Ineffective Erythropoiesis in β-Thalassemia Is Characterized by Enhanced Survival Mechanisms That Reduce P21-Mediated Cell Death.

Ineffective erythropoiesis (IE) in β-thalassemia is described as increased expansion of erythroid progenitor cells in combination with accelerated apoptosis and intramedullary hemolysis. However, evidence for this assumption is not particularly strong. In this study we evaluated the kinetics of red blood cell proliferation and survival in thalassemic mice that exhibit levels of anemia consistent with thalassemia intermedia (th3/+) and major (th3/th3), as we described previously. Th3/+ mice show anemia and increased reticulocyte counts (8.9±1.1 g/dL and 17.7±2.6x10⁵/ul compared to 14.9±1.1 g/dL and 2.6±0.4x10⁵/ul in +/+), whereas th3/th3 mice show more severe anemia than th3/+ mice and reduced production of reticulocytes (3.0±1.2 g/dL and 1.8±0.7x10⁵/ul). As soon as two months of age, EPO levels in thalassemic mice are significantly increase over normal mice, 10 times in th3/+ and up to three orders of magnitude in th3/th3 mice. The total number of nucleated erythroid cells (spleen + bone marrow) increased from 3.3±0.9x10⁸ in +/+ to 1.2±0.5x10⁹ (or 3.6 folds) in th3/+ and 1.6±0.6x10⁹ (or 4.8 folds) in th3/th3 mice (N=5 per group), whereas the level of apoptotic cells increased only 2 to 3 folds in percentage in th3/+ and th3/th3, respectively, as observed using the apoptotic Annexin-V and erythroid specific Ter119 markers (9.9±5.0, 14.3±2.5, 24.2±6.7 in BM and 10.6±3.5, 14.8±6.5, 25.4±6.2 in spleen of +/+, th3/+ and th3/th3, respectively; n=3 per genotype). This data was confirmed by TUNEL, cleaved caspase-3/7 and by immunostaining assays. Bilirubin and LDH levels were not different between thalassemic and +/+ mice. Altogether these observations indicated that in thalassemia there is a disproportion between number of proliferating and dying cells with a net increase of erythroid cells. Furthermore, microarray analysis on erythroid cells indicated increased expression of cell cycle promoting genes such as Ki67, Mcm3, Cyclin-A, CDK2 and BclXL (2 to 6 folds compared to +/+ mice, n=5 per genotype). This data was confirmed by Q-PCR, Western blot, immunostaining, clonogenic assay and by analysis of the percentage of erythroid cells in S-phase after in vivo BrdU injection (22, 30 and 44% BrdU⁺ in +/+, th3/+ and th3/th3, respectively). On the other hand, in th3/th3, which show more apoptosis than th3/+ mice, the cyclin-dependent kinase inhibitor p21 was upregulated both at the RNA (50-folds) and protein level. P53 was also analyzed in th3/th3 mice, showing no expression. In order to investigate the function of p21 in thalassemic erythroid cells, its expression was analyzed on purified erythroid cells isolated from th3/th3 that were transfused (10.4±0.4 g/dL of Hb) or thalassemic mice showing different levels of anemia (6.2±0.2 and 2.1±0.8 g/dL of Hb, respectively; N=3 per each group). We observed that the level of p21 increased with anemia and the severity of the pathology. However, in th3/th3 mice that were injected with BrdU, immunostaining analysis indicated that a large amount of p21⁺ erythroid cells were also BrdU⁺. In conclusion, we propose that erythropoiesis in β-thalassemia is characterized by enhanced expression of cell cycle promoting and survival factors that are able to overcome or mitigate p21 cell cycle block and, probably, apoptosis.