The AML1/ETO(MTG8) and AML1/Evi-1 Leukemia-Associated Chimeric Oncoproteins Accumulate PEBP2β(CBFβ) in the Nucleus More Efficiently Than Wild-Type AML1

Decreased intranuclear mobility of acute myeloid leukemia 1-containing fusion proteins is accompanied by reduced mobility and compartmentalization of core binding factor beta

Acute myeloid leukemia 1 (AML1) gene on chromosome 21 is involved in several chromosomal translocations, including t(8;21) and t(16;21), that produce chimeric fusion proteins AML1-eight twenty-one (ETO) and AML-myeloid transforming gene chromosome 16 (MTG16), which contribute to leukemogenesis. The molecular basis for the leukemogenic effects of these fusion proteins is incompletely understood. Using gel-shift assay, we showed that AML1-ETO and AML1-MTG16 bound to a series of AML1 consensus DNA-binding sites with different affinities. Using fluorescence recovery after photobleaching (FRAP), we demonstrated that a fusion of AML1 with ETO or MTG16 exhibits reduced intranuclear mobility compared with wild-type AML1 or either fusion partner. The dimerization domain (nervy homology region 2) of ETO is responsible for the reduced mobility of AML1-ETO. Dual FRAP studies revealed that CBFbeta colocalized with AML1-ETO within the nucleus, resulting in reduced mobility of CBFbeta. Therefore, AML1 fusion proteins may interfere with normal AML1 function due to aberrant nuclear dynamics, which leads to spatial and temporal sequestration of CBFbeta and perhaps other coregulators critical for myeloid differentiation.

Molecular mechanisms of leukemogenesis by AML1/EVI-1

The AML1/EVI-1 chimeric gene is generated by the t(3;21)(q26;q22) translocation and plays a pivotal role in progression of hematopoietic stem cell malignancies such as chronic myelocytic leukemia and myelodysplastic syndrome. In AML1/EVI-1, an N-terminal half of AML1 including a runt homology domain is fused to the entire zinc-finger EVI-1 protein. AML1 is essential for hematopoietic cell development in fetal liver and its lineage-specific differentiation in adult. In contrast, EVI-1 is barely expressed in normal hematopoietic cells, but it is overexpressed in chronic myelocytic leukemia in blastic crisis and myelodysplastic syndrome-derived leukemia. There are at least four mechanisms identified in AML1/EVI-1 fusion protein that possibly lead into malignant transformation of hematopoietic stem cells. Firstly, AML1/EVI-1 exerts dominant-negative effects over AML1-induced transcriptional activation. Although target genes repressed by AML1/EVI-1 are still not known, binding competition to a specific DNA sequence and histone deacetylase recruitment through a co-repressor CtBP in EVI-1 part are conceivable underlying mechanisms for the dominant-negative effects. Secondly, AML1/EVI-1 interferes with TGF beta signaling and antagonizes the growth-inhibitory effects of TGF beta. The first zinc-finger domain of EVI-1 associates with Smad3, a TGF beta signal transducer, and represses its transcriptional activity by recruiting histone deacetylase through CtBP that interacts with EVI-1. Thirdly, AML1/EVI-1 blocks JNK activity and prevents stress-induced apoptosis. AML1/EVI-1 associates with JNK through the first zinc-finger domain of EVI-1 and disturbs the association between JNK and its substrates. Lastly, AML1/EVI-1 enhances AP-1 activity by activating the c-Fos promoter depending on the second zinc-finger domain of EVI-1, and promotes cell proliferation. All these functions cooperatively contribute to the malignant transformation of the hematopoietic stem cells by AML1/EVI-1.

The t(3;21) fusion product, AML1/Evi-1 blocks AML1-induced transactivation by recruiting CtBP

AML1/Evi-1 is a chimeric protein that is derived from t(3;21), found in blastic transformation of chronic myelogenous leukemia. It is composed of the N-terminal AML1 portion with the DNA-binding Runt domain and the C-terminal Evi-1 portion. It has been shown to dominantly repress AML1-induced transactivation. The mechanism for it has been mainly attributed to competition with AML1 for the DNA-binding and for the interaction with PEBP2beta (CBFbeta), a partner protein which heterodimerizes with AML1. It was recently found that Evi-1 interacts with C-terminal binding protein (CtBP) to repress TGFbeta-induced transactivation. Here, we demonstrate that AML1/Evi-1 interacts with CtBP in SKH1 cells, a leukemic cell line which endogenously overexpresses AML1/Evi-1 and that AML1/Evi-1 requires the interaction with CtBP to repress AML1-induced transactivation. The association with CtBP is also required when AML1/Evi-1 blocks myeloid differentiation of 32Dcl3 cells induced by granulocyte colony-stimulating factor. Taken together, it is suggested that one of the mechanisms for AML1/Evi-1-associated leukemogenesis should be an aberrant recruitment of a corepressor complex by the chimeric protein.

Nuclear import and subnuclear localization of the proto-oncoprotein ETO (MTG8)

ETO (MTG8) was first described due to its involvement in the (8;21) translocation frequently observed in acute myeloid leukemias. In the t(8;21) the AML1 gene on chromosome 21 is fused to ETO on chromosome 8. The resultant hybrid protein is comprised of the DNA binding domain of AML-1 and the majority of ETO. This study examines the subnuclear distributions of ETO, AML-1B and AML-1/ETO proteins fused to green fluorescence protein in living cells using fluorescence microscopy. Further, we identified a 40 amino acid portion of ETO (amino acids 241-280) that was sufficient to cause nuclear import of green fluorescent protein. Mutational analysis demonstrated that lysine 265 and/or arginine 266 were required for nuclear import of ETO, but that the surrounding basic residues were not critical. ETO interacted with the nuclear import proteins importin-alpha and beta in vitro, and mutations in ETO that abolish nuclear localization also abolished the in vitro interaction with importin-alpha and beta. These data suggest that ETO enters the nucleus via an importin-mediated pathway. Additionally, ETO and AML-1/ETO co-localized to punctate nuclear bodies distinct from those containing promyelocytic leukemia protein. Nuclear body formation was dependent upon a region of ETO N-terminal to the nuclear localization signal. Thus, ETO and AML-1/ETO reside in potentially novel subnuclear compartments.