1

The new state of the art: CRISPR for gene activation and repression 1 Marie F La Russa1,2, Lei S Qi2,3,4 2 3 1Biomedical Sciences Graduate Program 4 University of California, San Francisco, CA 94143, USA 5 2Department of Bioengineering 6 3Department of Chemical and Systems Biology 7 4Stanford ChEM-H (Chemistry, Engineering & Medicine for Human Health) 8 Stanford University, Stanford, CA 94305, USA 9 *Correspondence: stanley.qi@stanford.edu 10

MCB Accepted Manuscript Posted Online 14 September 2015Mol. Cell. Biol. doi:10.1128/MCB.00512-15Copyright © 2015, American Society for Microbiology. All Rights Reserved.

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

2

Abstract 11 CRISPR-Cas9 technology has rapidly changed the landscape for how biologists and 12 bioengineers study and manipulate the genome. Derived from the bacterial adaptive 13 immune system, CRISPR-Cas9 has been co-opted and repurposed for a variety of new 14 functions including the activation or repression of gene expression (termed CRISPRa or 15 CRISPRi, respectively). This represents an exciting alternative to previously used 16 repression or activation technologies such as RNA interference (RNAi) or gene 17 overexpression vectors. We have only just begun exploring the possibilities that CRISPR 18 technology offers for gene regulation and the control of cell identity and behavior. In this 19 review, we describe the recent advances of CRISPR-Cas9 for gene regulation and outline 20 advantages and disadvantages of CRISPRa/i relative to alternative technologies. 21

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

3

Introduction 22 The ability to regulate expression is essential to the study of biology, from basic 23 biological research to clinical applications for the treatment of disease. Since the 24 elucidation of the central dogma of molecular biology, we have been searching for ways to 25 manipulate and perturb gene expression. In recent years, new technological breakthroughs 26 have provided greater precision, ease, and throughput of the manipulation of gene 27 regulation. 28 One such technology, CRISPR-Cas9 (clustered regularly interspaced palindromic 29 repeats/CRISPR-associated) has rapidly shifted the landscape for studying and 30 manipulating the genome. Repurposed from the bacterial immune system for cleaving 31 foreign DNA (1), this technology consists of the Cas9 endonuclease and a target-identifying 32 CRISPR RNA duplex made up of two RNA components: the crRNA and tracrRNA (Fig. 1A) 33 (2). These two RNAs can be engineered into a chimeric single guide RNA (sgRNA), 34 simplifying its use (3). The sgRNA basepairs with the DNA target and can be easily 35 programmed to target an 18-25 basepair (bp) sequence of interest. The only constraint is 36 that the sgRNA binding site must be adjacent to a short DNA motif, termed the protospacer 37 adjacent motif (PAM) (3, 4). In the most commonly used Cas9, derived from Streptococcus 38

pyogenes, the PAM sequence is NGG (where N is any nucleotide and G is the base guanine), 39 although NAG (where A is adenine) also functions sporadically with a lower efficiency than 40 NGG (5). NGG can be found on average every 8bp in the human genome, rendering S. 41 pyogenes Cas9 an extremely versatile genetic scissor (6). 42

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

4

This Cas9-sgRNA complex has proven to be incredibly useful as a genome-editing 43 tool. The simplicity of designing the 20-nucleotide (nt) DNA basepairing portion of an 44 sgRNA and Cas9’s natural RNA-directed endonuclease activity makes it straightforward to 45 target Cas9 to new DNA sites. Once targeted to the DNA, Cas9 will create a blunt-ended 46 double-stranded break (DSB) within the target sequence (3, 4). This DSB can be used to 47 facilitate generation of indel mutations that cause a frameshift within the coding sequence 48 of a gene due to imperfect repair by the native host DNA repair pathway (6-8). 49 Alternatively, by supplying a repair template that contains homology to the cut site, Cas9 50 can facilitate targeted integrations of precise mutations or insertions into the genome (9, 51 10). CRISPR-Cas9 has been used successfully in a wide variety of organisms, from bacteria 52 and yeast to plants and animals, both invertebrates and vertebrates (1, 8, 11-14). CRISPR’s 53 potential to expand genome engineering of previously intractable organisms cannot be 54 overstated, but will not be reviewed here as there are already several high-quality reviews 55 on the subject (2, 15-18). 56 In addition to its editing potential, the CRISPR-Cas9 system offers exciting 57 possibilities for genetic and epigenetic regulation. One strength of CRISPR technology lies 58 in the fact that it brings together DNA, RNA, and protein in a predictable and easily 59 programmable manner. This means that the Cas9-sgRNA complex can act as a scaffold to 60 recruit a broad range of effectors or markers to specific DNA sequences. It is this property 61 of CRISPR-Cas9 that has been exploited to regulate gene expression at the transcriptional 62 level, either to activate genes (CRISPRa) or repress genes (CRISPRi). Herein, we will review 63 the characteristics and use of CRISPRa/i and how these tools compare to alternative gene 64 regulation systems. 65

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

5

Repurposing CRISPR-Cas9 for Gene Activation 66 To convert Cas9 from a DNA scissor into a gene activator, it is necessary to disrupt 67 its nuclease activity. Cas9’s two nuclease domains, the RuvC and HNH domains, are 68 conserved amongst several types of nucleases and are each responsible for cutting one 69 strand of DNA upon binding (3, 4, 19). We and others have introduced mutations into these 70 two domains to create a nuclease-deactivated Cas9 (dCas9) (Fig. 1A) (3, 4, 20-22). This 71 converts the Cas9 nuclease into a generic RNA-guided DNA-binding protein. It is then 72 possible to fuse effectors directly to dCas9, which essentially transforms the dCas9-effector 73 fusion into an easily programmable artificial transcription factor when paired with a 74 target-specific sgRNA. As the RuvC and HNH domains are conserved amongst Cas9s from 75 other bacterial species, this approach provides a general strategy for repurposing 76 orthogonal Cas9s into RNA-guided DNA binding proteins. 77 One type of effector that can be fused to dCas9 is a transcriptional activator. There 78 are different forms of these dCas9-activator fusions. For example, Bikard et al. fused the ω 79 subunit of RNA polymerase to dCas9 for use in Escherichia coli (Fig. 1B) (5, 23). This fusion 80 was able to activate reporter gene expression up to 3-fold. Currently this is the only study 81 on the use of CRISPRa in bacteria, and further development and optimization is likely 82 needed before it can be broadly applied to endogenous genes. 83 In eukaryotic cells, the first generation of dCas9 activators consisted of dCas9 fused 84 to the activation domain of p65 or a VP64 activator, an engineered tetramer of the herpes 85 simplex VP16 transcriptional activator domain (Fig. 1C) (6, 24). The dCas9-VP64 fusion 86 proved more effective than the p65 fusion, and has been used more ubiquitously. A number 87

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

6

of studies have demonstrated that dCas9-VP64 is able to activate silent endogenous genes 88 and reporters, or upregulate already active genes (3, 4, 22, 24-27). This CRISPRa complex 89 functions in eukaryotic organisms such as budding yeast or mammalian cells. However, the 90 activation seen in mammalian cells is usually moderate, about 2-5 fold on average using a 91 single sgRNA. This activation can be enhanced by using multiple sgRNAs tiled across the 92 promoter (25, 27), suggesting that recruiting additional activators to the target gene 93 enhances activation. Therefore, the second generation of CRISPRa made use of strategies to 94 co-recruit multiple activators. 95 There have been several attempts to improve the direct fusion design for the second 96 generation of CRISPRa. One strategy, demonstrated by Gilbert et al. and Tanenbaum et al., 97 is to amplify activation by transforming dCas9 into a scaffold capable of recruiting many 98 copies of an activator (9, 10, 28, 29). This is done by fusing dCas9 to a tandem array of 99 peptides, called a SunTag array, which recruits many copies of the VP64 effector (Fig. 1D). 100 The recruitment strategy involves fusing VP64 to an scFv (single chain variable fragment), 101 an engineered portion of an antibody that binds to the peptide repeats in the SunTag array. 102 Compared to ~2-fold increase with dCas9-VP64 alone (30), we observed a 50-fold increase 103 at the protein level with dCas9-SunTag for endogenous genes such as the chemokine 104 receptor gene CXCR4 in human erythroleukemia K562 cells. Activating endogenous CXCR4 105 using dCas9-SunTag was sufficient to produce significant increases in cell migration. This 106 system represents a major improvement in activation efficiency, as one dCas9 can now 107 recruit up to 24 copies of the scFv-VP64 fusion protein, rather than just delivering one 108 VP64 via a dCas9-VP64 fusion. This is especially important given that simply increasing the 109

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

7

number of copies of VP16 in a direct protein fusion (i.e. using VP160) has limited 110 effectiveness (27). 111 Another strategy for CRISPR-dependent gene activation, reported by Chavez et al., 112 employs multiple different activators to synergistically amplify activation (31). The authors 113 created a tripartite effector fused to dCas9, composed of activators VP64, p65, and Rta 114 (VPR) linked in tandem (Fig. 1E). These three activators were joined in a defined order to 115 strongly activate genes. The dCas9-VPR system was successfully employed in human, 116 mouse, Drosophila melanogaster, and Saccharomyces cerevisiae cells. Additionally, it can 117 upregulate endogenous gene expression from 5- to 300-fold higher at the mRNA level than 118 a single dCas9-VP64 fusion. It should be noted that this was achieved using pools of 3-4 119 different sgRNAs per endogenous gene, which has been shown to greatly increase 120 activation for the 1st-generation dCas9-VP64 fusion (25, 27). In the future, it will be useful 121 to test additional activators and see if an even greater effect can be achieved, and also look 122 at endogenous gene activation using a single sgRNA. 123 A third approach, from Konermann et al., is termed the Synergistic Activation 124 Mediator (SAM) system (32). Like the VPR activator, the SAM system employs multiple 125 transcriptional activators to create a synergistic effect. This tool makes use of the 1st-126 generation dCas9-VP64, but the authors engineered additional features into the sgRNA to 127 enhance activator recruitment. This new sgRNA contains two copies of an RNA hairpin 128 from the MS2 bacteriophage, which interacts with the RNA-binding protein MCP (MS2 coat 129 protein) (Fig. 1F). An additional activation module was created by fusing MCP to the 130 transcriptional activator p65 as well as the activating domain of human heat-shock factor 1 131

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

8

(HSF1). MCP binds to MS2 as a dimer, so up to four additional copies of the activation 132 module can be recruited per dCas9-VP64. The SAM system can produce a wide range of 133 enhanced activation of endogenous genes at the mRNA level compared to dCas9-VP64 134 (two- to thousands-fold), depending on baseline expression. This includes both protein 135 coding genes and long noncoding RNAs (lncRNAs). In the future, it will be informative to 136 try a similar approach to repress gene expression (see below). 137 Finally, Hilton et al. were able to fuse a histone acetyltransferase to dCas9, creating a 138 dCas9-p300core fusion activator capable of acting as an epigenome editing platform (33). 139 This fusion was able to enable activation at both proximal and distal enhancers of genes. 140 This is in contrast to the dCas9-VP64 activator, which must be targeted to a promoter to 141 achieve significant gene activation. Furthermore, the dCas9-p300core fusion achieved higher 142 activation than dCas9-VP64 alone. In the future, it will be interesting to fuse other 143 epigenome modifiers to dCas9. Such tools could be used to specifically probe the effects 144 that epigenetic changes have on gene expression levels. 145 Together, these transcriptional activation systems function across a range of cell 146 types and species and provide many options for transcriptional and epigenetic 147 manipulation. Each strategy comes with its own advantages and disadvantages. For 148 example, while the VPR activator relies on fewer components, it has not yet been validated 149 for larger-scale screens like the SunTag and SAM activators. The high activation levels of 150 the VPR system depended on using a pool of 3-4 sgRNAs, making it more difficult to use 151 effectively in genome-wide screens. In addition, there may be cell type-specific efficiency or 152 toxicity issues with each of these technologies. All of these tools are relatively new, and so it 153

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

9

will be interesting to compare their efficiency and specificity directly in a range of cell types 154 and for a variety of genes. 155 Transforming dCas9 into a transcriptional repressor 156

In addition to being fused to transcriptional activators, dCas9 can also function as a 157 repressor. This was first demonstrated in bacterial cells, where dCas9 alone was able to act 158 as a transcriptional repressor by sterically hindering the transcriptional activity of RNA 159 polymerase (Fig. 1B) (20, 23). This provides a very efficient way to silence transcription in 160 bacteria, usually in the range of 1,000-fold. Repression is tunable, as the choice of sgRNA 161 site determines the strength of its repressive effect. It is also rapidly reversible using 162 inducible promoters to control expression of dCas9. This system is advantageous as genes 163 can be efficiently repressed without the addition of specific effectors, making the 164 repression system simpler and more transferable across genes, species, and cell types than 165 the activation system. 166 This steric hindrance strategy for repression has been employed in yeast and 167 mammalian cells (20, 24). While the simple dCas9 transcriptional blockade has been found 168 to work in these cells, the efficiency of repression is much lower. This is likely because the 169 binding of dCas9 to DNA is not sufficient to disrupt the action of eukaryotic RNA 170 polymerases. One strategy to improve the efficiency of repression in mammalian cells has 171 been to fuse transcriptional repressors to dCas9 (Fig. 1C) (24, 32). These repressors 172 include the KRAB (Krüppel-associated box) domain of Kox1, the CS (chromo shadow) 173 domain of HP1α, the WPRW domain of Hes1, or four concatenated copies of the mSin3 174 interaction domain (SID4X) (21, 24). Of these, the KRAB-dCas9 fusion has proven to be the 175

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

10

most effective. The most active sgRNAs can achieve repression levels in the range of 90-176 99%, although it may be necessary to screen through 5-10 sgRNAs to find 1-2 of the most 177 highly active guides (29). The efficacy of sgRNAs for CRISPRa/i may be further improved by 178 bioinformatically modeling efficacy of large pools of sgRNAs. While this has been done for 179 both CRISPR KO and CRISPRa/i, iteratively testing and analyzing shRNA efficiency as 180 described above has enormously improved the system. 181 While these strategies have proven quite effective at repressing transcription, there 182 are improvements that can be made. For example, we have found that using an N-terminal 183 KRAB fusion is more effective at repression than a C-terminal fusion (compare repression 184 in (29) to (24); also data not published). In addition, the repression system might also be 185 improved by combining several synergistic repressors in a manner similar to the activation 186 systems described above (31, 32). 187

Engineering complex regulatory patterns using CRISPRa/i 188 These CRISPRa/i systems are remarkably versatile. The activating or repressing 189 dCas9 fusions can regulate a single target or be multiplexed to regulate multiple targets at 190 once (Fig. 2A) (24, 27, 32). In mammalian cells, multiple sgRNAs can be used in the same 191 cell while still efficiently regulating any one target (24, 32). This ability can be used to 192 upregulate or downregulate multiple genes within the same pathway. 193 One exciting use of CRISPRa/i is to regulate multiple genes in multiple ways (i.e. 194 activation and repression) within a single cell. One disadvantage of the direct fusion of 195 effectors to dCas9 as described above is that only one type of perturbation can occur within 196

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

11

a given cell: either dCas9 will activate or repress genes, but not both. To work around this, 197 Zalatan et al. turned the sgRNA into a scaffold to recruit different types of effectors (30) 198 (Fig. 1G). This is made possible by fusing the effectors to RNA-binding proteins (RBPs) 199 from bacteriophages, which recognize specific RNA hairpin structures. These RNA hairpins 200 can be fused to the sgRNA, creating a scaffold RNA (scRNA). By fusing different RNA 201 hairpins to the sgRNA, different RBP-effector combinations can be recruited. Thus the 202 scRNA will encode both the target gene location (through the DNA basepairing region) and 203 the type of gene regulation (through the additional RBP-recruiting RNA hairpin). This 204 strategy has been used both in yeast and mammalian cells to regulate genes in orthogonal 205 directions simultaneously (30) (Fig. 2B). With three distinct sets of RBP-scRNA pairs, there 206 are many possibilities for synthetic biology using this technology. 207 The varied dCas9 and Cas9-mediated regulatory strategies can be combined in 208 diverse ways to create unprecedented levels of control. This is particularly relevant to 209 synthetic biology and cellular engineering studies. This has been illustrated by various 210 groups that have used dCas9 or Cas9 to create logic gates that influence cellular outcomes. 211 Zalatan et al. reprogrammed a branched metabolic pathway in yeast to control the 212 production of various product metabolites (30). Liu et al. used a nuclease Cas9 to create a 213 promoter-based “AND” logic gate to identify and control a specific type of cancer cell (34). 214 Promoter-based logic gates (e.g. AND, OR, NOT) could also be combined with the scRNA 215 components to create complex regulatory patterns that will only be induced under certain 216 conditions. Tuning defined sets of genes with such precision will allow for extraordinary 217 control over cell behavior and identity. 218

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

12

Large-scale screens with CRISPRa/i 219 In addition to its use for multiplexed regulation, the CRISPR-Cas9 system can be 220 adapted for use in pooled genetic screens to interrogate the function of many genes at once. 221 This was first shown by several groups using nuclease-active Cas9 for genome-wide 222 knockout (KO) screens in mammalian cells (35-37). These groups were able to pool sets of 223 tens of thousands of sgRNAs, with a coverage of ~3-10 sgRNAs per gene, to investigate a 224 range of phenotypes from cell growth to drug resistance to host factors influencing viral 225 susceptibility. While these have been powerful demonstrations of CRISPR-Cas9 technology, 226 we have been able to broaden our ability to perform genome-wide screens by adapting 227 CRISPRa/i to screening technology as well. 228 Recently, both CRISPRa and CRISPRi have been employed in pooled genetic screens 229 in mammalian cells (29, 32). This is a particularly revolutionary technique for gene 230 activation, as CRISPRa overcomes limitations of previous gene overexpression methods 231 (discussed further below). The CRISPR KO and CRISPRa/i screening systems can be 232 complementary to each other, as they each can enrich for different sets of genes 233 responsible for a certain phenotype. For example, genes that are most highly enriched with 234 CRISPRa are likely those that “drop out” of a CRISPRi or KO screen and vice-versa. Since 235 there is less sensitivity and more noise in sgRNAs that drop out of a screen, up- and down- 236 regulation may offer complementary sensitivities [see examples in refs (29) and (36) vs. 237 (32)], although this is not necessarily always true. 238 It will be important to carefully consider all of these aspects when designing 239 genome-wide functional screens. Investigators who wish to thoroughly probe as many 240

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

13

genes as possible involved in a given process may need to perform 2-3 different types of 241 CRISPR screens. 242 Specificity of CRISPRa/i 243

While there have been several studies examining the off-target effects for the 244 nuclease version of Cas9, investigation into the specificity of the CRISPRa/i system is still in 245 its nascent stages. One study used RNA sequencing in cells expressing an sgRNA targeting 246 an exogenously-added GFP, compared to a non-targeting sgRNA control (24). While GFP 247 was the only gene which was significantly repressed genome-wide, only a single sgRNA 248 was investigated. Other studies, using both the first generation dCas9-VP64 and the second 249 generation SAM system, have used a similar technique to show the specificity of gene 250 activation using CRISPRa (32, 33, 38). 251 It is difficult to directly compare nuclease Cas9 and dCas9-effectors, since the 252 readout of activity is different between the two. However, there are some indications that 253 dCas9-effector function may be more sensitive to mismatches (and thus less prone to off-254 target effects) than Cas9. This is supported by work from Gilbert et al., where dCas9-KRAB 255 and Cas9 nuclease were tested for their ability to function with sgRNAs containing 256 mismatches to a given target site (29). In this study, sgRNAs containing 1-5bp mismatches 257 were systematically tested for activity and normalized to fully on-target sgRNAs for both 258 nuclease Cas9 and dCas9-KRAB. Across the entire panel of sgRNAs tested, dCas9-KRAB 259 repression was more affected by mismatches than nuclease Cas9 cleavage. In the future, it 260 will be important to continue comparing dCas9-effectors with Cas9 nuclease, so that we 261 can better assess which technology to use for a given application. 262

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

14

In one genome-wide ChIP-seq study by Polstein et al., an HA-tagged dCas9 was 263 found to bind quite specifically (between 4 and 31 off-target sites) (39). The authors also 264 correlated this with dCas9-VP64 activation as assayed by RNA-seq, and found that the 265 dCas9-VP64 activation was also quite specific. This is in contrast to two genome-wide ChIP-266 seq studies by Kuscu et al. and Wu et al. using an HA-tagged dCas9, where dCas9 was found 267 to bind promiscuously (up to 100s to 1000s of off-target sites) to sequences matching the 268 “seed” region in the sgRNA portion adjacent to the PAM (40, 41). Interestingly, nuclease 269 Cas9 using the same sgRNAs was quite specific, only cleaving the few off-target sites with 270 extensive basepairing between the sgRNA and off-target DNA. 271 One possible explanation for the apparent promiscuity of dCas9 in Kuscu et al. and 272 Wu et al. could be that while dCas9 and Cas9 interrogate many sites transiently, prolonged 273 interactions only occur with extensive matching between the sgRNA and target DNA (42). 274 Transient interrogations would be captured in the above assays, since formaldehyde 275 crosslinking was used to fix samples. If the three papers used different fixation protocols 276 and peak-calling or thresholding methods to process the ChIP-seq data, they could 277 conceivably get quite different results. The studies above are consistent with the model 278 that Cas9 is only fully functional as a nuclease after extensive basepairing between the 279 sgRNA and target DNA (43). The specificity of nuclease Cas9 function due to the necessity 280 of prolonged binding may also apply to dCas9-effectors, especially those that need to 281 recruit other proteins to be active. More work needs to be done to continue probing dCas9 282 off-target binding and effector function. 283

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

15

Altogether, while research on dCas9-effector specificity is still in its early stages, the 284 initial studies are quite promising for the future use of CRISPRa/i. In the future, it will be 285 important to continue to probe for off-target effects for a wide variety of sgRNAs and for all 286 of the new types of dCas9-effector combinations, as different mechanisms of action for each 287 effector may result in different levels of functional promiscuity. With greater knowledge 288 about off-target effects from the use of CRISPRa/i, we can better judge the rate of false 289 positives when using these technologies for screens. Furthermore, such knowledge can be 290 used to inform us what level of sgRNA coverage per gene is needed to extract the maximum 291 amount of information out of as small a library as possible, which is particularly important 292 when working with systems where it is difficult to scale up the number of cells used. 293 CRISPRa vs. previous activation methods 294

CRISPRa offers many advantages over alternative gene overexpression or activation 295 methods (Table 1). One technique to overexpress genes is to clone the open reading frame 296 (ORF) or cDNA of the gene of interest [reviewed in (44)]. For longer or GC-rich genes, this 297 alone can be technically difficult. If cloning many genes at once using this method, there 298 will be a bias towards smaller and easier-to-amplify genes. Additionally, when cloning from 299 the cDNA, one may be missing physiologically relevant splice variants. However, there are 300 some applications where it is more suitable to use ORF overexpression constructs. For 301 example, it is possible to overexpress a certain splice variant or a mutant version of an 302 allele (44). This is particularly useful when studying disease-associated mutations or 303 variants. Additionally, ORF overexpression is useful for introducing fluorophore- or 304

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

16

peptide-tagged versions of proteins, which can be invaluable for tracking the location or 305 binding partners of the given protein when no good antibody is available (45). 306 Another, alternative approach is to use other engineered transcriptional activators 307 to turn on gene expression. These include Zinc Finger (ZF) effectors (46-48) and 308 Transcriptional Activator-Like Effectors (TALEs) (38, 49, 50) fused to a similar set of 309 activation domains as described above for CRISPRa. As with dCas9 activators, ZF and TALE 310 activators target specific sequences of DNA and recruit transcriptional machinery to 311 activate transcription. However, Cas9 finds its DNA target through simple Watson-Crick 312 basepairing interactions between the sgRNA and target DNA, whereas ZFs and TALEs rely 313 on protein-DNA interactions. Typical engineered zinc finger tandem arrays possess 314 between 3 and 6 individual zinc finger motifs for binding target DNA ranging from 9 to 18 315 bp in length (47). An individual zinc finger motif targets 3-4 nucleotides, and creating a 316 composite ZF protein to target novel DNA sites can require much engineering and testing 317 (51). TALEs are much easier to program than ZFs. A TALE contains a series of Repeat 318 Variable Domains (RVDs), each targeting a single DNA nucleotide. The RVD code for all 319 DNA nucleotides has been well characterized (52). Thus a TALE can easily be programmed 320 by creating an array of RVDs that bind to the nucleotides in the target DNA. 321 While there are limited reports directly comparing ZF, TALE, and CRISPR activators, 322 they seem to achieve similar levels of activation (31, 33, 39). The main disadvantage of 323 TALEs and ZF effectors compared to CRISPR-Cas9 is that they require more complicated 324 cloning to assemble, making them less user-friendly overall and not amenable to genome-325 wide screens. However, TALEs in particular can be advantageous when a particular target 326

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

17

area lacks a PAM, since TALEs can be programmed to target any sequence. TALEs are able 327 to distinguish between methylcytosine (mC) and cytosine (C), which may be advantageous 328 or disadvantageous, depending on the situation (53-55). The methylation status of the 329 targeted site must be taken into account when designing TALE effectors, which is not the 330 case for Cas9 targeting. Finally, some of the CRISPRa systems require many components, 331 making them potentially much more complicated to deliver than the one-component ZF or 332 TALE activators. These characteristics must all be taken into account when deciding which 333 activation system to use. 334 CRISPRi vs. previous repression methods 335

CRISPRi also offers several advantages over previous forms of gene repression, 336 although it is not necessarily the best choice for every assay (Table 2). The most widely-337 used repression tool predating CRISPRi is RNAi technology (56). RNAi can refer to either to 338 siRNAs or shRNAs, which make use of the cell’s endogenous siRNA and miRNA pathways 339 for processing and function (57). These pathways are naturally-occurring cellular 340 processes used to regulate mRNA translation and stability, and both employ small (~18-24 341 nt) double-stranded RNAs (dsRNA). During siRNA and miRNA processing, the precursor 342 dsRNA is loaded into the RNA-induced silencing complex (RISC), where one strand of the 343 RNA is degraded, while the other (the guide strand) remains. The mature RISC complex is 344 then guided to its target mRNA, where it causes repression in various ways [reviewed in 345 more detail in references (58, 59)]. 346 When the guide strand shares perfect complementarity with its mRNA target, the 347 target is directed down the siRNA pathway and is degraded through endonucleolytic 348

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

18

cleavage (58). When there is imperfect complementation between the guide and the target 349 mRNA, but a match in the guide strand’s seed region (nucleotides 2-8), the mRNA is 350 directed down the miRNA pathway (60). Repression through the miRNA pathway can 351 cause degradation by mRNA deadenylation or suppression of translation initiation or 352 elongation (58, 59). 353 Since RNAi makes use of the cell’s endogenous small RNA processing machinery, 354 only a single, small component must be added to achieve repression: the shRNA or siRNA. 355 This may make RNAi more advantageous in some systems, as other repression mechanisms 356 may be more cumbersome to deliver. Some caution must still be taken, as there is evidence 357 that hijacking the RISC complex using RNAi may prevent endogenous miRNAs from 358 functioning (61). 359 RNAi-based screens have revolutionized our ability test the function of many genes 360 at once through pooled, genome-wide knockdown screens. However, there are many 361 concerns about off-target effects and reproducibility. For example, meta-analyses of RNAi 362 screens have shown little reproducibility between different studies (62, 63). This is in large 363 part due to the prevalence of RNAi off-target effects. siRNAs and shRNAs act through a 364 miRNA-like mechanism when the seed region of the dsRNA binds to an mRNA (64, 65). 365 Hundreds of transcripts with a given seed may exist, and there is not currently a high-366 confidence method to rank these alternative targets for likelihood of being a true off-target 367 (60). It is possible that a large number of RNAi screen “hits” are due to off-target effects 368 from such a mechanism. While this can be accounted for, it requires testing with mutated 369

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

19

versions of the siRNA or shRNA and looking for a matching phenotype, which can be 370 burdensome (66). 371 One way to reduce the rates of false negatives and false positives is to have many 372 unique shRNAs (or sgRNAs in the case of CRISPR) that target each gene in the library pool,. 373 To find the ideal shRNA or sgRNA coverage and design, multiple studies have performed 374 small-scale pooled screens with very high coverage of shRNAs or sgRNAs per gene (29, 67-375 69). These studies each identified high-confidence hits and then computationally 376 subdivided their library to 1) discover the number of sgRNAs required to distinguish true 377 hits from background and 2) define rules that make more effective shRNAs/sgRNAs. Thus, 378 bioinformatic modeling and iterative analysis and testing large pools of sgRNAs have made 379 significant improvements in both RNAi and CRISPRi (29, 68). It is now possible to use a 380 library with 10 shRNAs or 10 sgRNAs per gene to produce robust results in a screen, 381 although both RNAi and CRISPRi will no doubt benefit from further refinement in the 382 future. 383 384 Ultimately, the choice of whether to use CRISPRi or RNAi will depend on the 385 requirements of a given user. For small-scale use targeting only a few genes, CRISPRi has 386 simpler design rules can achieve very high levels of knockdown (29). However, RNAi can be 387 advantageous in that one can target specific splice variants over others, which is not 388 possible with CRISPRi unless the different variants have different transcription start sites. 389 Additionally, it has been shown that off-target effects from siRNAs can result in cell toxicity 390 in a cell-type dependent manner (70). This has not yet been seen with CRISPRi, although it 391 has not been systematically investigated. It may be certain cell types will tolerate RNAi or 392

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

20

CRISPRi better, which will need to be determined empirically. Finally, CRISPRi is a two-393 component system, whereas RNAi is a one-component system. In assays where delivering 394 two components may be an issue, it may be more desirable to use RNAi. 395 In the study of noncoding RNAs (ncRNAs), CRISPRi also offers many advantages. 396 Noncoding RNAs such as microRNAs (miRNAs) and lncRNAs can be targeted by CRISPRi in 397 the same manner as coding genes (29, 71). Since many miRNAs are redundant, one can 398 potentially make use of the multiplexing capability of CRISPRi to hit all miRNAs in the same 399 targeting “family” at once. One alternative approach to CRISPRi is antagomir miRNA 400 inhibitors, which are modified oligonucleotides that are antisense to the target miRNA (72, 401 73). They bind to the miRNA with high affinity and prevent it from acting on its target 402 mRNA. However, these miRNA inhibitors can be expensive, and are specific for a single 403 miRNA. An alternative approach is to create miRNA “sponges”, an array of tandem repeats 404 of miRNA seed sequences, which act by sequestering active miRNAs and preventing them 405 from acting on their true targets (74). Since sponges consist of an array of repeats, they can 406 be difficult to synthesize or clone. However, they may be the better choice if a user wants to 407 repress all miRNAs of the same family, which share the same seed. 408

Conclusions 409 The present is an exciting time for biologists, bioengineers, and clinicians – anyone 410 who has an interest in the effect of genes and how to control them. CRISPR is ushering us 411 into an unprecedented era of biological control. Our toolkit keeps expanding, with each 412 new addition bringing greater precision and power. While we have advanced so much in 413 such a short time, much work remains to be done to continuously refine this technology. 414

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

21

We have only just begun to use these tools, and many basic technical, biological, and 415 biomedical questions remain. With CRISPRa/i, we have a powerful new means of 416 answering them. 417

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

22

Acknowledgements 418 The authors thank the members of the Qi lab for helpful discussions and critical 419 reading of this manuscript, as well as Gabriel McKinsey, Nicole Stone, and Osama Ahmed. 420 M.L acknowledges support from the Genentech Predoctoral Fellowship. L.S.Q. 421 acknowledges support from the NIH Office of the Director (OD), and National Institute of 422 Dental & Craniofacial Research (NIDCR). This work was supported by NIH R01 DA036858 423 and NIH DP5 OD017887. 424

425

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

23

Figure Legends 426 Figure 1. Engineered CRISPRa/i systems. (A) In the endogenous Type II-B CRISPR systems 427 like that of S. pyogenes (left), there are three essential components for nuclease activity. 428 These components are the Cas9 protein, and the crRNA and tracrRNA, which basepair with 429 each other. This Cas9-RNA complex is able to cleave DNA targeted by the crRNA and which 430 is adjacent to a PAM site (red). Cleavage sites are indicated by red “X”s. To simplify and 431 adapt CRISPR for gene regulation (right), mutations in the nuclease domains have been 432 introduced into the Cas9, rendering it a dCas9. Additionally, the crRNA and tracrRNA have 433 been combined into an sgRNA. (B) To control gene expression in bacterial cells, dCas9 can 434 be fused with the ω subunit of RNA polymerase for activation (left), or can repress 435 transcription by sterically blocking RNA polymerase (right). (C) To turn dCas9 into an 436 artificial transcription factor in mammalian cells, it can be fused with a VP64 activator (left) 437 or a KRAB repressor (right). (D) The SunTag activation system (left) consists of dCas9 438 fused to several tandem repeats of a short peptide sequence separated by linkers. The 439 SunTag activator module (right) is an scFv, which specifically binds the SunTag peptide. 440 The scFv is fused to sfGFP and VP64. (E) The VPR activation system is dCas9 fused to 441 VP64, p65, and Rta linked in tandem. (F) In the SAM activation system (left), dCas9 is fused 442 to VP64. In addition, the sgRNA has been modified such that it contains two MS2 hairpins 443 (green). An additional activator module (right) binds to an MS2 hairpin via the RNA-444 binding protein MCP. The MCP is fused to the activators p65 and HSF1. (G) The scRNA 445 system can be adapted such that it can act as an activator or repressor (left). The activator 446 and repressor modules (right) consist of an RNA-binding protein fused to VP64 or KRAB, 447 respectively. The activating and repressing systems can be used orthogonally when 448

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

24

different scRNAs that recruit different modules are used. Here, the MS2 scRNA recruits the 449 MCP activator module and the COM scRNA recruits the COM repressor module. 450 Figure 2. Multiplexed gene regulation with CRISPR. (A) A dCas9-effector fusion is capable 451 of perturbing expression of multiple genes simultaneously. In the example here, multiple 452 genes are repressed simultaneously. (B) With the scRNA system, it is possible to use dCas9 453 to activate and repress multiple distinct genes simultaneously. The diversity of function 454 arises from the different RNA hairpins attached to the sgRNA, which recruit orthogonal 455 effectors. Here, the COM RNA hairpin (purple) recruits a repressive effector module and 456 the MS2 RNA hairpin (green) recruits an activating effector module. 457 458

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

25

Table References 459 Table 1 460 CRISPRa: (3, 4, 24, 26, 29, 30, 32, 75). ORF Overexpression: (44, 76-79). TALE or ZF 461 activators: (39, 49, 51, 53-55, 80-82). 462 463 Table 2 464 CRISPRi: (3, 4, 20, 24, 26, 29, 30, 71). RNAi: (56, 61, 62, 64, 68). TALE or ZF repressors: (39, 465 49, 51, 80-82). miRNA sponges or antagomirs: (59, 60, 72-74). 466

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

26

References 467 1. Horvath P, Barrangou R. 2010. CRISPR/Cas, the Immune System of Bacteria and 468 Archaea. Science 327:167–170. 469 2. Deltcheva E, Chylinski K, Sharma CM, Gonzales K, Chao Y, Pirzada ZA, Eckert MR, 470

Vogel J, Charpentier E. 2012. CRISPR RNA maturation by trans-encoded small RNA 471 and host factor RNase III. Nature 471:602–607. 472 3. Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E. 2012. A 473 Programmable Dual-RNA-Guided DNA Endonuclease in Adaptive Bacterial Immunity. 474 Science 337:816–821. 475 4. Gasiunas G, Barrangou R, Horvath P, Siksnys V. 2012. Cas9-crRNA 476 ribonucleoprotein complex mediates specific DNA cleavage for adaptive immunity in 477 bacteria. Proceedings of the National Academy of Sciences 109:E2579–86. 478 5. Hsu PD, Scott DA, Weinstein JA, Ran FA, Konermann S, Agarwala V, Li Y, Fine EJ, 479

Wu X, Shalem O, Cradick TJ, Marraffini LA, Bao G, Zhang F. 2013. DNA targeting 480 specificity of RNA-guided Cas9 nucleases. Nat Biotechnol 31:827–832. 481 6. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, 482

Marraffini LA, Zhang F. 2013. Multiplex Genome Engineering Using CRISPR/Cas 483 Systems. Science 339:819–823. 484 7. Cho SW, Kim S, Kim JM, Kim J-S. 2013. Targeted genome engineering in human cells 485 with the Cas9 RNA-guided endonuclease. Nat Biotechnol 31:230–232. 486

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

27

8. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, Norville JE, Church GM. 487 2013. RNA-Guided Human Genome Engineering via Cas9. Science 339:823–826. 488 9. Yang H, Wang H, Shivalila CS, Cheng AW, Shi L, Jaenisch R. 2013. One-Step 489 Generation of Mice Carrying Reporter and Conditional Alleles by CRISPR/Cas-490 Mediated Genome Engineering. Cell 154:1370–1379. 491 10. Hou Z, Zhang Y, Propson NE, Howden SE, Chu L-F, Sontheimer EJ, Thomson JA. 492 2013. Efficient genome engineering in human pluripotent stem cells using Cas9 from 493 Neisseria meningitidis 110:15644–15649. 494 11. Jiang W, Cox D, Zhang F, Bikard D, Marraffini LA. 2013. rNA-guided editing of 495 bacterial genomes using crisPr-cas systems. Nat Biotechnol 31:233–239. 496 12. DiCarlo JE, Norville JE, Mali P, Rios X, Aach J, Church GM. 2013. Genome 497 engineering in Saccharomyces cerevisiae using CRISPR-Cas systems. Nucleic Acids 498 Research 41:4336–4343. 499 13. Bortesi L, Fischer R. 2015. The CRISPR/Cas9 system for plant genome editing and 500 beyond. Biotechnol Adv 33:41–52. 501 14. Gratz SJ, Cummings AM, Nguyen JN, Hamm DC, Donohue LK, Harrison MM, 502

Wildonger J, O'Connor-Giles KM. 2013. Genome engineering of Drosophila with the 503 CRISPR RNA-guided Cas9 nuclease. Genetics 194:1029–1035. 504 15. Mali P, Esvelt KM, Church GM. 2013. Cas9 as a versatile tool for engineering 505 biology. Nat Meth 10:957–963. 506

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

28

16. Hsu PD, Lander ES, Zhang F. 2014. Development and Applications of CRISPR-Cas9 507 for Genome Engineering. Cell 157:1262–1278. 508 17. Cox DBT, Platt RJ, Zhang F. 2015. Therapeutic genome editing: prospects and 509 challenges. Nature Publishing Group 21:121–131. 510 18. Doudna JA, Charpentier E. 2014. The new frontier of genome engineering with 511 CRISPR-Cas9. Science 346:1258096–1258096. 512 19. Sapranauskas R, Gasiunas G, Fremaux C, Barrangou R, Horvath P, Siksnys V. 513 2011. The Streptococcus thermophilus CRISPR/Cas system provides immunity in 514 Escherichia coli. Nucleic Acids Research 39:9275–9282. 515 20. Qi LS, Larson MH, Gilbert LA, Doudna JA, Weissman JS, Arkin AP, Lim WA. 2013. 516 Repurposing CRISPR as an RNA-Guided Platform for Sequence-SpecificControl of 517 Gene Expression. Cell 152:1173–1183. 518 21. Konermann S, Brigham MD, Trevino AE, Hsu PD, Heidenreich M, Cong L, Platt RJ, 519

Scott DA, Church GM, Zhang F. 2013. Optical control of mammalian endogenous 520 transcription and epigenetic states. Nature 500:472–476. 521 22. Mali P, Aach J, Stranges PB, Esvelt KM, Moosburner M, Kosuri S, Yang L, Church 522

GM. 2013. CAS9 transcriptional activators for target specificity screening and paired 523 nickases for cooperative genome engineering. Nat Biotechnol 31:833–838. 524 23. Bikard D, Jiang W, Samai P, Hochschild A, Zhang F, Marraffini LA. 2013. 525 Programmable repression and activation of bacterial gene expression using an 526

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

29

engineered CRISPR-Cas system. Nucleic Acids Research 41:7429–7437. 527 24. Gilbert LA, Larson MH, Morsut L, Liu Z, Brar GA, Torres SE, Stern-Ginossar N, 528

Brandman O, Whitehead EH, Doudna JA, Lim WA, Weissman JS, Qi LS. 2013. 529 CRISPR-Mediated Modular RNA-Guided Regulationof Transcription in Eukaryotes. 530 Cell 154:442–451. 531 25. Maeder ML, Linder SJ, Cascio VM, Fu Y, Ho QH, Joung JK. 2013. CRISPR RNA–532 guided activation of endogenous human genes. Nat Meth 10:977–981. 533 26. Perez-Pinera P, Kocak DD, Vockley CM, Adler AF, Kabadi AM, Polstein LR, 534

Thakore PI, Glass KA, Ousterout DG, Leong KW, Guilak F, Crawford GE, Reddy 535 TE, Gersbach CA. 2013. RNA-guided gene activation by CRISPR-Cas9–based 536 transcription factors. Nat Meth 10:973–976. 537

27. Cheng AW, Wang H, Yang H, Shi L, Katz Y, Theunissen TW, Rangarajan S, 538 Shivalila CS, Dadon DB, Jaenisch R. 2013. Multiplexed activation of endogenous 539 genes by CRISPR-on, an RNA-guided transcriptional activator system. Cell Res 540 23:1163–1171. 541

28. Tanenbaum ME, Gilbert LA, Qi LS, Weissman JS, Vale RD. 2014. A Protein-Tagging 542 System for Signal Amplification in Gene Expression and Fluorescence Imaging. Cell 543 159:635–646. 544

29. Gilbert LA, Horlbeck MA, Adamson B, Villalta JE, Chen Y, Whitehead EH, 545 Guimaraes C, Panning B, Ploegh HL, Bassik MC, Qi LS, Kampmann M, Weissman 546

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

30

JS. 2014. Genome-Scale CRISPR-Mediated Control of Gene Repression and Activation. 547 Cell 159:647–661. 548 30. Zalatan JG, Lee ME, Almeida R, Gilbert LA, Whitehead EH, LaRussa M, Tsai JC, 549

Weissman JS, Dueber JE, Qi LS, Lim WA. 2015. Engineering Complex Synthetic 550 Transcriptional Programs with CRISPR RNA Scaffolds. Cell 160:339–350. 551 31. Chavez A, Scheiman J, Vora S, Pruitt BW, Tuttle M, Iyer EPR, Lin S, Kiani S, 552

Guzman CD, Wiegand DJ, Ter-Ovanesyan D, Braff JL, Davidsohn N, Housden BE, 553 Perrimon N, Weiss R, Aach J, Collins JJ, Church GM. 2015. Highly efficient Cas9- 554 mediated transcriptional programming. Nat Meth 1–5. 555

32. Konermann S, Brigham MD, Trevino AE, Joung J, Abudayyeh OO, Barcena C, Hsu 556 PD, Habib N, Gootenberg JS, Nishimasu H, Nureki O, Zhang F. 2015. Genome-scale 557 transcriptional activation by an engineered CRISPR-Cas9 complex. Nature 517:583–558 588. 559

33. Hilton IB, D'Ippolito AM, Vockley CM, Thakore PI, Crawford GE, Reddy TE, 560 Gersbach CA. 2015. epigenome editing by a crisPr-cas9-based acetyltransferase 561 activates genes from promoters and enhancers. Nat Biotechnol 1–10. 562

34. Liu Y, Zeng Y, Liu L, Zhuang C, Fu X, Huang W, Cai Z. 2014. Synthesizing AND gate 563 genetic circuits based on CRISPR-Cas9 for identification of bladder cancer cells. Nat 564 Commun 5:5393. 565 35. Wang T, Wei JJ, Sabatini DM, Lander ES. 2013. Genetic Screens in Human Cells 566

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

31

Using the CRISPR/Cas9 System. Science 343:80–84. 567 36. Shalem O, Sanjana NE, Hartenian E, Shi X, Scott DA, Mikkelsen TS, Heckl D, Ebert 568

BL, Root DE, Doench JG, Zhang F. 2013. Genome-Scale CRISPR-Cas9 Knockout 569 Screening in Human Cells. Science 343:84–87. 570 37. Zhou Y, Zhu S, Cai C, Yuan P, Li C, Huang Y, Wei W. 2014. High-throughput 571 screening of a CRISPR/Cas9 library for functional genomics in human cells. Nature 1–572 16. 573 38. Perez-Pinera P, Ousterout DG, Brunger JM, Farin AM, Glass KA, Guilak F, 574

Crawford GE, Hartemink AJ, Gersbach CA. 2013. Synergistic and tunable human 575 gene activation by combinations of synthetic transcription factors. Nat Meth 10:239–576 242. 577 39. Polstein L, Perez-Pinera P, Kocak D, Vockley C, Bledsoe P, Song L, Safi A, 578

Crawford G, Reddy T, Gersbach C. 2015. Genome-wide specificity of DNA-binding, 579 gene regulation, and chromatin remodeling by TALE- and CRISPR/Cas9-based 580 transcriptional activators. Genome Research. 581 40. Wu X, Scott DA, Kriz AJ, Chiu AC, Hsu PD, Dadon DB, Cheng AW, Trevino AE, 582

Konermann S, Chen S, Jaenisch R, Zhang F, Sharp PA. 2014. Genome-wide binding 583 of the CRISPR endonuclease Cas9 in mammalian cells. Nat Biotechnol. 584 41. Kuscu C, Arslan S, Singh R, Thorpe J, Adli M. 2014. Genome-wide analysis reveals 585 characteristics of off-target sites bound by the cas9 endonuclease. Nat Biotechnol 1–586

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

32

9. 587 42. Sternberg SH, Redding S, Jinek M, Greene EC, Doudna JA. 2014. DNA interrogation 588 by the CRISPRRNA-guided endonuclease Cas9. Nature 1–17. 589 43. Jiang F, Zhou K, Ma L, Gressel S, Doudna JA. 2015. A Cas9-guide RNA complex 590 preorganized for target DNA recognition. Science 348:1477–1481. 591 44. Prelich G. 2012. Gene Overexpression: Uses, Mechanisms, and Interpretation. 592 Genetics 190:841–854. 593 45. Conrad C, Erfle H, Warnat P, Daigle N, Lörch T, Ellenberg J, Pepperkok R, Eils R. 594 2004. Automatic identification of subcellular phenotypes on human cell arrays. 595 Genome Research 14:1130–1136. 596 46. Beerli RR, Segal DJ, Dreier B, Barbas CF. 1998. Toward controlling gene expression 597 at will: specific regulation of the erbB-2/HER-2 promoter by using polydactyl zinc 598 finger proteins constructed from modular building blocks. Proc Natl Acad Sci USA 599

95:14628–14633. 600 47. Beerli RR, Dreier B, Barbas CF. 2000. Positive and negative regulation of 601 endogenous genes by designed transcription factors. Proc Natl Acad Sci USA 602

97:1495–1500. 603 48. Blancafort P. 2004. Designing Transcription Factor Architectures for Drug 604 Discovery. Molecular Pharmacology 66:1361–1371. 605 49. Zhang F, Cong L, Lodato S, Kosuri S, Church GM, Arlotta P. 2011. Efficient 606

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

33

construction of sequence-specific TAL effectors for modulating mammalian 607 transcription. Nat Biotechnol 29:149–153. 608 50. Maeder ML, Linder SJ, Reyon D, Angstman JF, Fu Y, Sander JD, Joung JK. 2013. 609 Robust, synergistic regulation of human gene expression using TALE activators. Nat 610 Meth 10:243–245. 611 51. Choo Y, Isalan M. 2000. Advances in zinc finger engineering. Curr Opin Struct Biol 612

10:411–416. 613 52. Yang J, Zhang Y, Yuan P, Zhou Y, Cai C, Ren Q, Wen D, Chu C, Qi H, Wei W. 2014. 614 Complete decoding of TAL effectors for DNA recognition. Cell Res 24:628–631. 615 53. Deng D, Yin P, Yan C, Pan X, Gong X, Qi S, Xie T, Mahfouz M, Zhu J-K, Yan N, Shi Y. 616 2012. Recognition of methylated DNA by TAL effectors. Cell Res 22:1502–1504. 617 54. Bultmann S, Morbitzer R, Schmidt CS, Thanisch K, Spada F, Elsaesser J, Lahaye T, 618

Leonhardt H. 2012. Targeted transcriptional activation of silent oct4 pluripotency 619 gene by combining designer TALEs and inhibition of epigenetic modifiers. Nucleic 620 Acids Research 40:5368–5377. 621 55. Valton J, Dupuy A, Daboussi F, Thomas S, Maréchal A, Macmaster R, Melliand K, 622

Juillerat A, Duchateau P. 2012. Overcoming transcription activator-like effector 623 (TALE) DNA binding domain sensitivity to cytosine methylation. Journal of Biological 624 Chemistry 287:38427–38432. 625 56. Mello CC, Conte D. 2004. Revealing the world of RNA interference. Nature 431:338–626

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

34

342. 627 57. Carthew RW, Sontheimer EJ. 2009. Origins and Mechanisms of miRNAs and siRNAs. 628 Cell 136:642–655. 629 58. Hutvagner G, Simard MJ. 2008. Argonaute proteins: key players in RNA silencing. 630 Nature Reviews Molecular Cell Biology 9:22–32. 631 59. Ha M, Kim VN. 2014. Regulation of microRNA biogenesis. Nature Reviews Molecular 632 Cell Biology 15:509–524. 633 60. Bartel DP. 2009. MicroRNAs: Target Recognition and Regulatory Functions. Cell 634

136:215–233. 635 61. Khan AA, Betel D, Miller ML, Sander C, Leslie CS, Marks DS. 2009. Transfection of 636 small RNAs globally perturbs gene regulation by endogenous microRNAs. Nat 637 Biotechnol. 638 62. Sigoillot FD, Lyman S, Huckins JF, Adamson B, Chung E, Quattrochi B, King RW. 639 2012. A bioinformatics method identifies prominent off-targeted transcripts in RNAi 640 screens. Nat Meth 9:363–366. 641 63. Bhinder B, Djaballah H. 2013. Systematic analysis of RNAi reports identifies dismal 642 commonality at gene-level & reveals an unprecedented enrichment in pooled shRNA 643 screens. Combinatorial chemistry & high throughput screening 16:665. 644 64. Birmingham A, Anderson EM, Reynolds A, Ilsley-Tyree D, Leake D, Fedorov Y, 645

Baskerville S, Maksimova E, Robinson K, Karpilow J, Marshall WS, Khvorova A. 646

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

35

2006. 3′ UTR seed matches, but not overall identity, are associated with RNAi off-647 targets. Nat Meth 3:199–204. 648 65. Franceschini A, Meier R, Casanova A, Kreibich S, Daga N, Andritschke D, Dilling 649

S, Ramo P, Emmenlauer M, Kaufmann A, Conde-Alvarez R, Low SH, Pelkmans L, 650 Helenius A, Hardt WD, Dehio C, Mering von C. 2014. Specific inhibition of diverse 651 pathogens in human cells by synthetic microRNA-like oligonucleotides inferred from 652 RNAi screens. Proceedings of the National Academy of Sciences 111:4548–4553. 653

66. Buehler E, Chen Y-C, Martin S. 2012. C911: A bench-level control for sequence 654 specific siRNA off-target effects. PLoS ONE 7:e51942. 655 67. Bassik MC, Kampmann M, Lebbink RJ, Wang S, Hein MY, Poser I, Weibezahn J, 656

Horlbeck MA, Chen S, Mann M, Hyman AA, Leproust EM, McManus MT, 657 Weissman JS. 2013. A systematic mammalian genetic interaction map reveals 658 pathways underlying ricin susceptibility. Cell 152:909–922. 659

68. Kampmann M, Horlbeck MA, Chen Y, Tsai JC, Bassik MC, Gilbert LA, Villalta JE, 660 Kwon SC, Chang H, Kim VN, Weissman JS. 2015. Next-generation libraries for 661 robust RNA interference-based genome-wide screens. Proceedings of the National 662 Academy of Sciences 112:E3384–E3391. 663

69. Doench JG, Hartenian E, Graham DB, Tothova Z, Hegde M, Smith I, Sullender M, 664 Ebert BL, Xavier RJ, Root DE. 2014. Rational design of highly active sgRNAs for 665 CRISPR-Cas9-mediated gene inactivation. Nat Biotechnol 32:1262–1267. 666

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

36

70. Fedorov Y, Anderson EM, Birmingham A, Reynolds A, Karpilow J, Robinson K, 667 Leake D, Marshall WS, Khvorova A. 2006. Off-target effects by siRNA can induce 668 toxic phenotype. RNA 12:1188–1196. 669

71. Zhao Y, Dai Z, Liang Y, Yin M, Ma K, He M, Ouyang H, Teng C-B. 2014. Sequence-670 specific inhibition of microRNA via CRISPR/CRISPRi system. Sci Rep 4. 671 72. Meister G, Landthaler M, Dorsett Y, Tuschl T. 2004. Sequence-specific inhibition of 672 microRNA- and siRNA-induced RNA silencing. RNA 10:544–550. 673 73. Krützfeldt J, Rajewsky N, Braich R, Rajeev KG, Tuschl T, Manoharan M, Stoffel M. 674 2005. Silencing of microRNAs in vivo with 'antagomirs'. Nature 438:685–689. 675 74. Ebert MS, Neilson JR, Sharp PA. 2007. MicroRNA sponges: competitive inhibitors of 676 small RNAs in mammalian cells. Nat Meth 4:721–726. 677 75. Larson MH, Gilbert LA, Wang X, Lim WA, Weissman JS, Qi LS. 2013. CRISPR 678 interference (CRISPRi) for sequence-specific control of gene expression. Nat Protoc 679

8:2180–2196. 680 76. Reboul J, Vaglio P, Rual J-F, Lamesch P, Martinez M, Armstrong CM, Li S, Jacotot 681

L, Bertin N, Janky R, Moore T, Hudson JR, Hartley JL, Brasch MA, Vandenhaute J, 682 Boulton S, Endress GA, Jenna S, Chevet E, Papasotiropoulos V, Tolias PP, Ptacek 683 J, Snyder M, Huang R, Chance MR, Lee H, Doucette-Stamm L, Hill DE, Vidal M. 684 2003. C. elegans ORFeome version 1.1: experimental verification of the genome 685 annotation and resource for proteome-scale protein expression. Nature Genetics 686

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

37

34:35–41. 687 77. Rual J-F, Hirozane-Kishikawa T, Hao T, Bertin N, Li S, Dricot A, Li N, Rosenberg J, 688

Lamesch P, Vidalain P-O, Clingingsmith TR, Hartley JL, Esposito D, Cheo D, 689 Moore T, Simmons B, Sequerra R, Bosak S, Doucette-Stamm L, Le Peuch C, 690 Vandenhaute J, Cusick ME, Albala JS, Hill DE, Vidal M. 2004. Human ORFeome 691 version 1.1: a platform for reverse proteomics. Genome Research 14:2128–2135. 692

78. Lamesch P, Li N, Milstein S, Fan C, Hao T, Szabo G, Hu Z, Venkatesan K, Bethel G, 693 Martin P, Rogers J, Lawlor S, McLaren S, Dricot A, Borick H, Cusick ME, 694 Vandenhaute J, Dunham I, Hill DE, Vidal M. 2007. hORFeome v3.1: A resource of 695 human open reading frames representing over 10,000 human genes. Genomics 696 89:307–315. 697

79. Yang X, Boehm JS, Yang X, Salehi-Ashtiani K, Hao T, Shen Y, Lubonja R, Thomas 698 SR, Alkan O, Bhimdi T, Green TM, Johannessen CM, Silver SJ, Nguyen C, Murray 699 RR, Hieronymus H, Balcha D, Fan C, Lin C, Ghamsari L, Vidal M, Hahn WC, Hill 700 DE, Root DE. 2011. A public genome-scale lentiviral expression library of human 701 ORFs. Nat Meth 8:659–661. 702

80. Yang J, Yuan P, Wen D, Sheng Y, Zhu S, Yu Y, Gao X, Wei W. 2013. ULtiMATE 703 system for rapid assembly of customized TAL effectors. PLoS ONE 8:e75649. 704 81. Cermak T, Doyle EL, Christian M, Wang L, Zhang Y, Schmidt C, Baller JA, Somia 705

NV, Bogdanove AJ, Voytas DF. 2011. Efficient design and assembly of custom 706 TALEN and other TAL effector-based constructs for DNA targeting. Nucleic Acids 707

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

38

Research 39:e82–e82. 708 82. Sera T. 2009. Advanced Drug Delivery Reviews. Advanced Drug Delivery Reviews 709

61:513–526. 710 711

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

C

VP64

KRAB

Mammalian RepressorMammalianActivator

GscRNA scRNA

MS2 Hairpin COM Hairpin

VP64

MCP RBP COM RBP

KRAB

ActivatorModule

RepressorModule

Activating scRNA complex Repressing scRNA complex

F

MS2 Hairpin

MCP RBP

ActivatorModule

VP64

p65

HSF1

SAM Activator

D

VP64

ActivatorModule

SunTag Activator

sfGFP

scFv

scFv recognitionpeptide

SunTag chain(10X or 24X)

La Russa and Qi Figure 1

E

VPR Activator VP64

p65

Rta

VPR

B

ω subunit

Bacterial Repressor

RNAPolymerase

Bacterial Activator

A

Nuclease Cas9

tracrRNA

crRNATargetDNA

PAM

dCas9

sgRNA

Nuclease-deactivatingmutation

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

A

La Russa and Qi Figure 2

Activation scRNAs

Repression scRNAs

B RepressorModules

ActivatorModules

dCas9

Repression

Activation

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

Table 1: Comparison of CRISPRa and other activation methods CRISPRa ORF overexpression TALE or ZF activators What is activated? Endogenous gene Exogenously-added gene Endogenous gene Ease of production? Simple sgRNA cloning Entire ORF of gene must be cloned Requires complicated cloning (TALEs) or protein engineering (ZFNs) Throughput of production? High Low Low Pooled genome-wide screens? Yes Yes, but these are burdensome to create and can contain biases towards smaller ORFs or certain splice variants

No Off-target effects? Minimal Nonexistent Minimal Affected by methylated DNA? No N/A TALEs can be blocked by methylated DNA but can also recognize it specifically Can express mutant alleles or specific splice variants? Can target specific variants only if different variants have different TSSs; not ideal for expressing mutant alleles

Yes Can target specific variants only if different variants have different TSSs; not ideal for expressing mutant alleles Limitations in use? Requires an “NGG” PAM adjacent to a target sequence Longer or GC-rich genes can be difficult to clone TALEs have no sequence limitations. ZFNs may require some engineering to target a given sequence Number of components required? Two to three (dCas9, sgRNA, optional activation module); these may be collected on one vector in some cases One One

References (3, 4, 24, 26, 29, 30, 32, 75) (44, 76-79) (39, 49, 51, 53-55, 80-82)

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from

Table 2: Comparison of CRISPRi and other repression methods CRISPRi RNAi TALE or ZF repressors miRNA sponges or antagomirs Mechanism of action? Transcriptional elongation of mRNA is sterically blocked; the KRAB fusion recruits repressive chromatin marks

The target mRNA is sequestered or degraded Transcriptional elongation of mRNA is sterically blocked; the KRAB fusion recruits repressive chromatin marks Both act as a dominant negative by binding to miRNAs and preventing them from acting on their target mRNAs Pooled genome-wide screens? Yes Yes No No Off-target effects? Minimal High Minimal miRNAs in a family sharing the same seed will bind to sponges; antagomirs can distinguish between family members Can target small RNAs? Yes No Yes Yes Can target specific splice variants? Only if different variants have different TSSs Yes Only if different variants have different TSSs N/A Limitations in targeting? Requires an “NGG” PAM adjacent to a target sequence No TALEs have no sequence limitations; ZFNs may require some engineering to target a given sequence Sponges are not ideal for selectively targeting one miRNA out a family sharing the same seed Number of components required? Two to three (dCas9, sgRNA, optional repression module); these may be collected on one vector in some cases

One One One; also, a single sponge can simultaneously repress multiple miRNAs References (3, 4, 20, 24, 26, 29, 30, 71) (56, 61, 62, 64, 68) (39, 49, 51, 80-82) (59, 60, 72-74)

on February 12, 2018 by guest

http://mcb.asm

.org/D

ownloaded from