New CRISPR gene editing tool could overcome problems

Switching enzymes from Cas9 to Cas12a would mean safer and more effective gene editing.

The gene editing tool CRISPR-Cas9 is one of the most significant scientific advances in recent years. But recent studies have found that it may not be as precise as previously thought. In a new study, researchers found that the problem lies with Cas9. Cas9 acts as CRISPR’s most popular “molecular scissors” to cut the DNA at a specific location so that bits of DNA can then be added or removed. To fix the problem, researchers from the University of Texas at Austin are making the case to switch from Cas9 to one of the lesser-known CRISPR enzymes called Cas12a.

We spoke to two of the study’s authors Ilya Finkelstein and Isabel Strohkendl about the work.

ResearchGate: What motivated this study?

Ilya Finkelstein & Isabel Strohkendl: This study was motivated by the growing excitement and potential for using Cas12a as a genetic engineering tool.

Both Cas9 and Cas12a are programmed by a short RNA molecule to nearly any genomic location, and both enzymes are active in humans and other model organisms. Yet Cas9 has been reported to bind and cut at off-target sites, leading to unanticipated and unwanted gene editing. When we entered this field, several studies reported anecdotal evidence that Cas12a is more selective than Cas9 in human cells. But these studies all failed to address a deep mystery about why Cas12a is more accurate than Cas9. At the core, both enzymes use the same recognition method—a short RNA—to recognize a genomic site. We were motivated by answering this key question and in the process discovering the rules that govern specificity by Cas12a and other emerging CRISPR enzymes.

RG: Where do the concerns about Cas9 come from?

Finkelstein & Strohkendl: A number of recent publications raised significant concerns with Cas9, including its propensity to bind and cut at off-target sites, its propensity to illicit an immune response in some humans, and the potential for Cas9 to cause large chromosomal rearrangements near the target site.

The most significant concern about Cas9 (as with all CRISPR enzymes) is that it will often cut DNA sequences that resemble the intended target sequence but that are incorrect. These unintended “off-target” cleavage events in a cell could be detrimental. This is why several groups are reporting engineered and natural Cas9 variants that supposedly improve targeting specificity. Whether these engineered variants are indeed better remains to be tested broadly by an independent third party. Nonetheless, off-target cleavage is still a pervasive problem.

“Cas9 was catapulted to prominence by a confluence of luck and historical precedent.”


 RG: How did Cas9 become the preferred method in the first place?

Finkelstein & Strohkendl: As with many key discoveries in biology, Cas9 was catapulted to prominence by a confluence of luck and historical precedent. In the early 2000s, several groups began to discover and characterize a heretofore unknown adaptive immune system. This system—now known as CRISPR-cas—protected bacteria and archea against infection by viruses and other foreign DNAs and RNAs.

Cas9 was unique because it encapsulated DNA recognition (targeting) and DNA cleavage into a single multi-functional polypeptide. In short, Cas9 functioned as a single protein rather than a complex of many subunits. This simplicity meant that Cas9 was simpler to port into a human cell. An additional key discovery was that the Cas9 CRISPR RNA could be simplified into a single molecule. Thus, Cas9 became the first easily programmable CRISPR nuclease that could also function in human cells.

This illustration shows the protein cas12a bound to a DNA helix (red and white). Credit: T. Yamano, H. Nishimasu, & James Rybarski.

RG: Can you tell us about Cas12a? What’s the difference to Cas9?

Finkelstein & Strohkendl: Cas12a has several key advantages that in my opinion will make this nuclease a cut above Cas9 for real-world applications in biomedicine. The biggest, most important difference is the specificity of these enzymes. Both Cas9 and Cas12a nucleases rely on a 20-nucleotide long guide RNA to identify target DNA. But Cas9 tolerates sequence errors in half of the guide, whereas Cas12a only tolerates sequence errors at very few positions.

In addition, Cas12a processes its own guide CRISPR RNA and uses it to scan for complementary sequences within DNA. This means Cas12a is more autonomous than Cas9 and can more easily target two or more different DNA sequences at the same time within the cell.

Different parts of the genome are inaccessible to Cas9 and Cas12a. In this sense, the two enzymes are complementary. Cas9 and Cas12a also cut the DNA in slightly different ways so that the newly-produced DNA ends are different and might favor certain DNA repair responses over others in the cell. The consequences of this difference in DNA cutting patterns is still being determined by us and other research groups.

Finally, we and others are actively working on engineering better Cas12a variants that will further improve on an already excellent enzyme. In short, we still don’t know how much better Cas12a can get but the future is bright indeed.

“A toolbox of different and diverse CRISPR enzymes will broaden our ability to safely edit the human genome.”


 RG: Does Cas12a have any disadvantages to Cas9?

Finkelstein & Strohkendl: Cas12a has a few minor disadvantages relative to Cas9. For example, after target recognition, its cutting activity can cleave unprotected single stranded DNA that is not immediately adjacent to the DNA binding site. Although this activity has been observed in the test tube, it’s not clear whether it has a significant impact on cellular health. More work clearly needs to be done to determine how this additional activity, not present in Cas9, will manifest in real-world applications.

On the flip side, this novel activity is already being exploited in several nucleic acid diagnostics kits. So the jury is still out as to whether this is a feature or a bug. But I’m confident that if necessary, this activity can be mitigated by further Cas12a protein engineering and optimization.

Another difference between the two proteins is that Cas12a remains tightly associated with one of the two DNA molecules after the genome is cut. We don’t know how this impacts gene editing and repair. More broadly, these open questions reflect the sheer volume of information available about how Cas9 works and how little we still know about Cas12a. This gap will continue to shrink rapidly as Cas12a gains prominence and starts to eclipse Cas9 as a more precise gene-editing tool.

RG: Does Cas12a bring us closer to using CRISPR in humans?

Finkelstein & Strohkendl: Both Cas9 and Cas12a are already used in human cells and many model organisms, including mice. Clinical trials are planned for both enzymes. One puzzling finding is that in some gene targets, Cas9 appears more efficient, whereas in others Cas12a is the tool of choice over Cas9. The practical reality is that a toolbox of different and diverse CRISPR enzymes will broaden our ability to edit across the entire genome. As with most other biomedical applications, I don’t anticipate that a single enzyme will fit all use cases. The more the merrier!

RG: What’s next for your research?

Finkelstein & Strohkendl: More broadly, we are continuing to probe the mechanisms of new and emerging CRISPR nucleases. Nature has already built an extensive toolkit and we are just scratching the surface of what’s out there!

Featured image courtesy of T. Yamano, H. Nishimasu, & James Rybarski.