Evolutionary biologist Dr. Katie Peichel has spent much of her career studying two types of fish.
One lives in the ocean, traversing through empty swathes over flat, open sand, where larger, predatory fish can easily spot it. This ocean fish has evolved to school in large groups with its fellows, decreasing the chances it will be the one selected for dinner.
The other lurks in rivers and streams, a very different environment. In its watery home, there are ample plants, rocks, twigs to hide behind. There are no open spaces. The freshwater fish is a solitary creature. It does not know how to school, nor is it interested in learning.
The fascinating part, for Peichel and her laboratory team at Fred Hutchinson Cancer Research Center, is that these fish are the same species — at least, technically speaking. Known as marine sticklebacks (the ocean dwellers) and benthic sticklebacks (the freshwater denizens), these two fish will happily mate and produce fertile offspring, if thrown together in a lab environment.
In the real world, the two groups never — or almost never — meet.
In the culmination of an seven-year project, Peichel and her team have now identified the gene that drive those behavioral differences — schooling or non-schooling — between marine and freshwater sticklebacks. The research team describes their findings in a study published earlier this month in the journal Genetics. This is one of the first times scientists have uncovered the gene that underlies a behavioral change in natural populations of vertebrate animals, Peichel said.
Drawing that explicit line between gene and behavior helps researchers understand the true nature of evolution — behavioral changes among groups of animals can eventually lead to the evolution of new species.
“Understanding the genes that are important for evolution tells us something about how evolution happens,” she said. “We know that evolution is a change in [genes] over time … I think it’s inherently interesting to know, are there particular types of genes that are more utilized in evolution?”
The particular gene they found could also explain why sticklebacks are so adaptable to new environments, Peichel said.
Peichel’s research team had a daunting task ahead of them. Studying “natural” animals in the lab is difficult by any measure — laboratory animals are typically chosen and bred for their relative ease of working with in an artificial setting — but fish schooling added an extra layer of difficulty.
To understand the genetics at play, the research team needed to study each fish in isolation. But schooling is a group activity. Dr. Anna Greenwood, formerly a Fred Hutch behavioral biologist working in Peichel’s lab, and Dr. Abigail Wark, who was pursuing her doctorate in the lab, set out to solve that problem — starting in 2009 — by tricking the fish. They thought if they could get the stickleback to “school” with a set of fake fish, they’d be able to study this social behavior in individual animals.
But they weren’t sure the fish would be fooled. To see if their idea had legs, Wark made several gray plastic fish models from casts of real stickleback fish and attached them to a string. Then she dragged the string through a fish tank containing a single stickleback fish.
“We were just amazed when this little fish, a live stickleback, really started to approach and follow the school. We could tell even in that moment — when it was clearly scared by Abby’s hand and it was under this bright light — that it was probably going to work,” said Greenwood, who is now a program manager at Amazon. “That was a really exciting moment, one of those rare moments that you get in science. It’s just a feeling of pure excitement and fascination.”
The researchers set to work to make their system more rigorous and repeatable. They wanted to build something that wouldn’t look like a predator to the fish (as Wark’s hand did) and that would cause the plastic fish to always “school” in the same way. That engineering task itself was a challenge, described previously in the story “MacGyvering Lab Equipment,” but at the end, Greenwood, Wark and their colleagues ended up with a mechanical fish schooling robot that spun eight of the fake fish models in a circle through a round fish tank.
They then looked at the sticklebacks’ tendencies to join the fake pod of fish, and their ability to school proficiently once they join — whether they were able to keep their bodies parallel to those of the models.
The robot allowed the researchers to determine that stickleback schooling has a genetic basis and is not a learned behavior. In a previous study, published in 2013, they showed that two components of schooling — how well fish school with the models and whether the sticklebacks are motivated to school in the first place — are linked to two separate regions of the genome.
In the current study, Peichel, Greenwood and their colleagues found the specific gene responsible for sticklebacks’ schooling skills. It’s called Ectodysplasin, or Eda for short. It’s also involved in sticklebacks’ bony plates (marine sticklebacks are heavily armored, their freshwater cousins less so) and in the formation of “lateral lines” — lines of sensory hairs, similar to the hairs in our ears, that run down each side of the fish’s bodies and help them orient themselves through touch, much like a cat’s whiskers. Marine sticklebacks have a pair of these sensory hairs on each bony plate. Freshwater sticklebacks just have a single line of the hairs.
The freshwater sticklebacks — the fish with fewer bony plates, fewer sensory hairs and that can’t school — have much lower amounts of the Eda protein in their bodies. So the researchers decided to see what happens when they turn up the gain on the Eda gene. When they genetically manipulated the freshwater fish with high levels of Eda, the fish were suddenly able to school — like their marine counterparts, although not quite as well.
“They’re not totally marine-like. This is not a single gene that controls behavior,” Peichel said.
The history of biological research is filled with studies of genetic changes, or mutations, and their effect on some physical trait — although the list of genetic changes linked to behavior is much shorter. What makes their study stand out, Greenwood said, is that they found a genetic change that affects a natural behavior in wild, not lab-bred and raised, creatures.
“You can’t create a huge mutation that messes up an entire organism. This is a small change that’s going on during the natural evolution of a species that has to keep on living as this change is going on,” Greenwood said. “We just don’t have as many examples of the types of genes that are acting during evolution.”
Peichel thinks the freshwater sticklebacks may have evolved to suppress the Eda protein because fresh water has very low levels of the minerals needed for bone formation, so it’s taxing for the fish to build bony plates when they can find other ways of protecting themselves. The researchers still aren’t sure how the plates or sensory hairs might affect schooling — there was a weak correlation between increased sensory cells and better schooling in the genetically manipulated fish, but they need more evidence to be convinced these cells are involved in the social behavior, Peichel said.
They do know that the freshwater version of the Eda gene “lurks” in marine populations though, which could explain why these fish can so quickly adapt to new environments, Peichel said. If the marine sticklebacks find themselves suddenly stranded in fresh water, that hidden gene may allow some members of the group to survive.
“They’ve got the raw genetic variation that they need to adapt,” she said. “We learn lessons about how evolution can happen by asking questions about what genes are involved.”
Rachel Tompa is a former staff writer at Fred Hutchinson Cancer Research Center. She has a Ph.D. in molecular biology from the University of California, San Francisco and a certificate in science writing from the University of California, Santa Cruz. Follow her on Twitter @Rachel_Tompa.