Project description:Synthetic DNA-binding proteins have found broad application in gene therapies and as tools for interrogating biology. Engineered proteins based on the CRISPR/Cas9 and TALE systems have been used to alter genomic DNA sequences, control transcription of endogenous genes, and modify epigenetic states. Although the activity of these proteins at their intended genomic target sites have been assessed, the genome-wide effects of their action have not been extensively characterized. Additionally, the role of chromatin structure in determining the binding of CRISPR/Cas9 and TALE proteins to their target sites and the regulation of nearby genes is poorly understood. Characterization of the activity these proteins using modern high-throughput genomic methods would provide valuable insight into the specificity and off-target effects of CRISPR- and TALE-based genome engineering tools. We have analyzed the genome-wide effects of TALE- and CRISPR-based transcriptional activators targeted to the promoters of two different endogenous human genes in HEK293T cells using a variety of high-throughput DNA sequencing methods. In particular, we assayed the DNA-binding specificity of these proteins and their effects on the epigenome. DNA-binding specificity was evaluated by ChIP-seq and RNA-seq was used to measure the specificity of these activators in perturbing the transcriptome. Additionally, DNase-seq was used to identify the chromatin state at target sites of the synthetic transcriptional activators and the genome-wide chromatin remodeling that occurs as a result of their action. Our results show that these genome engineering technologies are highly specific in both binding to their promoter target sites and inducing expression of downstream genes when multiple activators bind to a single promoter. Moreover, we show that these synthetic activators are able to induce the expression of silent genes in heterochromatic regions of the genome by opening regions of closed chromatin and decreasing DNA methylation. Interestingly, the transcriptional activation domain was not necessary for DNA-binding or chromatin remodeling in these regions, but was critical to inducing gene expression. This study shows that these CRISPR- and TALE-based transcriptional activators are exceptionally specific. Although we detected limited binding of off-target sites in the genome and changes to genome structure, these off-target event did not lead to any detectable changes in gene regulation. Collectively, these results underscore the potential for these technologies to make precise changes to gene expression for gene and cell therapies or fundamental studies of gene function. HEK293T cells were transfected in triplicate with plasmids expressing synthetic transcription factors. The synthetic TFs were either (a) dCas9-VP64 fusion protein and a targeting guide RNA (gRNA), or (b) a TALE-VP64 fusion protein engineered to bind to a specific target site in the genome. As a control, cells were transfected with plasmids expressing GFP. After transfection, RNA-seq was used to identify both on-target and off-target binding sites for the synthetic TFs. The data in this submission were generated using the TALE transfection experiments.

Project description:Synthetic DNA-binding proteins have found broad application in gene therapies and as tools for interrogating biology. Engineered proteins based on the CRISPR/Cas9 and TALE systems have been used to alter genomic DNA sequences, control transcription of endogenous genes, and modify epigenetic states. Although the activity of these proteins at their intended genomic target sites have been assessed, the genome-wide effects of their action have not been extensively characterized. Additionally, the role of chromatin structure in determining the binding of CRISPR/Cas9 and TALE proteins to their target sites and the regulation of nearby genes is poorly understood. Characterization of the activity these proteins using modern high-throughput genomic methods would provide valuable insight into the specificity and off-target effects of CRISPR- and TALE-based genome engineering tools. We have analyzed the genome-wide effects of TALE- and CRISPR-based transcriptional activators targeted to the promoters of two different endogenous human genes in HEK293T cells using a variety of high-throughput DNA sequencing methods. In particular, we assayed the DNA-binding specificity of these proteins and their effects on the epigenome. DNA-binding specificity was evaluated by ChIP-seq and RNA-seq was used to measure the specificity of these activators in perturbing the transcriptome. Additionally, DNase-seq was used to identify the chromatin state at target sites of the synthetic transcriptional activators and the genome-wide chromatin remodeling that occurs as a result of their action. Our results show that these genome engineering technologies are highly specific in both binding to their promoter target sites and inducing expression of downstream genes when multiple activators bind to a single promoter. Moreover, we show that these synthetic activators are able to induce the expression of silent genes in heterochromatic regions of the genome by opening regions of closed chromatin and decreasing DNA methylation. Interestingly, the transcriptional activation domain was not necessary for DNA-binding or chromatin remodeling in these regions, but was critical to inducing gene expression. This study shows that these CRISPR- and TALE-based transcriptional activators are exceptionally specific. Although we detected limited binding of off-target sites in the genome and changes to genome structure, these off-target event did not lead to any detectable changes in gene regulation. Collectively, these results underscore the potential for these technologies to make precise changes to gene expression for gene and cell therapies or fundamental studies of gene function.

Project description:Transcription factors are often regarded as having two distinct components: a DNA binding domain (DBD) to allow binding to the regulatory region of a target gene and a trans-acting functional domain (FD) to modulate gene expression. Recently, it is becoming clear that the DBDs of transcription factors alone are incapable of providing sufficient specificity to account for the highly complex genomic structures in eukaryotes. Other regions, outside of the DBD, may play a role in transcription factor DNA binding specificity in vivo. In order to test this hypothesis, we examined a model zinc finger transcription factor, KrÃ¼ppel-like factor 3 (KLF3) with a FD that recruits partner proteins for its repressive activity and a three zinc finger DBD. Previously, we showed that deletion of the entire KLF3 FD reduced DNA binding across the genome. This implied the importance of the FD for in vivo DNA binding specificity. In the current study, we fused the KLF3 FD onto an unrelated, but well characterised Artificial Zinc Finger (AZF) targeting a standard target gene, VEGF-A. We compared the genomic DNA binding profile of this chimeric KLF3 construct, termed KLF3FD-AZF, to that of the AZF alone. Chromatin Immunoprecipitation followed by next generation sequencing was used to assess genomic wide DNA-binding of these AZFs. Remarkably, in these gain-of-function experiments, we again observed that the KLF3 FD is critical for in vivo DNA binding specificity. The addition of KLF3 FD increased the number of binding sites by more than two fold compared to the AZF alone. KLF3 FD also directed more binding to the promoter regions. Further investigations of these acquired sites identified a substantial portion of these sites as the known endogenous KLF3 bound regions, whilst others contain sequences that are similar but not identical to the predicted AZF target sequence. These results clearly demonstrate that the specific localisation of this model transcription factor to target genes is not solely dependent on the DBD and that the regions outside of this domain may also play an important role. ChIP-Seq experiments were performed on four HEK293 cell lines, two each stably expressing AZF or KLF3FD.AZF or KLF3FD only (two biological replicates). Input samples were included as controls.

Project description:Remarkably, tens of thousands of piwi-interacting RNAs (piRNAs) arise from a minimal number of discrete clustered regions in many organisms. However, it remains unclear if organization into these domains contributes to the highly coordinated, germline-specific expression of these small RNAs. Here, we show that SNPC-4, the DNA-binding subunit of the small nuclear RNA activating protein complex (SNAPc), binds piRNA clusters in a germline-specific manner and is required for global piRNA expression in C. elegans. Within piRNA domains, SNPC-4 not only binds at discrete sites through its established sequence motif, but also exhibits widely distributed binding across the region. Intriguingly, discrete peaks are not found at every piRNA, but instead occur frequently at Pol III-occupied tRNA genes, which have been previously implicated in chromatin organization. Given its unique binding pattern, we suggest that SNPC-4 promotes robust piRNA expression by establishing a permissive chromatin environment at clusters specifically in the germ line. This data submission includes a total of 11 ChIP-seq samples and 2 RNA-seq samples. The ChIP-seq samples include six timecourse samples (Emb-YA) and three tissue-specific samples (WA, GL, SO) for one factor called SNPC-4 (previously called GEI-11). The remaining two ChIP-seq samples are for pol II (AMA-1) and pol III (RPC-1) at one developmental stage. All of these samples were done in replicate, with inputs, yielding at least four sequencing files per sample. Some of the ChIP-seq samples already have fastq files submitted to GEO: (SNPC-4 EMB _GSE48744: GSM1183805, GSM1183806, GSM1183807, GSM1183808), (SNPC-4 L1_GSE48702: GSM1183629, GSM1183630, GSM1183631, GSM1183632), (SNPC-4 L2_GSE37809: GSM928364, GSM928365, GSM928366, GSM928367), (SNPC-4 L3_GSE35276: GSM864851, GSM864852, GSM864853, GSM864854), and (AMA-1_GSE15535: GSM389309, GSM389310, GSM389312, GSM389313). We are submitting fastq files for the remaining ChIP-seq datasets: SNPC-4 L4 - 4 files, SNPC-4 YA - 5 files, SNPC-4 WA - 4 files, SNPC-4 GL - 4 files, SNPC-4 SO - 4 files, and RPC-1 WA - 5 files. Thus we are submitting an additional 26 fastq files. Additionally, we are submitting processed files for all 11 datasets, consisting of one normalized file for the combined duplicate IP and one normalized file for the input sequence. In the case of the three tissue-specific samples for SNPC-4, we are submitting an additional processed (but not normalized) file that represents the IP data with input-subtracted. Thus, we are submitting a total of 25 processed files. NOTE: The processed files for the ChIP-seq samples that already have fastq files submitted to GEO (SNPC-4 EMB _GSE48744, etc.) are linked below as supplementary files at the foot of the Series record. The RNA-seq samples represent one control (control_RNAseq) and one treatment (SNPC-4(RNAi)_RNA-seq)and were not duplicated. We are submitting one fastq file for each, as well as one processed data file for each, for a total of 2 fastq and 2 processed data files.

Dataset Information

Genome-Wide Specificity of DNA-Binding, Gene Regulation, and Chromatin Remodeling by TALE- and CRISPR/Cas9-Based Transcription Factors

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets