Functional Analysis of the GATA2 Promoter Shows That Mutations of GATA2 Impair Its Own Transcriptional Regulation

Abstract 1233

GATA2 encodes a transcription factor with essential functions in hematopoiesis. Somatic mutations of GATA2 have been reported in patients with chronic myeloid leukemia (CML) with blast crisis, and in bi-CEBPA-positive acute myeloid leukemia (AML); moreover, our group and others have shown that overexpression of this gene is a recurrent event associated with poor prognosis in AML. Several recent studies report mutations in the GATA2 gene in three different familial syndromes characterized by predisposition to myelodysplastic syndrome (MDS) and AML. Despite some differences, these mutations are very similar, and in some cases identical. This implies that individual mutations, although located in similar regions, may differentially affect GATA2 function. Therefore, additional research is required to explain why similar mutations lead to different syndromes (Hyde and Liu, 2011). On the other hand, it has been extensively studied in murine models that GATA2 activates its own transcription by binding to regions located at −2.8 and −1.8 kb from the transcription start site (TSS). We hypothesized that these reported GATA2 mutations could alter the GATA2 autoregulatory loop, affecting the transcription of GATA2. With this aim, we first aligned the murine and human GATA2 promoters in search for homologous GATA2 binding sites. Regions containing the cis-regulatory elements located at −2.8 and −1.8 kb from IS exon TSS in the murine promoter were highly homologous to two regions in the human promoter, with two putative GATA2 binding sites located at −3.4 and −2.4 kb from IS TSS, respectively. ChIP-qPCR assays showed that GATA2 binds to these sites in the human GATA2 promoter. To assess the ability of both wild-type and GATA2 mutations to regulate its own transcription, we transfected these GATA2 gene variants along with different GATA2 promoter constructs into HEK293T cells, and performed luciferase reporter assays. Wild-type GATA2 activated its transcription through the −2.4 kb site; however, it was not able to activate the full length promoter construct containing both the −3.4 and −2.4 sites. CEBPA binding sites near the −3.4 site could explain these results, since it has been reported that expression of GATA2 is transcriptionally repressed by CEBPA in a DNA binding-dependent manner. The T354M mutant activated GATA2 transcription in a similar manner than the GATA2 wild-type, raising the question about the complex function of T354M. On the contrary, del355T was totally incapable of sustain any activation of GATA2. Finally, the L359V mutation, present in 10% of CML cases with blast crisis, was able to activate the GATA2 promoter, even the full length promoter construct that contains both −3.4 and −2.4 sites, supporting that L359V is a gain-of-function mutation. In summary, GATA2 mutations had different effects on the GATA2 promoter that could affect the dose of GATA2. Expression of GATA2 is critical at various stages of hematopoiesis and since it in part determines the fate of distinct myeloid lineages, this could alter normal hematopoiesis. Moreover, as happened with GATA2, mutant GATA2 proteins could affect the expression of other targets of GATA2, as SCL, BMP4, PU.1, WT1 and others. Studies to further clarify these questions are in progress.