We regularly invite scientists to present their research to the Mammoth team. This facilitates collaborations and expands our knowledge of important topics including CRISPR, genome editing, diagnostics, and more. Rather than keep what we learn to ourselves, we’ve decided to share it with you through our Mammoth interviews series. This series features short Q&As with the many interesting scientists who speak at Mammoth. Get ready for some fascinating science and even more fascinating people!
What is your research background and why did you start working with CRISPR?
I am a molecular biologist by training. I received my PhD in Merlin Crossley’s lab at the University of New South Wales in Sydney. It was during my PhD that I discovered the power of CRISPR as a research tool.
In the Crossley Lab, I was working on promoter mutations that naturally cause high expression of fetal hemoglobin – a protein whose expression can be beneficial in patients with sickle cell disease or thalassemia. At the time, it was difficult to model these mutations in cells. Researchers commonly used artificial reporter constructs to investigate the effects of point mutations as opposed to making mutations in the genome.
When CRISPR emerged, it suddenly became much easier to generate cellular models with nothing but small changes to their genomic DNA. I was the first to use CRISPR in our institute and soon convinced many others to jump on board.
I was so fascinated by CRISPR that I decided to move to the US for a postdoc in the field. I ended up working with Jacob Corn at the Innovative Genomics Institute (IGI). In the Corn Lab, I delved more into the mechanisms behind CRISPR’s function. I got very interested in the biological processes that happen when CRISPR does what it does: cuts DNA. After my time at the IGI, I moved to my current position in Bruce Conklin’s lab at the Gladstone Institutes.
At both organizations, I’ve worked at the intersection of genome editing and DNA repair. As part of this work, I develop techniques that help researchers detect off-target genome editing. That is, detecting when CRISPR makes changes to the wrong sections of DNA. In particular, I’ve developed a method called DISCOVER-Seq. This method can identify off-target editing in a wide range of cell types as well as in vivo.
Why is off-target detection important in CRISPR research?
This is a great question. I actually believe that for many in vitro applications of CRISPR, such as knocking out a gene in a cancer cell line, off-target detection is not that important. An easier way to deal with off-targets in that context is to develop three independent editing strategies and use these strategies to create three different cell lines. If all three cell lines show the same phenotype, it’s good evidence that the observed effect is real.
However, this point of view dramatically changes once you move into applications of CRISPR that will be used in a clinical context. Like any other therapeutic, we must know the side effects of a CRISPR treatment before using it on patients. Thus, before moving CRISPR into the clinic, we need to make sure we know what will happen in the cells that we modify. Part of this is having a good understanding of what DNA sequences get altered during the therapeutic process and knowing what their potential side-effects might be.
What’s wrong with current off-target detection methods?
Nothing is wrong with current off-target methods. My colleagues who have developed these methods have come up with very creative ways to detect off-targets. However, any tool has strengths and weaknesses. For example, a hammer is a great tool, but you wouldn’t try to use it as a drill and vice versa. The off-target method you should use really depends on the question you are asking.
For example, some off-target detection methods involve extracting DNA from cells and cutting this DNA with a CRISPR system in vitro. These are highly sensitive. They will likely report back any DNA sequence that could potentially be cut. However, we know that off-target editing depends on multiple factors. It’s not just the DNA sequence that matters. Things like chromatin accessibility and DNA repair machinery can make a big difference in target cells. So, if you want to know what off-target sites are modified in mouse liver cells after injecting CRISPR in vivo, in vitro methods will not provide a sufficient answer. They’ll likely give you too many false positives.
Other methods that use cultured cells also provide high sensitivity but rely on ligation of an adapter sequence into the double-strand breaks. This ligation process can be difficult in sensitive cell types and will often kill such cells.
How does your DISCOVER-seq provide solutions to the off-target problem?
I see DISCOVER-Seq as an addition to the off-target tool kit, not as a replacement for existing methods. As a biologist I like to think about problems from a biological point of view. DISCOVER-Seq works by identifying sequences within the genome that are bound by endogenous DNA repair proteins, in particular a protein called MRE11. Gene editing is a result of DNA repair. By looking for DNA repair protein binding, we can identify sites that have suffered from a double-strand break induced by CRISPR.
Importantly, DISCOVER-Seq allows researchers to determine which off-target edits will occur in specific tissues. This is because it takes cell-type-specific chromatin accessibility into account. It also has the advantage that it does not require the introduction of any other reagents into the cell – all you need to introduce is the CRISPR editing reagents. That means that in theory it can be used in any cell type. This includes sensitive primary cells and even animal tissues after in vivo CRISPR editing. However, I truly believe that no method is perfect. Thus, when determining off-targets for clinical applications of CRISPR, one should thoroughly investigate using any orthogonal methods that are available.