Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
To mangle the old saw, there’s more than one way to edit a genome. Whether to fix a mutation for therapeutic purposes or insert a new one to see what it does, researchers have essentially three options: zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), and the new kid on the molecular block, CRISPR/Cas.
In all three cases, a targeted nuclease introduces a double-stranded break at a predetermined location in a cell’s DNA. In attempting to repair that damage, the cell will sometimes inadvertently kill the gene via nonhomologous end-joining (NHEJ) or rewrite its sequence to researchers’ specifications via homology-directed repair (HDR).
In either case the result is a quick-and-easy surgical genomic strike, and researchers are using that power to create gene knockouts, probe disease etiology, develop new pharmaceuticals and even as a form of genetic therapeutic.
Fundamentally, any of the three editing approaches will do. But as a practical matter, the CRISPR/Cas system—short for “clustered regularly interspaced short palindromic repeats”/CRISPR-associated—has caught fire in a way ZFNs and TALENs have not. Why? Simplicity and multiplexing.
Powered by those two advantages, the CRISPR/Cas system has been put to work in a flurry of publications across species from Anopheles to zebrafish. Naturally, molecular-tool developers have been paying attention, and there now exists a rich set of reagents and tools to help researchers implement the system in their own labs.
Molecular tools
ZFNs and TALENs require laborious and sometimes lengthy cloning, engineering and optimization, but all that’s required to use the CRISPR/Cas system is the Cas9 nuclease and an inexpensive 20-nucleotide “single-guide RNA” (sgRNA), which tells the nuclease where to cut. Several companies now offer predesigned expression tools and reagents to express one or both of these elements.
Sigma Aldrich, for instance, sells predesigned plasmids that express both Cas9 and sgRNAs targeting every human, mouse or rat gene—users simply select the sgRNA they want to use with an online ordering tool. Custom designs also are available.
“At least on the face of it, the design [of a genome-editing experiment] is more straightforward” with CRISPRs than with ZFNs, explains Shawn Shafer, functional genomics market segment manager at Sigma Aldrich. “If you have a 20-bp target, you’re just synthesizing the complement to that. Whether that’s a good clean target or not is a different story, but the design is more straightforward and easy because it’s an RNA-DNA interaction.”
Sigma also offers libraries of purified sgRNAs (which can be transfected into cells directly or used in combination with purified Cas9 enzyme), as well as a “paired nickase” design involving two plasmids, one to express the Cas9 and a second for the guide RNA. A nickase is simply a mutant Cas9 protein that cuts only one strand of DNA. The paired nickase strategy minimizes off-target targeting by requiring two independent but closely spaced binding and cutting events to initiate editing.
Also supporting CRISPR/Cas is Thermo Fisher Scientific. The company’s GeneArt® CRISPR Nuclease Vectors come in two formats, one for flow-cytometric enrichment of transfected cells and the other for purification on beads. But Thermo Fisher also offer reagents to express Cas9 from an mRNA rather than a plasmid. In this case, the mRNA is delivered alongside either an in vitro transcribed sgRNA or a “CRISPR Strings™” construct, which is a short DNA that expresses the guide RNA in cells from a U6 promoter.
According to Helge Bastian, vice president and general manager of the synthetic biology business at Thermo Scientific, these CRISPR Strings reagents are part of the company’s efforts to make it easier for genome-editing neophytes to implement the technology. Normally, Bastian explains, users deliver the nuclease and/or guide sequences as RNAs. “But of course, the people handling this RNA need to be trained and aware that you shouldn’t work in an RNAse-rich environment, so that your most important molecule … isn’t degraded.” By using DNA-based constructs instead, he says, these researchers “can execute on these very straightforward experiments, because the material they get from us has been stabilized and is provided in a ready-to-go format.”
Of rAAV and Purified Cas
Horizon Discovery also offers a range of CRISPR/Cas reagents and tools, including QuickStart cell lines that constitutively express the Cas9 nuclease. All that is required to perform editing in that case is to supply one or more sgRNAs.
The company also offers recombinant adeno-associated viral (rAAV) vectors that can be used to deliver donor template sequences for efficient HDR-based editing. According to Eric Rhodes, the company’s chief technical officer, rAAV, as a single-stranded DNA, is easier to get into the nucleus than traditional plasmids, and thus provides more robust HDR-mediated repair—albeit with a size limit of about 5 kb. “That, in our hands, has proven to be the most effective donor,” he says.
Another newly available reagent comes from New England Biolabs, which offers purified Cas9 nuclease. According to Brett Robb, the company’s head of RNA research, the purified Cas9 enzyme has two key applications. First, when combined with an sgRNA in vitro, it can be used as a custom restriction enzyme. “We know that people are … using it to create basically long recognition sequence restriction enzymes for cloning,” he says.
Alternatively, researchers can deliver the Cas9 protein directly to cells to effect genome editing without delivering either a Cas9 expression construct or mRNA, a strategy that provides finer control of the Cas9 dose than nucleic acids can provide. Indeed, several recent studies have demonstrated the efficacy of this approach. In one, South Korean researchers delivered preformed Cas9/sgRNA ribonucleoprotein (RNP) complexes into cultured fibroblasts and embryonic stem cells. The strategy, they write, produced “site-specific mutations at frequencies of up to 79%, while reducing off-target mutations associated with plasmid transfection at off-target sites that differ by one or two nucleotides from on-target sites” [1].
The second study, led by Alex Schier’s lab at Harvard University, microinjected preformed Cas9-sgRNA RNPs into zebrafish embryos to enhance mutagenesis frequency compared to codelivered Cas9 mRNA and sgRNA. In one case, the efficiency of NHEJ-mediated mutation increased from about 8% to 47%, a six-fold improvement, which the authors suggest may be due to overcoming limitations caused by RNP complex formation and/or sgRNA instability in vivo [2].
Assessing efficacy and accuracy
After an editing experiment is concluded, you need to determine whether it worked. Thermo Fisher's GeneArt Genomic Cleavage Detection Kit enables users to quantify efficiency by PCR amplifying across the putative editing site. The amplified DNA is then denatured and allowed to reanneal. If editing occurred, those amplicons will have a different sequence than unedited (or differently edited) ones; upon reannealing to other sequences, they will produce small single-stranded regions of noncomplementarity. These regions are cut by an enzyme included with the kit, and the products run on a gel.
(Thermo Fisher's manual [PDF] for this kit doesn’t name the enzyme used to digest the DNA following annealing. But a protocol to perform essentially the same process using T7 endonuclease I is described on the New England Biolabs web site.)
Alternatively, researchers can try to sequence genomic DNA to see what changed in edited cells. Whole-genome sequencing is for the most part too costly and impractical for that purpose, says Jonas Korlach, chief scientific officer at Pacific Biosciences, but more targeted approaches will work, too.
Earlier this year, Ayal Hendel and colleagues used Pacific Biosciences’ PacBio RS to deep sequence the on-target consequences of genome editing using ZFNs, TALENs or CRISPR/Cas [3].
The long read length of PacBio’s SMRT sequencing chemistry (approximately 8.5 kb per read, on average) was key to the study, Hendel says, because the goal was to sequence across the full length of the donor sequence, which could be up to 800 bases long per side—far too long for Illumina’s comparatively shorter reads.
The results suggest CRISPR/Cas is generally more efficient than TALENs in both NHEJ and HDR-mediated editing, though with site-to-site variation. More importantly, they document the occurrence of what Hendel calls “mini-translocations,” in which unintended donor sequences, such as the plasmid DNA expressing the nuclease or even other cellular chromosomal DNA, is improperly copied into an editing site in place of the desired donor sequence. In one case, the team observed a 289-nucleotide insertion in the beta-hemoglobin locus arising from a TALEN-expression construct.
Hendel says his lab is using this tool as a metric with which to optimize HDR-mediated editing efficiency, which his team hopes to apply to genetic blood disorders. Right now, he says, the rate of HDR-mediated editing with hematopoietic stem cells is about 3% to 5%. “We are working to improve that [rate], to move forward with this technology for gene-therapy applications.”
If you’d like to move your own research forward, the tools are available to get started. All you really need to do is decide on a guide RNA. Predesigned sgRNA libraries are available (from Addgene and Sigma-Aldrich, for instance). But if you find yourself needing to design new targeting molecules, there are tools for that, as well, including CHOPCHOP, for both sgRNA and TALEN design, and Feng Zhang’s CRISPR Design.
Now, what are you waiting for?
References
[1] Kim, S, et al., “Highly efficient RNA-guided genome editing in human cells via delivery of purified Cas9 ribonucleoproteins,” Genome Res, 24:1012-9, 2014. [PubMed ID: 24696461]
[2] Gagnon, JA, et al., “Efficient mutagenesis by Cas9 protein-mediated oligonucleotide insertion and large-scale assessment of single-guide RNAs,” PLoS ONE, 9:e98186, 2014. [PubMed ID: 24873830]
[3] Hendel, A, et al., “Quantifying genome-editing outcomes at endogenous loci with SMRT sequencing,” Cell Reports, 7:293-305, 2014. [PubMed ID: 24685129]
Correction (Aug. 22): The original version of this document referred to the GeneArt portfolio as a product of Life Technologies. Life Technologies has been acquired by Thermo Fisher Scientific. The article has been updated to reflect that change.