Role of the chromatin remodeling ATPase Snf2h in erythroleukemic blasts

Abstract

Expression of a major mammalian ISWI family chromatin remodeling ATPase, Snf2h, has been observed in patients with acute myelocytic leukemia. Our laboratory uses murine erythroleukemia (MEL) cells as an inducible model system of the molecular mechanisms responsible for leukemic proliferation and erythroid differentiation. Studies in our lab have shown that the dysregulated expression of, PU.1, is critical for blocking differentiation in MEL cells. PU.1 interacts with GATA-1, the major transcription factor that drives erythroid differentiation, and represses erythroid gene activation. Interestingly, Snf2h is present in the promoter regions, along with GATA-1 and PU.1, of several repressed GATA-1 target genes in undifferentiated MEL cells. The occupancy of Snf2h, along with PU.1, at these GATA-1 binding sites declines as MEL cells differentiate and induce globin expression. To investigate the role of Snf2h in proliferation and the block to differentiation in MEL cells, we studied its ability to bind GATA-1 and PU.1 as well as its role in the regulation of globin gene expression, chromatin structure, differentiation, and proliferation. Protein-protein interaction studies suggest that Snf2h interacts with PU.1 and GATA-1. Snf2h expression is down regulated during MEL cell differentiation. MEL cells blocked in their ability to differentiate by exogenous PU.1 failed to downregulate Snf2h. These observations suggested that Snf2h might participate in the PU.1-mediated block to differentiation in MEL cells. However, overexpression of exogenous Snf2h failed to block MEL cell differentiation or alter their proliferation. Likewise, reducing Snf2h expression levels with siRNA neither altered proliferation nor induced spontaneous differentiation or derepression of globin gene expression. These data suggest that while Snf2h may be associated with PU.1 and GATA-1 at repressed GATA-1 target genes in undifferentiated MEL cells, it may function primarily during initiation of repression and may not be required to maintain the repressed state. Additionally, reduction of Snf2h levels in D. rerio results in a pleiotropic phenotype affecting the circulatory systems including heart, vasculature, and blood development. This observation in light of our other results raises the possibility that Snf2h is regulating differentiation at an early point in time and remains at loci it has been recruited to, possibly to initiate repression.