Acute myeloid leukemia (AML) is a clonal malignant disorder originating from a small number of leukemic stem cells (LSCs). AML relapse after conventional chemotherapy is caused by a remaining population of drug-resistant LSCs. Selective targeting of LSCs is a promising strategy for the prevention and treatment of AML relapse. Polycomb repressive complexes 1 (PRC1) and 2 (PRC2) are important epigenetic regulators that maintain the stemeness of ES cells and hematopoietic stem cells. Enhancer of zeste homolog 1 and 2 (EZH1/2) is a catalytic component of PRC2 that trimethylates histone H3 at lysine 27 (H3K27) to repress the transcription of target genes. Mutations and overexpression of EZH1/2 are associated with cancers including hematopoietic malignancies. Here, we used genetic deletion of EZH1/2 or a novel dual inhibitor of EZH1/2 activity to show that loss or inhibition of EZH1/2 eradicates dormant AML stem cells.

To examine the effects of genetic deletion of EZH1/2 on AML cells, Ezh1-null, Ezh2-conditional and double knock-out mice were generated. Hematopoietic stem/progenitor cells prepared from single knock-out mice or double knock-out mice were transduced with various types of AML fusion-genes, such as MOZ-TIF2, MLL fusions, AML1-ETO, and others by retroviral infection, and cultured in vitro or transplanted into irradiated recipient mice to induce AML in vivo. When the cells were cultured in vitro, double deletion of Ezh1/2 induced cell differentiation and apoptosis more severely than single deletions in all subtypes of AML tested, resulting in complete loss of cells. In AML mice, deletion of Ezh1/2 induced AML cell differentiation and complete remission of AML, which was not achieved by single deletion of Ezh1 or Ezh2. Genetic deletion of both Ezh1 and Ezh2 on the LSC fraction dramatically reduced the number of LSCs (Lin-c-Kit+ CD16/32+ CD34+, L-GMP), especially quiescent LSCs, whereas deletion of either Ezh1 or Ezh2 did not have such a strong effect. The transcriptional profiles of LSCs deficient in both Ezh1 and Ezh2 were characterized by the upregulation of cell cycle-related genes such as Cyclin D1/D2, which is the main regulator of G0/G1 transition, along with differentiation-related genes. These results suggested that deletion of both Ezh1 and Ezh2 is required for eradication of LSCs.

To investigate whether pharmacologic inhibition of EZH1/2 could serve as a therapeutic strategy in AML, we developed a novel EZH1/2 dual inhibitor with potent inhibitory activity against both EZH1 and EZH2. The drug induced cell differentiation and apoptosis in most subtypes of AML tested in vitro and its effects were similar to those of genetic depletion of EZH1/2. A selective EZH2 inhibitor did not affect the growth and survival of AML cells to the same extent as the dual inhibitor. Oral administration of the EZH1/2 dual inhibitor reduced the number of LSCs effectively in AML mice in a manner similar to the effect of genetic deletion of EZH1/2.

Taken together, these results strongly suggest that dual inhibition of EZH1 and EZH2 is a promising therapeutic strategy to eradicate LSCs in a wide range of AMLs, which could lead to important advances in the treatment of AML.


Honma:Daiichi Sankyo Co., Ltd: Employment. Adachi:Daiichi Sankyo Co., Ltd: Employment. Araki:Daiichi Sankyo Co., Ltd.: Employment. Kitabayashi:Daiichi Sankyo Co., Ltd.: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution