Epigenetic dysregulation of CEBPE expression in MLL-rearranged acute myeloid leukemia

Abstract

Mixed lineage leukemia (MLL)-rearranged acute myeloid leukemia (MLLr AML) is characterized by the chromosomal translocations between the MLL gene and its partner genes. The resultant MLL fusion proteins are capable of blocking cell differentiation at early hematopoietic progenitor stage through inducing aberrant epigenetic patterns at gene regulatory element. CEBPE, which is a member of CCAAT enhancer binding protein (CEBP) family transcription factor, plays an important role in terminal granulocytic differentiation. I noticed that Cebpe expression was markedly repressed in murine hematopoietic stem and progenitor cells expressing MLL-EEN owing to gain of DNA methylation at its promoter, which could be associated with differentiation blockage. Nonetheless, the biological consequence of CEBPE repression in MLLr AML and the detailed mechanism of how CEBPE is dysregulated in MLLr AML remains unclear. Gene expression analysis showed that CEBPE expression was significantly reduced in MLLr AML cell lines when compared to non-MLLr AML cell lines, suggesting that CEBPE repression is a common observation mediated by MLL fusion proteins. Importantly, 5´-aza-cytidine (5´-aza-C) treatment re-activated CEBPE expression in MLLr AML cells, further confirming the involvement of DNA methylation machinery in CEBPE dysregulation. It is observed that re-activation of CEBPE by combined treatment of 5´-aza-C and all-trans retinoic acid or direct forced expression of CEBPE can trigger MLLr AML cell differentiation. The above findings indicate that CEBPE repression contributes to cell differentiation blockage in MLLr AML. To investigate the epigenetic mechanism of CEBPE dysregulation in MLLr AML, I identified CEBPE enhancer in human AML cells. Low activity of CEBPE enhancer was detected in MLLr AML cells by luciferase reporter assay and H3K27ac ChIP experiments. Further evidence showed that the low enhancer activity was associated with the loss of binding of RUNX1 and MLL proteins and DNA hypermethylation. However, chromosomal conformation capture (3C) assay demonstrated the interaction between CEBPE promoter and enhancer in both MLLr and non-MLLr AML cells, suggesting the looping between CEBPE enhancer and promoter is stable and pre-existed before CEBPE expression or lineage maturation in AML cells. Interestingly, I found that CEBPE enhancer possessed transcriptional function to produce enhancer RNA (eRNA), which is predominantly localized in the nucleus of non-MLLr AML cells only. The positive correlation of CEBPE mRNA and eRNA was observed in both MLLr AML cell lines and patient samples. CRISPR interference assays demonstrated that transcriptional repression of CEBPE enhancer activity could lead to down-regulation of both CEBPE eRNA and mRNA. The Chromatin Isolation by RNA Purification (ChIRP) assay provided further evidence that CEBPE eRNA was involved in the CEBPE promoter-enhancer complex. Finally, I showed that removal of DNA methylation by 5´-aza-C treatment in MLLr AML cells re-activated CEBPE eRNA transcription, proposing a mechanistic link between aberrant DNA methylation and CEBPE regulation by MLL fusion proteins.

In summary, my study provides novel insight into the epigenetic mechanism of how the MLL fusion proteins impede cell differentiation through dysregulation of CEBPE enhancer function, which could serve as foundation for the development of novel therapeutic strategies to MLLr AML.