C/EBPα and MiR-182 Generate a Negative Feedback Loop Which Is Dysregulated in Acute Myeloid Leukemia

The transcription factor CCAAT enhancer binding protein alpha (C/EBPα) is a master regulator of granulopoiesis and is silenced in approximately 50% of all acute myeloid leukemia (AML) cases. There are several mechanisms known how C/EBPα is inactivated in AML, including promoter hypermethylation, posttranslational modifications and mutations in the ORF of the CEBPA gene. MicroRNAs, a class of small non-coding RNAs, were identified as important regulators of normal hematopoiesis and leukemia development. We have already shown that microRNAs, such as miR-223, miR-34a and miR-30c, are essential elements in C/EBPα triggered granulocytic differentiation. But to our knowledge nothing is known about inactivation of C/EBPα by microRNAs in acute myeloid leukemia.

In this study, we identified a novel network between C/EBPα and miR-182. In a next generation sequencing approach based on inducible K562-C/EBPα-ER cell line, we found miR-182 strongly downregulated by wildtype C/EBPα. We could further demonstrate an inverse correlation between C/EBPα protein amount and miR-182 expression level in several in vitro systems, including leukemic cell lines and G-CSF treated primary human CD34⁺progenitor cells. Additionally, C/EBPα and miR-182 showed reciprocal expression in sorted murine bone marrow subpopulations in vivo. To discover the mechanism how miR-182 is blocked by C/EBPα, we analyzed the minimal promoter region of miR-182 and performed chromatin immunoprecipitation (ChIP). Here, we could demonstrate a strong binding of C/EBPα to the miR-182 promoter, particularly to a conserved E2F binding site. Because E2F is a well known inhibitor of C/EBPα function, we tested whether E2F also effects miR-182 expression. An overexpression of E2F1 in U937 cells leads to an elevated miR-182 expression level. In addition, we measured the expression of miR-182 in bone marrow from AML patients regarding to their CEBPA mutation status. We could show that only patients with mutations in the C-terminal region of C/EBPα showed elevated miR-182 expression, while patients with N-terminal CEBPA mutations revealed no abnormal miR-182 expression compared to healthy donors or AML patients with no CEBPA mutation. The C-terminal domain of C/EBPα is necessary for E2F inhibition. These findings illustrate the importance of C/EBPα-E2F interaction during miR-182 regulation. Next, we found a highly conserved binding site of miR-182 in the 3’UTR of CEBPA itself, suggesting a possible negative feedback loop. To test this, we performed overexpression of miR-182 in U937 cells, umbilical cord blood mononuclear cells (UCB-MNCs) and primary blasts from AML patients. Here, we observed a strong reduction of C/EBPα protein after miR-182 overexpression in all cell types. Furthermore, we could demonstrate a direct binding of miR-182 to the 3’UTR of CEBPA via luciferase activity assay. Finally, we were interested in the functional impact of miR-182 in myeloid differentiation and leukemia development. We showed that enforced expression of miR-182 in U937 cells reduced the percentage of Mac-1 positive myeloid cells after treatment with all-trans retinoid acid (ATRA). Additionally, lentiviral overexpression of miR-182 induces a block of differentiation and hyperproliferation in G-CSF treated 32D cells and an enhanced replating capacity of primary mouse bone marrow mononuclear cells.

Taken together, we identified miR-182 as novel oncogenic microRNA that directly blocks C/EBPα during myeloid differentiation and leukemia development. Thus, our data display a potential new strategy for therapeutics in C/EBPα dysregulated AML.