Originally published on GMWatch

Scientists at the University of Uppsala have found that CRISPR-Cas gene editing in zebrafish caused large structural changes at on-target and off-target sites (both at the intended edit site and elsewhere in the genome). Off-target sites can be very similar to the target site, which means that the “gene scissors” can also cut at these sites and cause specific unintended mutations. The publication, which is currently in pre-print format and has not yet been peer reviewed, shows that major unintended DNA changes are possible.

In the study, the scientists examined the genome of zebrafish after using CRISPR-Cas. They found that large structural changes (large insertions and deletions) had occurred at both the target sequence and at so-called off-target sites.

In the medical field, in which the authors were conducting their research, such large changes had previously only been found at the target site (on-target). However, large off-target changes have been found to occur in plants. Biswas et al (2020), in experiments on CRISPR-edited rice plants, found an unintended large deletion “far downstream of the target” site – in other words, off-target.

In the zebrafish experiment, the major structural changes at on- and off-target regions were found both in the fish that developed from the injected eggs and their offspring.

The results are highly relevant for clinical applications of CRISPR-Cas, as such unexpected effects could have serious consequences for patients.

Overall, it is important that the genome of the target organisms is examined in detail after CRISPR/Cas experiments, at both on- and off-target sites to detect unintended changes that may affect other gene sites.

Most screening for unintended CRISPR effects is inadequate

The authors of the study in zebrafish also draw attention to the inadequacy of the methods generally used to screen for unintended effects in gene-edited organisms – short-read sequencing and/or short-range PCR. They emphasise that long-read sequencing must be used (something that GMWatch has repeatedly advised in the context of gene-edited plants and farm animals).

While the researchers did use long-range PCR, they warn that this method alone “is unable to capture large genome aberrations and complex genomic rearrangements, such as chromothripsis and whole chromosome deletions”. Detecting such events, they state, requires an additional analysis that studies the whole genome of the edited organism.

They have developed their own in vitro screening method, called Nano-OTS, which they suggest should be used as a first step in the analysis of unintended effects, in combination with long-read sequencing and other methods to try to spot as many as possible unintended effects of CRISPR gene editing.

The latest study adds to a long list of studies showing unintended effects of gene editing, as collected by GMWatch. This list has been recently updated and will be further updated in the coming days.

Thanks to Testbiotech for some elements of this commentary. Additional analysis by GMWatch.