CRISPR-Cas9 has provided scientists with a powerful and versatile gene-editing tool, capable of eliciting targeted changes to gene expression in a myriad of different cell lines and models, both in vitro and in vivo. Originally discovered as part of an endogenous bacterial immune system, the CRISPR-Cas9 system comprises a Cas9 endonuclease that can be programmed to target the genomic locus of interest with just a short guide RNA (sgRNA). The simplicity and the programmable nature of the technology has seen it rapidly adopted across multiple research areas and applications, including studying gene function, the creation of genetically modified cell lines and animal models, and research and potential treatment of genetic disorders.

But as with all gene-editing tools, there is a potential for off-target activity and editing at sites other than the desired loci. This remains a concern for users of the technology as off-target activity adds an element of uncertainty to any scientific discoveries—and is of particular importance when considering using CRISPR-Cas9 in therapeutic applications, where cleavage at even a single off-target site could have a major and deleterious effect to the patient. Considerable effort has been put into exploring the off-target activity of CRISPR-Cas9—this article discusses the different approaches that can be deployed to improve the precision and specificity of CRISPR-Cas9 gene-editing technology.

Choose your sgRNA wisely

Targeting the Cas9 endonuclease is directed by Watson-Crick based pairing of the sgRNA —the most commonly used Cas9 from S.pyogenes binds the 3’-NGG-5’ PAM site (protospacer adjacent motif), which is ubiquitous throughout the genome. If the sgRNA shares sufficient similarity with the adjacent sequence, this activates the endonuclease function of Cas9, and a double-strand break is introduced. So, any sites that share significant sequence homology to the sgRNA throughout the genome may be at risk of off-target cleavage—and there are plenty of sites within the human genome that are not unique, such as pseudogenes, duplications, or transposable elements. To get around this, several computational models have been developed to help predict the off-target activity of a given sgRNA and help you select the most efficient and specific one for your experiment. The different models are constantly evolving as more data is generated, but the general consensus is that off-target activity decreases with an increase in mismatch between sgRNA and target sequence—although the degree to which mismatch is tolerated does seem to vary across the length of the sgRNA, with a higher level of mismatch tolerated outside of a “seed sequence” that is directly proximal to the PAM site.

Check your edits with validation tools

Algorithms can predict which guides will be most specific in your CRISPR-Cas9 experiments, but it’s important to validate the off-target activity of the recommended sgRNA, as this can vary depending on whether it is utilized in vitro or in vivo, or due to polymorphisms within individual genomes.1-3 The gold standard for validating an edit is whole-genome sequencing (WGS) but despite it being the most comprehensive approach, the high cost can make it prohibitive. As an alternative, validation techniques such as GUIDE-Seq and DISCOVER-Seq have been developed that use next-generation sequencing to identify cleavage events from DNA fragments generated from double-strand breaks.4,5 These approaches have been shown to reliably and efficiently detect on- and off-target activity, thereby validating those chosen sgRNA for a particular experiment.

Modulate Cas9 delivery

As only a single Cas9-sgRNA complex is required to bind the target site and facilitate the edit, any additional complexes will be free to bind off-target sequences. Several studies have shown that by decreasing the concentration of CRISPR-Cas9 reagents, the level of off-target activity can be reduced. Different methods of controlling the expression of Cas9 were used to decrease off-target activity—instead of encoding the Cas9 reagents on plasmids, Kim et al used a ribonucleotide protein (RNP) complex to deliver the sgRNA and Cas9 molecules, which are rapidly turned over by cellular proteases.6 Because expression of Cas9 is transient, this reduces the time for any off-target cleavage events to occur—in addition, delivery by RNP can be more efficient and exert less cellular stress. Alternatively, Davis et al created a Cas9 protein with inducible expression, where cleavage can only occur in the presence of a small molecule.7 It should be noted, however, that while modulating Cas9 expression using these approaches can limit the ability of Cas9 to elicit off-target activity, they can also result in a decrease to on-target editing frequency.

Modified Cas9: Nickase variants and high fidelity Cas9

Inactivating either of the RuvC or HNH nuclease domains of Cas9 creates a “nickase” that will only introduce a single-strand break at the target site. A double-strand break is introduced by using a pair of offset nickase variants with two independent sgRNA present simultaneously—as the likelihood of this occurring at off-target sites is low, this approach has been shown to offer improved specificity.8,9 Unfortunately, this can also come with reduced on-target mutagenesis, and the requirement for two properly positioned sgRNA could potentially limit the selection of target sites.

Another approach has been to engineer Cas9 to reduce the DNA binding energy of the sgRNA-Cas9 complex that is required for recognition of the DNA target site. By substituting key amino acids, a high fidelity Cas9 (SpCas9-HF) has been created that disrupts nonspecific interactions between the sgRNA-Cas9 complex and DNA, thereby reducing its ability to cleave target sequences that are mismatched, significantly improving on target specificity while maintaining on-target cleavage.10  The DNA targeting capabilities of the CRISPR system have also been expanded with the creation of other high-fidelity Cas9 variants, such as SpCas9-NG and xCas9, thereby adding to the toolkit for precision gene editing.11,12

Engineered sgRNA improves specificity

In a similar vein to high-fidelity Cas9 variants, work has shown that interfering with the energetics of sgRNA-Cas9-DNA binding can also reduce off-target DNA cleavage—but this time, by engineering the sgRNA. Kocak et al introduced a hairpin spacer region in the sgRNA that impedes the formation of an R-loop, which is crucial for Cas9 endocnuclease function. When the guide binds to the target site, there will be sequence homology between the sgRNA and DNA and so binding is energetically favorable—the hairpin unwinds and Cas9 can introduce a double-strand break. But in off-target sites, there may be a mismatch—and so the hairpin retains its secondary structure as unfolding is energetically unfavorable.13

From blunt instrument to laser focus

Almost from its debut as a gene-editing tool, scientists have been working on ways to improve the specificity of CRISPR-Cas9 and reduce off-target activity. The primary underlying reason for this is to ensure its safety for in situ gene editing in patients, where off-target edits are highly undesirable and potentially dangerous. However, off-target effects are also undesirable in the research context, with associated false positives potentially leading to wasted time, effort, and money in research labs. At present no single approach completely mitigates off-target activity of CRISPR-Cas9 but work continues—and a combination of strategies can ensure that your CRISPR-Cas9 experiments are as efficient as possible and maximize the likelihood of the next exciting and biologically relevant discovery.

Key takeaways

  • CRISPR-Cas9 gene editing has revolutionized biological sciences, allowing scientists to make targeted changes to the genetic sequence within the technical capabilities of most labs
  • The CRISPR-Cas9 system is composed of a Cas9 endonuclease guided to the target site by single guide RNA (sgRNA), which contains a 20-nucleotide portion complementary to the sequence of interest
  • The auto-inhibition of unbound Cas9 means that random cleaving events are unlikely to occur—but sites that share significant sequence homology with the sgRNA may be at risk of off-target cleavage
  • Several strategies have been developed to improve the precision of CRISPR-Cas9 gene editing, including prediction models for selecting sgRNA with minimal off-target capacity, experimental validation of off-target cleavage, modulating expression levels of Cas9, Cas9 modifications including nickase and high-fidelity variants, and engineered sgRNA

References

1. Hsu, P. D. et al. DNA targeting specificity of RNA-guided Cas9 nucleases. Nature Biotechnology 31, 827–832 (2013).

2. Fu, Y. et al. High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells. Nature Biotechnology 31, 822–826 (2013).

3. Pattanayak, V. et al. High-throughput profiling of off-target DNA cleavage reveals RNA-programmed Cas9 nuclease specificity. Nature Biotechnology 31, 839–843 (2013).

4. Tsai, S. Q. et al. GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases. Nature Biotechnology 33, 187–198 (2015).

5. Wienert, B. et al. Unbiased detection of CRISPR off-targets in vivo using DISCOVER-Seq. Science364, 286–289 (2019).

6. Kim, S., Kim, D., Cho, S. W., Kim, J. & Kim, J. S. Highly efficient RNA-guided genome editing in human cells via delivery of purified Cas9 ribonucleoproteins. Genome Research 24, 1012–1019 (2014).

7. Davis, K. M., Pattanayak, V., Thompson, D. B., Zuris, J. A. & Liu, D. R. Small molecule-triggered Cas9 protein with improved genome-editing specificity. Nature Chemical Biology 11, 316–318 (2015).

8. Cho, S. W. et al. Analysis of off-target effects of CRISPR/Cas-derived RNA-guided endonucleases and nickases. Genome Research 24, 132–141 (2014).

9. Ran, F. A. et al. Double nicking by RNA-guided CRISPR cas9 for enhanced genome editing specificity. Cell 154, 1380–1389 (2013).

10. Kleinstiver, B. P. et al. High-fidelity CRISPR-Cas9 nucleases with no detectable genome-wide off-target effects. Nature 529, 490–495 (2016).

11. Hu, J. H. et al. Evolved Cas9 variants with broad PAM compatibility and high DNA specificity. Nature 556, 57–63 (2018).

12. Kulcsár, P. I. et al. Crossing enhanced and high fidelity SpCas9 nucleases to optimize specificity and cleavage. Genome Biology 18, (2017).

13. Kocak, D. D. et al. Increasing the specificity of CRISPR systems with engineered RNA secondary structures. Nature Biotechnology 37, 657–666 (2019).