a complementary approach named CRISPR activation (CRISPRa) was developed by fusing transactivator domains of VP64 or NFkB (Perez‐Pinera et al. 2013; Gilbert et al. 2014) (Figure 3.7a). Both of these systems have been improved further, namely by identifying protein domains with optimal activities (Chavez et al. 2016) or improving the output by recruiting multiple synergistic effectors (Tanenbaum et al. 2014; Zalatan et al. 2015). In parallel, dCas9 system for introducing specific epigenetic modifications at programmed genomic loci was also developed by fusing dCas9 to catalytic domains of DNA methyltransferases and various histone‐modifying enzymes (Figure 3.7c, d), allowing one to enforce a heritable modulation of gene expression (Amabile et al. 2016; Liu et al. 2016; Vojta et al. 2016). While these systems are not as nearly as sophisticated as CRISPRi and CRISPRa, they present a promising basis to therapeutic epigenome editing. dCas9 systems have also been used for a number of other purposes used to elucidate biology, such as imaging and proteomics Figure 3.7e; we refer the reader to excellent reviews on these recently developed systems (Xu and Qi 2019). It should be also noted that dCas9 can be used to tether DNA‐modifying enzymes (such as cytidine or adenosine deaminases) which can alter the underlying sequence; this is the basis for DNA base editing, which is addressed in detail in Chapter 14.
Figure 3.7 Applications of CRISPR systems beyond genome editing. Fusing catalytically inactive Cas9 (dCas9) to a transcriptional activator (a) or repression (b) domains can directly regulate gene expression. By fusing dCas9 to DNA‐modifying enzymes, such as DNA methyltransferases (DNMT) or TET enzymes (c), or histone deacetylases (HDACs), histone acetyltransferases (HATs), or histone methyltransferases (HMTs) (d), one can modify the epigenetic marks leading to indirect albeit potentially permanent modulation of gene expression. If fused to a fluorescent protein, dCas9 can be used to visualize targeted locus in imaging studies (e). RNA targeting enzymes such as Cas13a (f) are able to specifically degrade target RNA molecules, or by fusing a catalytically inactive version of this enzyme to RNA‐modifying enzymes (such as ADAR deaminases), one can perform RNA editing; here, deamination of adenosine to inosine (decoded as a guanosine) is depicted (g). Cas13‐based detection methods rely on the activation of Cas13a by recognizing the target RNA molecule, which is then unspecifically able to digest the reporter RNA molecule. Cleavage of the reporter molecule separates fluorophore (F) from the quencher dye (Q), generating a diagnostic fluorescent signal (h). A similar approach can be used for DNA‐targeting enzymes exhibiting collateral activity, such as some Cas12 enzymes.
3.4.2.2 RNA Targeting
So far, we have discussed how various CRISPR systems can be used for genome editing. The discovery of type VI systems and their ability to target RNA by Cas13 effectors has led to the rise of novel approaches to manipulate the transcriptome of a given cell without altering the underlying genetic component. As discussed in previous sections, Cas13 proteins can degrade target RNA molecules by nucleolytic activity of their HEPN domain. Heterologous expression of Cas13 orthologs, such as those of Leptotrichia wadei (LwaCas13a), Leptotrichia shahii (LshCas13a) (Abudayyeh et al. 2017), Prevotella sp. P5‐125 (PspCas13b) (Cox et al. 2017) or Ruminococcus flavefaciens (RfxCas13d, also known as CasRx) (Konermann et al. 2018), and cognate crRNA in human cells leads to knockdown of specific RNA transcripts (Figure 3.7f) without substantial off‐target effects usually associated with short‐hairpin RNA (shRNA). The knockdown efficiency is ortholog and transcript dependent, but comparable to reduction observed with genome editing approaches (50–90%). As discussed previously, nearly all Cas13 proteins exhibit an indiscriminate RNase activity, meaning that they can degrade any bystander RNA. While this is an important mechanism of conferring population‐level immunity in prokaryotes (Meeske et al. 2019), the collateral activity of tested proteins has not been shown when expressed in human cells, encouraging further use of this system in mammalian models. It should be noted that some Cas9 proteins are also able to target RNA (Sampson et al. 2013; Dugar et al. 2018; Strutt et al. 2018), but their activities have not been tested yet in human cells.
Mutating the key catalytic residues of Cas13 converted this protein to a binding‐proficient but nuclease‐deficient protein (dCas13). Subsequent targeting to key regulatory pre‐mRNA elements (such as splicing acceptor or donor sites) permits one to alter the splicing pattern of target transcript (Konermann et al. 2018). Importantly, dCas13 can be used as a programmable RNA‐binding protein (analogous to dCas9), and fusing it to a suitable protein domain allows one to modulate or analyze the properties of target RNA. For example, fusing a domain of Adenosine deaminases acting on RNA (ADAR), key enzymes involved in RNA editing, allows one to post‐transcriptionally change the sequence of RNA from adenine to inosine (decoded during the translation as guanine), altering the protein primary sequence without affecting the genome (Cox et al. 2017) (Figure 3.7g).
Overall, the adaptation of RNA‐targeting type VI CRISPR systems now allows researchers to manipulate the transcriptome, a powerful and complementary tool to manipulating the genome by other systems.
3.4.2.3 Biochemical Detection
Finally, we would like to point that CRISPR systems can be used for molecular diagnostics. The collateral activity of Cas13 proteins, while conceptually undesirable, has been exploited to develop highly sensitive detectors for the presence of specific RNA. This approach relies on in vitro activation of Cas13’s indiscriminate RNase activity by an on‐target recognition of specific RNA. Once activated, Cas13 is able to degrade other RNA molecules in the reaction mix; if a fluorescent RNA reporter molecule is included, then Cas13 is able to detect a specific RNA molecule with attomolar sensitivity (Gootenberg et al. 2017), and improvements claiming to bring sensitivity down to zeptomolar (10−21 mol/l) range (Gootenberg et al. 2018) (Figure 3.7h). This system, termed SHERLOCK, has been used since to develop fast and sensitive methods for diagnostics of a number of highly pathogenic RNA viruses (Myhrvold et al. 2018; Patchsung et al. 2020). A similar approach exploits a collateral activity of Cas12a proteins on ssDNA, allowing one to detect specific DNA (or cDNA) sequences; indeed a flurry of different variants of this concept have been developed to detect pathogen DNA or cDNA or to perform SNP profiling (Chen et al. 2018; Li et al. 2018b; Teng et al. 2019b). These developments and innovation are momentous and provide a major progress for molecular diagnostics (Li et al. 2019b).
3.5 Concluding Remarks
The discoveries made in the field of CRISPR biology in the last two decades have paved a way for efficient, cost‐effective, and precise genome editing. Efforts in biochemical and structural characterization of a great number of Cas proteins expanded the toolset of CRISPR proteins one can use for genome manipulation and other purposes. Initially, SpyCas9 was the only Cas protein used for genome editing, but the development of a multiple Cas9 variants with expanded PAM targeting landscape, the discovery of Cas9 orthologs and finally characterization of new class 2 proteins, has in a manner removed constraints imposed by the biochemistry of Cas9. Thanks to an ever‐expanding collections of available Cas nucleases, in the years to come we can expect to perform gene editing with the best tool for a given sequence and application, and without compromising precision and efficiency. Indeed, one can draw a parallel between the democratization of restriction enzymes for molecular cloning and CRISPR enzymes
