Epigenome editing by a CRISPR/Cas9-based acetyltransferase activates genes from promoters and enhancers (original) (raw)

Nat Biotechnol. Author manuscript; available in PMC 2015 Nov 1.

Published in final edited form as:

PMCID: PMC4430400

NIHMSID: NIHMS672873

Isaac B. Hilton,1,2 Anthony M. D’Ippolito,2,3 Christopher M. Vockley,2,4 Pratiksha I. Thakore,1,2 Gregory E. Crawford,2,5 Timothy E. Reddy,2,6,* and Charles A. Gersbach1,2,7,*

Isaac B. Hilton

1Department of Biomedical Engineering, Duke University, Durham, North Carolina, United States of America

2Center for Genomic & Computational Biology, Duke University, Durham, North Carolina, United States of America

Anthony M. D’Ippolito

2Center for Genomic & Computational Biology, Duke University, Durham, North Carolina, United States of America

3University Program in Genetics and Genomics, Duke University Medical Center, Durham, North Carolina, United States of America

Christopher M. Vockley

2Center for Genomic & Computational Biology, Duke University, Durham, North Carolina, United States of America

4Department of Cell Biology, Duke University Medical Center, Durham, North Carolina, United States of America

Pratiksha I. Thakore

1Department of Biomedical Engineering, Duke University, Durham, North Carolina, United States of America

2Center for Genomic & Computational Biology, Duke University, Durham, North Carolina, United States of America

Gregory E. Crawford

2Center for Genomic & Computational Biology, Duke University, Durham, North Carolina, United States of America

5Department of Pediatrics, Division of Medical Genetics, Duke University Medical Center, Durham, North Carolina, United States of America

Timothy E. Reddy

2Center for Genomic & Computational Biology, Duke University, Durham, North Carolina, United States of America

6Department of Biostatistics & Bioinformatics, Duke University Medical Center, Durham, North Carolina, United States of America

Charles A. Gersbach

1Department of Biomedical Engineering, Duke University, Durham, North Carolina, United States of America

2Center for Genomic & Computational Biology, Duke University, Durham, North Carolina, United States of America

7Department of Orthopaedic Surgery, Duke University Medical Center, Durham, North Carolina, United States of America

1Department of Biomedical Engineering, Duke University, Durham, North Carolina, United States of America

2Center for Genomic & Computational Biology, Duke University, Durham, North Carolina, United States of America

3University Program in Genetics and Genomics, Duke University Medical Center, Durham, North Carolina, United States of America

4Department of Cell Biology, Duke University Medical Center, Durham, North Carolina, United States of America

5Department of Pediatrics, Division of Medical Genetics, Duke University Medical Center, Durham, North Carolina, United States of America

6Department of Biostatistics & Bioinformatics, Duke University Medical Center, Durham, North Carolina, United States of America

7Department of Orthopaedic Surgery, Duke University Medical Center, Durham, North Carolina, United States of America

Address for correspondence: Timothy E. Reddy, Ph.D., Center for Genomic & Computational Biology, Box 3382, 101 Science Drive, Duke University, Durham, NC 27708, 919-684-3286, ude.ekud@ydder.mit. Charles A. Gersbach, Ph.D., Department of Biomedical Engineering, Room 136 Hudson Hall, Box 90281, Duke University, Durham, NC 27708-0281, 919-613-2147, ude.ekud@hcabsreg.selrahc

*Co-corresponding authors

Abstract

Technologies that facilitate the targeted manipulation of epigenetic marks could be used to precisely control cell phenotype or interrogate the relationship between the epigenome and transcriptional control. Here we have generated a programmable acetyltransferase based on the CRISPR/Cas9 gene regulation system, consisting of the nuclease-null dCas9 protein fused to the catalytic core of the human acetyltransferase p300. This fusion protein catalyzes acetylation of histone H3 lysine 27 at its target sites, corresponding with robust transcriptional activation of target genes from promoters, proximal enhancers, and distal enhancers. Gene activation by the targeted acetyltransferase is highly specific across the genome. In contrast to conventional dCas9-based activators, the acetyltransferase effectively activates genes from enhancer regions and with individual guide RNAs. The core p300 domain is also portable to other programmable DNA-binding proteins. These results support targeted acetylation as a causal mechanism of transactivation and provide a new robust tool for manipulating gene regulation.

A primary challenge of functional genomics is to develop technologies that directly and precisely manipulate genome function at individual loci. Projects such as ENCODE1 and the Roadmap Epigenomics Project2 have identified millions of epigenetic marks across the human genome for many human cell types and tissues. Studying the function of those marks, however, has been largely limited to statistical associations with gene expression. Technologies for targeted direct manipulation of these epigenetic properties are necessary to transform such association-based findings into mechanistic principles of gene regulation. Such advances have the potential to benefit human health, as they could lead to gene therapies that modify the epigenetic code at targeted regions of the genome, strategies for regenerative medicine and disease modeling based on the epigenetic reprogramming of cell lineage specification, and the engineering of epigenome-specific drug screening platforms.

Manipulation of the epigenome is possible by treating cells with small molecule drugs, such as inhibitors of histone deacetylases or DNA methyltransferases, or differentiating cells into specific lineages. However, small molecule-based methods globally alter the epigenome and transcriptome, and are not suitable for targeting individual loci. Epigenome editing technologies, including the fusion of epigenome-modifying enzymes to programmable DNA-binding proteins such as zinc finger proteins and transcription activator-like effectors (TALEs), have been effective at achieving targeted DNA methylation, DNA hydroxymethylation, and histone demethylation, methylation, and deacetylation3–8. However a strategy for targeted histone acetylation, which is strongly associated with active gene regulatory elements and enhancers, has not been described. Additionally, the recent emergence of the CRISPR/Cas9 system as a robust and facile genome engineering platform has the potential to broadly enable epigenome editing across science and technology.

The type II CRISPR/Cas9 system from Streptococcus pyogenes (Cas9) is a versatile technology for genome engineering9, 10. The Cas9 nuclease can be directed to specific genomic loci via complementarity between an engineered guide RNA (gRNA) and the target site11–13. The enzymatic activity of the Cas9 nuclease can be abolished by mutation of the RuvC and HNH domains, generating the nuclease-null deactivated Cas9 (dCas9)12. Fused to repression domains, such as the KRAB domain, or activation domains, such as oligomers of the herpes simplex viral protein 16 (VP16), dCas9 can function as a synthetic transcriptional regulator.14–21 However, limitations in the use of dCas9 activators remain, including the need for multiple activation domains14, 20, 22, 23 or combinations of gRNAs16, 17 to achieve high levels of gene induction by synergistic effects between activation domains24, 25. The conventional activator domains used in these engineered transcriptional factors, such as the VP16 tetramer VP6426, function as a scaffold for recruiting multiple components of the preinitiation complex27, 28 and do not have direct enzymatic function to specifically modulate the chromatin state. This indirect method of epigenetic remodeling does not allow for testing the role of specific epigenetic marks and may not be as potent as the direct programming of epigenetic states. We hypothesized that recruitment of acetyltransferase function by dCas9 and a gRNA to the genomic target site would directly modulate epigenetic structure, allowing for more effective gene activation. To this end, we rationally designed an effector fusion protein of dCas9 with the catalytic histone acetyltransferase (HAT) core domain of the human E1A-associated protein p30029. This is an easily programmable approach that facilitates robust control of the epigenome and downstream gene expression.

Results

A dCas9 fusion to the p300 HAT domain activates target genes

p300 is a highly conserved acetyltransferase involved in a wide range of cellular processes30, 31. We fused the full-length p300 protein to dCas9 (dCas9FLp300; Fig. 1a, b) and assayed its capacity for transactivation by transient co-transfection of human HEK293T cells with four gRNAs targeting the endogenous promoters of IL1RN, MYOD1 (MYOD), and POU5F1/OCT4 (OCT4) (Fig. 1c). We used a combination of four gRNAs targeting each promoter based on our and others’ previous observations that multiple gRNAs at a single promoter are necessary for robust gene activation14–22. dCas9FLp300 was well expressed and induced modest activation above background compared to the canonical dCas9 activator fused to the VP64 acidic activation domain (dCas9VP64) (Figs. 1a–c). The full-length p300 protein is a promiscuous acetyltransferase which interacts with a multitude of endogenous proteins, largely via its termini29, 31. In order to mitigate these interactions we isolated the contiguous region of full-length p300 (2414 aa) solely required for inherent HAT activity (amino acids 1048-1664), known as the p300 HAT core domain (p300 Core)29. When fused to the C-terminus of dCas9 (dCas9p300 Core, Fig. 1a, b) the p300 Core domain induced high levels of transcription from endogenous gRNA-targeted promoters (Fig. 1c). When targeted to the IL1RN and MYOD promoters, the dCas9p300 Core fusion displayed significantly higher levels of transactivation than dCas9VP64 (_P-_value 0.01924 and 0.0324 respectively; Fig. 1c). These dCas9-effector fusion proteins were expressed at similar levels (Fig. 1b, Supplementary Fig. 1) indicating that the observed differences were due to differences to transactivation capacity. Additionally, no changes to target gene expression were observed when the effector fusions were transfected without gRNAs (Supplementary Fig. 2). To ensure that the p300 Core acetyltransferase activity was responsible for gene transactivation using the dCas9p300 Core fusion, we screened a panel of dCas9p300 Core HAT-domain mutant fusion proteins (Supplementary Fig. 1)29. A single inactivating amino acid substitution within the HAT core domain (WT residue D1399 of full-length p300) of dCas9p300 Core (dCas9p300 Core (D1399Y); Fig. 1a) abolished the transactivation capacity of the fusion protein (Fig. 1c), demonstrating that intact p300 Core acetyltransferase activity was required for dCas9p300 Core-mediated transactivation.

An external file that holds a picture, illustration, etc. Object name is nihms672873f1.jpg

The dCas9p300 Core fusion protein activates transcription of endogenous genes from proximal promoter regions. (a) Schematic of dCas9 fusion proteins dCas9VP64, dCas9FL p300, and dCas9p300 Core. Streptococcus pyogenes dCas9 contains nuclease inactivating mutations D10A and H840A. The D1399 catalytic residue in the p300 HAT domain is indicated. (b) Western blot showing expression levels of dCas9 fusion proteins and GAPDH in co-transfected cells (full blot shown in Supplementary Fig. 1c). (c) Relative mRNA expression of IL1RN, MYOD, and OCT4, determined by qRT-PCR, by the indicated dCas9 fusion protein co-transfected with four gRNAs targeted to each promoter region (Tukey-test, *_P_-value < 0.05, n = 3 independent experiments each, error bars: s.e.m.). Numbers above bars indicate mean expression. FLAG, epitope tag; NLS, nuclear localization signal; HA, hemagglutinin epitope tag; CH, cysteine-histidine-rich region; Bd, bromodomain; HAT, histone acetyltransferase domain.

dCas9p300 Core activates genes from proximal and distal enhancers

Although there are many published examples of activating genes with engineered transcription factors targeted to promoters, inducing gene expression from other distal regulatory elements has been limited, particularly for dCas9-based activators32–35. Given the role and localization of p300 at endogenous enhancers36, 37, we hypothesized that the dCas9p300 Core would effectively induce transcription from distal regulatory regions with appropriately targeted gRNAs. We targeted the distal regulatory region (DRR) and core enhancer (CE) of the human MYOD locus38 through co-transfection of four gRNAs targeted to each region and either dCas9VP64 or dCas9p300 Core (Fig. 2a). Compared to a mock-transfected control, dCas9VP64 did not show any induction when targeted to the MYOD DRR or CE region. In contrast, dCas9p300 Core induced significant transcription when targeted to either MYOD regulatory element with corresponding gRNAs (_P_-value 0.0115 and 0.0009 for the CE and DRR regions respectively). We also targeted the upstream proximal (PE) and distal (DE) enhancer regions of the human OCT4 gene39 by co-transfection of six gRNAs and either dCas9VP64 or dCas9p300 Core (Fig. 2b). dCas9p300 Core induced significant transcription from these regions (_P_-value ≤ 0.0001 and _P_-value ≤0.003 for the DE and PE, respectively), whereas dCas9VP64 was unable to activate OCT4 above background levels when targeted to either the PE or DE regions.

An external file that holds a picture, illustration, etc. Object name is nihms672873f2.jpg

The dCas9p300 Core fusion protein activates transcription of endogenous genes from distal enhancer regions. (a) Relative MYOD mRNA production in cells co-transfected with a pool of gRNAs targeted to either the proximal or distal regulatory regions and dCas9VP64 or dCas9p300 Core; promoter data from Fig. 1c (Tukey-test, *_P_-value <0.05 compared to mock-transfected cells, Tukey test †_P_-value <0.05 between dCas9p300 Core and dCas9VP64, n = 3 independent experiments, error bars: s.e.m.). The human MYOD locus is schematically depicted with corresponding gRNA locations in red. CE, MyoD core enhancer; DRR, MyoD distal regulatory region. (b) Relative OCT4 mRNA production in cells co-transfected with a pool of gRNAs targeted to the proximal and distal regulatory regions and dCas9VP64 or dCas9p300 Core; promoter data from Fig. 1c (Tukey-test, *_P_-value <0.05 compared to mock-transfected cells, Tukey test †_P_-value <0.05 between dCas9p300 Core and dCas9VP64, n = 3 independent experiments, error bars: s.e.m.). The human OCT4 locus is schematically depicted with corresponding gRNA locations in red. DE, Oct4 distal enhancer; PE, Oct4 proximal enhancer. (c) The human β-globin locus is schematically depicted with approximate locations of the hypersensitive site 2 (HS2) enhancer region and downstream genes (HBE, HBG, HBD, and HBB). Corresponding HS2 gRNA locations are shown in red. Relative mRNA production from distal genes in cells co-transfected with four gRNAs targeted to the HS2 enhancer and the indicated dCas9 proteins. Note logarithmic y-axis and dashed red line indicating background expression (Tukey test among conditions for each β-globin gene, †_P_-value <0.05, n = 3 independent experiments, error bars: s.e.m.). n.s., not significant.

The well-characterized mammalian β-globin locus control region (LCR) orchestrates transcription of the downstream hemoglobin genes; hemoglobin epsilon 1 (HBE, from ~ 11 kb), hemoglobin gamma 1 and 2 (HBG, from ~ 30 kb), hemoglobin delta (HBD, from ~ 46 kb), and hemoglobin beta (HBB, from ~54 kb) (Fig. 2c)35, 40. DNase hypersensitive sites within the β-globin LCR serve as docking sites for transcriptional and chromatin modifiers, including p30041, which coordinate distal target gene expression. We designed four gRNAs targeting the DNase hypersensitive site 2 within the LCR enhancer region (HS2 enhancer). These four HS2-targeted gRNAs were co-transfected with dCas9, dCas9VP64, dCas9p300 Core, or dCas9p300 Core (D1399Y), and the resulting mRNA production from HBE, HBG, HBD, and HBB, was assayed (Fig. 2c). dCas9, dCas9VP64, and dCas9p300 Core (D1399Y) were unable to transactivate any downstream genes when targeted to the HS2 enhancer. In contrast, targeting of dCas9p300 Core to the HS2 enhancer led to significant expression of the downstream HBE, HBG, and HBD genes (_P_-value ≤ 0.0001, 0.0056, and 0.0003 between dCas9p300 Core and mock-transfected cells for HBE, HBG, and HBD respectively). Overall, HBD and HBE appeared relatively less responsive to synthetic p300 Core-mediated activation from the HS2 enhancer; a finding consistent with lower rates of general transcription from these two genes across several cell lines (Supplementary Fig. 3). Nevertheless, with the exception of the most distal HBB gene, dCas9p300 Core exhibited a capacity to activate transcription from downstream genes when targeted to all characterized enhancer regions assayed, a capability not observed for dCas9VP64. Together, these results demonstrate that dCas9p300 Core is a potent programmable transcription factor that can be used to regulate gene expression from a variety of promoter-proximal and promoter-distal locations.

Gene activation by dCas9p300 Core is highly specific

Recent reports indicate that dCas9 may have widespread off-target binding events in mammalian cells in combination with some gRNAs42, 43, which could potentially lead to off-target changes in gene expression. In order to assess the transcriptional specificity of the dCas9p300 Core fusion protein we performed transcriptome profiling by RNA-seq in cells co-transfected with four _IL1RN_-targeted gRNAs and either dCas9, dCas9VP64, dCas9p300 Core, or dCas9p300 Core (D1399Y). Genome-wide transcriptional changes were compared between dCas9 with no fused effector domain and either dCas9VP64, dCas9p300 Core, or dCas9p300 Core (D1399Y) (Fig. 3). While both dCas9VP64 and dCas9p300 Core upregulated all four IL1RN isoforms, only the effects of dCas9p300 Core reached genome-wide significance (Fig. 3a–b, Supplementary Table 1; _P_-value 1.0 × 10−3–5.3 × 10−4 for dCas9VP64; _P_-value 1.8 × 10−17–1.5 × 10−19 for dCas9p300 Core. In contrast, dCas9p300 Core (D1399Y) did not significantly induce any IL1RN expression (Fig. 3c; _P_-value > 0.5 for all 4 IL1RN isoforms). Comparative analysis to dCas9 revealed limited dCas9p300 Core off-target gene induction, with only two transcripts induced significantly above background at a false discovery rate (FDR) < 5%: KDR (FDR = 1.4 × 10−3); and FAM49A (FDR = 0.04) (Fig. 3b, Supplementary Table 1). We also found increased expression of p300 mRNA in cells transfected with dCas9p300 Core and dCas9p300 Core (D1399Y). This finding is most likely explained by RNA-seq reads mapping to mRNA from the transiently transfected p300 core fusion domains. Thus the dCas9p300 Core fusion displayed high genome-wide targeted transcriptional specificity and robust gene induction of all four targeted IL1RN isoforms.

An external file that holds a picture, illustration, etc. Object name is nihms672873f3.jpg

dCas9p300 Core targeted transcriptional activation is specific and robust. (a–c) MA plots generated from DEseq2 analysis of genome-wide RNA-seq data from HEK293T cells transiently co-transfected with dCas9VP64 (a) dCas9p300 Core (b) or dCas9p300 Core (D1399Y) (c) and four IL1RN promoter-targeting gRNAs compared to HEK293T cells transiently co-transfected with dCas9 and four IL1RN promoter-targeting gRNAs. mRNAs corresponding to IL1RN isoforms are shown in blue and circled in each panel. Red labeled points in panels b–c correspond to off-target transcripts significantly enriched after multiple hypothesis testing (KDR, (FDR = 1.4 × 10−3); FAM49A, (FDR = 0.04); p300, (FDR = 1.7 × 10−4) in panel b; and p300, (FDR = 4.4 × 10−10) in panel c.

dCas9p300 Core acetylates H3K27 at enhancers and promoters

Activity of regulatory elements correlates with covalent histone modifications such as acetylation and methylation1, 2. Of those histone modifications, acetylation of lysine 27 on histone H3 (H3K27ac) is one of the most widely documented indicators of enhancer activity29–31, 36, 37. Acetylation of H3K27 is catalyzed by p300 and is also correlated with endogenous p300 binding profiles36, 37. Therefore we used H3K27ac enrichment as a measurement of relative dCas9p300 Core-mediated acetylation at the genomic target site. To quantify targeted H3K27 acetylation by dCas9p300 Core we performed chromatin immuno-precipitation with an anti-H3K27ac antibody followed by quantitative PCR (ChIP-qPCR) in HEK293T cells co-transfected with four HS2 enhancer-targeted gRNAs and either dCas9, dCas9VP64, dCas9p300 Core, or dCas9p300 Core (D1399Y) (Fig. 4). We analyzed three amplicons at or around the target site in the HS2 enhancer or within the promoter regions of the HBE and HBG genes (Fig. 4a). Notably, H3K27ac is enriched in each of these regions in the human K562 erythroid cell line that has a high level of globin gene expression (Fig. 4a). We observed significant H3K27ac enrichment at the HS2 enhancer target locus compared to treatment with dCas9 in both the dCas9VP64 (_P_-value 0.0056 for ChIP Region 1and _P_-value 0.0029 for ChIP Region 3) and dCas9p300 Core (_P_-value 0.0013 for ChIP Region 1and _P_-value 0.0069 for ChIP Region 3) co-transfected samples (Fig. 4b). A similar trend of H3K27ac enrichment was also observed when targeting the IL1RN promoter with dCas9VP64 or dCas9p300 Core (Supplementary Fig. 4). In contrast to these increases in H3K27ac at the target sites by both dCas9VP64 or dCas9p300 Core, robust enrichment in H3K27ac at the HS2-regulated HBE and HBG promoters was observed only with dCas9p300 Core treatment (Fig. 4c–d). Together these results demonstrate that dCas9p300 Core uniquely catalyzes H3K27ac enrichment at gRNA-targeted loci and at enhancer-targeted distal promoters. Therefore the acetylation established by dCas9p300 Core at HS2 may catalyze enhancer activity in a manner distinct from direct recruitment of preinitiation complex components by dCas9VP64(refs 27, 28), as indicated by the distal activation of the HBE, HBG, and HBD genes from the HS2 enhancer by dCas9p300 Core but not by dCas9VP64 (Fig. 2c, Supplementary Fig. 3).

An external file that holds a picture, illustration, etc. Object name is nihms672873f4.jpg

The dCas9p300 Core fusion protein acetylates chromatin at a targeted enhancer and corresponding downstream genes. (a) The region encompassing the human β-globin locus on chromosome 11 (5,304,000 – 5,268,000; GRCh37/hg19 assembly) is shown. HS2 gRNA target locations are indicated in red and ChIP-qPCR amplicon regions are depicted in black with corresponding green numbers. ENCODE/Broad Institute H3K27ac enrichment signal in K562 cells is shown for comparison. Magnified insets for the HS2 enhancer, HBE, and HBG1/2 promoter regions are displayed below. (b–d) H3K27ac ChIP-qPCR enrichment (relative to dCas9; red dotted line) at the HS2 enhancer, HBE promoter, and HBG1/2 promoters in cells co-transfected with four gRNAs targeted to the HS2 enhancer and the indicated dCas9 fusion protein. HBG ChIP amplicons 1 and 2 amplify redundant sequences at the HBG1 and HBG2 promoters (denoted by ‡). Tukey test among conditions for each ChIP-qPCR region, *_P_-value <0.05 (n = 3 independent experiments, error bars: s.e.m.).

dCas9p300 Core activates genes with a single gRNA

Robust transactivation using dCas9-effector fusion proteins currently relies upon the application of multiple gRNAs, multiple effector domains, or both14–22, 24, 25. Transcriptional activation could be simplified with the use of single gRNAs in tandem with a single dCas9-effector fusion. This would also facilitate multiplexing distinct target genes and the incorporation of additional functionalities into the system. We compared the transactivation potential of dCas9p300 Core with single gRNAs and four pooled gRNAs targeting the IL1RN, MYOD and OCT4 promoters (Fig. 5a–c). Substantial activation was observed upon co-transfection of the dCas9p300 Core and a single gRNA at each promoter tested. For the IL1RN and MYOD promoters, there was no significant difference between the pooled gRNAs and the best individual gRNA (Fig. 5a–b; IL1RN gRNA “C”, _P_-value 0.78; MYOD gRNA “D”, _P_-value 0.26). Although activation of the OCT4 promoter produced additive effects when four gRNAs were pooled with dCas9p300 Core, the most potent single gRNA (gRNA “D”) induced a statistically comparable amount of gene expression to that observed upon co-transfection of dCas9VP64 with an equimolar pool of all four promoter gRNAs (_P_-value 0.73; Fig. 5c). Compared to dCas9p300 Core, levels of gene activation with dCas9VP64 and single gRNAs were substantially lower. Also, in contrast to dCas9p300 Core, dCas9VP64 demonstrated synergistic effects with combinations of gRNAs in every case, (Fig. 5a–c) as reported previously16, 17.

An external file that holds a picture, illustration, etc. Object name is nihms672873f5.jpg

The dCas9p300 Core fusion protein activates transcription of endogenous genes from regulatory regions with a single gRNA. (a–c) Relative (a) IL1RN, (b) MYOD or (c) OCT4 mRNA produced from cells co-transfected with dCas9p300 Core or dCas9VP64 and gRNAs targeting respective promtoters (n = 3 independent experiments, error bars: s.e.m.). Relative MYOD (d) or OCT4 (e) mRNA produced from cells co-transfected with dCas9p300 Core and indicated gRNAs targeting the indicated MYOD or OCT4 enhancers (n = 3 independent experiments, error bars: s.e.m.). DRR, MYOD distal regulatory region; CE, MYOD core enhancer; PE, OCT4 proximal enhancer; DE, OCT4 distal enhancer. (Tukey test between dCas9p300 Core and single OCT4 DE gRNAs compared to mock-transfected cells, *_P_-value <0.05, Tukey test among dCas9p300 Core and OCT4 DE gRNAs compared to All, †_P_-value <0.05,). Relative HBE (f) or HBG (g) mRNA production in cells co-transfected with dCas9p300 Core and the indicated gRNAs targeted to the HS2 enhancer (Tukey test between dCas9p300 Core and single HS2 gRNAs compared to mock-transfected cells, *P_-value <0.05, Tukey test among dCas9p300 Core and HS2 single gRNAs compared to All, †_P<0.05, n = 3 independent experiments, error bars: s.e.m.). HS2, β-globin locus control region hypersensitive site 2; n.s., not significant using Tukey test.

Based on the transactivation ability of dCas9p300 Core at enhancer regions and with single gRNAs at promoter regions, we hypothesized that dCas9p300 Core might also be able to transactivate enhancers via a single targeted gRNA. We tested the MYOD (DRR and CE), OCT4 (PE and DE), and HS2 enhancer regions with equimolar concentrations of pools or single gRNAs (Fig. 5d–g). For both MYOD enhancer regions, co-transfection of dCas9p300 Core and a single enhancer-targeting gRNA was sufficient to activate gene expression to levels similar to cells co-transfected with dCas9p300 Core and the four pooled enhancer gRNAs (Fig. 5d). Similarly, OCT4 gene expression was activated from the PE via dCas9p300 Core localization with a single gRNA to similar levels as dCas9p300 Core localized with a pool of six PE-targeted gRNAs (Fig. 5e). dCas9p300 Core-mediated induction of OCT4 from the DE (Fig. 5e) and HBE and HBG genes from the HS2 enhancer (Fig. 5f–g) showed increased expression with the pooled gRNAs relative to single gRNAs. Nevertheless, there was activation of target gene expression above control for several single gRNAs at these enhancers (Fig. 5e–g).

The p300 HAT domain is portable to other DNA-binding proteins

The dCas9/gRNA system from Streptococcus pyogenes has been widely adopted due to its robust, versatile, and easily programmable properties9, 10. However, several other programmable DNA-binding proteins are also under development for various applications and may be preferable for particular applications, including orthogonal dCas9 systems from other species44, TALEs, and zinc finger proteins. To determine if the p300 Core HAT domain was portable to these other systems, we created fusions to dCas9 from Neisseria meningitidis (_Nm_-dCas9)44, four different TALEs targeting the IL1RN promoter24, and a zinc finger protein targeting ICAM1 (Fig. 6). Co-transfection of _Nm_-dCas9p300 Core and five _Nm_-gRNAs targeted to the HBE or the HBG promoters led to significant gene induction compared to mock-transfected controls (_P-_value 0.038 and 0.0141 for HBE and HBG respectively) (Fig. 6b–c). When co-transfected with five _Nm_-gRNAs targeted to the HS2 enhancer, _Nm_-dCas9p300 Core also significantly activated the distal HBE and HBG, globin genes compared to mock-transfected controls (p= 0.0192 and _p_=0.0393, respectively)(Fig. 6d–e). Similar to dCas9p300 Core, _Nm_-dCas9p300 Core activated gene expression from promoters and the HS2 enhancer via a single gRNA. _Nm_-dCas9VP64 displayed negligible capacity to transactivate HBE or HBG regardless of localization to promoter regions or to the HS2 enhancer either with single or multiple gRNAs (Fig. 6b–e). Transfection of the expression plasmids for a combination of four TALEp300 Core fusion proteins targeted to the IL1RN promoter (IL1RN TALEp300 Core) also activated downstream gene expression, although to a lesser extent than four corresponding TALEVP64 fusions (IL1RN TALEVP64) (Fig. 6f–g). However, single p300 Core effectors were much more potent than single VP64 domains when fused to IL1RN TALEs. Interestingly, dCas9p300 Core directed to any single binding site generated comparable IL1RN expression relative to single or pooled IL1RN TALE effectors and direct comparisons suggest that dCas9 may be a more robust activator than TALEs when fused to the larger p300 Core fusion domain (Supplementary Figure 5). Finally, the ZFp300 Core fusion targeted to the ICAM1 promoter (ICAM1 ZFp300 Core) also activated its target gene relative to control and at a similar level as ZFVP64 (ICAM1 ZFVP64) (Fig. 6h–i). The versatility of the p300 Core fusion with multiple targeting domains is evidence that this is a robust approach for targeted acetylation and gene regulation. The various p300 core fusion proteins were expressed well, as determined by western blot (Supplementary Fig. 6), but differences in p300 Core activity between different fusion proteins could be attributable to binding affinity or protein folding.

An external file that holds a picture, illustration, etc. Object name is nihms672873f6.jpg

The p300 Core can be targeted to genomic loci by diverse programmable DNA-binding proteins. (a) Schematic of the Neisseria meningitidis (Nm) dCas9 fusion proteins _Nm_-dCas9VP64 and _Nm_-dCas9p300 Core. Neisseria meningitidis dCas9 contains nuclease-inactivating mutations D16A, D587A, H588A, and N611A. (b–c) Relative (b) HBE or (c) HBG mRNA in cells co-transfected with five individual or pooled (A–E) Nm gRNAs targeted to the HBE or HBG promoter and _Nm_-dCas9VP64 or _Nm_-dCas9p300 Core. (d–e) Relative (d) HBE or (e) HBG mRNA in cells co-transfected with five individual or pooled (A–E) Nm gRNAs targeted to the HS2 enhancer and _Nm_-dCas9VP64 or _Nm_-dCas9p300 Core. (f) Schematic of TALEs with domains containing _IL1RN_-targeted repeat variable diresidues (Repeat Domain). (g) Relative IL1RN mRNA in cells transfected with individual or pooled (A–D) IL1RN TALEVP64 or IL1RN TALEp300 Core encoding plasmids. (h) Schematic of ZF fusion proteins with zinc finger helices 1–6 (F1–F6) targeting the ICAM1 promoter. (i) Relative ICAM1 mRNA in cells transfected with ICAM1 ZFVP64 or ICAM1 ZFp300 Core. Tukey-test, *_P_-value <0.05 compared to mock-transfected control, n = 3 independent experiments each, error bars: s.e.m. NLS, nuclear localization signal; HA, hemagglutinin tag; Bd, bromodomain; CH, cysteine-histidine-rich region; HAT, histone acetyltransferase domain.

Discussion

These results establish the dCas9p300 Core fusion protein as a potent and easily programmable tool to synthetically manipulate acetylation at targeted endogenous loci, leading to regulation of proximal and distal enhancer-regulated genes. The intrinsic p300 Core acetyltransferase activity is crucial for this efficacy, as demonstrated by the consistent lack of chromatin modification and transactivation potential of the dCas9p300 Core (D1399Y) acetyltransferase-null mutant (Figs. 1c, ​2c, 4b–d, Supplementary Fig. 1). Fusion of the catalytic core domain of p300 to dCas9 resulted in substantially higher transactivation of downstream genes than the direct fusion of full-length p300 protein despite robust protein expression (Fig. 1b, c). This finding may be due to differences in protein structure and function or interactions with other cellular proteins, suggesting that isolation of catalytic core regions is a useful strategy for future programmable epigenome editing tools. The dCas9p300 Core fusion protein also exhibited an increased transactivation capacity relative to dCas9VP64 (Figs. 1c, ​2, ​3, ​5), including in the context of the _Nm_-dCas9 scaffold (Fig. 6b–e). This was especially evident at distal enhancer regions, at which dCas9VP64 displayed little, if any, measurable downstream transcriptional activity (Figs. 2, 6d, e), an important finding for applications involving synthetic transcriptional and epigenomic control. Additionally, the dCas9p300 Core displayed precise and robust genome-wide transcriptional specificity (Fig. 3, Supplementary Fig. 2).

The observation that targeted acetylation is sufficient for gene activation at an endogenous locus and enhancer is a significant finding of our study. While it is possible that activation or recruitment of other co-factors is involved in the targeted dCas9p300 Core-mediated epigenomic control, only dCas9p300 Core was capable of potent transcriptional activation and co-enrichment of acetylation at promoters targeted by the epigenetically modified enhancer (Figs. 2c, 3c, d, Supplementary Fig. 3). This distinction is likely the result of differing modes of activity between the two effector domains. While p300 has inherent acetyltransferase activity29, 30, VP64 relies upon the sequential recruitment of co-factors, including endogenous p300, for HAT activity at sites of transactivation27, 28. Hence dCas9VP64-mediated H3K27 acetylation is likely an indirect secondary effect, whereas dCas9p300 Core likely acetylates target loci directly and in a manner that is dependent upon an intact acetyltransferase domain. Together, these results support a model in which acetylation is a sufficient step in activating enhancer activity at the loci tested here.

The unique activity of dCas9p300 Core supports its use in elucidating key steps of gene regulation, including dissection of the interplay between the epigenome, regulatory element activity, and gene regulation. For example, the GATA1 transcription factor that is essential for globin gene expression erythroid cells45 is not expressed in the HEK293T cells used in this study16. The observed potent activation of globin gene expression via dCas9p300 Core-mediated acetylation in the absence of GATA1 suggests that the acetylation at the enhancer occurs downstream of GATA1 but upstream of globin gene activation in erythroid cells. Importantly, the greater transactivation potency of primary and direct acetylation by dCas9p300 Core relative to dCas9VP64 suggests that acetylation of additional factors at the target site may also play a substantial role, and this should be a subject of future investigation.

Although synergy has been widely observed with other synthetic transcriptional activators, including dCas9VP64 (refs. 8, 15–17, 19–22, 24, 25), the p300 Core effector domain did not display similar synergy with either additional gRNAs or TALEs (Figs. 5, ​6, Supplementary Figs. 3, 5) or in combination with VP64 (Supplementary Fig. 7). Moreover, the p300 Core effector was capable of robustly activating gene expression through a single gRNA at promoters and characterized enhancers (Figs. 5, ​6, Supplementary Fig. 3). This effector was also capable of potent gene activation when targeted by a single TALE recognition site (Fig. 6, Supplementary Fig. 5). Interestingly, certain loci appear to be less responsive to transactivation by the localization of a single dCas9p300 Core effector to a corresponding regulatory region (Fig. 5e–g, Supplementary Fig. 3). This does not appear to be related to chromatin accessibility based on ENCODE data (Supplementary Fig. 8), but may be related to transcription factor occupancy or competition (including endogenous p300; Supplementary Figs. 3, 8), gRNA and genetic composition46, TSS proximity14, and/or the underlying local epigenetic circuitry; none of which are mutually exclusive. These factors are relevant to the function of all programmable DNA-binding proteins and would benefit from future efforts to design optimal gRNAs.

Collectively our results demonstrate that the p300 Core acetyltransferase effector domain is a versatile tool for robust and specific targeted transcriptional control. This effector domain is more potent than existing engineered transcription factors made with single activator domains and demonstrates the ability to catalyze specific epigenetic marks with a CRISPR/Cas9-based fusion protein. The results indicate a causal relationship between directed acetylation and subsequent target gene activation. This technology also affords the ability to synthetically transactivate distal genes from putative and known regulatory regions and simplifies transactivation via the application of a single programmable effector and single target site. These capabilities could be significant in enabling genome-wide screens of regulatory element activity and in multiplexing to target several promoters and/or enhancers simultaneously14, 20, 47. Furthermore combining dCas9p300 Core with light-inducible48 or chemically inducible49 control will enable the dynamic control of gene activation in space and time. The ability to target the p300 Core domain with orthogonal dCas9s, TALEs, and zinc finger proteins should also facilitate studies of independent targeting of particular effector functions to distinct loci (Fig. 5, Supplementary Fig. 3). This could include multiplexing various activators, repressors, and epigenetic modifiers to precisely control cell phenotype23, 33, 50 or decipher complex networks of gene regulation. Additionally the mammalian origin of p300 may provide advantages over virally-derived effector domains for in vivo applications by minimizing potential immunogenicity.

The synthetic control of transcription and chromatin is a critical component of cellular engineering. Our results suggest that the unique capabilities of dCas9p300 Core will afford a greater versatility in controlling the epigenetic and transcriptional states of designed biological systems. This system takes advantage of the simple programmability of the CRISPR/Cas9 system to target acetyltransferase activity and complements other recently described epigenetic editing tools, including fusions of demethylases, methyltransferases, and deacetylases3–7 to generate a more complete set of epigenome editing tools.

Online Methods

Cell lines and transfection

HEK293T cells were procured from the American Tissue Collection Center (ATCC, Manassas VA) through the Duke University Cell Culture Facility. Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% FBS and 1% penicillin/streptomycin and maintained at 37°C and 5% CO2. Transfections were performed in 24-well plates using 375 ng of respective dCas9 expression vector and 125 ng of equimolar pooled or individual gRNA expression vectors mixed with Lipofectamine 2000 (Life Technologies, cat. #11668019) as per manufacturer’s instruction. For ChIP-qPCR experiments, HEK293T cells were transfected in 15 cm dishes with Lipofectamine 2000 and 30 μg of respective dCas9 expression vector and 10 μg of equimolar pooled gRNA expression vectors as per manufacturer’s instruction.

Plasmid constructs

pcDNA-dCas9VP64 (dCas9VP64; Addgene, plasmid #47107) has been described previously16. An HA epitope tag was added to dCas9 (no effector) by removing the VP64 effector domain from dCas9VP64 via AscI/PacI restriction sites and using isothermal assembly51 to include an annealed set of oligos containing the appropriate sequence as per manufacturers instruction (NEB cat. #2611). pcDNA-dCas9FLp300 (dCas9FLp300) was created by amplifying full-length p300 from pcDNA3.1-p300 (Addgene, plasmid #23252)52 in two separate fragments and cloning these fragments into the dCas9VP64 backbone via isothermal assembly. A substitution in the full-length p300 protein (L553M), located outside of the HAT Core region, was identified in dCas9FLp300 and in the precursor pcDNA3.1-p300 during sequence validation. pcDNA-dCas9p300 Core (dCas9p300 Core) was generated by first amplifying amino acids 1048 – 1664 of human p300 from cDNA and then subcloning the resulting amplicon into pCR-Blunt (pCR-Bluntp300 Core) (Life Technologies cat. #K2700). An AscI site, HA-epitope tag, and a PmeI site were added by PCR amplification of the p300 Core from pCR-Bluntp300 Core and subsequently this amplicon was cloned into pCR-Blunt (pCR-Bluntp300 Core + HA) (Life Technologies cat. #K2700). The HA-tagged p300 Core was cloned from pCR-Bluntp300 Core + HA into the dCas9VP64 backbone via shared AscI/PmeI restriction sites. pcDNA-dCas9p300 Core (D1399Y) (dCas9p300 Core (D1399Y)) was generated by amplification of the p300 Core from dCas9p300 Core in overlapping fragments with primer sets including the specified nucleic acid mutations, with a subsequent round of linkage PCR and cloning into the dCas9p300 Core backbone using shared AscI/PmeI restriction sites. All PCR amplifications were carried out using Q5 high-fidelity DNA polymerase (NEB cat. #M0491). Protein sequences of all dCas9 constructs are shown in Supplementary Note 1.

IL1RN, MYOD, and OCT4 promoter gRNA protospacers have been described previously16, 53. Neisseria meningitidis dCas9VP64 (_Nm_-dCas9VP64) was obtained from Addgene (plasmid #48676)44. _Nm_-dCas9p300 Core was created by amplifying the HA-tagged p300 Core from dCas9p300 Core with primers to facilitate subcloning into the AleI/AgeI-digested _Nm_-dCas9VP64 backbone using isothermal assembly (NEB cat. # 2611). IL1RN TALEp300 Core TALEs were generated by subcloning the HA-tagged p300 Core domain from dCas9p300 Core into previously published24 IL1RN TALEVP64 constructs via shared AscI/PmeI restriction sites. IL1RN TALE target sites are shown in Supplementary Table 2. ICAM1 ZFVP64 and ICAM1 ZFp300 Core were constructed by subcloning the ICAM1 ZF from pMX-CD54-31Opt-VP6454 into dCas9VP64 and dCas9p300 Core backbones, respectively, using isothermal assembly (NEB cat. #2611). Protein sequences of ICAM1 ZF constructs are shown in Supplementary Note 2. Transfection efficiency was routinely above 90% as assayed by co-transfection of PL-SIN-EF1α-EGFP (Addgene plasmid #21320)55 and gRNA empty vector in all experiments. All Streptococcus pyogenes gRNAs were annealed and cloned into pZdonor-pSPgRNA (Addgene plasmid # 47108)16 for expression as described previously13 with slight modifications using NEB BbsI and T4 ligase (Cat. #s R0539 and M0202). _Nm_-dCas9 gRNA oligos were rationally designed using published PAM requirements44, and then cloned into pZDonor-Nm-Cas9-gRNA-hU6 (Addgene, plasmid #61366) via BbsI sites. Plasmids are available through Addgene (Supplementary Table 3). All gRNA protospacer targets are listed in Supplementary Table 4.

Western Blotting

20 μg of protein was loaded for SDS PAGE and transferred onto a nitrocellulose membrane for western blots. Primary antibodies (α-FLAG; Sigma-Aldrich cat. #F7425 and α-GAPDH; Cell Signaling Technology cat. #14C10) were used at a 1:1000 dilution in TBST + 5% Milk. Secondary α-Rabbit HRP (Sigma-Aldrich cat. #A6154) was used at a 1:5000 dilution in TBST + 5% Milk. Membranes were exposed after addition of ECL (Bio-Rad cat. #170-5060).

Quantitative reverse-transcription PCR

RNA was isolated from transfected cells using the RNeasy Plus mini kit (Qiagen cat. #74136) and 500 ng of purified RNA was used as template for cDNA synthesis (Life Technologies, cat. #11754). Real-time PCR was performed using PerfeCTa SYBR Green FastMix (Quanta Biosciences, cat. #95072) and a CFX96 Real-Time PCR Detection System with a C1000 Thermal Cycler (Bio-Rad). Baselines were subtracted using the baseline subtraction curve fit analysis mode and thresholds were automatically calculated using the Bio-Rad CFX Manager software version 2.1. Results are expressed as fold change above control mock transfected cells (No DNA) after normalization to GAPDH expression using the ΔΔCt method as previously described56. All qPCR primers and conditions are listed in Supplementary Table 5.

RNA-seq

RNA-seq was performed using three replicates per experimental condition. RNA was isolated from transfected cells using the RNeasy Plus mini kit (Qiagen cat. #74136) and 1 μg of purified mRNA was used as template for cDNA synthesis and library construction using the PrepX RNA-Seq Library Kit (Wafergen Biosystems, cat. #400039). Libraries were prepared using the Apollo 324 liquid handling platform, as per manufacturer’s instruction. Indexed libraries were validated for quality and size distribution using the Tapestation 2200 (Agilent) and quantified by qPCR using the KAPA Library Quantification Kit (KAPA Biosystems; KK4835) prior to multiplex pooling and sequencing at the Duke University Genome Sequencing Shared Resource facility. Libraries were pooled and then 50 bp single-end reads were sequenced on a Hiseq 2500 (Illumina), de-multiplexed and then aligned to the HG19 transcriptome using Bowtie 257. Transcript abundance was calculated using the SAMtools58 suite and differential expression was determined in R using the DESeq2 analysis package. Multiple hypothesis correction was performed using the method of Benjamini and Hochberg with a FDR of < 5%. RNA-seq data is deposited in the NCBI’s Gene Expression Omnibus and is accessible through GEO Series accession number GSE66742.

ChIP-qPCR

HEK293T cells were co-transfected with four HS2 enhancer gRNA constructs and indicated dCas9 fusion expression vectors in 15 cm plates in biological triplicate for each condition tested. Cells were cross-linked with 1% Formaldehyde (final concentration; Sigma F8775-25ML) for 10 min at RT and then the reaction was stopped by the addition of glycine to a final concentration of 125 mM. From each plate ~2.5e7 cells were used for H3K27ac ChIP-enrichment. Chromatin was sheared to a median fragment size of 250 bp using a Bioruptor XL (Diagenode). H3K27ac enrichment was performed by incubation with 5 μg of Abcam ab4729 and 200 μl of sheep anti-rabbit IgG magnetic beads (Life Technologies 11203D) for 16 hrs at 4°. Cross-links were reversed via overnight incubation at 65°C with sodium dodecyl sulfate, and DNA was purified using MinElute DNA purification columns (Qiagen). 10ng of DNA was used for subsequent qPCR reactions using a CFX96 Real-Time PCR Detection System with a C1000 Thermal Cycler (Bio-Rad). Baselines were subtracted using the baseline subtraction curve fit analysis mode and thresholds were automatically calculated using the Bio-Rad CFX Manager software version 2.1. Results are expressed as fold change above cells co-transfected with dCas9 and four HS2 gRNAs after normalization to β-actin enrichment using the ΔΔCt method as previously described56. All ChIP-qPCR primers and conditions are listed in Supplementary Table 5.

Supplementary Material

1

2

Acknowledgments

Pablo Perez-Pinera, Pratiksha Thakore, Dewran Kocak, David Ousterout, and Darrin Lim provided assistance with gRNA design, plasmid cloning, PCR primer validation, and/or RNA isolations. The gene encoding the _ICAM1_-targeted zinc finger protein was provided by Carlos Barbas, III. This work was supported by a US National Institutes of Health (NIH) grants to G.E.C., C.A.G., and T.E.R. (R01DA036865 and U01HG007900), a NIH Director’s New Innovator Award (DP2OD008586) and National Science Foundation (NSF) Faculty Early Career Development (CAREER) Award (CBET-1151035) to C.A.G., and NIH grant P30AR066527.

Footnotes

Author Contributions

I.B.H., A.M.D., C.M.V., G.E.C., T.E.R. and C.A.G. designed experiments. I.B.H., A.M.D., and C.M.V. performed the experiments. I.B.H., A.M.D., C.M.V., G.E.C., T.E.R. and C.A.G. analyzed the data. I.B.H. and C.A.G. wrote the manuscript with contributions by all authors.

Conflict of interest statement: C.A.G. and I.B.H. have filed patent applications related to genome engineering technologies. C.A.G. is a scientific advisor to Editas Medicine, a company engaged in therapeutic development of genome engineering technologies.

References

1. Dunham I, et al. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489:57–74. [PMC free article] [PubMed] [Google Scholar]

2. Bernstein BE, et al. The NIH Roadmap Epigenomics Mapping Consortium. Nat Biotechnol. 2010;28:1045–1048. [PMC free article] [PubMed] [Google Scholar]

3. Snowden AW, Gregory PD, Case CC, Pabo CO. Gene-specific targeting of H3K9 methylation is sufficient for initiating repression in vivo. Curr Biol. 2002;12:2159–2166. [PubMed] [Google Scholar]

4. Maeder ML, et al. Targeted DNA demethylation and activation of endogenous genes using programmable TALE-TET1 fusion proteins. Nat Biotechnol. 2013;31:1137–1142. [PMC free article] [PubMed] [Google Scholar]

5. Mendenhall EM, et al. Locus-specific editing of histone modifications at endogenous enhancers. Nat Biotechnol. 2013;31:1133–1136. [PMC free article] [PubMed] [Google Scholar]

6. Rivenbark AG, et al. Epigenetic reprogramming of cancer cells via targeted DNA methylation. Epigenetics. 2012;7:350–360. [PMC free article] [PubMed] [Google Scholar]

7. Konermann S, et al. Optical control of mammalian endogenous transcription and epigenetic states. Nature. 2013;500:472–476. [PMC free article] [PubMed] [Google Scholar]

8. Keung AJ, Bashor CJ, Kiriakov S, Collins JJ, Khalil AS. Using targeted chromatin regulators to engineer combinatorial and spatial transcriptional regulation. Cell. 2014;158:110–120. [PMC free article] [PubMed] [Google Scholar]

9. Hsu PD, Lander ES, Zhang F. Development and applications of CRISPR-Cas9 for genome engineering. Cell. 2014;157:1262–1278. [PMC free article] [PubMed] [Google Scholar]

10. Doudna JA, Charpentier E. Genome editing. The new frontier of genome engineering with CRISPR-Cas9. Science. 2014;346:1258096. [PubMed] [Google Scholar]

12. Jinek M, et al. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science. 2012;337:816–821. [PMC free article] [PubMed] [Google Scholar]

13. Cong L, et al. Multiplex genome engineering using CRISPR/Cas systems. Science. 2013;339:819–823. [PMC free article] [PubMed] [Google Scholar]

14. Konermann S, et al. Genome-scale transcriptional activation by an engineered CRISPR-Cas9 complex. Nature. 2015;517:583–588. [PMC free article] [PubMed] [Google Scholar]

15. Mali P, et al. CAS9 transcriptional activators for target specificity screening and paired nickases for cooperative genome engineering. Nat Biotechnol. 2013;31:833–838. [PMC free article] [PubMed] [Google Scholar]

16. Perez-Pinera P, et al. RNA-guided gene activation by CRISPR-Cas9-based transcription factors. Nature methods. 2013;10:973–976. [PMC free article] [PubMed] [Google Scholar]

17. Maeder ML, et al. CRISPR RNA-guided activation of endogenous human genes. Nat Methods. 2013;10:977–979. [PMC free article] [PubMed] [Google Scholar]

18. Qi LS, et al. Repurposing CRISPR as an RNA-guided platform for sequence-specific control of gene expression. Cell. 2013;152:1173–1183. [PMC free article] [PubMed] [Google Scholar]

19. Gilbert LA, et al. CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell. 2013;154:442–451. [PMC free article] [PubMed] [Google Scholar]

20. Cheng AW, et al. Multiplexed activation of endogenous genes by CRISPR-on, an RNA-guided transcriptional activator system. Cell Res. 2013;23:1163–1171. [PMC free article] [PubMed] [Google Scholar]

21. Farzadfard F, Perli SD, Lu TK. Tunable and multifunctional eukaryotic transcription factors based on CRISPR/Cas. ACS synthetic biology. 2013;2:604–613. [PMC free article] [PubMed] [Google Scholar]

22. Tanenbaum ME, Gilbert LA, Qi LS, Weissman JS, Vale RD. A protein-tagging system for signal amplification in gene expression and fluorescence imaging. Cell. 2014;159:635–646. [PMC free article] [PubMed] [Google Scholar]

23. Chavez A, et al. Highly efficient Cas9-mediated transcriptional programming. Nature methods. 2015 [PMC free article] [PubMed] [Google Scholar]

24. Perez-Pinera P, et al. Synergistic and tunable human gene activation by combinations of synthetic transcription factors. Nat Methods. 2013;10:239–242. [PMC free article] [PubMed] [Google Scholar]

25. Maeder ML, et al. Robust, synergistic regulation of human gene expression using TALE activators. Nat Methods. 2013;10:243–245. [PMC free article] [PubMed] [Google Scholar]

26. Beerli RR, Dreier B, Barbas CF., 3rd Positive and negative regulation of endogenous genes by designed transcription factors. Proc Natl Acad Sci U S A. 2000;97:1495–1500. [PMC free article] [PubMed] [Google Scholar]

27. Choy B, Green MR. Eukaryotic activators function during multiple steps of preinitiation complex assembly. Nature. 1993;366:531–536. [PubMed] [Google Scholar]

28. Memedula S, Belmont AS. Sequential recruitment of HAT and SWI/SNF components to condensed chromatin by VP16. Current biology: CB. 2003;13:241–246. [PubMed] [Google Scholar]

29. Delvecchio M, Gaucher J, Aguilar-Gurrieri C, Ortega E, Panne D. Structure of the p300 catalytic core and implications for chromatin targeting and HAT regulation. Nature structural & molecular biology. 2013;20:1040–1046. [PubMed] [Google Scholar]

30. Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y. The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell. 1996;87:953–959. [PubMed] [Google Scholar]

31. Chen J, Li Q. Life and death of transcriptional co-activator p300. Epigenetics: official journal of the DNA Methylation Society. 2011;6:957–961. [PubMed] [Google Scholar]

32. Gao X, et al. Comparison of TALE designer transcription factors and the CRISPR/dCas9 in regulation of gene expression by targeting enhancers. Nucleic Acids Res. 2014;42:e155. [PMC free article] [PubMed] [Google Scholar]

33. Gao X, et al. Reprogramming to Pluripotency Using Designer TALE Transcription Factors Targeting Enhancers. Stem cell reports. 2013;1:183–197. [PMC free article] [PubMed] [Google Scholar]

34. Ji Q, et al. Engineered zinc-finger transcription factors activate OCT4 (POU5F1), SOX2, KLF4, c-MYC (MYC) and miR302/367. Nucleic acids research. 2014;42:6158–6167. [PMC free article] [PubMed] [Google Scholar]

35. Deng W, et al. Reactivation of developmentally silenced globin genes by forced chromatin looping. Cell. 2014;158:849–860. [PMC free article] [PubMed] [Google Scholar]

36. Rada-Iglesias A, et al. A unique chromatin signature uncovers early developmental enhancers in humans. Nature. 2011;470:279–283. [PMC free article] [PubMed] [Google Scholar]

37. Visel A, et al. ChIP-seq accurately predicts tissue-specific activity of enhancers. Nature. 2009;457:854–858. [PMC free article] [PubMed] [Google Scholar]

38. Chen JC, Love CM, Goldhamer DJ. Two upstream enhancers collaborate to regulate the spatial patterning and timing of MyoD transcription during mouse development. Dev Dyn. 2001;221:274–288. [PubMed] [Google Scholar]

39. Nordhoff V, et al. Comparative analysis of human, bovine, and murine Oct-4 upstream promoter sequences. Mammalian genome: official journal of the International Mammalian Genome Society. 2001;12:309–317. [PubMed] [Google Scholar]

40. Carter D, Chakalova L, Osborne CS, Dai YF, Fraser P. Long-range chromatin regulatory interactions in vivo. Nature genetics. 2002;32:623–626. [PubMed] [Google Scholar]

41. Kim YW, Kim A. Histone acetylation contributes to chromatin looping between the locus control region and globin gene by influencing hypersensitive site formation. Biochimica et biophysica acta. 2013;1829:963–969. [PubMed] [Google Scholar]

42. Kuscu C, Arslan S, Singh R, Thorpe J, Adli M. Genome-wide analysis reveals characteristics of off-target sites bound by the Cas9 endonuclease. Nature biotechnology. 2014;32:677–683. [PubMed] [Google Scholar]

43. Wu X, et al. Genome-wide binding of the CRISPR endonuclease Cas9 in mammalian cells. Nature biotechnology. 2014;32:670–676. [PMC free article] [PubMed] [Google Scholar]

44. Esvelt KM, et al. Orthogonal Cas9 proteins for RNA-guided gene regulation and editing. Nat Methods. 2013;10:1116–1121. [PMC free article] [PubMed] [Google Scholar]

45. Su MY, et al. Identification of biologically relevant enhancers in human erythroid cells. The Journal of biological chemistry. 2013;288:8433–8444. [PMC free article] [PubMed] [Google Scholar]

46. Doench JG, et al. Rational design of highly active sgRNAs for CRISPR-Cas9-mediated gene inactivation. Nature biotechnology. 2014;32:1262–1267. [PMC free article] [PubMed] [Google Scholar]

47. Gilbert LA, et al. Genome-Scale CRISPR-Mediated Control of Gene Repression and Activation. Cell. 2014;159:647–661. [PMC free article] [PubMed] [Google Scholar]

48. Polstein LR, Gersbach CA. A light-inducible CRISPR-Cas9 system for control of endogenous gene activation. Nature chemical biology. 2015;11:198–200. [PMC free article] [PubMed] [Google Scholar]

49. Zetsche B, Volz SE, Zhang F. A split-Cas9 architecture for inducible genome editing and transcription modulation. Nature biotechnology. 2015;33:139–142. [PMC free article] [PubMed] [Google Scholar]

50. Chakraborty S, et al. A CRISPR/Cas9-Based System for Reprogramming Cell Lineage Specification. Stem cell reports. 2014;3:940–947. [PMC free article] [PubMed] [Google Scholar]

51. Gibson DG, et al. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nature methods. 2009;6:343–345. [PubMed] [Google Scholar]

52. Chen LF, Mu Y, Greene WC. Acetylation of RelA at discrete sites regulates distinct nuclear functions of NF-kappaB. The EMBO journal. 2002;21:6539–6548. [PMC free article] [PubMed] [Google Scholar]

53. Hu J, et al. Direct activation of human and mouse Oct4 genes using engineered TALE and Cas9 transcription factors. Nucleic Acids Res. 2014;42:4375–4390. [PMC free article] [PubMed] [Google Scholar]

54. Magnenat L, Blancafort P, Barbas CF., 3rd In vivo selection of combinatorial libraries and designed affinity maturation of polydactyl zinc finger transcription factors for ICAM-1 provides new insights into gene regulation. Journal of molecular biology. 2004;341:635–649. [PubMed] [Google Scholar]

55. Hotta A, et al. Isolation of human iPS cells using EOS lentiviral vectors to select for pluripotency. Nature methods. 2009;6:370–376. [PubMed] [Google Scholar]

56. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparative C(T) method. Nature protocols. 2008;3:1101–1108. [PubMed] [Google Scholar]

57. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nature methods. 2012;9:357–359. [PMC free article] [PubMed] [Google Scholar]