Genome-wide Screening And Functional Verification Of CTCF Bound Cis-regulatory Elements

Eukaryotic gene expression is a sophisticated biological process, which can be regulated at multiple levels by cis-regulatory elements, transcription fator binding, histone modification, DNA methylation, miRNAs and RNA modification. Among them, cis-regulatory elements could be the core issue since they supply the basic sequence requirement for transcription factors and other regulatory factors. Interactions between cis-regulatory elements or cis-regulatory elements and trans-regulatory factors can alter local chromatin conformation to promote or surpress the expression of target genes. Identification of transcriptional regulatory elements in human genome is crucial for elaborating the delicate molecular mechanism of gene expression. In the mean time, regulatory elements are widely used in translational medicine studies, for example, insulators, which can block the interaction between promoters and enhancers and prevent the spread of heterochromatin, are used for reducing genotoxicity in gene therapy.The strategies of human gene therapy contain gene addition and gene repair. The typical approach is importing a functional gene into a patient’s genome to replace the mutated one, which is a really promise therapeutic process. However, when the delivered genes integrated into the genome, they may cause genotoxicity representing a major barrier to realize the promise of gene therapy. Adding insulator elements flanking inserted genes could be used to mitigate or eliminate this risk. Yet, no human insulator with high functional potency has been identified.CTCF, a multifunctional transcription factor, is involved in multiple cellular processes, including transcription activation or repression, chromatin conformation alteration. Besides, functioning as the sole insulator protein in mammal, CTCF is also crucial for the insulating activity of insulators. For example, cHS4 (Chicken, DNase I Hypersensitive Site 4), a classical insulator element used for gene therapy, only can function as an insulator when bound by CTCF. Therefore, mining genome-wide CTCF recognition sites may contribute to identification of new chromatin insulators. Here we described a genomic approach for identification of compact sequence elements that function as insulators. We first scanned the whole human genome for CTCF motifs using an informatics approach, then grouped these potential sites into classes based on shared indentical 14 bp core motifs and computed CTCF occupancy by data from CTCF ChIP-seq and DNase I-seq in K562 cells. Candidates were randomly selected from classes with both high (over 98.6%) and low (9.6%-9.8%) CTCF occupancy. Within high CTCF occupancy class, all 27 candidates exhibited enhancer-blocking activity and 20 of them displayed enhancer-blocking activity that was superior to that of cHS4. Only one of seven elements from low CTCF occupancy classes had a moderate degree of enhancer-blocking activity (lower than cHS4). To further test the enhancer-blocking activity of elements with high CTCF occupancy, we selected another 23 elements from class A with 100% CTCF occupancy. All the 23 candidates showed enhancer-blocking activity and 16 of them exhibited enhancer-blocking activity surpassing cHS4. We identified 36 (72%) elements in total with better enhancer-blocking activity than cHS4, and they are smaller than cHS4 (119 bp to 284 bp vs.1.2 kb), can be efficiently accommodated by viral vectors and have no detrimental effects on viral titers. Our data can serve as a resource for exploring potent insulators in the human genome. The insulators we describe here are expected to increase the safety of gene therapy of genetic diseases.During the study of insulators, we also screened another type of cis-regulatory elements, silencers. Sequences of the genome that are capable to silence gene expression are also thought to play a key role in gene regulation. However, very few functional silencer elements in mammalian cells have been described, and only a fraction of these silecers have been tested for their functions in an autonomous fashion. We reported here the characterization and functional validation of a constitutive autonomous silencer element from the human genome called T39, and the comparison of T39 to three other putative silencer elements previously described by others. Functional analysis included one assay for enhancer-blocking insulator activity and two independent assays for silencer activity, all based on stable transfection and comparison to a neutral spacer control. In erythroid K562 cells, T39 exhibited potent silencer activity, the previously described element PRE2-S5 exhibited modest silencer activity, and the two other previously described elements exhibited no silencer activity. T39 was further found to be capable of silencing three disparate promoters, and of silencing gene expression in three disparate cell lines, and of functioning as a single copy in a topology-independent manner. Of the four elements analyzed, only T39 exhibits a constitutive pattern of DNase hypersensitivity and binding by CTCF. In its native location the T39 element also exhibits a unique interaction profile with a subset of distal putative regulatory elements. Taken together, these studies validate T39 as a constitutive autonomous silencer, identify T39 as a defined control for future studies of other regulatory elements such as insulators, provide a basic chromatin profile for one highly potent silencer element, and predict potential functional mechanism of T39 in gene regulation.