Clustered Regularly InterSpaced Palindromic Repeats (CRISPR) is a new technology for gene editing, with promising potential in medicine and basic research. CRISPR can be found in the DNA of bacteria. There are many different types of CRISPR DNA, but they all make proteins that protect bacteria against viruses, much like the human immune system functions to kill off viruses. The way CRISPR proteins do this is to cut the DNA of viruses — think of them as molecular scissors — preventing the virus from growing.
CRISPR has been in the news a lot recently. Some of the news coverage has been positive (promising new therapies) and some of it has been negative (CRISPR babies).
Scientists discovered that CRISPR proteins could be adapted to cut the DNA of other organisms, including humans. This led to demonstrations that CRISPR could be used for gene editing and gene therapy. The most commonly used CRISPR protein is called Cas9.
As with all game-changing technologies, attention has focused on ethically responsible uses of the technology, and rightly so. What is often missing from many CRISPR debates is an understanding of technical limitations that could impact how and when CRISPR is used in human gene-editing applications.
Read more: Opening Pandora's Box: Gene editing and its consequences
My laboratory became interested in CRISPR because we were working on technologies for gene editing. It quickly became evident that CRISPR and Cas9 were the future of gene editing because Cas9 was significantly easier to use than other technologies but, like every new technology, was ripe for improvement.
One of the biggest issues when trying to modify human DNA is the problem of so-called off-target effects. Off-target effects arise when Cas9 cuts a piece of DNA that it wasn’t programmed to do.
In describing off-target effects, I like the analogy of a programming a car GPS unit with the address “Tim Hortons.” In any Canadian city or town, this search will result in multiple individual Tim’s locations. But what is the correct location?
In much the same way, Cas9 is guided to its DNA target by a small piece of ribonucleic acid (RNA), appropriately called a guideRNA. If the address specified by the guideRNA is not unique, Cas9 will be guided to multiple locations where it will cut the DNA. The address in this case is a continuous DNA sequence in the human genome, typically 20 base pairs in length.
Why are off-target effects an issue? Not all 20 base pair stretches are unique in the genome, and Cas9 has the added problem of being able to recognize sites that are not perfect matches to the 20 base pair address (these are called off-target sites). This means that there is always the potential for Cas9 to cut DNA at off-target sites that could lead to unwanted and serious side effects, including cancer. For any human therapeutic application using gene editing, minimizing the potential for off-target effects is paramount.
Locating off-target sites
Off-target effects are not a new issue in the gene-editing field, and the potential for Cas9 to cause off-target cuts was recognized as a serious issue soon after Cas9 was identified.
Significant efforts by multiple research groups resulted in the development of engineered versions of Cas9 with greater specificity and fewer off-target effects. Scientists are finding new CRISPR-related proteins with novel applications in gene therapy.
A related problem is predicting off-target sites and identifying where Cas9 cuts in the human genome. In theory, finding potential Cas9 off-target sites should be straightforward. Simply search the three billion base pairs of the human genome for matches to the 20 base pair target (this is easier than it sounds and can be done on a laptop with basic programming experience).
Some studies have found that predicted off-target sites don’t always match up with experimentally identified Cas9 off-target sites, whereas other studies have found low numbers of Cas9 off-target sites. These findings underscore the need to understand how Cas9 recognizes DNA so that improvements can be made to software that predicts Cas9 sites.
For gene editing and gene therapy applications done in somatic tissues, such as blood cells, DNA changes are not heritable. In these cases, a higher amount of off-target effects may be tolerated. Any off-target effects are unacceptable simply because they are difficult to detect and reverse. Until these issues are resolved, the question of off-target effects should not be ignored in debates about CRISPR gene editing.
[ You’re smart and curious about the world. So are The Conversation’s authors and editors. You can read us daily by subscribing to our newsletter. ]