Functional Differences in GATA-1 Affinity for Double Versus Single GATA-1 Binding Sites Dictate Synergistic Versus Antagonistic Interactions of PU.1 and GATA-1 in Myeloid Gene Transcription.

Abstract Instructive roles for both GATA-1 and PU.1 have been demonstrated in hematopoiesis, and recent studies have identified both antagonistic and synergistic interactions between them in myeloid gene transcription and lineage development. In prior studies, we reported that PU.1 synergizes with rather than antagonizes GATA-1 for transactivation of a hallmark eosinophil gene, the major basic protein P2 promoter (MBP-P2), which possesses a novel dual (double) GATA-binding site, similar to the palindromic double site in the murine GATA-1 control locus that may specify eosinophil lineage-specific expression of GATA-1 and eosinophil development. To address the transcriptional mechanism for PU.1-GATA-1 synergy through the MBP-P2 dual GATA site, we investigated GATA-1 and PU.1 physical and functonal interactions via their binding sites in the MBP-P2 promoter. DNA binding affinities of GATA-1 and its C- versus N-terminal zinc fingers were assessed for single versus double GATA sites in the presence or absence of PU.1. Our results show that the dual GATA site strongly binds full length GATA-1 with higher affinity than either of the single sites, using both zinc fingers, but that mutant GATA-1 proteins with C-finger or N-finger deletions retain their ability to bind, albeit at lower affinity, to the dual site. DNA binding activities of the two zinc fingers with the dual GATA site were confirmed using peptides containing only the C-finger or N-finger region. Of note, formation of GATA-1 complexes with the dual GATA site was not inhibited by the addition of PU.1, whereas formation of binding complexes for mutants of GATA-1 containing only the C- or N-finger region could be completely inhibited in a dose-response fashion by PU.1. These unique features of PU.1/GATA-1 interactions on a dual versus single GATA-1 site were confirmed using peptides containing only the C- or N-finger regions of GATA-1. Our findings indicate that both zinc fingers of GATA-1 are involved in formation of the high-affinity GATA-1 complex with the dual site. Importantly, we show that the higher affinity dual GATA-1 site complex is not affected by the addition of PU.1, whereas formation of the binding complex with a single GATA-1 site is eliminated by PU.1, emphasizing the different mechanisms of GATA-1/PU.1 interactions on dual versus single GATA binding sites. Functional analyses by transactivation confirmed that synergistic activation of the MBP-P2 promoter by GATA-1 and PU.1 is mediated by their protein-protein interactions through this unique high affinity dual GATA-1 binding site. We suggest two possible mechanisms for PU.1/GATA-1 synergy on dual GATA sites: (1) PU.1 may change GATA-1 conformation and its high affinity for the dual site, enhancing its availability for interaction with the basal transcriptional machinery. Alternatively, (2) PU.1 could impede interactions of GATA-1 with a co-repressor, e.g. FOG-1, which we and others have shown represses GATA-1 function in the eosinophil lineage.