Runt-related transcription factor (RUNX) transcription factors are essential regulators of diverse developmental processes. In mammals, there are three RUNX genes, RUNX1, RUNX2, and RUNX3. All RUNX proteins contain a highly conserved DNA-binding domain, called the runt-homology domain (RHD), which is responsible for DNA-binding and interaction with a partner, core binding factor subunit β (CBFβ). They regulate transcription of target genes, involving hematopoietic differentiation, cell cycle regulation, p53 pathways, and so on. From our previous studies, we assume that compensation mechanism is present among the RUNX family members. RUNX plays pivotal roles in leukemogenesis and inhibition of RUNX has now been widely recognized as a novel strategy in anti-leukemic therapies. However, common mechanism via RUNX in diverse acute myeloid leukemia (AML) remains elusive. Here, we demonstrate that targeting RUNX-nuclear factor of activated T cells 2 (NFATC2) axis is an effective strategy to suppress drug-resistant (DR)-acute promyelocytic leukemia (APL) cells. Silencing of RUNX and NFATC2 in DR-APL cells suppressed cell growth and induced apoptotic cell death. Next, by RNA-seq analysis of several AML patient cohorts, we confirmed that a strong positive correlation between RUNX family (RUNX1,2,3: Pan RUNX) and NFAT family (NFATC1,2,3,4, NFAT5: Pan NFAT) exists not only in APL but also in all hematopoietic malignancies and that AML forms the Pan RUNX high-Pan NFAT high expression cluster. Inspection of the NFATC1-3 promoter revealed the RUNX binding sequence, and direct transcriptionally regulation of NFATC1-3 by RUNX family was confirmed in both chromatin immunoprecipitation (ChIP)-seq analysis and dual luciferase reporter assay. We believe that RUNX-NFAT axis could be an important target in diverse AML.

Next, considering the well-established role of RUNX and NFATC2 in T cell immunity, we also apply targeting RUNX-NFATC2 strategy to suppress T cell activation and xenogeneic graft-versus-host disease (GVHD).The expansion of donor T cells requires IL-2, and aGVHD has been defined as a Th1-mediated disease. It is now well known that RUNX, especially RUNX1 and RUNX3 , are highly expressed in T cells, and directly regulate Th1 cytokine genes. As immunosuppressive approach for the prevention or treatment of aGVHD, calcineurin inhibitors, cyclosporine A and tacrolimus, inhibit GVHD by preventing the activation of NFAT, and steroid inhibits transcription of proinflammatory genes. We suppose that targeting RUNX can downregulate NFAT and also cytokine genes in T cell. RUNX1 knockdown and PanRUNX knockdown led to deceased NFATC2 and cytokine gene expression in cytokine-producing Jurkat cell line. It was also confirmed that by inhibiting the RUNX family and suppressing the NFATC2 family at the transcriptional level, the amount of the total NFATC family was significantly reduced compared with the drug that suppresses the nuclear translocation of NFATc2.The importance of RUNX-NFATC2 axis in T cell immunity was also exactly confirmed by the rescue experiments. Finally, to achieve "cluster regulation of RUNX (CROX)" strategy, we have been developing a novel RUNX inhibitor: chlorambucil-conjugated pyrrole-imidazole (PI) polyamides (Chb-M') that targets consensus RUNX-binding sequences, and specifically inhibits binding of RUNX family members. So, Chb-M' can switch off the RUNX target genes efficiently. In diverse AML including APL, core binding factor (CBF)-AML, mixed lineage leukemia (MLL)-rearranged AML, and AML-M0 and so on, Chb-M' was remarkably effective, and suppressed the expression of NFAT family in the protein level and induced apoptotic cell death. ChbM' also had a prominent effect in the AMLPDX model.The importance of RUNX-NFAT axis in AML was confirmed by the pharmacological rescue experiments using phorbol 12-myristate 13-acetate (PMA) and Ionomycin stimulation. Chb-M' also suppressed NFATC2 and cytokine gene expression in peripheral blood mononuclear cells (PBMC) and ameliorated GVHD for xenogeneic GVHD mouse model by transplanting human PBMC into immunodeficient mice. Taken together, we show RUNX could be a novel therapeutic target against diverse AML and GVHD through targeting RUNX-NFAT axis.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.