β-thalassemia arises as a result of mutations in the β-globin gene. As a consequence erythropoiesis, the process that insures the daily generation of billions of red blood cells (RBCs), becomes disrupted. Ineffective erythropoiesis is a major contributor to the β-thalassemic anemia and is partially due to aberrant apoptosis during late stages of erythroid maturation. Despite the importance of apoptosis, the underlying molecular mechanisms regulating this process in β-thalassemia erythroblasts are not fully elucidated.
One potential mechanism involves the transcription factor Foxo3, which under specific contexts can act as a positive regulator of apoptosis, but is also an essential transcriptional regulator of terminal erythroblast maturation. Foxo3 has a range of outputs that it can execute from sustaining cellular integrity by mitigating oxidative stress to inducing apoptosis under conditions of overwhelming stress. Given these functions, we sought to determine if Foxo3 played a role in maintaining RBC maturation in β-thalassemic mice. To address this, we used Hbbth3/+ (th3/+) mice that display a phenotype similar to β-thalassemia intermedia, and produced double mutant Foxo3-/-/Th3/+ mice. The th3/+ mice display a mild erythroblast apoptotic phenotype. We hypothesized that loss of Foxo3 may exacerbate the β-thalassemic phenotype. On the contrary, we found that loss of Foxo3 in a β-thalassemic background improved RBC numbers and hemoglobin concentration (by 1g/dl, n=10 mice) in double mutant mice compared to th3/+ mice. Furthermore, double mutant mice had a statistically significant lower frequency of apoptosis (2 fold less) during bone marrow erythroblast maturation as measured by flow cytometry analysis of annexin V-binding and 7AAD staining in distinct erythroblast stages resolved by TER119, CD44 and cell size (n=3 mice per genotype). We predicted that high levels of oxidative stress may prematurely activate FOXO3 during erythroblast maturation in β-thalassemic mice. In turn, activated FOXO3 may potentially promote apoptosis in these cells. To evaluate this, we examined FOXO3 levels by qRT-PCR and immunofluorescence in FACS sorted populations of erythroblasts (TER119+,CD44,FSC) or erythroid progenitors (TER119-,c-KIT+,CD71HI) acquired from bone marrow of at least 3 mice per genotype. Our data show increased mRNA levels of Foxo3 in early erythroblasts, corresponding to increased FOXO3 protein expression in erythroid progenitors from β-thalassemic mice relative to wild-type mice. We also examined the activation status of p53, as it is also a major regulator of apoptosis that can be triggered by oxidative stress. Nuclear p53 levels were greater in β-thalassemic as compared to wild-type erythroid progenitors based on immunofluorescence analysis of sorted cells from bone marrow of 3 mice per genotype. These results suggest a higher level of active p53 in β-thalassemic erythroid progenitors. Our results provide evidence that FOXO3, a factor normally critical for erythroblast maturation, may cooperate with aberrantly active p53 to induce apoptosis in β-thalassemic erythroblasts. In support of this, downstream p53 targets including Gadd45a and p21 that are also Foxo3 targets were significantly upregulated in β-thalassemic erythroblasts relative to wild-type erythroblasts as determined by qRT-PCR of cDNA produced from 3 mice per genotype. To more closely examine the mechanism of decreased apoptosis in double mutant Foxo3-/-/Th3/+ erythroblasts, we compared the expression of multiple genes involved in apoptosis by qRT-PCR of sorted erythroblast populations from at least 3 mice per genotype. We found multiple pro-apoptotic genes including, Cycs, Tnfsf10, Puma, and Bim expressed at significantly lower levels at various erythroblast stages in double mutant compared to β-thalassemic erythroblasts. Together, our data suggests Foxo3 becomes inappropriately and prematurely activated in erythroid progenitors and early erythroblasts in the context of β-thalassemia and cooperates with p53 to promote apoptosis. These findings raise the possibility that cooperation of Foxo3 and p53 in β-thalassemic erythroblasts might contribute to the ineffective erythropoiesis of β-thalassemic mice. They also suggest the possibility that as a homeostatic maintaining factor, Foxo3 behaves differently in the context of disease.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.