Our objective is the development of a detailed model for the activation of globin gene expression in murine erythroleukemia (MEL) cells. The specific goal is to identify and characterize nuclear proteins from MEL cells which directly interact with cloned globin gene DNA fragments. A rapid and unambiguous binding assay will be used to identify factors capable of forming stable complexes with Alpha- and Beta-globin gene promoter regions. The binding assay will be used to evaluate enrichment of specific binding activities by conventional chromatography. Using this approach we show in preliminary results that a protein which binds to the CCAAT-box of the Alpha-globin gene has been extensively enriched. We propose use the binding assay to identify additional factors that interact with the globin genes. DNA sequences required for factor binding will be determined by DNase I footprinting. Binding sites will be extensively characterized by producing a series of deletion and point mutations in order to identify minimal binding sites. Minimal binding sites will be transferred into non-binding domains to confirm the identity of the binding site. Protein factors responsible for binding activity will be determined by extensive chromatographic protocols and by isolating proteins-nucleic acid complexes and determining their protein content. We will use in vitro transcription assays to determine if fractionated nuclear extracts can be reconstituted, stimulated or repressed by the factors we will enrich. To determine if binding sites indentified in vitro are occupied within the nucleus, in vivo footprinting experiments will be performed. The isolation and characterization of proteins that bind directly to the globin genes will provide novel insights into the regulation of these genes. It is of interest to note that a naturally occurring mutation in the CCAAT box of a human fetal globin gene has recently been implicated in the inappropriate expression of this gene during adult life. Identification of proteins that bind to these regions in vitro might provide insights into how these mutations enhance fetal globin gene expression, and could significantly advance our understanding of these inherited hematologic disorders. It is also not unreasonable to speculate that the expression of cellular onc-genes might be regulated by specific protein-nucleic acid interactions. A general understanding of these interactions could thus lead to significant increases in our understanding of the regulation of cell growth.
