The transcription factor CCAAT-enhancer-binding protein alpha (C/EBPα) is a master regulator of granulopoiesis and regulates the switch between proliferating, uncommitted progenitors and cell-cycle-arrested, differentiated myeloid cells. Usage of two alternative translation initiation sites in the CEBPA mRNA results in expression of a full-length C/EBPα protein p42 (42 kDa) and a shorter p30 isoform (30 kDa). CEBPA mutations are found in 9–15% of Acute Myeloid Leukemia (AML) patients. N-terminal frameshift mutations in the CEBPA gene lead to selective ablation of p42 expression, while C-terminal mutations disrupt the dimerization and DNA-binding ability of C/EBPα. AML patients harbor either mono- or biallelic CEBPA mutations (CEBPAmo or CEBPAbi) and both genotypes are frequently associated with concurrent mutations in other genes. The most commonly co-occurring mutations in both groups are loss-of-function mutations in the methylcytosine dioxygenase TET2 (44.4% in CEBPAmo / 34.8% in CEBPAbi). While CEBPA mutations typically confer good prognosis, we found that the presence of mutated TET2 resulted in significantly reduced overall survival. Thus, we hypothesize that combinatorial effects of CEBPA mutations with TET2 loss specifically rewire transcriptional and epigenetic circuitries in AML cells, thereby strongly influencing disease outcome.
We aim to elucidate the molecular mechanisms that underlie the cooperative effects of CEBPA and TET2 mutations by combining novel in vitro and in vivo AML models with integrated transcriptomic and epigenetic analyses.
Cebpa-mutated AML mouse models were used to generate cell line models for N-terminal CEBPA mutations (Cebpap30/p30) as well as mimicking biallelic CEBPA mutations (Cebpap30/C−mut.). We used the CRISPR-Cas9 technology to introduce Tet2 mutations into Cebpap30/p30 and Cebpap30/C−mut. cell lines to study the functional cooperation of these mutations. RNA- and ATAC-Seq analysis of >30 clones representing the spectrum of CEBPA and/or TET2 mutations yielded a comprehensive dataset for in-depth comparative analysis and correlation with patient data. In parallel, we establish fetal liver transplant models harboring Cebpa and Tet2 mutations to study their cooperative effect during leukemogenesis in vivo.
Introduction of Tet2 mutations into Cebpap30/p30 and Cebpap30/C−mut. cell lines conveyed a strong selective advantage on Tet2-targeted cells over Tet2-wild-type cells. ATAC-Seq analysis revealed that Tet2 loss caused a significant decrease in the accessibility of chromatin regions that are associated with cellular differentiation. Integration of these results with RNA-Seq data showed that Tet2 deletion led to a strong deregulation of genes involved in RNA binding and splicing. Furthermore, Cebpa-Tet2 co-mutation induced significant down-regulation of components of the p53 pathway (including Rprm and Zmat3).
Our results from these novel models provide deeper insights into the global epigenetic and transcriptomic changes that depend on CEBPA and TET2 aberrations in a physiologically relevant mutational context. The data imply that loss of TET2 reinforces the differentiation block imposed by CEBPA mutations in AML. In addition, TET2-deficiency interferes with components of the p53 pathway, thereby possibly compromising its activity. Integration of these findings with in vivo models and available patient data will enhance our understanding of gene cooperativity in AML and will provide entry points for the development of novel patient management strategies for patients with CEBPA and TET2 mutations.