JMJD1C Interacts with the Ncor Corepressor Complex and Is Required for Human MLL-AF9 Leukemogenesis

MLL rearranged leukemia (MLLr) is a subtype of Acute Myeloid Leukemia (AML) in which the N-terminus of the Mixed Lineage Leukemia (MLL1) gene fuses with one of over 60 fusion partner genes. The most common fusion in AML is MLL-AF9. MLLr leukemia has an intermediate to poor prognosis.

Previous work has identified JMJD1C, a jumonji containing lysine demethylase, as a target of MLL-AF9 fusion oncoprotein, and studies in mice have shown JMJD1C is required for MLL-AF9 and HOXA9/MEIS1 leukemogenesis. Further, our lab utilized CRISPR(clustered regularly interspaced short palindromic repeats))/Cas9(CRISPR-associated protein-9 nuclease) to show the Jumonji (JmjD) and Zinc finger domains (ZFD) of JMJD1C are important for its function in mouse MLL-AF9 leukemia.

While the role of JMJD1C in mouse MLL-AF9 leukemia is well documented, its role in human disease is incompletely understood. We studied the role of JMJD1C in human AML using human CD34+ cord blood derived MLL-AF9 (CD34+ MA9) model. We transfected CD34+ MA9 cells with a combination of three sgRNAs targeting genomic regions of JMJD1C, or sgRNA against Renilla as control to knockout JMJD1C and assessed the effect on leukemia cells in vitro. We observed significant increases in apoptosis and levels of mature markers CD13 and CD14 in JMJD1C knockout versus control cells by flow cytometry. Experiments are ongoing to assess knockout of JMJD1C on CD34+ MA9 leukemogenesis in vivo.

To study the specific requirement of JmjD and ZFD domains in human leukemia, we transduced CD34+ MA9 leukemia cells with single sgRNAs targeting the JmjD or ZFD. We performed competitive transplants and found significant depletion of cells harboring JmjD and ZFD sgRNAs compared to controls, suggesting they are functionally important for human MLL-AF9 leukemogenesis in vivo. Finally, we performed RNA-seq on CD34+ MA9 cells transduced with sgRNA against JmjD or ZFD domain, which showed positive enrichment of apoptotic pathways and significant negative enrichment for human MLLr leukemia, AML, and Hematopoietic Stem Cell pathways in ZFD and JmjD mutated cells.

To further understand the mechanism behind the requirement of JMJD1C in human AML, we sought to identify novel interacting proteins of JMJD1C by using BioID, an unbiased proximity ligation approach. Thirteen unique proteins were identified by Mass Spectrometry as JMJD1C interacting proteins including three members of the nuclear co-repressor complex: NCOR1, TBL1X, and TBL1XR1. NCOR1 has been shown to be required for MLL-AF9 leukemogenesis and, together with NCOR complex member HDAC3, binds promoters of myeloid genes to regulate myeloid cell differentiation during leukemogenesis.

A phage display assay of JMJD1C was also conducted and confirmed an interaction between JMJD1C and NCOR1. We further examined the interaction in MLLr leukemia cell lines, MOLM13 and THP1, as well as, non MLLr AML cell lines, NB4 and Kasumi-1, which harbor oncogenic fusion PML-RARa and AML-ETO respectively to determine if the interaction is specific to MLLr or applies to other AML types. JMJD1C and components of the NCOR complex reciprocally co-immunoprecipitated in all four cell lines. Moreover, JMJD1C co-fractionated with the NCOR complex in size exclusion chromatography in all lines. Taken together, these results prove an interaction between JMJD1C and NCOR1 that is found in diverse AML cell lines.

<Overall, this study demonstrates JMJD1C is required for human AML MLL-AF9 leukemogenesis and uncovers an interaction between JMJD1C and the NCOR complex in human AML cells. Currently, we are studying the functional importance of this interaction.

No relevant conflicts of interest to declare.