Abstract 179


MEF/ELF4 belongs to an ETS family of transcription factors that has roles in hematopoietic stem cells (HSCs): in a Mef-null mouse model, HSC increased residence in G0 status of the cell cycle. It was also reported that Mef/Elf4 promoted the transformation of fibroblasts by inhibiting two major tumor suppressor pathways, the p53 and p16/Rb pathways, inducing the expression of the Mdm2 gene. Our previous study on AML samples showed that the expression of MEF/ELF4 was significantly lower in cases with t(8;21) and t(15;17) compared with those with normal karyotype (AML-NK). However, little has been investigated about the role of MEF/ELF4 in AML-NK.


The aim of the present study is to elucidate how MEF/ELF4 works and is controlled in AML-NK.


We identified the associated protein with MEF/ELF4 using the Tandem affinity purification (TAP) method followed by MASS analysis. Transforming activity of MEF/ELF4 was tested by colony-formation assay using NIH3T3 cells. The transcriptional activity of MEF/ELF4 and its binding to the HDM2 promoter region, with or without wild type NPM1 (wtNPM1) and its mutants (mutNPM1), were examined using luciferase analysis, EMSA and ChIP assay. We also examined the expression of MEF/ELF4 and HDM2 in CD34-positive AML-NK cells obtained from 22 patients.


Nucleophosmin (NPM1) was included in the twenty-six proteins that were found in the TAP analysis, and the direct binding between MEF/ELF4 and NPM1 was confirmed by immunoprecipitation, GST pull down, and in vitro translation assays. Transforming activity of MEF/ELF4 was decreased 3-fold by the overexpression of wild-type NPM1, whereas the activity was enhanced by mutNPM1. Transcriptional activity of MEF/ELF4 measured using luciferase assay (159-fold by arbitral unit) was increased by the co-expression of mutNPM1 (315-fold), and decreased by wtNPM1 (109-fold) (Figure1). Forced expression of siRNA for NPM1 enhanced the luciferase activity of MEF/ELF4. These results indicated that the transactivating activity of MEF/ELF4 was enhanced by mutNPM1, and decreased by wtNPM1. EMSA and ChIP assay for the HDM2 promoter demonstrated that MEF/ELF4 was bound to the specific binding sites of the HDM2 gene. wtNPM1 weakened the binding of MEF/ELF4 to the promoter of the HDM2 gene in EMSA and ChIP assay. However, co-expression of mutNPM1 increased its binding to the HDM2 promoter in ChIP assay (Figure2). These data suggested that NPM1 regulated the binding of MEF/ELF4 to the HDM2 promoter, which influenced the activity of MEF/ELF4. In clinical samples, AML-NK cells with the high expression of MEF/ELF4 showed significantly higher expression of HDM2 than those with the low expression of MEF/ELF4 (P=0.009). In AML-NK cases with the mutated-NPM1 gene, the expression of HDM2 was higher than those with the wild-type NPM1 (P=0.03).


These data suggested that MEF/ELF4 activates the expression of the HDM2 gene under the influence of the mutational status of the NPM1 gene in AML-NK.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.