The Role of Gata2 in Murine Models of Acute Myeloid Leukemia

The importance of stem cell and self-renewal programs in Acute Myeloid Leukemia (AML) is generally accepted, but the molecular details are incompletely understood. The master transcriptional regulator GATA2 is highly expressed in hematopoietic stem cells (HSCs) and has critically important roles in the hematopoietic system. Gata2 is required for murine HSC development and maintenance, and heterozygous loss of Gata2 compromises murine HSC- and progenitor cell-function. High levels of GATA2-expression have been correlated with adverse prognosis in human AML. GATA2 is also overexpressed in human chronic myeloid leukemia. These data suggest an important role for GATA2 in normal stem cells and in leukemia. However, genetic lesions resulting in compromised GATA2 function can lead to MDS and in some cases AML. In a murine AML model driven by Flt3-ITD and inactivation of Tet2, Gata2 is strongly downregulated. Furthermore, mouse models of leukemia suggest that high-level forced expression of Gata2 can have a tumor suppressor role. To clarify the role of Gata2in AML we used homozygous genetic inactivation in established murine models of leukemia, using a a conditional allele.

We initially tested the role of Gata2 in a murine leukemia mediated by forced expression of Meningioma1 (MN1). This model has a HoxA9/Meis1 transcriptional program. We recently found that MN1-driven leukemia depends on the histone methyltransferase Dot1l (J Clin Invest. 2016 Feb 29. pii: 80825). Lineage marker negative (Lin-), Sca1+, Kit-positive (LSK) bone marrow cells from mice with a floxed exon 5 in the Gata2 gene, and a ROSA26-YFP Cre-reporter allele were transduced with an MSCV-based ecotropic retroviral vector expressing MN1 and linked via an internal ribosomal entry site (IRES) the selectable marker GFP. Floxed Gata2-sequences were excised using transduction with a self-excising Cre-expressing vector (HR-Cre). Cells were sorted and plated in methylcellulose. The GFP/YFP double positive Gata2^ko cells showed a replating defect compared to GFP single positive Gata2-floxed cells, both with regard to colony number and colony size. Next, we tested the role of Gata2 in disease maintenance in vivo. We established MN1 Gata2^ff leukemias in primary recipients. Primary leukemias were transduced with Cre-expressing vector and Gata2^ko and Gata2^ff MN1 cells were transplanted. While all mice in the Gata2^ffgroup developed leukemia with a median survival of 35 days, the mice in the Gata2^ko cohort developed leukemia with incomplete penetrance with a latency of 249 days (p=0.0005). These data suggest an important role for Gata2in MN1 leukemia in vitro and in vivo. Genetic inactivation of Gata2 resulted in increased protein levels of p53 in vitro as detected by Western blot. Furthermore, MN1-transduced cells showed accentuated p53 stabilization and apoptosis in response to the Mdm2-antagonist and p53 stabilizer Nutlin3.

We next tested the role of Gata2 in leukemia driven by the oncogenic fusion MLL-AF9. In contrast to the MN1 model, recipients of Gata2^koleukemias only showed a trend towards prolonged latency in secondary recipients (median survival Gata2^ff=48 days vs. Gata2^ko=62 days, p=0.09). In this model, we also did not observe a substantial effect of Gata2-inactivation on p53 activation. We are currently characterizing the underlying molecular mechanisms.

Our data document an important role for Gata2 in AML mediated by MN1 and to a lesser degree, MLL-AF9. The role of Gata2 in leukemia is complex and depends on expression levels and cellular context. A more detailed understanding of leukemic self-renewal, including the role of Gata2, will inform the development of more efficacious and less toxic therapies for this difficult-to-treat malignancy.