The Impacts of Transcriptional Regulatory Mechanism and Protein Interaction Relationship between RUNX1 and KLF4 on the Biological Functions of t (8;21) Leukemia Cells

Our previous study has reported that sodium phenylbutyrate (PB), one of the HDAC inhibitors, could induce t (8;21) leukemia cells to undergo differentiation and apoptosis. However, genes that promote differentiation and apoptosis in this process are not fully identified. Identification of these genes and elucidation of their interaction relationships are essential to develop novel therapy strategies for t (8;21) leukemia.

In this study, we found that the expressions of RUNX1, KLF4 and P57 at both mRNA and protein level were significantly up-regulated in Kasumi-1 cells after PB treatment. Bioinformatics analysis further demonstrated that there might be some transcriptional regulatory mechanism and protein interaction relationship among these three genes.

For the study of transcriptional regulatory mechanism, a KLF4 promoter/enhancer reporter plasmid containing RUNX1 binding sites and a P57 promoter/enhancer reporter plasmid containing KLF4 binding sites were constructed. Luciferase assay showed that RUNX1 exhibited strong transcriptional activation effect on KLF4 promoter/enhancer reporter plasmid in a dose-dependent manner, while RUNX1-ETO showed slight effect. KLF4 exhibited dose-dependent transcriptional activation effect on P57 promoter /enhancer reporter plasmid. These transcriptional regulatory relationships were further confirmed by Western blot assay. Transient expression of exogenous RUNX1 up-regulated the expression of KLF4 and P57 in 293T cells and Kasumi-1 cells, while RUNX1-ETO had no effect.

In the study of protein interaction relationship, co-immunoprecipitation (CO-IP) and immunofluorescence confocal imaging showed that RUNX1 interacted with KLF4 and co-localized with it in nucleus, and Runt homology domain (RHD) was responsible for this interaction. RUNX1-ETO competed with RUNX1 for binding to KLF4 in a dose-dependent manner through RHD and the ETO domains. PB has no effect on the interactions between RUNX1/RUNX1-ETO with KLF4. To further investigate the effects of RUNX1 and RUNX1-ETO on KLF4-dependent transactivation of target genes, a luciferase reporting plasmid containing KLF4 binding sites was constructed. Luciferase assay was then performed and showed that RUNX1 could coactivate KLF4 target genes while RUNX1-ETO had little effect.

Finally,RUNX1, KLF4 and P57 were overexpressed in Kasumi-1 cells by a pCDH lentivirus system, and their biological functions were investigated by MTS assay and flow cytometry. Overexpression of either of the three genes could inhibit cell proliferation, block cell cycle and induce apoptosis of t (8;21) leukemia cells. Moreover, in vivo experiment demonstrated that overexpression of KLF4 and P57 in mouse bone marrow leukemic cells could reduce the incidence of leukemia development in recipient mice.

In conclusion, we reported "RUNX1-KLF4-P57" as a novel signaling pathway in t (8;21) leukemia cells in this study. RUNX1 transactivated KLF4 and then P57 in leukemia cells to inhibit cell proliferation, block cell cycle and induce cell apoptosis, while oncoprotein RUNX1-ETO had little effect. Moreover, RUNX1 interacted with KLF4 and coactivated its target genes, which could be blocked by the competed binding of RUNX1-ETO with KLF4. Our study revealed the regulatory relationship between RUNX1 and KLF4 and the impacts of RUNX1-ETO on this relationship, which helps to elucidate the mechanism of t (8;21) leukemogenensis.