For all the furious hype around the gene-editing tool Crispr/Cas9, no one has ever really seen it in action. Like really seen it. How the protein Cas9 unzips a strand of DNA, how it slips in the molecule that guides it to a target---and finally, how it goes snip snip on the DNA. The power of Crispr/Cas9 is its ability to do this all so precisely and reliably.
How can you see something as small as a protein anyway? For decades, that has meant coaxing proteins to grow into crystal structures. Scientists then shoot X-rays through the crystal, and the diffraction pattern elucidates the protein’s structure. Today, for the first time, a study in Science led by Crispr/Cas9 pioneer Jennifer Doudna uses that technique to capture the structure of activated Cas9, in the moment it's primed to cut DNA.
Knowing how Cas9 works in such magnificent molecular detail matters because while the system is good for editing genes---even very good---it is not perfect. Sometimes it cuts the wrong stretch of DNA. Sometimes it doesn’t cut the stretch it is supposed to. Insights from the new study could lead to “more efficient design of Cas9 mutant with high specificity,” says Osamu Nureki, a biologist at the University of Tokyo who has also worked on the structure of Cas9.
But here’s the thing. Even without knowing the structure of active Cas9, scientists have already started modifying the protein. Such is the pace of Crispr/Cas9 research, which has exploded since the first paper to show its DNA-editing potential in 2012. As scientists have raced to use the system to modify pigs, mosquitoes, mice, and even in one case, nonviable human embryos, others have been working on making it better---so good that it could one day be used to cure diseases in humans.
The big hang up is specificity. Cas9 finds its target with the help of a guide RNA, a molecule whose letters pair up with the target DNA sequence. Occasionally, though, the guide RNA pairs up with sequences it does not match perfectly---the so-called off-target problem. In December, a team led by MIT and the Broad Institute’s Feng Zhang, another Crispr pioneer, tweaked the molecules in a groove of Cas9 that holds DNA to improve specificity 25-fold for certain sites.
Zhang and fellow Broad researcher George Church have worked on another strategy to combat off-target mutations, too. Cas9 is often compared to a pair of scissors, but it’s actually two pairs of scissors fused together, each of which cuts one of DNA’s two strands. Zhang and Church have mutated Cas9 to blunt one of those scissors, so it only cuts one strand. Now you need a second Cas9 with second guide RNA to cut the second strand---with redundancy comes less error.
The downside is that these single-scissor Cas9s can still “nick” DNA individually and cause potential mutations. So yet another group led by Harvard’s Keith Joung have fused the guide RNA-binding part of Cas9 to the the scissors of another a DNA-cutting protein called FokI. Not only do you need two FokI-Cas9 to cut a whole piece of DNA, but the two individual hybrid proteins need to actually combine into one mega protein before either will cut DNA, so you don’t get any nicking either.
But what happens if you blunt both of Cas9’s scissors and don’t give it any replacements? That’s where things get really interesting. Jonathan Weissman, a biochemist at the University of California, San Francisco, and collaborators including Doudna have fused that dead Cas9 to molecules that can turn genes on and off.
Every cell in your body has the same genome, but the epigenome turns genes on or off to turn skin cells into skin cells or brain cells into brain cells. “Cas9 has been a great tool for engineering the genome,” he says. “The dead Cas9 is great too for engineering the epigenome.” Weissman calls the system Crispr-i or Crispr-a (for interference and activation, respectively), and his collaborators are using it to manipulate the activation of genes in mice. The technique is good for investigating the function of genes, but it could also, conceivably, be useful therapy. For example, you might turn off the genes for receptors that the Ebola virus uses to enter human cells.
All of this research into modifying Cas9 has plowed ahead while scientists are still figuring out exactly how the protein works. With a higher resolution molecular map of Cas9 now available, that work is only going to speed up.
