In 1996, French scientists showed that two genes in this group—Hoxa13 and Hoxd13—control the development of hands and feet in mice. When they deleted these genes, rodents would grow up with limbs that ended in stumps. No paws. No digits.
For Shubin, that discovery was thrilling. He knew that the legs of mice and other four-legged animals evolved from the fins of fish, and he knew that fish have counterparts of Hoxa13 and Hoxd13. What do those fishy genes do?
To find out, Gehrke used a technique that automatically adds glowing red molecules to any cell that activates Hoxa13—essentially, painting them for easy identification. In mice, this method only paints bone-making cells in the wrist—all of these cells, and only these cells, go on to build the bones of hands and digits. What about a zebrafish? Which part of its fin would turn red? Shubin had some very specific expectations.
Think about a fish’s pectoral fin—the one that sits behind its head. The fleshy base contains several bones that are related to those in your arms. But there’s nothing that quite matches your hands and fingers. The long, thin, parallel bones in the rest of the fin—the fin rays—look metaphorically like fingers, but they’re made from a different kind of bone, and they grow in a totally different way. So scientists, Shubin included, believed that our hands and digits did not evolve from fin rays. Instead, they either evolved from some small bones in the fleshy base of the fin, or were a totally new evolutionary innovation with no predecessors.
So, when Shubin looked at Gehrke’s image, he wasn’t expecting to see that the fin rays were red. “That was shocking,” he says. “It ran counter to everything we thought.”
It seemed unbelievable. To confirm what he was seeing, he needed to do the fish version of that classic French experiment with mice: delete the relevant Hox genes and see what happened. He had wanted to do that for years, but never had the right tools for the job. But in 2013, he heard about a new gene-editing technique called CRISPR, which could modify and delete genes with unparalleled ease and precision. While other scientists wrangled about the ethics of using CRISPR to change humans, Shubin realized that it could solve his fishy fin mystery.
Another team member, Tetsuya Nakamura, did the experiment, using CRISPR to delete the fish versions of Hoxa13 and Hoxd13. And those fish ended up with much shorter fin rays.
“It’s an unexpected finding that really adds a new level of clarity to the fin-to-limb transition,” says Clifford Tabin from Harvard Medical School. “But it doesn’t mean that fin rays are the equivalent of digits, or the ancestral structures that gave rise to digits.” The reality, as with much of biology, is more complicated.
Shubin thinks that as a fish’s fin grows, cells in its base activate Hox genes and move outwards, creating the fin rays. Delete those genes and the cells don’t migrate, making shorter rays. And that’s sort of what happened as fish made the move to land and fins evolved into limbs: those Hox-activating cells stayed put and shifted from making fin rays into making digits.