In our series about the use of CRISPR/Cas9 used by NCCR Chemical Biology members, we have met Takeshi Harayama, Maître assistant in the Riezman lab, Department of Biochemistry, University of Geneva. He is currently working on understanding the diversity of membrane lipid composition, how it is generated, sensed, adapted and maintained and how disrupting membrane lipid homeostasis is linked to various diseases. He pursues innovative applications of CRISPR-Cas9 for studying mammalian gene functions in lipid metabolism.
How do you apply CRISPR/Cas9 in your research?
We use it in its simplest way, to generate indels in the genome and (hopefully) disrupt genes in cultured cells. Most often, our target genes encode enzymes that metabolize lipids. By mutating them, we understand better their contribution on lipid metabolism. In addition, we can mutate multiple enzymes to manipulate lipid metabolism artificially. By doing this, we can orient the metabolism of, for example, lipid analogs that were synthesized during the projects of the NCCR Chemical Biology. We also develop new strategies to improve the efficiency of current protocols.
How does the tool speed up discoveries? Do you manage to do things that you were not able to do? Or just better?
Fifteen years ago, I was working with embryonic stem cells to generate knockout mice. It took me more than one year (doing every day Southern blot) to isolate a single clone having a heterozygous mutation that I wanted. Now, I generate dozens of homozygous mutants within a month. This is something that I could do 15 years ago only if I was able to work something like 72 hours per day. So yes, this tool speeds up the discoveries.
I never saw any other technology that could transform both basic and clinical research so drastically and so quickly. It clearly changed the way I am doing research, and I can easily imagine that it did the same for many others. I can only underestimate its impact, because there are new ways to use this technology (that I could never imagine) appearing daily.
How the CRISPR tool could be improved?
There are problems such as off-target effects, deleterious on-target mutations (e.g. extremely large deletions), or low knock-in efficiency, that are already discussed repeatedly, but many researchers are working hard to overcome them. For my personal research, the biggest problem that I meet is that indels do not always disrupt gene functions. Thus, if we could predict perfectly (better than existing prediction tools) the indel patterns to be created, this would be a great advance for my research.