Expression of the five active human beta-like globin genes is controlled by a developmental program that relies on interactions between the promoter of each gene and regulatory elements located upstream in the Locus Control Region (LCR). Elucidation of the mechanisms underlying these interactions and the sequences and proteins through which they occur has important implications for the further understanding of developmental stage-specific gene regulation. The gamma globin genes are particularly important targets for study since identification of the elements responsible for the stage-specific expression of gamma could suggest strategies to reactivate this gene in individuals with severe hemoglobinopathies (beta thalassemia and sickle cell anemia). Three hypotheses based on recent data form the basis for the present proposal: 1) The human gamma gene contains one or more dominant elements that insure its fetal expression. These elements may strengthen LCR:gamma interactions or may insure a chromatin structure that is permissive for fetal expression. A series of hybrid constructs will be produced in which portions of the xi gene are replaced by gamma gene sequences. These constructs will be tested in transgenic mice in order to localize and further characterize the putative fetal elements. 2) A YY1 binding site at -1086 bp upstream from both gamma genes is important for fetal activation of gamma (and possibly also for adult repression of this gene). This site may represent a structural element, rather than an activator or respressor per se. Structural and functional aspects of this site will be explored in transgenic mice and in cell based assays. Chromatin structure surrounding this site will be examined by restriction endonuclease accessibility. The galago gamma gene (an embryonic gamma gene) will be used to determine whether the -1086 YY1 site is sufficient to confer fetal activation on an otherwise embryonic gamma gene. 3) By utilization of sequences that are highly conserved between human and galago, is possible to design a 35 kb midilocus construct that contains the complete LCR and xi, gamma and beta genes; such a construct is expected to recapitulate the correct expression pattern of the human genes in a position-independent fashion. The wild type construct and three mutant derivatives will be analyzed in transgenic mice and in cell-based assays to establish that the midilocus can recapitulate all aspects of beta-like globin gene regulation. This should provide a valuable tool for the further dissection of the regulators of hemoglobin switching.
