GATA-1 Deacetylation and Interaction with HDAC1 Is Critical for GATA-1 Mediated Gene Transcription

The transcription factor GATA-1 is an essential factor for globin gene transcrption and is required for erythroid and megakaryocytic cell differentiation and maturation. GATA-1 can be acetylated by p300/CBP, and the acetylation modulates GATA-1 chromatin binding activity. However, it is not clear whether GATA-1 acetylation can be reversed by a deacetylase. It is showed that GATA-1 can indirectly interact with histone deacetylase 1 (HDAC1) associated NURD corepressor complexes through binding to FOG-1. However, we found that the NURD complex does not deacetylate GATA-1. We discovered that GATA-1 can directly interact with HDAC1 in a FOG-1 independent manner. The interaction results in deacetylation of GATA-1. We have identified two arginine sites within GATA-1 that are important for the interaction with HDAC1. The arginine to alanine mutation on these sites (2RA) blocks the interaction of HDAC1, but doesn't affect its DNA binding in vitro. Importantly, the mutation does not affect the interaction with FOG-1, indicating that GATA-1 direct interaction and indirect association via FOG-1 with HDAC1 are separate events. To further understand the role of the HDAC1-GATA-1 direct interaction in gene transcription and erythropoiesis, we introduced the 2RA mutant of GATA-1 fused with estrogen receptor ligand binding domains into G1E cells, a GATA-1-null erythroid progenitor cell line. Interestingly, upon estradiol induction, GATA-1 2RA does not promote b-globin gene transcription and erythroid differentiation of G1E cells, although GATA-1 2RA is highly acetylated. Chromatin immunoprecipitation assay (ChIP) shows that GATA-1 2RA binds poorly to HS3 and β-globin promoter. Similar binding defect is also detected on GATA-1 promoter, indicating defective GATA-1 recruitment on chromatin. Interestingly, HDAC1 binding to these regions are also significantly reduced, suggesting HDAC1-GATA-1 interaction may be important for GATA-1 deacetylation, as well as stabilizing GATA-1 binding. In order to investigate the effect of GATA-1 2RA in vivo, we generated a GATA-1 2RA knock in mice. The knock in mice are viable but suffered from anemia and thrombocytopenia. β-globin expression reduced at least 50% in knock in mice compare to wild type litter mate. To further identify and compare gene expression profiles regulated by HDAC1 direct or indirect associated GATA-1 during erythroid differentiation, we performed RNA sequencing assays to study the effects of GATA-1 2RA in gene expression in comparison with wild type GATA-1 or GATA-1-V205M (a mutation abolished binding with FOG-1). Expression of GATA-1 2RA largely affects gene expression profile in both GATA-1 activated and repressed genes compared to cells expressing wild type GATA-1. The gene expression pattern in 2RA cells also is largely different from cells expressing GATA-1 V205M, indicating the direct and indirect interaction with HDAC1 may mediate differential functions. Our results indicate that HDAC1 is required for GATA-1 recruitment and GATA-1 mediated transcription regulation. Thus, this study unveils a novel regulation of GATA-1 by its direct interaction with HDAC1.