The disappearance of snakes’ limbs is more than a story of loss—it is a complex history detailed in their DNA. Hoping to understand how and why evolution shaped the snake as it did—and what happened to its genome when it stopped walking—a team of scientists is using the gene-editing CRISPR/Cas9 system to produce the same change in mice.

Advances in genetic technology have accelerated the study of evolution via genomics, says Axel Visel, a geneticist at the Lawrence Berkeley National Laboratory. His team hopes to better understand the evolution of morphology, or the way animals physically look. “We decided to look at one of the most dramatic morphological adaptations that happened in vertebrate evolution,” he says: “The loss of limbs in snakes as they evolved from an ancestor that looked like lizards.”

His team took advantage of prior work sequencing the genomes of several snake species including boas, pythons and cobras. Snake genomes, like those of humans, have so-called enhancer sequences that regulate how other sequences should function. Visel’s team found that one enhancer called the zone of polarizing activity regulatory sequence, or ZRS, had degraded in snake species; some of the protein-making instructions in the snake ZRS looked like they had been copied so many times that numerous imperfections had appeared, changing some of the DNA’s base pairs (sets of molecules that determine the genetic instructions).

ZRS happens to enhance the sonic hedgehog gene, which is in charge of creating a protein crucial to embryonic limb development. In the latest research the scientists first swapped out the mouse ZRS from mouse embryos using a more common genetic replacement methods. They replaced the sequence with ZRS specific to different animals including horses, humans and snakes, combined with a special gene that would cause the tips of a mouse embryo’s developing limbs to turn blue after chemical treatment if the new ZRS was doing its job. When the researchers looked at the animal embryos the tips of all of the mice’s limbs had turned blue—except for those with snake ZRS, where either nothing happened or the ZRS seemed to behave in a way that did not imply limb development.

Essentially, snake ZRS did not seem to be doing its job. This was made vividly clear when Visel’s team used CRISPR to create live mice with their ZRS swapped with that of snakes. The “serpentized” mice had severely underdeveloped limbs; their lower arms and legs were each serverely reduced—in most cases to stunted bone. But the mouse limbs formed normally when the team replaced the enhancer with human or even fish DNA. Visel’s team released its results in Cell last week.

Visel emphasized that his team has not discovered how snakes’ limbs disappeared. Rather, they concluded that once snakes’ limbs were gone, evolution did not require the ZRS to actually work. A mutation in a snake’s ZRS would not have harmed snakes as they evolved—because they do not use their legs. Working separately, another team happened upon complementary results when looking at limbs in python embryos. This study’s authors said they found that python embryos still develop cartilage precursors to all of the parts of a leg—but the snake’s body sheds the lower leg cartilage and keeps just a rudimentary leg bone. That is probably because the ZRS degraded while another important set of limb enhancers stuck around. These enhance HOXD. (short for Homeobox-D, the proteins they code for) the same way that ZRS enhances sonic hedgehog. HOXD regulate legs and genitalia simultaneously, so evolution maintained its enhancers in order for snakes to be able to keep mating, says Martin Cohn, professor at the University of Florida. He and his co-author recently published their results in Current Biology.

Biologists have experimented with enhancers and their effects on limb development in mice as early as 2005, says Kimberly Cooper, assistant professor in Cell and Developmental Biology at the University of California, San Diego. Cooper was not involved in the new study, but notes that Visel’s use of CRISPR technology allowed for far cheaper and faster gene replacement in mice. “It’s really exciting,” Cooper says. “The end result of this engineering is no different than stuff we’ve done before. But we’ve made it so much more technologically simple.”

Lawrence Berkeley National Laboratory’s Animal Welfare and Research Committee reviewed and approved the animal use protocol for the latest study, says Antoine Snijders, the committee chair. “All the work described in the protocol is what the principal investigator is allowed to perform.” Although deliberately producing limbless mice might sound macabre, the work is ethically justifiable in this case, says Carolyn Neuhaus, postdoctoral fellow at New York University School of Medicine’s Division of Medical Ethics, who was not involved in the study. “Refining a technique and using a new technology is an important contribution to science and publicly valuable,” she says. Visel adds that there is no alternative to altering mice genes in these kinds of studies. “There’s no way you can do these kinds of experiments in computation models and be confident with the results,” he says.

Using CRISPR in animal experiments will hopefully lead to productive debates regarding the gene-editing technology in general, says Arthur Caplan, N.Y.U. bioethicist. Caplan believes such a dialogue is especially important in the context of eventually using the technique in medicine. “I think it’s really fascinating that most of the ethical attention is focused on what gene editing means to humans,” he says. “You want to debate CRISPR use at the level of mice, not men, first.”

Visel does not feel his work has a place in such a debate, and expects the field will see more studies using CRISPR to understand limb development. “These kind of studies were possible before, but now they’re within reach for many of these experiments in a more targeted way,” Visel says. “That excites me scientifically.”