CRISPR enzyme can boost growth of cells with cancer mutations
Cas9 can cut DNA without any guide RNA and select for cells with inactivated p53.
By Sarah C.P. Williams
Uri Ben-David is senior author of a study that looked at the effect of Cas9 expression in cell lines. Credit: Maria Nemchuk, Broad Communications
CRISPR-Cas9 gene editing technology has revolutionized biomedical research, allowing scientists to alter DNA rapidly and at low cost. This has led to new ways of screening cells for the effects of genetic mutations, and potential new treatments now in development for various diseases.
But the CRISPR-Cas9 system can make unintended cuts in DNA. The approach couples the DNA-cutting Cas9 protein with a guide RNA that brings the molecular scissors to the desired spot in the genome.
New research published in Nature Genetics shows that Cas9 can cut DNA even without any guide RNA. Moreover, Cas9, when introduced into many cell lines, ultimately favored the growth of cells with mutations in the p53 gene — the same mutations found in many cancers. This gene is important for repairing DNA damage and blocking the growth of cells with damaged DNA, such as cancer cells.
To learn more about what these findings mean for CRISPR-Cas9 gene editing, we spoke with Uri Ben-David, who initiated the work during a postdoctoral research fellowship in the labs of Todd Golub, director of the Cancer Program and chief scientific officer at the Broad Institute of MIT and Harvard, and Rameen Beroukhim, associate professor at Dana-Farber Cancer Institute and associate member at the Broad, who are both authors of the study.
Ben-David, senior author, is now an assistant professor at Tel Aviv University. Oana Enache of Broad and Veronica Rendo of Dana-Farber are co-first authors of the study and they worked in collaboration with other colleagues from Broad and Dana-Farber.
Q. What did you set out to study?
A. We wanted to know the effect of Cas9 expression alone on cells. It’s become pretty common when you’re using CRISPR that you first generate a stable cell line with Cas9, and then on top of that you add any guide RNA to direct the Cas9 to your gene of interest. Generally speaking, Cas9 alone is considered to be pretty neutral, and a Cas9-expressing cell line is considered more or less equivalent to the parental cell line. But the question we had was: Are there any undesired consequences to expressing Cas9 in cells?
Q. Were you surprised by what you found?
A. Yes. Cas9 shouldn’t cut DNA when it’s not directed to do so. But we found elevated levels of DNA damage in cells expressing Cas9, even in the absence of guide RNAs. And a high fraction of the 165 cell lines we looked at had activation of the p53 pathway—we weren’t expecting that to occur in the absence of guide RNAs. In a smaller subset of these cell lines, we found mutations that inactivated p53. These are the same mutations that are found in cancer cells, and allow them to grow unchecked even in the presence of DNA damage.
Q. What’s the importance of the p53 findings?
A. p53 does a lot of things; it helps keep DNA-damaged cells from dividing. What happens when cells have a mutation in p53 is that when they acquire DNA damage, the cells keep dividing; they don’t have that brake anymore.
If Cas9 is triggering DNA breaks and p53 activation, then cells without functioning p53 will have a competitive advantage because their proliferation isn’t limited. What probably happened in the cell lines we studied was not that the p53 gene was mutated somehow by Cas9, but that there was already a very small, undetectable subpopulation of cells with p53 mutations that quickly expanded due to this selection pressure.
Keep in mind that we already knew that p53 could be activated when you introduce Cas9 together with guide RNAs. We’ve just shown that this can also occur without the guide RNAs, and can result in the emergence or expansion of p53-inactivating mutations.
Q. Does this change what scientists thought they knew about Cas9 and how to use it?
A. The findings beg the question of why and how Cas9 is cutting DNA in the absence of guide RNAs. Could there be natural RNAs in the cell that somehow direct the Cas9?
The big takeaway for researchers is that you can’t assume Cas9-expressing cell lines are otherwise identical to parental cell lines. You need to use Cas9-expressing cells as your controls and you need to make sure you have a handle on what kinds of mutations Cas9 might have introduced to your cells.
In some cases, it might mean that it’s preferable to add guide RNA at the same time as Cas9 rather than introducing Cas9 first before the guide RNA. You also could use a system where you express Cas9 only transiently rather than have it stably expressed through the lifetime of a cell.
Q. What does this mean for CRISPR-Cas9-based therapeutics that are in development?
A. We don’t know yet whether there are clinical implications. Our findings mostly suggest that we need to be on the lookout for p53 activation and p53 mutations and test for it in therapeutic settings.
Many of these experimental CRISPR therapeutics are being done ex vivo; you take cells out of the body and correct a gene mutation and put them back in the body. A next phase of research will be to determine whether this process selects for p53 mutations in patients.
Q. Is there any need for concern for the future of CRISPR-Cas9?
A. No. For both lab work and in therapeutics, the CRISPR-Cas9 system is an amazing system. It’s very useful and it works very well. All of us involved in this work still think this is a very important tool. Now, by being aware of the potential side effects, we can control for these off-target effects and we can improve upon the technology.
Our recommendation is certainly not to avoid CRISPR-Cas9 editing, but to bear in mind that these mutations can occur. There are some relatively easy recommendations coming out of this work that researchers can now incorporate into their genome editing pipeline to avoid or mitigate these off-target effects.
Support for this work was provided in part by the National Institutes of Health (R01 CA18828, CA215489, CA219943), the Gray Matters Brain Cancer Foundation, Pediatric Brain Tumor Foundation, Howard Hughes Medical Institute, Human Frontiers Science Program and the Azrieli Foundation.