Understanding the complex regulation of beta-globin genes is critically important to design therapeutic approaches to beta-thalassemia and sickle cell disease. An erythroid-specific protein complex containing LDB1/GATA-1/TAL1/LMO2 (LDB1 complex) is directly involved in activation of globin gene expression through providing long-range interaction between the beta-globin LCR enhancer and globin gene promoters. The LDB1 complex regulates erythroid enhancers and mechanistic studies of the beta-globin locus indicate that LDB1 dimerization brings the beta-globin gene into proximity with the LCR for transcription activation. The role of co-activators in establishing this long range interaction is poorly understood. We eliminated the RNA Pol II pre-initiation complex from the beta-globin promoter by deleting the TATA-box and initiator using CRISPR/Cas9 editing. While transcription was lost, enhancer-promoter interaction was unaffected. Additional deletion of the promoter GATA1 site eliminated LDB1 complex and mediator occupancy and resulted in loss of enhancer-promoter interaction. However, a mutant form of LDB1 was able to support LCR/beta-globin promoter interaction without mediator core occupancy. Moreover, ENCODE data and ChIP showed that cohesin was almost completely absent from LDB1-regulated erythroid enhancer-gene pairs. We conclude that erythroid lineage specific transcription factors largely mediate enhancer-promoter looping independent of mediator and cohesin. Lineage-specific transcription factors are critical for long-range enhancer interactions but direct or indirect contributions of architectural proteins such as CTCF to enhancer function remain less clear. The LDB1 complex mediates enhancer-gene interactions at the beta-globin locus through LDB1 self-interaction. We find that a novel LDB1-bound enhancer upstream of carbonic anhydrase 2 (Car2) activates its expression by interacting directly with CTCF at the gene promoter. Both LDB1 and CTCF are required for enhancer-Car2 looping and the domain of LDB1 contacted by CTCF is necessary to rescue Car2 transcription in LDB1 deficient cells. LDB1 dimerization is dispensable for this collaboration. Genome wide studies and CRISPR/Cas9 genome editing indicate that LDB1-CTCF enhancer looping underlies activation of a substantial fraction of erythroid genes. Our results provide a mechanism by which long-range interactions of architectural protein CTCF can be tailored to achieve a tissue-restricted pattern of chromatin loops and gene expression. The principles underlying the architectural landscape of chromosomes in living cells remain largely unknown despite its potential to play a role in mammalian gene regulation. We found that the 3-dimensional conformation of a 1 Mbp region of human chromosome 11 containing the beta-globin locus functionally contributes to globin gene expression according to a polymer model of CTCF-dependent chromosome folding. To test this function in vivo, we are using CRISPR/Cas9 technology to delete CTCF sites such as HS5 and 3'HS1 (flanking the beta-globin locus) and CTCF sites between or in the boundaries of CTCF/cohesin-mediated chromatin domains in erythroid K562 and non-erythroid 293T cells. PCR amplification and sequencing were used to validate mono- and bi-allelic deletions. Western blots showed that CTCF protein level remained the same in CTCF binding site deleted cells. ChIP indicated that CTCF binding site deletion resulted in loss of CTCF binding at these sites. To understand how CTCF binding site deletion alters beta-globin gene expression, RNA-seq analysis is being carried out in these two cell lines. Further study using Chromosome Conformation Capture Hi-C will bring us a closer look at the function of CTCF in cell type specific chromosome architecture organization.
Showing the most recent 10 out of 21 publications
