In June, several dozen scientists flew to Big Sky, Montana, to discuss the latest in CRISPR research. They had a lot to talk about, given that CRISPR—a tool that allows scientists to cut DNA to disable genes or insert new ones—is currently the hottest topic in biology, mentioned in the same breath as pronouncements like “changing the world” and “curing humanity of disease.”
On the second day in Big Sky, a Japanese researcher named Osamu Nureki got up to play a short movie clip. “I was sitting in the front, and I just heard this gasp from everyone behind me,” says Sam Sternberg, who worked in the CRISPR pioneer Jennifer Doudna’s lab at the University of California, Berkeley. It was, he says, the biggest reaction to data he’s ever seen at a conference.
Nureki’s paper was published in Nature Communications Friday, and by early morning, the video that astonished the room in Big Sky was making the rounds on science Twitter, too. I watched it, still bleary-eyed from sleep, and I jolted awake immediately.
Single-molecule movie of DNA search and cleavage by CRISPR-Cas9. pic.twitter.com/3NQxmbvzJF— hnisimasu (@hnisimasu) November 10, 2017
The video is grainy, blobular, and dark, but for a molecule-scale movie, it is remarkably clear. You can see CRISPR, in real time, cleaving a strand of DNA in two. There is nothing that surprising in the clip given what scientists already knew, but that is exactly what makes it so astonishing: Scientists had figured out so much about CRISPR without ever seeing it.
It had the satisfying snap of things falling into place—like the first time a telescope sighted Neptune years after a French astronomer had predicted its existence from perturbations in the orbit of a neighboring planet.
CRISPR obviously exists, and it obviously works. Scientists have used the technique to do everything from ridding mice of HIV to beefing up dogs. But the evidence for how it works had always been indirect. Modern biochemistry research is a series of elaborate workarounds to infer the behavior of molecules until now too small to see.
Take, for example, X-ray crystallography, which Nureki had previously used to study the structure of CRISPR-Cas9. (“CRISPR” is the popular shorthand, but Cas9 is the name of the actual enzyme that cuts DNA.) Scientists spend months perfecting the art of coaxing Cas9 to grow into crystals. Then they take these precious crystals to a particle accelerator to shoot X-rays through them, producing a pattern unintelligible to the average person but which experts can measure to infer the structure of Cas9. The end result is a computer model of Cas9, resembling a clump of curly ribbons. Repeat all of this using Cas9 frozen at different stages of editing DNA, and you can after many months or even years get a handful of static, computer-generated snapshots.
Compare that to the fluidity and directness of the real-time video. His team used a technique called high-speed atomic-force microscopy, in which a tiny needle moves back and forth probing the shape of Cas9. The needle moves so fast that it produces a movie. “The result is fairly easy to understand,” says Hiroshi Nishimasu, one of Nureki’s collaborators on the paper. “People say, ‘Wow!’ It’s very simple.”
Nishimasu posted the video to Twitter, and that dark, grainy clip has since gotten more than 2,500 likes. The response has been far and beyond that to any other paper he’s published, including ones in the most prestigious journals like Nature and Cell.
In fact, he says, the team did submit the videos to top-tier journals, who rejected it citing a lack of novelty—which, perhaps, is true. Scientists are supposed to be dispassionate creatures of data. But they are also only human, and humans believe their eyes. “I realized that seeing is believing,” says Nishimasu.