The role of GATA1 acetylation during erythroid development

Post-translational modifications such as acetylation and phosphorylation play key roles in controlling protein function. The hematopoietic transcription factor GATA1 is acetylated at several lysine residues near its zinc finger domains. To understand the functional importance GATA1 acetylation, we examined the activity of an acetylation-defective mutant of GATA1 in maturing erythroid cells. We found that removal of the acetylation sites in GATA1 does not impair its nuclear localization, steady-state protein levels, or its ability to bind naked GATA elements in vitro. However, chromatin immunoprecipitation (ChIP) experiments revealed that mutant GATA1 was dramatically impaired in binding to all examined cellular target sites in vivo, including genes that are normally activated and repressed by GATA1. These results suggest that acetylation regulates chromatin occupancy of GATA1 and point to a novel function for transcription factor acetylation, perhaps by facilitating protein interactions required for stable association with chromatin templates in vivo. To test this hypothesis, we performed a peptide affinity screen and identified Brd3 as an acetylated GATA1 interacting partner. Brd3 belongs to the BET protein family and is characterized by tandem bromodomains (BD1 and BD2) and an extraterminal (ET) domain. We show that Brd3 and GATA1 physically interact in an acetylation-dependent manner in vitro and in vivo. Mapping studies revealed that the interaction depends on BD1 of Brd3 and one of the two major acetylation sites that resides near the C terminal zinc finger of GATA1. By ChIP-seq and ChIP-qPCR, endogenous Brd3 is recruited to virtually all GATA1-occupied regulatory elements in erythroid cells, including both GATA1 activated and repressed genes. In agreement with our biochemical data, an intact BD1 is essential for the in vivo recruitment of Brd3 to GATA1-occupied elements, further demonstrating that acetylation of GATA1 is essential for Brd3 association in vivo. Notably, a pharmacological compound that targets acetyl lysine binding sites in BD1 and BD2 disrupts the Brd3/GATA1 interaction in vitro, diminishes Brd3 and GATA1 association at key erythroid genes in vivo, and impairs GATA1 target gene expression and erythroid maturation. In concert, these findings suggest a mechanism by which the first bromodomain of Brd3 “reads” acetyl-lysines on GATA1 to facilitate GATA1 chromatin occupancy