CRISPR Nucleases: The Ultimate Guide To Selection

CRISPR nucleases continue to dominate the gene-editing landscape because they are powerful and easy-to-use gene-editing tools. Their rapid discovery and development, however, make choosing the right nuclease for your experiment difficult. This guide highlights the broad range of available CRISPR nucleases and distils the major factors you should consider when choosing one for your next gene-editing experiment.

New to CRISPR? See here and here for articles on getting started with CRISPR.

S. pyogenes Cas9: “The Original (CRISPR) Gene Editor”

Cas9 from S. pyogenes (SpCas9) was the first, and remains the most popular, CRISPR nuclease for gene editing. SpCas9 is a large CRISPR nuclease (1368 amino acids) and possesses a relatively relaxed protospacer adjacent motif (PAM; 5’-NGG) compared to its orthologs (1-3). Because it was the first to be repurposed as a gene editor, SpCas9 is the most well-characterized of all CRISPR systems. This means it’s readily available in virtually any format (plasmid, mRNA, and protein) and amenable to most genome engineering applications. In short, SpCas9 is a solid “entry-level” CRISPR nuclease for labs new to gene editing.

Cas9 From Other Species

Although SpCas9 is the most popular, many naturally-occurring Cas9 orthologs have been discovered (4-7). For gene-editing purposes, there are two main characteristics that differentiate these orthologs from SpCas9: overall size and PAM specificity.

Size matters for hard-to-transfect cell lines (vectors can easily exceed 10kb) and AAV applications (AAV can only package ~4.5 kb). Compact Cas9 orthologs include those isolated from S. aureus (SaCas9) and C. jejuni (CjCas9), both of which can be packaged in AAV.

PAM specificity, on the other hand, dictates the available “target space” for each ortholog, or how many guide RNA (gRNA) binding sites there are in a given genome for a given PAM. Orthologs with relaxed PAMs can interact with more genomic sites than those with stringent PAMs. As you can imagine, there are more instances of “5’-NGG” (PAM of SpCas9) in any given genome than “5’-NGGNG” (PAM of St3Cas9; N = any nucleobase). Relaxed PAMs are good if you need more flexibility in gRNA design, but you should avoid them if you are trying to avoid off-target activity. (More gRNA binding sites means more potential off-target activity!) See Table 1 for more details of the various Cas9 orthologs.

Other CRISPR Nucleases

CRISPR gene editing does not end with Cas9. The realization of Cas9’s gene-editing capabilities prompted researchers to search for other naturally-occurring CRISPR systems that could be repurposed similarly. This led to the discovery of Cpf1 (also known as Cas12a), Cas12b, and CasX (also known as Cas12e) — all of which are RNA-guided nucleases capable of gene editing (8-10). Together, these CRISPR nucleases possess features — such as small size and unique PAMs — that diversify the CRISPR toolbox, allowing researchers to select CRISPR nucleases that will work best for their specific experimental system or application. Table 1 provides details of the features of these alternative CRISPR nucleases.

Engineered Nucleases

Researchers have devoted significant effort to mitigate some of the unfavourable aspects of various CRISPR nucleases, including altering PAM specificity, reducing size, limiting off-target activity, and repurposing CRISPR nucleases for other genomic engineering tasks (e.g., gene activation and repression). This has been done largely by introducing various point mutations in mechanistically-important residues. Some examples include, but are not limited to, the following.

High-fidelity SpCas9 (SpCas9-HF1, eSpCas9-1.1) (11, 12)

Four alanine substitutions (N497A, R661A, Q695A, Q926A) eliminate nonspecific interactions between Cas9 and target DNA and drastically reduce Cas9’s off-target activity. An independent group developed a similar variant called eSpCas9-1.1 and which bears the aforementioned mutations along with K848A, K1003A, and R1060A. (Note: Some commercial vendors offer their own high-fidelity Cas9’s; however, it is not clear what mutations have been introduced in these commercial variants.)

SpCas9 with relaxed PAM (SpCas9-NG) (13)

This engineered SpCas9 replaces R1335 with a valine to disrupt the interaction between Cas9 and the second G of the 5’-NGG PAM. However, because this substitution ablates Cas9’s nuclease activity as well, additional substitutions (L1111R, D1135V, G1218R, E1219F, A1322R, T1337R) were made based on molecular modeling to restore the cleavage activity of this “relaxed” Cas9 variant.

SpCas9 nickase (Cas9n) (14)

A single alanine substitution (D10A or H840A) transforms Cas9 into a “nickase” that can only cut one strand of target DNA. Two Cas9 nickases must be paired (one bearing D10A and the other H840A) in order to generate a double-strand break. This approach is associated with fewer off-target DSBs than SpCas9.

Nuclease-dead SpCas9 (dCas9) (15-17)

Two alanine substitutions (D10A and H840A) eliminate Cas9’s nuclease activity. Because it’s DNA-targeting ability is not affected, various effector domains can be fused to dCas9 to enable programmable transcriptional and/or epigenetic control. Other CRISPR nucleases have been adapted similarly.

The table below summarizes the CRISPR nucleases discussed above and offers further details on their use. All comparisons (e.g., “Smaller target space” or “More compact”) are relative to SpCas9. As discussed above, target space refers to the relative number of gRNA bindings sites for each CRISPR nuclease. Also, because off-target effects are specific to each guide RNA and genome, this characteristic has been omitted from this table except for Cas9n and high-fidelity variants of Cas9. That said, CRISPR nucleases with shorter PAMs have fewer off-target effects on average.

This table is by no means a complete list of available CRISPR nuclease reagents. In fact, you can find all of the CRISPR nucleases discussed above and more on Addgene. Addgene provides some good resources for navigating their extensive catalog of CRISPR-related plasmids and viral preparations.

How to Choose a CRISPR Nuclease

As mentioned above, size and PAM specificity are critical parameters that you should focus on when choosing a CRISPR nuclease. These features dictate the ease with which you’ll be able to deliver your nuclease to your cells and the available genomic sites that your nuclease will be able to target (GC- vs AT-rich sites), respectively.

Another factor you should consider is the format that you’ll use for your actual experiment — plasmid, mRNA, or protein? Ribonucleoprotein (RNP) complexes (i.e., CRISPR nuclease protein in complex with a guide RNA) offer the best gene-editing efficiencies of any delivery format and are generally associated with fewer off-target effects due to the transient nature of this format (18). That said, only Cpf1 and select orthologs of Cas9 are commercially available as purified protein for RNP delivery. More advice on CRISPR delivery formats can be found here.

References:

1. L. Cong et al., Multiplex genome engineering using CRISPR/Cas systems. Science 339, 819-823 (2013).
2. P. Mali et al., RNA-guided human genome engineering via Cas9. Science 339, 823-826 (2013).
3. M. Jinek et al., A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science 337, 816-821 (2012).
4. Z. Hou et al., Efficient genome engineering in human pluripotent stem cells using Cas9 from Neisseria meningitidis. Proc Natl Acad Sci U S A 110, 15644-15649 (2013).
5. B. P. Kleinstiver et al., Engineered CRISPR-Cas9 nucleases with altered PAM specificities. Nature 523, 481-485 (2015).
6. E. Kim et al., In vivo genome editing with a small Cas9 orthologue derived from Campylobacter jejuni. Nat Commun 8, 14500 (2017).
7. M. Yamada et al., Crystal Structure of the Minimal Cas9 from Campylobacter jejuni Reveals the Molecular Diversity in the CRISPR-Cas9 Systems. Mol Cell 65, 1109-1121 e1103 (2017).
8. J. J. Liu et al., CasX enzymes comprise a distinct family of RNA-guided genome editors. Nature 566, 218-223 (2019).
9. J. Strecker et al., Engineering of CRISPR-Cas12b for human genome editing. Nat Commun 10, 212 (2019).
10. B. Zetsche et al., Cpf1 is a single RNA-guided endonuclease of a Class 2 CRISPR-Cas system. Cell 163, 759-771 (2015).
11. I. M. Slaymaker et al., Rationally engineered Cas9 nucleases with improved specificity. Science 351, 84-88 (2016).
12. B. P. Kleinstiver et al., High-fidelity CRISPR-Cas9 nucleases with no detectable genome-wide off-target effects. Nature 529, 490-495 (2016).
13. H. Nishimasu et al., Engineered CRISPR-Cas9 nuclease with expanded targeting space. Science 361, 1259-1262 (2018).
14. F. A. Ran et al., Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity. Cell 154, 1380-1389 (2013).
15. L. A. Gilbert et al., CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes. Cell 154, 442-451 (2013).
16. M. L. Maeder et al., CRISPR RNA-guided activation of endogenous human genes. Nat Methods 10, 977-979 (2013).
17. A. W. Cheng et al., Multiplexed activation of endogenous genes by CRISPR-on, an RNA-guided transcriptional activator system. Cell Res 23, 1163-1171 (2013).
18. M. Kosicki et al., Dynamics of Indel Profiles Induced by Various CRISPR/Cas9 Delivery Methods. Prog Mol Biol Transl Sci 152, 49-67 (2017).