fb-pixel Skip to main content

Scientists untangle connection between genes, behavior

Mouse burrow-building helps elucidate connections

An oldfield mouse (Peromyscus polionotus) emerged from its burrow where it made its home in the sandy dunes of Florida’s Atlantic coast.
An oldfield mouse (Peromyscus polionotus) emerged from its burrow where it made its home in the sandy dunes of Florida’s Atlantic coast.Vera Domingues/Hopi Hoekstra, Harvard University

For years, scientists have been trying to untangle the complicated connections between genes and behavior, knowing that just as DNA gives rise to physical traits, it must also influence how people and animals act. But many of those efforts have been difficult, given that behaviors such as aggression or empathy are complicated, influenced by individuals’ environment and past experiences, as well as evolution.

Now, a team of Harvard University biologists has untangled a connection between genes and one simple behavior using a most unlikely set of tools: a large box filled with a ton-and-a-half of soil and sand, two closely related species of mice, and a polyurethane foam that can expand to fill crevices.


The researchers, led by zoology professor Hopi Hoekstra, had initially been focused on more straightforward questions, such as trying to understand the genetic basis of color variation and physical differences in mice. Then, Hoekstra’s interest was piqued by a distinctive behavior first described in a 1929 scientific paper: the tendency of oldfield mice to dig complex burrows. Hoekstra wasn’t so much interested in the burrows as she was in the consistency of oldfield mice, which appeared to follow the same basic blueprint again and again: a long entrance tunnel leading to a nest, with a second escape tunnel that tilted upwards, without quite breaching the surface.

Hoekstra and her team decided to check whether they could discern how genes contributed to this characteristic behavior. In a paper published Wednesday in the journal Nature, they report that they put laboratory oldfield mice that had never built a burrow in the wild (and whose parents had never built burrows in the wild) into a large box filled with soil. They let the mice up to their own devices for two days. When researchers returned, they used expanding polyurethane foam to make a mold of the burrows the mice had built. The oldfield mice, they found, repeatedly dug the same basic burrows, and the behavior appeared to be innate -- not a skill they learned.


Then, the researchers took a closely related species, deer mice, that made more “dinky burrows,” Hoekstra said, with a shorter entrance tunnel and no escape tunnel.

They mated the two mouse species together and saw what kinds of architects their offspring were. Left alone in the big box of soil, the offspring built burrows that resembled their big-burrowing parent. The complex home-making tendencies appeared to be a genetically dominant behavior.

To further probe the genetic basis of the behavior, the researchers then mated those big-burrow-making offspring with the deer mice who built dinky burrows. The burrow-making in the resultant offspring varied widely, with shorter and longer lengths of entrance tunnels. But one thing seemed to be a dominant trait that was either present or not: the tendency to build the upward-tilting escape hatch.

The researchers were able to locate three regions of the genome that seemed to contribute to the length of the entrance tunnel, and a single swath of genetic material that appeared to determine whether the mice would build the escape tunnel.

“The ultimate goal, now that we’ve found these regions, is to find genes in those regions,” Hoekstra said. The researchers can see what happens to mouse tunnel-making tendencies if they knock out those genes, or even see what happens if they insert a key gene from a big-burrowing mouse into one that digs a more modest home.


Ultimately, such research may help researchers not only understand rodent behavior, but give them clues as to where to start when looking for genes in people that may be involved in behavior. No one expects that an analogous gene in a human would cause people to dig more skilfully in the dirt, Hoekstra joked, but it could be that genes involved in such a fundamental behavior in mice also contribute to important behaviors in people.

And she isn’t only looking at rodent’s homemaking abilities: the oldfield mice that build the bigger burrows are monogamous, whereas deer mice are polygamous. Such social behavior, she thinks, is likely to have a genetic basis, too.

“Really, the insights into human behavior will come when we identify the genes [involved in behavior] and understand how they work,” Hoekstra said.

Carolyn Y. Johnson can be reached at cjohnson@globe.com. Follow her on Twitter @carolynyjohnson.