The AML1 gene: a transcription factor involved in the pathogenesis of myeloid and lymphoid leukemias

The AML1 gene was identified in 1991 by cloning the t(8;21) chromosome translocation associated with FAB M2 acute myeloid leukemia (AML). AML1 encodes a nuclear transcription factor (TF) which shows homology in its 5' part with the Drosophila melanogaster segmentation gene, runt, and contains a transactivation domain in the carboxyterminal portion. In the t(8;21), AML1 is fused to the ETO (MTG8) gene, resulting in a hybrid AML1/ETO mRNA, which in turn is translated into a chimeric protein. The objective of this article is to review here the main structural and biological features of AML1 and of its fusion products, with special focus on their clinical correlations and their potential usefulness for prognostic and monitoring studies in human leukemia.

The material examined in the present review includes articles and abstracts published in journals covered by the Science Citation Index and Medline.

The normal AML-1 protein forms the alpha-subunit of the heterodimeric TF core binding factor (or CBF), whose beta-subunit is encoded by the CBF beta gene on chromosome 16q22. CBF beta is rearranged and fused to the MYH11 gene in the AML M4Eo-associated inv(16) aberration. Thus, the two most common chromosome abnormalities of AML, i.e. t(8;21) and inv(16), affect the two subunits of the same target protein. This suggests that the wild type CBF must exert an important role in the control of myeloid cell growth and/or differentiation. Evidence that AML1 is a pivotal regulator of definitive hematopoiesis has been recently provided by analyzing AML1 knockout mice. The chromosome region 21q22, where AML1 maps, is involved in several other karyotypic aberrations, such as the t(3;21) translocation associated with a subset of therapy-related myelodysplastic syndromes and AML, and the blast phase of chronic myelogenous leukemia. In this abnormality, three distinct genes: EVI1, EAP, MDS1, located on chromosome band 3q26, have been identified that may recombine with AML1. Finally, the recently cloned t(12;21) translocation has been found to involve the TEL gene (coding for a novel TF) on 12p13, and AML1 on 21q22. This alteration, which results in the production of a TEL/AML1 chimeric protein, is restricted to pediatric B-lineage acute lymphoid leukemia (ALL), where it represents the most frequent molecular defect known to date (up to 25% of cases). Strikingly, the same t(12;21) is identified in only 0.05% of pediatric B-lineage ALL cases analyzed by conventional karyotyping. Other relevant characteristics of TEL/AML1-positive ALL are frequent deletion of the other TEL allele and association with an excellent prognostic outcome.

It is expected that future studies will provide more detailed information on the leukemogenic effect of AML1 alterations, and better define the prognostic relevance of detecting the hybrid proteins formed by this gene at diagnosis and during remission.