Target genes of the MDS-associated transcription factor EVI 1

Myelodysplastic syndromes (MDS) are a group of clonal malignant diseases of myeloid cells that are characterized by hypercellularity and dysplasia of the bone marrow with concomitant cytopenia of the peripheral blood. Currently available therapies are rarely curative, and most patients die either as a consequence of their cytopenia, or because their disease progresses to a therapy resistant form of acute myeloid leukemia (AML). Comparatively little is known about the genetic basis of MDS; however, a significant proportion of patients were found to overexpress the oncogene EVI1. The role of this gene in the pathogenesis of MDS is corroborated by recent findings that its experimental overexpression leads to an MDS-like disease in mice. In addition, EVI1 influences the proliferation, differentiation, and programmed death of cultured cells. Its nuclear localization, predicted protein structure, ability to bind to DNA, and interactions with transcriptional cofactors suggest that the EVI1 protein exerts is biological effects mainly by acting as a transcription factor. However, only a small number of EVI1- regulated genes have been described so far. To contribute to a better understanding of the role of EVI1 in MDS, additional EVI1 target genes shall be identified and characterized in the course of the presented project. We have established and begun to characterize a human myeloid cell line, U937T_EVI1-HA, in which the expression of EVI1 can be turned on by removing the antibiotic tetracycline from the culture media. Induction of EVI1 in these cells diminished their proliferation rate and viability. In addition, preliminary data indicated that the differentiation of induced U937T_EVI1-HA cells was disturbed in a manner typical of MDS: in response to pertinent signals, they died rather than differentiating. Taken together, therefore, expression of EVI1 in U937T_EVI1-HA cells brings forth cellular phenotypes that are characteristic of MDS. In the proposed project, the effects of EVI1 on cell death and differentiation of U937T_EVI1-HA cells shall be investigated in more detail. Furthermore, microarray analyses shall be employed in order to identify EVI1 target genes, both among `traditional` protein coding genes and among the only recently discovered microRNA genes. The regulation of some of the newly found EVI1 target genes shall be confirmed by quantitative RT-PCR, and further characterized by reporter gene assays and chromatin immunoprecipitation experiments. In addition, their role in the phenotypic effects of EVI1 expression shall be investigated. The results of these studies will yield new insights into the ways in which EVI1 contributes to the emergence and clinical properties of MDS.

The "ecotropic viral integration site 1" (EVI1) gene plays important roles in several developmental processes, but is inactive (not "expressed") in the mature cells of most tissues of the adult organism. However, it can be activated in a variety of tumors, and both contribute to tumorigenesis and reduce the sensitivity of malignant cells to chemotherapeutic agents. The association between overexpression of EVI1 and poor therapy response is particularly well documented for acute myeloid leukemia (AML), but a role for EVI1 has also been reported in myelodysplastic syndromes (MDS), a disease that precedes AML in some cases.The aim of project P20920 was to contribute to a better understanding of the way of action of EVI1 in AML and MDS through the use of a suitable experimental model system. To this end, a human myeloid cell line - the pathological alteration of myeloid cells being the basis for both AML and MDS - was manipulated to express or not express EVI1 according to experimental requirements. EVI1 expression in this system lead to a number of changes in cellular behaviour that corresponded to specific characteristics of AML and MDS. The clinically most relevant of these alterations was a reduced sensitivity towards chemotherapeutic drugs, a phenotype that despite of its great relevance to the role of EVI1 in AML had not previously described in a model system amenable to experimental analysis. Since EVI1 exhibits properties of a classical regulator of the activity of other genes (a "transcription factor"), the search for the molecular mechanisms through which EVI1 caused chemotherapy resistance focussed on the identification of EVI1 regulated genes. Indeed, genome wide analyses uncovered several dozens of genes whose activity was altered in response to EVI1 expression. Interestingly, these experiments also revealed a significant overlap between genes regulated by EVI1 and genes regulated by chemotherapeutic drugs as identified in a parallel approach, indicating that overexpression of EVI1 may cause a state of pre-adaptation of malignant cells that increases their ability to protect themselves during exposure to chemotherapeutic agents. In subsequent studies, three genes were subjected to detailed analyses of their mode of regulation by EVI1, and their role in EVI1 mediated therapy resistance. Even though in one case initial results require further confirmation, a role in chemotherapy resistance conferred by EVI1 could be demonstrated for all three genes.