Creation and Analysis of Two Mouse Models Revealing the Role of Gata2 as A Tumor Suppressor

Introduction: The hematopoietic transcription factor GATA2, is critical for the generation and maintenance of the hematopoietic stem cells pool. In human, germline heterozygous loss of GATA2, resulting in an inevitable development of MDS/AML (myelodysplastic syndrome/acute myeloid leukemia, respectively). Yet, how GATA2 heterozygosity promotes leukemia is elusive. Mechanistic studies are meager since GATA2 heterozygous animals do not generate malignancy. ERG, yet another pivotal hematopoietic transcription factor, is a direct GATA2 activator, highly expressed in poor prognosis AML. We have previously shown that transgenic hERG mice develop AML. In this study, we hypothesize that Gata2 tumor suppressive effect will manifest with an oncogenic pressure of hERG.

Materials and Method: To generate a GATA2 haploinsufficiency model, we crossed tgERG mice with 2 types of Gata2 mutated mice (Gata2+/- or Gata2+/L359V), assuming ERG/Gata2^mut compound mice will have an accelerated leukemogenesis comparing to tgERG/Gata2^wt. To study the pre-leukemic state in GATA2 haploinsufficiency, we tested for self-renewal, colony forming capacity and cell cycle components, in hematopoietic precursors of i. 4-6 week old mice bone marrow; ii. 14.5E embryos, fetal liver cells. To identify putative transcription pathways that could link the partial loss of Gata2 to leukemogenesis, we performed RNA sequencing and analysis in pre-leukemia and leukemia states.

Results and discussion: When tgERG mice were crossed with Gata2 mutated mice, 50% shortening in latency time to leukemia, and steep decreased in survival, were observed in both models (tgERG/Gata2^+/-and tgERG/Gata2^+/L359V) comparing to tgERG carrying WT-Gata2 (p<0.001). Immunophenotype analysis, however, revealed similar early myeloid markers of leukemia cells within the 3 groups, hence, cd150++, ckit+, sca1-, Fcgr -, Il7r -. In pre-leukemia stages –a hyper-proliferative state was demonstrated in tgERG/Gata2^+/-cells compared to tgERG: This was shown both by significant advantage in colony forming capacity (p=0.04 one-way ANOVA), as well as by a higher ratio of cells found in G₁/S, comparing to tgERG cells that were mostly quiescent. RNAseq analyses showed an enrichment of energetic /metabolic pathways, particularly a shift towards mitochondrial/ oxidative phosphorylation in ERG/Gata2^+/-, in pre-leukemia and in leukemia stages. A higher expression level of CD48 and CD117, both progenitor markers, was shown in pre-leukemia ERG/Gata2+/- cells, both in RNAseq and flow cytometry.

Conclusion: Our observations imply that partial loss of GATA2 in an early developmental stage, may redistribute subpopulations of the hematopoietic compartment. This, in turn, results in an expansion of a discrete subpopulation, that has unique metabolic features, and is more amenable to leukemogenic transformation.