Myeloid leukemic stem cells are maintained by programs that drive self-renewal and block myeloid differentiation through both genetic and epigenetic mechanisms. Previously, we found the chromatin remodeler IKZF2 as a target of RNA binding protein MSI2 which is a central regulator of translation in stem cell programs. In contrast to being commonly deleted in hypodiploid B-cell Acute Lymphoblastic Leukemia and acting as a tumor suppressor, we propose that IKZF2 is required for myeloid leukemia. Although IKZF2 is highly expressed in hematopoietic stem cells (HSC), we found that it is dispensable for HSC function utilizing IKZF2 deficient mice. IKZF2 is also highly expressed in leukemic stem cells (LSCs) in a murine MLL-AF9 model. Conditional deletion of Ikzf2 in the hematopoietic system with Vav-Cre system, significantly impaired LSC function as assessed through limiting dilution assays (LSC frequency is 1:7,697 in Ikzf2-deficient cells versus 1:122 cells in wildtype cells) and serial transplantations. IKZF2 deletion with a tamoxifen inducible Cre (Cre-ER) in established leukemias resulted in reduced colony formation, increased differentiation and apoptosis while delaying leukemogenesis. Furthermore, shRNA depletion of IKZF2 in another murine AML model using the oncogene AML1-Eto9a also showed reduced colony formation and delayed leukemogenesis, suggesting that IKZF2 is required for myeloid leukemia.

Similar to the mouse HSCs, shRNA depletion of IKZF2 in human CD34+ enriched cord blood HSPCs resulted in no overt phenotype in colony formation, differentiation and apoptosis. Intracellular flow cytometry for IKZF2 revealed that IKZF2 is highly expressed in the CD34+CD38- fraction compared to the CD34- fraction in nine AML patients. Notably, IKZF2 depletion with shRNAs resulted in reduced frequency of CD34+CD38- fraction and reduced colony formation in AML patient samples. Depletion of IKZF2 in five human AML cell lines (MOLM-13, KCL-22, KASUMI-1, NOMO-1 and NB-4) with different oncogenes also resulted in reduced proliferation, increased differentiation and increased apoptosis. These data suggest that IKZF2 is differentially required in myeloid leukemia cells compared to normal cells.

Mechanistically, ATAC-sequencing (assay for transposase-accessible chromatin with sequencing) in MLL-AF9 LSCs revealed that a substantial portion of the decreased accessibility changes occur in the intronic regions (34.65% for open peaks compared with 45.95% for closed peaks) whereas more promoter regions are opened than closed (21.26% for open peaks; 12.77% for closed peaks) when IKZF2 is lost. This suggests that IKZF2 loss leads to reduced accessibility preferentially occurring in intronic enhancers whereas increased accessibility was found at promoters. Motif enrichment analysis from the combinatorial assessment of RNA-sequencing, chromatin accessibility by ATAC-seq and direct binding of IKZF2 by the cut and run method in MLL-AF9 LSCs identified the C/EBPδ and C/EBPε as the most accessible motifs whereas HOXA9 motif became less accessible in the Ikzf2 deleted LSCs. More specifically, we found 13 genes bound by IKZF2 that contained C/EBP motifs that had also increased accessibility (Log2FC>1, pval<0.05) and increased gene expression (Log2FC>0.75, pval<0.05) in Ikzf2 deleted MLL-AF9 LSCs. Using the cre-ER expressing MLL-AF9 LSCs, we validated that C3, Fpr2, S100a8 and S100a9 were upregulated after acute deletion. These direct targets and CEBP expression correlated with IKZF2 expression in the TCGA AML patient cohort. Furthermore, forced HOXA9 expression could partially rescue the colony formation, differentiation and apoptosis effects after Ikzf2 was deleted by tamoxifen treatment. Additionally, CEBPE depletion by shRNAs partially rescued the effects of IKZF2 deletion. Thus, we demonstrate that IKZF2 is dispensable for normal hematopoiesis but required for maintaining LSC function. We find that IKZF2 can act as a chromatin remodeler that regulates the self-renewal HOXA9 gene expression program and inhibits C/EBP driven differentiation program in LSCs. Our study provides the rationale to therapeutically target IKZF2 in myeloid leukemia.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution