Uh-oh! 2 New Studies Emphasize CRISPR Off-Target Edits and Imperfections

iluustration CRISPR-Cas9 enzyme working on DNA

CRISPR, the gene editing technique that promises to revolutionize healthcare and medicine, is not perfect. Right from the beginning, there have been concerns that this technique, which makes it easy to select specific areas of the genome and quickly and easily snip out and replace pieces of DNA, may also make unintended cuts in other areas of the genome at the same time. And two new studies, one in mice embryos and the other in rice plants, seem to confirm this.

A 2017 article in Nature brought up the issue, then was later retracted when the authors admitted they had made a mistake. However, there have remained concerns about off-target edits.

There are ongoing human clinical trials in the U.S., Europe and China using CRISPR. They have been overshadowed by the November 2018 story of Chinese researcher He Jiankui, who used CRISPR-Cas9 gene editing to alter the DNA of embryos for seven couples, with the resultant birth of a set of twins and another pregnancy. This was met by global condemnation.

The mouse study evaluated two base editors. They are different from classic CRISPR in specific ways. In classic CRISPR, a guide RNA finds a specific DNA sequence and an enzyme is used to cut the DNA, slicing out the piece of the gene that causes a disease or inserting preferred DNA. Base editors also use guide RNA, but they only cut a single DNA strand instead of the double-strand, then they convert one target nucleotide into another. Many inherited diseases are caused by a single incorrect nucleotide.

But these “point mutations” are all throughout the genome. As a result, it’s hard to evaluate whether base editors are making only the changes they’re supposed to.

In the mouse research, Chinese researchers led by Hui Yang of the Shanghai Institutes for Biological Sciences utilized base editing in a single cell of a two-cell mouse embryo, leaving the second cell unedited. The two cells are genetically identical. As a result, the researchers could compare them and their descendant cells, allowing them to identify off-target edits without being confused by which ones occurred naturally and which ones came about because of the CRISPR edits.

What they found was that an adenine base editor, which transforms AT nucleotide pairs into GC nucleotide pairs, had almost no off-target edits. However, the cytosine base editor, which changes CG pairs into TA pairs, changed about one nucleotide in 20 million. In the mouse genome, which has about the same number of nucleotides as humans, there were about 150 off-target edits.

The research on rice found similar off-target edits in the cytosine base editing technique, but not the adenine type.

Click to sign up for job alerts

How big of a problem is a one-in-20 million mistake?

That’s not clear. DNA mutates constantly because of background radiation in the environment, other environmental factors and just random mistakes. The “natural” rate of mutation can be as high as one in one million. The body has repair mechanisms for many, and many don’t have any major effect. But when they do, they can lead to cancer or any number of other diseases and disorders.

David Liu, a biochemist with Harvard University, who invented base editing, and who co-founded a Cambridge, Mass. biotech company, Beam Therapeutics, told STAT, “One in 20 million is in that range.” So it’s possible it might “have little or no impact” on patients who might undergo base editing procedures.

The mouse used levels of the base editor that were much higher than what would likely be used for therapy. The co-author of the study, Lars Steinmetz of Stanford University, told STAT that using a lower concentration of the base editor created fewer off-target edits, “but they didn’t disappear.”

What the studies seem to suggest, at least from a broader overview, is there’s still quite a bit that needs to be understood about CRISPR before it’s routinely used safely in clinical studies. It’s important to note that researchers at both universities and biotech companies are constantly developing newer versions of CRISPR or using other enzymes than Cas9, variations on a theme that may prove to be more accurate and safer than the original techniques. The future is still bright, but caution is recommended.

Back to news