Abstract
Activation of both EpoR signaling and GATA-1 transcription factor is required for normal erythropoiesis. Whether any signal generated from Epo-stimulated EpoR regulates GATA-1 function is not known. In particular the function of the PI3-kinase-AKT signaling pathway downstream of EpoR is not clear. Retroviral (MSCV-IRES-GFP) transduction with a constitutively active but not wild type AKT induces red cell differentiation of both JAK2−/− and wild type fetal liver cells in the absence of Epo. The differentiation of fetal liver cells along the erythroid lineage was determined by the number of CFU-E-generated colonies in vitro, Realtime PCR analysis of red cell specific gene expression, FACS analysis of cell surface markers TER119 and CD71, morphological analysis and diaminobenzidine staining of hemoglobin of the transduced GFP+ cells. Consistent with a role for AKT serine threonine kinase in supporting erythroid differentiation, overexpression of a dominant negative AKT partially inhibited Epo-dependent erythroid differentiation of fetal liver cells and of cultured erythroid cells. Furthermore, the significant potential of the constitutively active AKT in inducing red cell differentiation of fetal liver cells could not be solely attributed to its survival signal. We have identified serine 310 (S310) within a putative AKT consensus phosphorylation sequence in GATA-1 transcription factor. This sequence is highly conserved among species and among hematopoietic GATA (1, 2, 3) members. Recombinant and immunocomplexes of activated AKT but not the related kinase SGK phosphorylated specifically GST-GATA-1 WT but not GST-GATA-1 S310A in vitro. The constitutively activate AKT transactivated wild type (WT) GATA-1 but not the mutated GATA-1 S310A in reporter gene assays. We raised an anti-GATA-1 pS310 antibody and analyzed by Western Blot GATA-1 phosphorylation in response to Epo in the Epo-starved erythroleukemic HCD57 cells. Phosphorylation of GATA-1 on S310 was detected within 30 minutes and up to several hours in nuclear extracts of Epo-stimulated HCD57 cells and was inhibited in the presence of the PI3-Kinase inhibitor LY294002 (10μM) but not the MAPkinase inhibitor. Interestingly, among the seven constitutively phosphorylated serines of GATA-1, serine 310 is the only residue that is hyperphosphorylated during DMSO-induced differentiation of murine erythroleukemia cells. To further investigate the role of GATA-1 phosphorylation, we retrovirally transduced GATA-1-deficient G1E cells with GATA-1 WT and mutants and assessed red cell differentiation by benzidine staining and Realtime RT-PCR analysis. G1E cells are arrested at a proerythroblast stage and differentiate to mature red cells when overexpressing WT GATA-1. Expression of GATA-1 Dephospho missing all seven phospho-serine residues in G1E cells induced only 40% of erythroid differentiation seen with WT GATA-1. Expression of GATA-1 DephosphoA310S mutant with S310 added back to GATA-1 Dephospho resulted in 70% of differentiation seen with GATA-1 WT. Similarly, retroviral expression of GATA-1 Dephospho blocked significantly erythroid differentiation of transduced GFP+ fetal liver cells in the presence of Epo whereas expression of GATA-1 DephosphoA310S did not have a significant inhibitory effect. Taken together, these data suggest that phosphorylation of GATA-1 is regulated by PI3-kinase downstream of EpoR and is important for red cell differentiation.
Chromatin occupancy and epigenetic analysis reveal new insights into the function of the GATA1 N terminus in erythropoiesis
