Transcriptional regulation and downstream targets of the COUP-TFII transcription factor in erythroid cells

COUP-TFII is an embryo/fetal transcription factor (TF) expressed in erythroid cells of yolk sac origin where it acts as a specific γ-globin activator. However, the regulation of COUP-TFII expression and its role in erythroid cells remains largely unknown. The understanding of COUP-TFII networks is a crucial step prior to the potential use of COUP-TFII in β-hemoglobinopathies’ therapies since the reactivation of the fetal gene γ in adult cells could substitute for the defective adult β-globin production. The aim of the project was the understanding of the transcriptional control of COUP-TFII gene, as well as the characterization of COUP-TFII downstream network in erythroid cells. In order to study how COUP-TFII expression is regulated in erythroid cells, and possibly in other cell types, long-range interactions with the promoter and upstream potential regulatory elements were first mapped by using the GeneHancer database (doi: 10.1093/database/bax028). After further analysis of the evolutionary conservation, epigenetic marks and transcription factor binding sites (TFBSs) within the interacting regions, I selected four candidate regulatory sequences. Next, I tested their regulatory potential in two cell types expressing COUP-TFII: erythroid K562 cells and hepatic HepG2 cells by using luciferase reporter assay. The region with the highest activity in erythroid cells is located at –52 kb from the COUP-TFII gene and contains a GATA1/TAL1 motif bound in vivo that suggests the importance of this sequence for the erythroid gene expression program. Interestingly, the same region is also bound by COUP-TFII itself, implying a potential autoregulatory mechanism. As a next step, I performed a CRISPR/Cas9 knockout of the –52 kb enhancer in K562 cells. Both the deletion of the whole sequence, and the mutation targeting the COUP-TFII binding site resulted in a significantly reduced mRNA level of COUP-TFII. This result suggests that the identified region, localized 52 kb upstream to the COUP-TFII gene is a novel enhancer element of COUP-TFII. Next, in collaboration with C. Cantù and G. Zambanini (Linkoping University, Sweden), a CUT&RUN experiment in HUDEP-2 cells (Human Umbilical Cord-Derived Erythroid Progenitor cells, adult-like) was performed to precisely map the COUP-TFII binding sites within the β-locus. The results confirmed that COUP-TFII binds in vivo to the Locus Control Region (LCR) of the β-locus. In collaboration with C. Pitsillidou and A. Roberto (FME, Bresso, Italy), I assessed the effect of different COUP-TFII levels on the K562 cell proliferation and erythroid differentiation by using flow cytometry. Both COUP-TFII overexpression and knockout in K562 led to only mild changes in cell phenotype. Moreover, to identify downstream direct targets of COUP-TFII in erythroid cells, in addition to the β-locus, the overexpression (OE) or knockout (KO) of COUP-TFII in K562 cells were performed and the integrative analysis of RNA-seq and ChIP-seq allowed me to obtain a list of 66 genes - COUP-TFII candidate direct targets. Next, I confirmed the effect of COUP-TFII OE or KO on the expression of the most differentially expressed genes (DEGs) by RT-qPCR in K562 cells and I selected 12 genes as the most promising targets of COUP-TFII in erythroid cells. By CRISPR/Cas9 knockout of these 12 genes, I evaluated their impact on erythroid phenotype (experiment performed in collaboration with F. Grebien and L. Proietti, Vetmeduni, Vienna, Austria). The most interesting direct target seems to be ApoE, a protein involved in lipid metabolism, has been shown to play a role in erythroid terminal differentiation.