Exogenous EPO is sufficient to initiate late-stage erythroid progenitor expansion and accelerate erythroid recovery after sublethal irradiation. (A) Erythroid progenitor recovery kinetics in the bone marrow ± intraperitoneal injection of 1000 IU/kg EPO at 1 hour post–4 Gy TBI. Recovery of d3 BFU-E is advanced by 2 to 3 days in EPO-injected mice (red bars) compared with mock-treated mice (blue bars) and leads to accelerated CFU-E recovery. (B) Erythroid bone marrow precursor recovery kinetics ± IP EPO injection at 1 hour after TBI. Erythroid precursors in EPO-injected mice (red bars) undergo a wave of recovery 2 to 3 days sooner than mock-treated mice (blue bars) with kinetics that mirror endogenous recovery. Erythroid progenitors and precursors are normalized per femur and expressed as a percent of unirradiated control marrow. (C) Circulating red cell recovery ± EPO injection at 1 hour post–4 Gy TBI. Advanced reticulocyte recovery beginning at 5 days after radiation (red line) leads to partial HCT normalization by 6 days after TBI (green line) in EPO-treated mice. Reticulocytes are calculated as absolute reticulocyte index (% Retic × total RBC and Retic number × %HCT) and expressed as a percent of unirradiated control levels. RBC levels are expressed as percent HCT. Dotted lines represent unirradiated control levels. Error bars represent SEM of at least 3 experiments, and 3 or more independently assayed mice were used to determine each data point. Statistical analyses were performed using a 2-tailed Student t test, (*P < .05; **P < .01; ***P < .001; significantly different from 4 Gy TBI mock-treated mice at matched timepoints). (D) Representative histologic sections of bone marrow ± EPO 4 days post–4 Gy TBI (H&E; 20-micron bars). EPO treatment leads to increased cellularity and decreased vascular dilation compared with mock-treated samples. Images were captured with a Nikon Digital Sight Ds-Fi1 camera using Nikon NIS-Elements software on a Nikon Eclipse 80i upright microscope using a 20× objective.