Mutations in TET2 and WT1 accelerate the development of CEBPA-mutant acute myeloid leukemia
Bi-allelic (bi) mutations in the myeloid transcription factor CEBPA are present in approximately 5-10% of patients with acute myeloid leukemia (AML). These mutations often co-occur with loss-of-function (LOF) mutations in TET2, WT1 and GATA2. Co-mutations with TET2 and WT1 mutations are associated with a worse overall survival, whereas co-mutations with GATA2 tend to improve overall survival rates.
To study the underlying molecular mechanisms, we used CRISPR/Cas9 gene editing to model these mutations in primary human bone marrow (BM) hematopoietic stem and progenitor cells (HSPCs).
Targeting the N-terminal region of CEBPA using Cas9 ribonucleoproteins resulted in the loss of the full-length CEBPA p42 protein, with marked upregulation of the shorter p30 isoform, similar to what is observed in AML patients. CEBPAbi mutations resulted in loss of monocytic colonies, increased serial replating capacity in CFU assays, and a pre-leukemic phenotype in a mouse stromal-5 co-culture assay. Introducing LOF mutations in TET2 or WT1, but not GATA2, simultaneously with CEBPAbi mutations, resulted in leukemic transformation and uncontrolled cell proliferation in vitro. Transplantation of CEBPAbi-mutated HSPCs into immunodeficient mice expressing human cytokines (NSGS mice) revealed a myeloid skewing, and when combined with TET2 or WT1 LOF mutations, leukemic transformation. Interestingly, while CEBPAbi- and co-mutated HSPCs engrafted in the BM of NSG mice, we did not observe dissemination to the peripheral blood and no leukemic transformation in these mice. Subsequent cytokine dropout experiments revealed a dependency on human stem cell factor in CEBPAbi leukemia development. Proteomic analysis of HSPCs revealed that expression differences were mainly driven by CEBPA p30, which were further amplified by loss of TET2 or WT1. ATAC-seq analysis on HSPCs showed distinct chromatin changes following the loss of TET2 and WT1, potentially enabling increased transcriptional activity of CEBPA p30 and promoting leukemic transformation.
Current efforts are focused on refining these analyses and validate our findings at specific loci. Overall, by introducing somatic mutations in healthy human BM HSPCs, we have been able to decipher the molecular consequences of single and combined leukemic mutations in a human model of CEBPA-mutant AML.