Specificity and Precision of the CRISPR System

In the past decade, CRISPR-Cas9 has revolutionized gene editing—researchers can make changes to an organism’s DNA, allowing genetic sequences to be removed, added, or altered at specific locations. Targeted changes to gene expression can now be made in vitro and in vivo, as well as in a wide range of cell lines and models. But despite the efficiency of the CRISPR-Cas9 system for homing in on a specific sequence of DNA, it can sometimes miss the mark, and off-target activity remains a cause for concern. This article discusses the limitations of CRISPR-Cas9 technology in terms of off-target activity and inefficient cleavage and explores some recent improvements to the precision and specificity of the system.

Missing the mark: off-target activity

The CRISPR-Cas9 system is inherently simple, requiring only a short guide RNA (sgRNA) to target the Cas9 endonuclease to the site of interest to introduce a double-strand break (DSB) at the target loci—and delivery of the components only requires standard molecular biology techniques. The most commonly used Cas9 is from S .pyogenes, which binds the 3’-NGG-5’ PAM site (Protospacer Adjacent Motif) found throughout the genome. The programmable nature of the technology comes from the sgRNA, which contains a 20-nucleotide portion that matches the genomic target site. If the sgRNA shares sufficient sequence homology with the adjacent sequence to the PAM site, the endonuclease activity of Cas9 is activated and a DSB is introduced.

But any sites that share significant sequence homology to the sgRNA can be targeted, resulting in potential cleavage at multiple sites in the genome. For this reason, a number of researchers have directed significant effort to predicting the off-target activity of sgRNAs. The sgRNA sequence itself has also been shown to affect the efficiency of Cas9 cleavage—although the PAM site is essential for initial binding of Cas9, the 3-nucleotide sgRNA seed sequence immediately adjacent to the PAM has been shown to be critical for activation of the endonuclease activity of Cas9.¹

Designing the right sgRNA

The specificity and efficiency of the sgRNA will therefore have an impact on the success of a CRISPR experiment. Several strategies can be utilized to increase guide efficiency, such as ensuring optimal GC content of the sgRNA sequence, truncated sgRNA, or the addition of chemical modifications.² Several computational models have been developed to predict on-target efficiency from existing sgRNA activity data—but there can be discrepancies as the number of datasets used for training is low, and experimental design and cleavage evaluation differs widely. To combat this, Xiang et al., generated sgRNA activity data for 10,592 guides and then used these with published data to train CRISPRon, a deep learning model that was able to significantly improve prediction performance compared to existing tools.³

Maximizing your knockout efficiency

The selection of specific and efficient guides is one element to the success of a CRISPR experiment. Once a DSB has been introduced at the target site, the cut activates endogenous cellular DNA repair pathways—the Non-Homologous End Joining (NHEJ) pathway. For gene knockouts, the desired outcome is that error-prone repair leads to the introduction of a frameshift mutation that ablates expression of the functional gene. However, this does not always happen as not all repairs are error-prone—or the resulting indel is not a frameshift and so the subsequent protein is still functional. The unpredictability of this edit reduces the efficiency with which gene editing can be performed and so requires additional validation.

“Accurate characterization of potential hits of interest can dramatically affect your downstream assays. Creating functional knockouts ensures you create the most robust data set and potential to shorten your experimental workflow,” explains Tanushi Sahai, Associate Product Manager, CRISPRevolution, Synthego.

An alternative strategy to using just a single guide for a knockout is a multi-guide approach, where several sgRNA are used for each target site to better guarantee knockout. For example, the Gene Knockout Kit from Synthego employs three guides working cooperatively to introduce specific fragment deletions resulting in a higher likelihood of protein depletion.⁴

Modulating Cas9 delivery

Delivery of the CRISPR components is usually via plasmid—but as gene editing only requires a single Cas9-sgRNA complex to facilitate the edit, this can lead to increased off-target activity due to the sustained expression of the construct and presence of additional complexes. Direct delivery of purified Cas9 protein and sgRNA results in transient expression due to rapid degradation by cellular proteases—and has been shown to result in reduced off-target effects compared to use of plasmid delivery.⁵

Working with modified Cas9 variants

As well as modifications to sgRNA to reduce off-target activity, other approaches have turned to alterations of the Cas9 protein. A ‘nickase’ can be created by inactivating either of the RuvC or HNH nuclease domains of Cas9, which results in the modified Cas9 only being able to introduce a single-strand break at the target site. To facilitate gene editing, a pair of offset nickase variants are used to introduce a DSB—and so the low likelihood of two independent sgRNA targeting off-target effects can improve specificity.⁶

Alternatively, by substituting key amino acids, a high-fidelity version of Cas9 (SpCas9-HF) has been created that disrupts non-specific interactions between the Cas9-sgRNA complex and DNA. By reducing the DNA binding energy of the sgRNA-Cas9 complex, this reduces the ability to cleave target sequences that are mismatched therefore improving on target specificity but maintaining on-target cleavage.⁷

Avoiding the cut—search and replace with prime editing

CRISPR-Cas9 genome editing relies upon the introduction of a DSB at the target site to facilitate editing—in the presence of an exogenous DNA template, the DSB triggers the Homologous Repair (HR) pathway, which results in the replacement of a section of DNA at the target site. But the HR process is often error-prone and inefficient—and when considering the use of CRISPR-Cas9 for clinical use, even a single off-target site can have a major impact.

To overcome these limitations, a novel method of CRISPR gene editing has been developed—prime editing is described as a "search and replace" technology, but one that does not rely upon the introduction of a DSB. Instead, prime editing utilizes a prime-editing guide RNA (pegRNA) that contains the required edit, alongside a Cas9 nickase fused to a reverse transcriptase. The non-complementary strand of DNA is cut 3 base pairs upstream of the PAM site creating a flap—the required edit is then reverse transcribed and incorporated into the DNA and the target sequence is repaired—the original DNA section is then removed by an endogenous nuclease. Not only can prime editing be used to make a variety of edits but has also comes with significantly reduced off-target activity.⁸

CRISPR-Cas9: Great, and still improving

The simplicity, versatility, and low cost of CRISPR-Cas9 gene editing has seen it rapidly adopted across a wide range of research areas and applications, surpassing previous site-directed nuclease systems, such as Zinc Fingers (ZFNs) and TALENs. But it can cause unintended effects due to inefficient binding and off-target activity. However, CRISPR-Cas9 technology is continually developing and improving, and there are several strategies that can be employed to minimize the limitations and increase the likelihood of success in your CRISPR experiments.

Key takeaways

• The CRISPR revolution now means that researchers can make targeted changes to gene expression in vitro and in vivo, as well as in a wide range of cell lines and models at low cost and relatively easily
• But off-target activity and inefficient cleavage at the target site remain an issue when performing CRISPR experiments
• Several strategies can be employed to minimize the limitations of CRISPR technology, such as accurate prediction models for selecting optimal sgRNA, using modified Cas9 proteins, altering the delivery of Cas9 components, and prime editing

References

1. Jinek, Martin et al. “Structures of Cas9 endonucleases reveal RNA-mediated conformational activation.” Science (New York, N.Y.) vol. 343,6176 (2014): 1247997. 

2. Naeem, Muhammad et al. “Latest Developed Strategies to Minimize the Off-Target Effects in CRISPR-Cas-Mediated Genome Editing.” Cells vol. 9,7 1608. 2 Jul. 2020, 

3. Xiang, Xi et al. “Enhancing CRISPR-Cas9 gRNA efficiency prediction by data integration and deep learning.” Nature communications vol. 12,1 3238. 28 May. 2021, 

4. https://www.synthego.com/help/synthego-multi-guide-design

5. Kim, Sojung et al. “Highly efficient RNA-guided genome editing in human cells via delivery of purified Cas9 ribonucleoproteins.” Genome research vol. 24,6 (2014): 1012-9. 

6. Ran, F Ann et al. “Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity.” Cell vol. 154,6 (2013): 1380-9. 

7. Kleinstiver, Benjamin P et al. “High-fidelity CRISPR-Cas9 nucleases with no detectable genome-wide off-target effects.” Nature vol. 529,7587 (2016): 490-5. 

8. Anzalone, Andrew V et al. “Search-and-replace genome editing without double-strand breaks or donor DNA.” Nature vol. 576,7785 (2019): 149-157. doi:10.1038/s41586-019-1711-4