The Science of Controlling CRISPR
As we move toward correcting somatic editing in humans, we need better controls and a better understanding of what exactly is being done to our genomes
CRISPR continues to make headlines, and rightfully so! CRISPR technology rapidly accelerates the timetable for creating knockout mice as well as other mutant animal models, and what was thought to be possible through gene editing in the far future, may be reality in only 5-10 years.
Earlier this year the National Academy of Sciences and National Academy of Medicine published a report on Human Genome Editing (brief highlights). In the report they discuss how CRISPR is cheaper, easier to use and more efficient than previous methods, and the rapid adoption of the CRISPR/Cas9 system makes the need to consider broader implications pressing.
The concept of doctors being able to instantly diagnose and treat diseases was a science fiction dream, but that horizon is closer than we thought. Technological leaps forward, like CRISPR, may soon make it a reality. This makes me think of watching Next Generation Star Trek in the ‘90s. Many of their fanciful gadgets now exist as smart phones, tablets, self-driving cars, even devices like Alexa and Google Home that allow you, literally, to talk to a computer. Medicine may not be too far away. "Whoosh,” goes a beam of light and your genome is projected on the screen, with all flaws noted. Next Gen Sequencing (NGS) and annotation is rapidly bringing that part of the fantasy into reality. Several companies can even mail you a kit (for under $100) allowing you to explore your own genome. You send back your sample and they give you a wealth of personalized information. While they are still very cautious about reporting medical information (such as Alzheimer’s and cancer susceptibility) clinical NGS is getting close, with numerous panels offered for different disease areas.
The next step was Doc poking you in the neck. After a pinch and a hiss, you were cured! With CRISPR, and better delivery mechanisms, that may not be too wild an idea. Treatments ex vivo (taking something out, fixing it, and putting it back in) are the low hanging fruit of gene editing, and are showing promise for diseases like sickle cell.
Controlling Off-Target Effects
As we move toward considering somatic editing in humans (correcting a disease in the adult affected tissue) and especially when considering heritable editing (removing a risk factors) we need better controls and we need a better understanding of what exactly is being done to our genomes. If treating in vitro or ex vivo, we need to be able to screen for unintended edits and not put those back into a patient. Since CRISPR editing has been so successful, interest is growing in how best to control and fine tune this technology. The goal is to keep risk of unintentional edits (off-target effects) is as low as possible.
Many approaches are being used to identify and optimize CRISPR controls. Crystal structures of Cas9, especially when bound to guide RNAs and DNA targets, are extremely helpful in understanding exactly how cutting occurs (and conversely how to control it). In fact, higher fidelity Cas9s were developed using structural insights making the complex less stable at incorrect targets.
Knowledge of structure and modeling lends not only to optimization, but identification of critical components and methods to attenuate activity. Small molecules have existed for some time, and still make up the majority of pharmaceutical pipelines. It is natural that many groups are pursuing the small molecule inhibition of Cas9 to control CRISPR.
Circular permutation is one approach that is helping us to better understand domains within a protein and how they interact. Proteins are encoded linearly, as one long string of amino acids (primary structure), but in reality, they don’t function that way. The long string folds in a particular way, forming various loops and pockets that interact with other elements within the protein (secondary and tertiary structures). To be fully functional, several proteins must form a complex based more on their three-dimensional shapes and interactions with each other (quaternary structure). These layers of structure can be quite complex, requiring careful engineering to understand. With circular permutation one takes a domain at the beginning of the sequence and moves it to the end, continuing in a loop to determine which domains can freely move, and what needs to be in order for appropriate folding. This approach is being used to understand Cas9 better, and has led to some interesting methods of control.
Another way to inhibit CRISPR is to find anti-CRISPRS. CRISPR technology arose from the microbial arms race between bacterial and the phage infecting them. What better place to look for molecules to control CRISPR (anti-CRISPRs) than in microbial genomes. Phage are very small viruses that infect bacteria, and as such don’t have a lot of room for DNA that isn’t critical. Multiple groups have looked for anti-CRISPRs by mining the genomes of bacteria. The reasoning is, if a phage successfully integrated itself into a bacteria’s genome, despite the bacteria having CRISPR as a defense mechanism, then novel sequence found with that phage genome may encode an anti-CRISPR. Numerous anti-CRISPRs have been found this way, including several that are effective across bacterial species.
The future of gene editing is bright, and the horizon not nearly as distant as it seemed a few years ago. There are significant and well founded concerns as the field moves so quickly. A challenge as the technology advances will be to fully understand and control the better, faster, cheaper tools we develop. I hope to see more dreams of science fiction become realities in my lifetime! Powerful tools like CRISPR, wielded with skill and precision, will bring us closer to realizing those dreams.