Since the CRISPR-Cas9 gene editing system debuted in 2012, it has been hailed as both evolutionary and cautionary. It’s undeniable promise as a treatment for genetic disorders clashes with its potential to induce harmful permanent edits to a human’s gene line, along with ethical concerns regarding its nefarious use to create “designer” babies.
In the almost decade since CRISPR-Cas9 gained mainstream attention, it has been the subject of thousands of published papers, some showing the system’s editing accuracy with others suggesting more research to ensure an off-target edit doesn’t harm a human’s germline. A new paper from researchers at the Children’s Hospital of Chicago falls into the former category with results on-par with previous data from animal models that suggest gene editing poses minimal risk to the rest of the genome.
Although CRISPR-Cas9 has been used in early stage clinical trials for cancer, sickle cell disease and childhood blindness in the United States, all editing has been done ex vivo. Researchers remove cells from the body, edit the target gene and then return it to the body. Thus far, gene editing has only been performed in vivo on animal models, including zebrafish, mice, dogs, goats and pigs. In the current study, Nico Katsanis and colleagues performed whole exome sequencing in over 50 individual organisms from three generations of zebrafish, allowing robust testing of long-term gene editing effects.
Using a trio sequencing workflow aligned with best practices for the Genome Analysis Toolkit (GATK), the researchers did not observe an increase in the number of transmissible variants in the two generations of offspring.
Importantly, the researchers specifically examined predicted off-target sites for genetic mutations using three algorithms: the MIT CRISPR design site, the CRISPR direct engine and CAS-OFFinder. Looking within 100 bp flanking of predicted off-target sites, Katsanis and his team found no support for variants in the first generation, and sporadic variants near predicted sites in the second generation. However, the number of variants was not significantly different than expected by chance.
Together with the data indicating no variants or inflated mutation count in the whole genome of neither the first generation nor the multigenerational cohort, the researchers concluded that off-target CRISPR events occur infrequently.
“Our data add to a growing and important body of work from the scientific community that unintended mutations from gene editing with CRISPR-Cas9 are extremely rare," said Katsanis, director of the Advanced Center for Translational and Genetic Medicine at Ann & Robert H. Lurie Children's Hospital of Chicago. “These findings help to alleviate concerns about harmful errors when gene editing is used in humans. Our results provide reassurance that this technology is a valuable and valid tool with great promise for the treatment of genetic disorders.”
Katsanis and co-authors noted several study limitations, such that they did not assess large structural variants or deletions at the on-target site. They urge further large-scale research projects with zebrafish, noting several that are currently ongoing.