The leukemogenic transcription factor EVI1 and its putative antagonist, MDSu/EVI1: Expression patterns, biological properties and search for target genes

The EVI1 gene is affected by chromosome rearrangements that are found in a group of patients with a particularly aggressive form of myeloid leukemia. One consequence of such rearrangements is overexpression, i.e. the formation of excessive amounts, of its gene product, the EVI1 protein. EVI1 is able to cause a number of cellular defects characteristic of leukemia, like aberrant maturation, accelerated proliferation, and prolonged survival of blood cells, and is therefore considered an oncogene. Because of its gene structure, it is assumed to act mainly by regulating the expression of other genes. However, these genes have not been identified yet, and, as a consequence, the mechanism of action of EVI1 is understood only incompletely. One of the major goals of the proposed project is therefore to identify genes regulated by this protein. This will lead not only to a better understanding of the molecular events taking place in cells overexpressing EVI1, but may also pinpoint molecular targets for drug development. A variant form of EVI1, termed MDS1/EVI1, has been shown in a small number of studies to oppose the action of EVI1, and has therefore been considered an anti-oncogene. Our own experiments, however, did not confirm these findings. In order to further elucidate the relationship between EVI1 and MDS1/EVI1, we plan to determine their expression patterns in leukemia samples using recently developed methodology for the accurate quantitation of gene expression. We will also compare their effects on cell maturation, proliferation, and death in an experimental system where these proteins can be switched on and off according to experimental requirements. Finally, we will search for genes regulated by MDS1/EVI1. Together, these studies will contribute to a better understanding of the roles that EVI1 and MDS1/EVI1 play in leukemogenesis. This is an important prerequisite for the development of drugs that are specifically targeted to the molecular defect underlying a particularly aggressive form of myeloid leukemia.

Chromosome rearrangements are frequently observed in human leukemia cells, and contribute to the formation and the specific properties of leukemias through alterations in the structure or activity of genes in the vicinity of the respective chromosome breakpoints. Specifically, some chromosome aberrations are associated with a particularly good responsiveness to therapy, while others are typically observed in aggressive and largely therapy-resistant disease. Rearrangements of chromosome subregion (`band`) 3q26, which lead to activation of the EVI1 (`ecotropic viral integration site 1`) gene, belong to the latter group. Even though EVI1 plays important roles in normal mammalian development, it is inactive in healthy blood and bone marrow cells, the nonmalignant counterparts of leukemia cells. EVI1 reactivation through chromosome rearrangements contributes to the acquisition of leukemogenic properties, as shown by experiments using cultured cells and animal model systems. The EVI1 gene gives rise to several variant messenger RNA (mRNA) forms. mRNAs are intermediates in the formation of an actual gene product, which is usually, and also in the case of EVI1, a protein. MDS1/EVI1, the first EVI1 mRNA variant to be isolated, is translated into a protein whose structure and biological properties differ from those of EVI1. In some experiments, its effects were even opposite to those of EVI1, which lead to the hypothesis that MDS1/EVI1 could act as an antagonist of EVI1 also in the development of leukemia. The goals of project P16084 were to investigate certain properties of EVI1 and MDS1/EVI1 in a suitable biological model system, and to compare their expression (~activity) in human leukemia samples. One of the most important aspects of the project was the exact quantification of the expression of EVI1 and MDS1/EVI1 in human leukemia cells through the use of a novel technology, RTQ-RT-PCR. The results of these experiments confirmed assumptions made on the basis of preliminary analyses carried out during the previous project, and showed that MDS1/EVI1 does not potently counteract EVI1 in a clinically relevant manner in human leukemia. Furthermore, we could demonstrate that quantitative measurement of EVI1 can be used to monitor responsiveness to therapy, as well as recurrence of disease, in leukemia patients overexpressing this gene. Another important line of investigations concerned the basic characterization of additional EVI1 mRNA variants that were previously identified in the project leader`s laboratory. Heterogeneity in mRNA structure is known to play important roles in the control of the activity of several genes. Our results showed for the first time that such regulatory mechanisms also operate on the medically highly relevant EVI1 gene.