The CRISPR system, which involves a Cas enzyme to cut DNA, is a powerful tool for gene editing. But the genetic scissors sometimes make changes at the wrong place, creating a major safety problem that could limit their therapeutic use.
Now, scientists at the University of Texas (UT) at Austin have refined the Cas9 protein used in the Nobel Prize-winning CRISPR-Cas9 tool. The new version, dubbed SuperFi-Cas9, was thousands of times less likely to perform off-target editing but just as efficient at on-target editing as the original version, the team said in a paper published in Nature.
“This really could be a game-changer in terms of a wider application of the CRISPR-Cas systems in gene editing,” Kenneth Johnson, Ph.D., the study’s co-senior author, said in a statement.
To locate its editing target, the CRISPR-Cas9 system uses a guide RNA of 20 nucleotides to search for a matching DNA sequence. Sometimes, though, the guide RNA takes Cas9 to imperfectly matched DNA sequences as differences in spots 18 through 20 are somehow ignored. The phenomenon, called mismatch tolerance, can cut DNA at the wrong places, potentially leading to dangerous mutations.
The UT Austin researchers set out to understand the mechanism for Cas9 activation during a mismatch. They used a cryoelectron microscope to capture the structural changes of Cas9 as it interacted with the mismatched DNA sequence.
When Cas9 encounters a mismatch in 18 through 20, it forms a fingerlike linear structure (cyan in image above) that latches onto the DNA, "locking" Cas9 in an active conformation, the team found. The structure stabilizes the guide RNA-DNA duplex, paving the way for Cas9 to make the cut.
“It’s like if you had a chair and one of the legs was snapped off and you just duct taped it together again,” Jack Bravo, Ph.D., a co-first author of the study, explained in a statement. “It could still function as a chair, but it might be a bit wobbly. It’s a pretty dirty fix.”
Armed with the knowledge, the UT Austin team designed a new Cas9 variant by introducing mutations to the fingerlike structure so that it doesn’t stabilize the DNA during a mismatch. Without that stabilization, Cas9 won't make the cut.
In test tubes, the resulting SuperFi-Cas9 showed a similar rate of cutting at the right DNA target as the natural Cas9 but was 4,000 times less likely to cleave off-target sites, the team reported.
CRISPR-Cas has been widely applied in genetic research thanks to its simplicity and low cost, and it holds great promise for use in therapies. Recognizing that off-target cleavage of DNA is a key weakness, other researchers have been working on tweaks, too. To improve the precision of the CRISPR system, a team at the University of California, Berkeley created Cas9 variants called ProCas9s that can only be activated in a specific type of cell.
Other Cas enzymes have also been explored. A Stanford University team recently designed variants of Cas12f that can perform editing in mammalian cells. The CasMINI proteins are smaller than Cas9. Mammoth Biosciences, a gene editing biotech co-founded by CRISPR pioneer Jennifer Doudna, Ph.D., is also working on small Cas enzymes, including Cas14 and Casɸ. A small enzyme size could potentially allow scientists to add other components to fine-tune the CRISPR-Cas system.
Insights into the structural mechanisms behind off-target effects of Cas9 “provide a molecular blueprint” for the design of next-generation Cas9 variants, the UT Austin researchers said. These new versions could reduce safety problems while retaining efficient, on-target gene editing. The team is now collaborating with other researchers to test SuperFi-Cas9 for gene editing in living cells.