The Prominent Characteristics of the Effective sgRNA for a Precise CRISPR Genome Editing

Abstract

Clustered regularly interspaced short palindromic repeats (CRISPRs) technique is the most effective and novelist technique for genome editing. CRISPR mechanism has been widely developed for gene editing, gene silencing, high-specific regulation of the transcription, and reducing off-target effects through double-strand breaks (DSBs) in the genomic DNA and then modifying nucleotide sequences of the target gene in diverse plant and animal species. However, the application may be restricted by a high rate of off-target effects. So, there are many studies on designing precise single-guide RNAs (sgRNAs) to minimize off-target effects. Thus, the high-efficiency design of a specific sgRNA is critical. First, in the chapter, the sgRNA origin and different types of gRNA will be outlined. Then, the off-target effect will be described. Next, the remarkable characteristics of the sgRNA will be highlighted to improve precise gene editing. Finally, some popular in silico tools will be introduced for designing sgRNA.

1. Introduction

Clustered regularly interspaced short palindromic repeats (CRISPRs) and their CRISPR-associated (Cas) proteins system is an effective immune system among bacteria and archaea. This system was first discovered in the E.coli genome. CRISPR/Cas is an acquired immunity mechanism in many bacteria and archaea against the genome of the infection factors such as viruses and plasmids [1]. CRISPR/Cas system is classified into three major groups (I, II, and III) with a specific functional mechanism and gene family encoding the specific Cas proteins. Types I and III apply several Cas proteins for endonuclease activity, while type II uses only one protein (Cas9) [2]. Evolutionary competition between the pathogens and host in the CRISPR/Cas system shows a very high variable rate in structures and functions. So, recent classification has stated that the CRISPR system has been categorized into two classes (I and II) and six types (I–VI) [3].

Most studies on genome engineering have been performed in system type II, derived from Streptococcus pyogenes (SpCas9). The advantage of system type II is that it needs only one protein (Cas9) for endonuclease activity. However, this system also needs the types of RNA, including CRISPR RNA (crRNA), which functions as a Cas protein guide for pairing with the target genome sequences, and trans-activating CRISPR RNA (tracrRNA), which play critical roles in crRNA maturation and directing Cas9 to the desired site [4, 5]. Merging of tracrRNA:crRNA sequences as a chimeric sequence known as single-guide RNA (sgRNA), which covers the features of both RNA types, makes it suitable and applicable for genome editing [6].

All CRISPR sites cover consecutive and spacer repetitions. These consecutive repetitions include identical sequences, while the spacer sequences originate from the genome of foreign factors [7, 8]. CRISPR sites and Cas proteins develop acquired immunity against the invading DNA. Suppose a microorganism survives the invasion of a pathogen. In that case, the integrated CRISPR system will be able to incorporate a piece of the pathogen DNA into its genome and then use it to fight against subsequent infections. This bacterial immune system degenerates the phage genome by integrating short fragments of the phage DNA in the spacer region of the CRISPR sites and transcribing the spacers (known as crRNAs) with associated Cas endonuclease in subsequent infections [9, 10, 11].

Briefly, the CRISPR system, as the RNA-mediated immune system in prokaryotes (bacteria and archaea), functions in four stages (Figure 1):

1. Admission: short viral or plasmid DNA fragments are recognized and then inserted as a spacer between two consecutive repetitions into the CRISPR sites [12, 13].

2. Expression: the CRISPR sites are transcribed to a long precursor crRNA (pre-crRNA) containing a complete array of CRISPR repetitions and sequences derived from infectious factors [14].

3. Insertion: Pre-crRNA is cleaved by a special endoribonuclease into small guide sequences known as crRNA [15].

4. Targeting: The crRNAs guide Cas endonuclease to cleave complementary DNA or RNA sequences flanked by a protospacer adjacent motif (PAM) (for DNA targets) or a protospacer flanking sequence (usually known as PFS) without significant complement to the crRNA repeats (for RNA targets) [16, 17].

Figure 1.

RNA-mediated CRISPR immune system in.

The classification represents the evolution of subtype-specific molecular defensive mechanisms for crRNA expression and maturation, as well as the inhibition of infectious factors [18]. The main known vital component of the CRISPR/Cas system is crRNA, which is common in types I and III. Pre-crRNAs are initially cleaved within the repeats by a Cas6 endoribonucleases family, and then intermediate crRNAs are further matured to generate shorter repeat-spacer crRNAs in type III. In both type I and type III, mature crRNAs direct a complex of multiple Cas proteins to the cognate-invading nucleic acids. Then, the target nucleic acids are cleaved by a Cas endonuclease of the ribonucleoprotein complex [19, 20, 21].

Pre-crRNA processing necessitates base-pairing of each pre-crRNA repeat with tracrRNA, a small noncoding RNA encoded near the Cas genes and spacer array [7]. The base-pairing drives cleavage and binding by RNase III and Cas9, respectively. Then, the crRNA:tracrRNA complex can direct Cas9 to bind target DNA sequences by matching PAM [16, 17]. In addition, type II CRISPR-Cas systems continuously utilize the crRNA:tracrRNA complex to identify and cleave double-stranded DNA (dsDNA). Recognition is driven by base-pairing between the guide sequence and the RNA in the crRNA:tracrRNA complex (Figure 2) [22, 23]. The tracrRNA is a common component of CRISPR-Cas systems among all three subtypes of type II systems and is required for crRNA biosynthesis [24]. These basic principles were discovered by following efforts to characterize tracrRNAs and crRNA biogenesis [25]. The biogenesis of crRNA, the structures of the crRNA:tracrRNA complex, and tracrRNA genomic location are variable among these subtypes [26]. The tracrRNA discovery as a key factor of crRNA biosynthesis allowed the sgRNA to be invented and Cas9 to be adopted as the core component of CRISPR technology [27].

Figure 2.

Schematic structure of the crRNA:tracrRNA complex.

The major function of the sgRNA is efficiently the target region detection through the PAM sequence to edit a gene precisely. However, two vital challenges include efficacy and specificity for designing an effective sgRNA [28]. According to the significance of the sgRNA function, the role and designing tools of sgRNA will be outlined in the following.

2. The role of sgRNA in CRISPR technology

CRISPR/Cas genome editing technology can cause double-stranded DNA breaks (DSBs) in predefined genomic loci [29]. These DSBs are then repaired by the DNA repair systems of the target organism, which inherently can cause a mutation in the target gene. Despite the general processes driving all these genome editing systems being similar, CRISPR/Cas technology has emerged as the preferred technique due to its easy usage, low cost, outstanding adaptability, and ability to target several genes at once [30, 31, 32].

This technology consists of a Cas endonuclease, responsible for eliciting the DSB, and a short noncoding about sgRNA (20-nt), directing Cas to the correct genomic region for targeted genome editing. A chimeric gRNA (complementary to the target area) and trans-activating CRISPR-RNA are usually included in this sgRNA. Most Cas systems require the predesigned sgRNA to anneal immediately upstream of a PAM, which in the case of SpCas9 (the most extensively employed Cas protein for genome editing) is 5` NGG3` [4, 5, 33]. In these cases, the PAM is required to cleave target DNA around 3-nt upstream of this region. DSBs are repaired by either nonhomologous end joining (NHEJ) or homology-directed repair (HDR) as two central intrinsic DNA repair systems [34]. The error-prone nature of NHEJ is the main DNA repair route in species and the most common and straightforward pathway in genome editing, causing small insertions or deletions (indels) to interrupt the target sequence (Figure 3) [35].

Figure 3.

CRISPR/Cas9 technology for gene editing. The Cas9 DNA endonuclease is recruited by a single-guide RNA (sgRNA) that detects a genomic sequence followed by a 5′-NGG-3′ PAM motif.

CRISPR/Cas technology relies on DNA-RNA interaction as well as simple design of RNA molecule for each specific sequence. However, protein-DNA interaction also depends on zinc-finger nucleases (ZFNs) and transcription activator-like effector nucleases (TALENs). So there is a required reconstruction for each target DNA sequence. This is a significant benefit of CRISPR/Cas technology [36]. In addition, the technology has three other advantages over TALENs and ZFNs, including the following (Table 1):

1. Simplicity: The 20 nt sequence of sgRNA can be simply designed to target any sequence [5].

2. Multifunction: As this technology’s main advantage over TALENs and ZFNs, several sgRNAs can function differently and simultaneously on variable genomic sites [37].

3. Non-specificity to the target DNA methylation: the CRISPR/Cas technology can edit some genome sites which are highly regulated by epigenetic changes [38], especially in plants that contain around 70% of CpG/CpNpG sites and CpG islands methylated in the proximal exons promoter [39].

2.1 The role of PAM

Although specific targeting significantly depends on the sgRNA sequence, PAM-specific sequence plays a crucial role in the effective enzymatic activity of Cas endonuclease. In this system, Cas9 endonuclease can cleave any genome sequence immediately located on five nucleotides (nt) downstream of PAM sequence [25]. PAM sequences recognized by SpCas9 and StCas9 (from Streptococcus thermophilus) is, respectively, 3`-NGG-5` and 5`-NNAGAAW-3` (Table 2) [40, 41]. The SpCas9-sgRNA complex first seeks the complement sequence of PAM in the target genome and then the sgRNA base pairs to the target DNA, then the DNA is cleaved by SpCas9 to create DSB. Generally, the length of the DNA detection sequence in the crRNA region is 20 nt, though the more base pairs bind between RNA and DNA, the more the specificity of the sgRNA function can enhance. Hence, the 20 nt sequence of sgRNA and 3 nt in PAM play key roles in the specific targeting of CRISPR/Cas technology. However, there are some limitations in using the 3`-NGG-5` motif, particularly at high AT sequences of the target genome [42, 43].

System	ZFNs	TALENs	CRISPR
Function	Cleavage mediated by DNA-protein interaction	Cleavage mediated by DNA-protein interaction	Cleavage mediated by DNA-RNA interaction
Nuclease designing and assembly	Hard and costly	Mostly possible in the laboratory, but highly difficult	Easy
Designing efficiency	Low	High	High
Assembly efficiency	Variable	High (%99<)	High (%90<)
Targeting range	Restricted, because of dependence on ZF modules	Unrestricted because of independence on PAM	Restricted by PAM, but generally unrestricted
Off-target effects	Yes	Yes	Yes
Sensitivity to DNA methylation	Undefined	Sensitive to CpG methylation	Nonsensitive to CpG methylation
High operating power	No	Restricted	Possible

Table 1.

Comparison between genome editing systems.

Bacterial species	Endonuclease	PAM-specific sequence
Streptococcus pyogenes	SpCas9	NGG
	SpCas9 D1135E	NAG
	SpCas9 VRER	NGCG
	SpCas9 EQR	NGAG
	SpCas9 VQR	NGNG/ NGAN
Staphylococcus aureus	SaCas9	NNGRR(N)/ NNGRRT
Neisseria meningitidis	NmCas9	NNNNGATT
Streptococcus thermophilus	StCas9	NNAGAAW
Treponema denticola	TdCas9	NAAAAC
S. aureus	SaCas12a	TTN

Table 2.

Different PAM sequences of some Cas endonucleases and their origins.

N = A,T,G,C; R = G,A; and W = A, T.

2.2 Variety of sgRNA types

According to the development of the CRISPR system as technology and sgRNA invention, there are some improvements of sgRNA to enhance the efficiency, precision, and specificity of the genome editing technology [44]. The improvements include as follows:

1. Truncated guide RNA (tuRNA): the RNA contains a homologous 17 nt sequence to the target gene. Hence, the specificity of Cas endonuclease activity can increase by reducing off-target sequences [45].

2. Polycistronic tRNA-gRNA (PTG/Cas9): in the system, the RNA is a frequent repetition of the tRNA-gRNA units and target-specific spacer sequences for targeting several sites in the genome sequence [46]. After PTG transcription, the primary copy is matured as sgRNAs through tRNA processing system and RNaseP and RNaseZ activity. In addition, the sgRNAs can interact with Cas endonuclease to target multiple genes specifically and simultaneously [47].

2.3 Off-target effect

Off-target mutation is a major challenge in CRISPR/Cas technology [48]. If the gRNA sequence contains less than three heterologous nucleotides to an off-target region, off-target effects will be observed [49]. Studies have indicated that mismatched pairs at the end of the 3` terminal of the target sequence are not tolerated (typically 8–14 nucleotides upstream of PAM sequence). In contrast, the mismatched pairs at the 5` terminal of the target sequence are more tolerable [50]. The sgRNA/Cas9 also can influence the off-target effects [49].

Generally, although Cas9 protein can be differently used according to high endonuclease activity and the wide targeting range of the enzyme, the high molecular weight of the Cas9 endonuclease and off-target effects can restrict the popularity of the enzyme. Nevertheless, some variants of Cas9, such as SpCas9-HF and eSpCas9 [51, 52], can be mutated. So, the mutation can reduce the nonspecific interaction between the Cas9 protein and the target sequence. Digenome-seq, GUIDE-Seq (Genome-wide, Unbiased Identification of DSBs Enabled by Sequencing), and HTGTS (High-Throughput Genome-Wide Translocation Sequencing) can be employed for detecting off-target regions [53]. However, precisely designing sgRNA can significantly decrease the rate of the off-target effects [44].

3. Characteristics of an effective sgRNA

In addition to directing Cas endonuclease, sgRNA can stimulate Cas endonuclease activity. These functions of sgRNA can clarify how to tackle on-target effects [54]. It has been demonstrated that the proximal and distal ends of the PAM sequence are highly responsible for improving on-target effects. Besides, genomic frameworks of the target sequence, GC content, sgRNA length, and secondary structure play significant roles in enhancing the on-target rate [44]. Also, 5` terminal 20 nt of the sgRNA is highly efficient for on-target efficiency. However, there is insufficient information on the correlation between structure and sequence properties of sgRNA influencing the on-target effect [55]. At least, enhancement of the on-target rate can improve the efficiency of the CRISPR technology and facilitate the statistical interpretation of the edited genes rate [56].

3.1 GC content of the sgRNA

As mentioned above, the GC content of the sgRNA is closely related to the on-target rate and the efficiency of the CRISPR gene editing. It has been indicated that too high or low GC content is unsuitable for achieving high rate of the on-target effects [57]. The knockout effectiveness of the CRISPR/Cas9 system was significantly improved by changing the sgRNA structure by expanding the duplex length (about 5 nt) and replacing the fourth T by C or G [58]. Many studies have reported that GC content plays a key role in improving the knockout efficiency of the CRISPR/Cas9 system. The effective rate of GC content is 40–60 percent. It is also recommended that sgRNA containing 50 percent GC content is efficient for CRISPR gene editing [59, 60, 61]. However, some studies suggest higher than 60 percent GC content for each organism, such as Escherichia coli (62.5%) and Vitis vinifera (65%) [62, 63]. In addition to GC content, purine residues and curvature in positions C3 and C16 of sgRNA could be effective in improving on-target activity [64]. It has also been reported that sgRNAs containing 4 GC in the 6 nt close to the PAM sequence can effectively reduce off-target effects [65].

3.2 The sgRNA length

The most common sgRNA length is about 100 nt. Therefore, 5` terminal 20 nt of the sgRNA can be designed as the complement sequence of the target gene to direct precisely Cas endonuclease for achieving effective gene editing [66]. According to several studies, the less the sgRNA length is, the higher the rate of off-target effects increase [67, 68, 69]. However, as mentioned above, 17 nt length for tuRNA can be highly effective in reducing off-target effects. However, the length of the sgRNA recognition site is less than 15 nt, and Cas endonuclease will not show any activity [70, 71]. In addition to the sgRNA length, the efficiency and specificity of Cas cleavage activity in the targeting sequence are significantly influenced by the distance between the PAM site and the start codon [72].

3.3 The sgRNA secondary structure

The sgRNA secondary structure is highly responsible for effective Cas-target sequence binding [73]. There are also many reports to indicate that the presence of the quad stem-loop structure of sgRNA is a key factor in improving the efficiency of the riboprotein function. The repeat and anti-repeat region (stem-loop RAR, GAAA) can activate sgRNA processing before Cas-sgRNA binding. Besides, loops 2 (GAAA) and 3 (AGU) are demanded to create a stable riboprotein complex, but there is no report on the possible loop 1 role in sgRNA efficiency [74, 75, 76, 77]. Besides, the hairpin structure of the sgRNA, particularly the inner side of the hairpin, plays a key role in cleaving target DNA by Cas9/sgRNA. In fact, the hairpin structure can provide a suitable conformation to bind Cas9 enzyme. If the loop-stem structure is elongated, the gene editing efficiency will be enhanced [78, 79]. It has been indicated that CRISPR efficiency can be improved when an engineered hairpin structure is inserted into the spacer region of sgRNA. The modified sgRNA can positively influence Cas-mediated transactivation and improve the function of the five different Cas9 and Cas12a variants. The evidence can demonstrate the effect of sgRNA secondary structure on the success rate of gene editing [80, 81]. There is also a correlation between the sgRNA secondary structure and the efficiency of the Cas9-mediated CRISPR [55]. It has been reported that sgRNA refolding can refine the destructive bonds of the deactivated sgRNAs. Also, heating or slowly chilling can thermodynamically activate these sgRNAs to improve Cas9 cleavage activity. At least, the sgRNA secondary structure can change the guide sequence activity and deactivated sgRNA can be recovered by refolding [82, 83].

3.4 The sgRNA sequence

In addition to these criteria, the sgRNA sequence features can show different efficiency levels. It has been reported that the functional sgRNA can be significantly accessed at certain nucleotide positions more than nonfunctional sgRNA. Particularly, 3` terminal nucleotides (positions 18–20) of the sgRNA can highly make a prominent difference in accessibility [84, 85]. The sgRNA 3` terminus, known as the seed region, is a key player in recognizing the target sequence. Therefore, accessibility of the final three bases of the seed region is a remarkable characteristic in distinguishing functional sgRNAs from nonfunctional ones based on structural analysis [86, 87]. Also, G at the 5` terminus of the spacer is demanded in the non-ribosomal and ribosomal complexes. Besides, G is extremely proper at positions −1 and − 2, close to PAM associated with the sequence preference for loading Cas9 endonuclease [88]. According to the finding that multiple U in the spacer results in low sgRNA expression, T is not preferred at the four positions nearest to the PAM. The downstream nucleotides of the PAM cooperate with the efficiency of sgRNA, while upstream sequences of the spacer do not show significant effects. The early termination of sgRNA transcription is mostly responsible for the reduced sgRNA expression rates caused by the high frequency of nonconsecutive T clusters in the protospacer. Cytosine is favored at the −3 position as the cleavage site of the sgRNA/Cas9 complex.

Additionally, guanines are favored from positions −14 to −17, while adenines are favored positions −5 to −12 [89, 90, 91, 92, 93]. Most molecular characteristics that govern sgRNA stability, loading, and targeting in vivo are yet unknown. While variable Cas9 off-target binding, positioning of the nucleosome, and sgRNA loading are not key factors, adenine depletion and guanine enrichment improved sgRNA activity and stability. There is also a close correlation between sgRNA efficiency and guanine enrichment PAM-proximal site, supposedly caused by G-quadruple structure increasing sgRNA stability [94, 95, 96, 97].

4. Computational tools for designing sgRNA

Generally, the potential off-target effect is still a vital concern for several applications of CRISPR technology. There are many strategies including Cas endonucleases engineering, transcriptome analysis, tunable systems (small-molecule induction of Cas9, light-activated and intein-inactivated Cas9), functional screening after dCas9 treatment, direct delivery (RNP complex), truncated sgRNAs (small-guide RNAs), and separate Cas9-binding approaches (paired Cas9 nickase) to reduce the off-target activity in CRISPR/Cas gene editing (Table 3) [105, 106, 107]. However, designing sgRNA can be the most simple, effective, low-cost, and time-saving approach [28].

There are some key factors for designing an efficient sgRNA for CRISPR editing and reducing off-target effects. First, the GC contact should be 40–80 per cent, although a higher percentage is more desirable and beneficial [28]. Second, the sgRNA length needs to be 17–20 nt, depending on the used Cas enzyme. The shorter the sequence is designed, the less the off-target effects are observed; however, too short sequences can increase the off-target effects [71]. As the third factor, off-target effects may result from mismatches between sgRNA and the target sequence, according to the mismatch numbers and positions [96]. ΔG calculation provides a significant benefit to assess sgRNA-DNA binding potential. The ΔG of a highly effective gRNA is from −64.53 to −47.09 kcal/mol, but higher ΔG can increase mismatching rates, causing a high rate of the off-target effect [93]. Although the more the negative ΔG would be, the more stable the secondary structure of the sgRNA is observed, higher ΔG (~ −30 to −20 kcal/mol) can positively influence sgRNA transcription and practically make more effective RNA-RNA binding for a more functional secondary structure of the sgRNA [28].

According to all these criteria, designing sgRNA is a crucial concern in CRISPR technology [72]. As CRISPR/Cas contain two key players, including Cas endonuclease and sgRNA, to cooperate with genome editing, the development of each component could be beneficial to enhancing CRISPR/Cas editing. However, enzyme engineering is a costly, time-consuming, and complicated strategy. So, sgRNA designing could be more effective [108, 109]. Furthermore, an effective sgRNA should simultaneously show the highest on-target efficiency and the lowest off-target activity. So, several well-developed computational tools can be found to design sgRNA for high-efficiency genome editing [110]. In addition to the simplicity, high efficiency, and cost-effectiveness, the in silico tool offers adaptability, automation, and fast processing to analyze many genes [111].

In the last decade, many in silico tools have been introduced to developing CRISPR technology because there has been an urgent demand to design an effective sgRNA to create precise mutations via CRISPR/Cas. Some tools have combined several scoring methods and/or algorithms to provide better design services [112, 113]. In addition to the different features of effectively designing sgRNA, these tools would be user-friendly [114]. The most popular computational sgRNA designing tools are outlined in the following and the other in silico tools are summarized in Table 4. The outlined tools are able to offer candidate sgRNAs and simultaneously score off-target activity. Besides, they are more user-friendly and fast-processing tools [28, 44].

4.1 CHOPCHOP

CHOPCHOP website, one of the most conventional in silico tools for detecting target sequences, includes a clear interface and complete functions. There are more than 200 reference genomes on the website so that the users can search for the target sequence, genomic coordinates, and name of the desired gene. The users are also able to choose two different methods to detect off-target and seven scoring approaches for on-target efficiency before sgRNA designing (Table 5) [117, 125, 126].

Strategies	Advantages	Limitation	Ref.
Cas engineering	PAMs improvement, shortening sgRNAs	Costly, time-consuming, reforming for each propose	[98]
Transcriptome analysis	Precisely detecting on- and off-target activity	Costly, time-consuming	[99]
Tunable system	Regulating Cas 9 working time, reducing undesirable DNA cleavage	Slow on-rate, decreased editing efficiency, time-consuming for inducing by light or chemicals	[100]
Functional screening	Validation of gene functions, controlling genetic disruption	Just applicable for the low-throughput formats	[101]
RNP complex	Reduced off-target effects, transient genome editing	High molecular weight, inducing phospholipid bilayer stress	[102]
Truncated sgRNA	Minimizing off-target effects without reducing on-target activity, decreasing sgRNA length	Editing in some new off-target sequences	[103]
Paired Cas9 nickase	High efficiency and specificity, using for insertion in target gene by NHEJ	Requiring two simultaneous sgRNAs	[104]

Table 3.

Benefits and limitations of the strategies predicting off-target effects.

Name	Services	URL	Ref.
sgRNA Designer	sgRNA designing	https://portals.broadinstitute.org/gpp/public/analysis-tools/sgrna-design	[115]
CRISPR-ERA	sgRNA designing	http://CRISPR-ERA.stanford.edu	[116]
CHOPCHOP	Detecting target sequence	https://chopchop.cbu.uib.no/about	[117]
CRISPRscan	sgRNA designing and analysis	https://www.crisprscan.org/	[118]
CRISPR-GE	sgRNA designing detecting target sequence	http://skl.scau.edu.cn/	[119]
CRISPR RGEN Tools	Detecting off-target sites	http://www.rgenome.net/	[120]
sgRNAcas9	sgRNA designing predicting off-target sites	http://biootools.com/	[121]
CRISPR MultiTargeter	Multiple sgRNA designing tool	http://multicrispr.net/	[122]
CRISPR-P	sgRNA designing for plants	http://crispr.hzau.edu.cn/CRISPR2/	[123]
CRISPOR	sgRNA designing; Detecting on- and off-target sites	http://crispor.tefor.net/	[124]

Table 4.

Common in silico tools for sgRNA designing in CRISPR technology.

Tool	Covered genome	Nucleases	Nickase	Off-target analysis	Cas type
ChopChop	>200	Yes	No	Yes	Different Type II
CRISPR RGEN	350	Yes	No	Yes	Different Type II
CRISPOR	417	Yes	No	Yes	Different Type II

Table 5.

Comparing three in silico tools.

4.2 CRISPR RGEN tools

CRISPR/Cas-derived RNA-guided engineered nuclease (CRISPR RGEN) can provide several computational tools and sgRNA/Cas libraries, including nine tools such as Cas-OFFinder, Cas-Designer, and Digenome-Seq. Compared with other tools, Cas-Designer can rapidly detect potential off-target sites containing a DNA or RNA bulge. Besides, Cas-Designer can offer an out-of-frame score for each sgRNA to find the proper sites for the gene knockout [120, 127, 128]. Cas-OFFinder is also used to seek potential off-target positions of NmCas9 endonuclease (from N.meningitides), recognizing 5′-NNNNGMTT-3′ PAM sequence (M = A or C) as well as a 24-bp target sequence specific to the design sgRNA in the target genome. Also, mixed bases can be used by Cas-OFFinder to analyze the degeneracy of PAM sequences. At least, Cas-OFFinder can provide quick scanning for potential off-target positions in any sequenced genome, regardless of the mismatched nucleotides numbers or the PAM sequence limitation (Table 5) [120, 129, 130].

4.3 CRISPOR

Among these in silico tools for effectively designing sgRNA, CRISPOR includes various useful tools to design sgRNA, 417 genomes, and 19 PAM types. This in silico tool can receive genome coordinates and sequences with more than 2000 bp length as the inputs. After processing, comprehensive information is provided as the output. The result can be, by default, presented in two sections; first, visualizing the PAM sites along the target sequence, available in different formats such as fasta, GenBank and SnapGene; second, providing a table containing all information such as 2 specificity scores and 10 efficiency scores for every predicted sgRNA (Table 5) [105, 124, 131, 132].

4.4 Challenges and limitations of in silico tools

Although the computational tools for sgRNA designing can facilitate on-target prediction and reduce off-target activity, some tools cannot cover all vital criteria. So, it is highly recommended to use different in silico tools. For example, some free, user-friendly, and reliable online tools include RNAfold and Mfold to predict sgRNA secondary structure [59, 79].

Moreover, all current prediction models struggle with four main challenges:

1. Data insufficiency: machine learning models (MLDs) operate better than other approaches due to their data-driven process. Nevertheless, they cannot accurately anticipate previously unseen data without sufficient data to fully extract features [133, 134].

2. Unclear mechanism: The mechanism of the CRISPR/Cas9 editing system has not been completely discovered and limits the characteristics used in the most recent cutting-edge algorithms. So, MDL-mediated approaches are not simply able to achieve significant advancements with sufficient data [133, 134].

3. Data heterogeneity: generated datasets from various platforms and cell types should be merged for data augmentation [133].

4. Data imbalance: the detection of the frequency of off-target sequences through high-throughput whole-genome sequencing is considerably less than the numbers detected by in silico prediction tools [133, 135].

5. Adenine base editors (ABEs)

Base editors (BEs), as chimeric proteins, contain a catalytic domain and a DNA targeting modules which is able to deaminate adenines and cytosines. There is no need to make DSBs in DNA bases editing when BEs are used for base editing. So, these proteins can reduce the off-target effects and random indels at the on-target sequences. The BEs have been introduced as novel promising tools to make precise gene modification [6, 136, 137]. ABEs are the fused Cas9 nickase with a deaminase domain converting A-G and C-T (C > T) at the target sequence [138]. In fact, ABEs can effectively and precisely convert A-G and C-T base pairs at the target site within the editing frame while producing few by-products, consequently reducing off-target activity significantly. The ABE variants can improve the precision of adenine base editing by reducing the off-target activities of RNA and DNA [105, 139]. It has been reported that ABEs cooperate to induce free-sgRNA transcriptome editing. ABEs have also been discovered to display RNA off-target activity and the capacity to edit their own transcripts [140, 141]. ABEs can generally produce significant off-target single-nucleotide variations (SNVs) in RNA sequences. Therefore, deaminase engineering enables ABE variants to decrease off-target mutation of SNVs in RNA sequences while increasing on-target efficiency with DNA [105, 142].

6. Conclusion

Over the last 30 years, genome editing technology, particularly CRISPR/Cas, has promoted biosciences by editing and targeting the genomic DNAs of any species. CRISPR/Cas is the most precise, effective, and affordable among all these genome editing technologies. Although there are diverse types and classes of CRISPR/Cas systems, they are not all applicable due to the high rate of off-target effects. Different approaches have been developed to decrease the off-target effects for enhancing the precision and efficiency of the different CRISPR/Cas techniques. According to a refined reference genome, a well-designed sgRNA can support high on-target efficiency to create a precise and desirable mutation. Finally, it is highly recommended to consider the criteria as mentioned above, including GC content, length, secondary structure, and sequence, for designing an effective sgRNA to achieve high-precision CRISPR genome editing.

Acknowledgments

The chapter was funded by Amin Techno Gene Private Virtual Lab (NGO), Tehran, Iran.

Conflict of interest

The authors declare no conflict of interest.