage Erythroblasts.

Bcl-x is a protein in the outer mitochondrial membrane. A member of the Bcl-2 family, Bcl-x protects developing erythroid cells from apoptosis. The exact stage of erythroid development at which Bcl-x exerts its anti-apoptotic effect is not known, but induction of Bcl-x has been proposed to be the mediator of erythropoietin’s (EPO) anti-apoptotic effect in erythroid differentiation. EPO is the principal trophic hormone that controls red blood cell production by regulating apoptosis of erythroid progenitor cells at the CFU-E and early erythroblast stages. Bcl-x has also been reported to be necessary for heme synthesis. In mice, Bcl-x deficiency is embryonically lethal; when Bcl-x deficiency is acquired postnatally by conditional knockout technology, it is associated with splenomegaly, thrombocytopenia, and a profound anemia that is thought to be hemolytic in origin.

Objectives: 1)To characterize the defect of erythroid differentiation in conditional Bcl-x −/− mice. 2)To determine whether Bcl-x is the mediator of EPO’s anti-apoptotic action. 3)To determine whether Bcl-x is necessary for heme synthesis.

Methods: Phlebotomized or unbled littermate controls and anemic adult Bcl-x −/− mice obtained by cre-lox conditional knockout were bled, sacrificed, and splenectomized. Purified populations of splenic erythroblasts were isolated by sedimentation at unit gravity, cultured with or without EPO, and harvested at 0, 8, 20, 32, and 44 hours for cell counts, cytospin preparations for morphology, flow cytometry analyses for apoptosis (TUNEL) and cell cycle phases, and ⁵⁹FeCl₃ incorporation into heme.

Results: Compared to littermate controls, Bcl-x −/− mice were severely anemic (Hgb 2.8 g/dL vs 15.4 g/dL in unbled controls and 7.2 g/dL in bled controls), thrombocytopenic (platelets 23x10³/microL vs 905x10³/microL in unbled controls and 984x10³/microL in bled controls), and reticulocytopenic (82.8x10³/microL vs 281x10³/microL in unbled controls and 1410x10³/microL in bled controls), while WBCs were unaffected. Expanded erythropoiesis led to massive splenomegaly (spleens =4.3gm vs 0.1gm in unbled controls and 0.3gm in bled controls). After 44 hours of culture with EPO, purified erythroblasts from bled controls proliferated 4-fold and differentiated such that the majority enucleated, producing 200–250 reticulocytes per 100 erythroblasts plated, whereas Bcl-x −/− erythroblast numbers doubled during the first 20 hours in culture, but the large majority died by apoptosis between 20 and 44 hours, producing only 9–12 reticulocytes per 100 erythroblasts plated. Bcl-x −/− erythroblast apoptosis occurred after the initiation of heme synthesis and proportionally in all phases of cell cycle. Compared to culture with EPO, Bcl-x −/− erythroblasts cultured without EPO underwent increased apoptosis at earlier times of culture-- at 8 hours (45% vs 29%), 20 hours (71% vs 42%) and 32 hours (83% vs 57%).

Conclusions: 1)Bcl-x is required for the survival and differentiation of the late-stage erythroblasts in all phases of cell cycle. Thus, Bcl-x deficiency results in ineffective erythropoiesis rather than hemolytic anemia. 2)Bcl-x is not required for heme synthesis, but has its anti-apoptotic effect during the stage of hemoglobin synthesis. 3)Bcl-x does not mediate EPO’s anti-apoptotic effect in early-stage erythroblasts.