Latest Issue of Science News

cover 3/7


A peek behind the science curtain

Bethany Brookshire

You smell, and mice can tell

Mice can tell if you are male or female, and they will change their behavior in response.  

Sponsor Message

When I was a lab scientist working with mice, I spent hours controlling variables. I stood on precarious chairs to tape tarps over lights to get the light level perfectly right. I made one undergraduate who wore perfume to the lab for animal training wear the same perfume for a whole semester. I was so worried about the mice “recognizing” me over long, overlapping experiments that I did not change the scents of any of my personal care products for nine years.

Many of these variables got reported in the methods sections of my papers. “All experiments conducted between 5:00 and 7:00 a.m. Maze dimensions: 4 inches wide, with walls 6 inches tall. Lighting held constant at 10 lux.” All of these variables are reported to allow other people to repeat my experiments, and hopefully get the same result.

Now, a new study suggests that maybe I should have included another element in my methods section: “All mice exposed to the scent of a woman.”

Jeffrey Mogil’s lab at McGill University in Montreal, Canada, reports April 28 in Nature Methods that mice respond differently to men and women, and that men in fact are a stressful influence. The results show that there’s yet another variable to control when doing sensitive mouse behavioral studies, a variable that could impact fields from pain to depression and beyond.

Every department that does animal research has stories about particular experimenters. I recall hearing a story of a lab technician who could get results no one else could, because mice just loved her strawberry-scented hair conditioner. Another colleague told of one experimenter who was so good at handling rats that no one believed her anxiety results. Her rats were just so relaxed. And Mogil’s lab had its own story. In their lab, the presence of human experimenters seemed to stop mice from showing pain. “It was a bit of a lab legend,” Mogil says. “Generally speaking this is the sort of thing that people whisper about in meetings.” But when your lab studies ways to combat pain, this is a legend worth investigating.

Mogil’s lab investigated the effect of human experimenters on mice by giving the animals injections of the inflammatory drug zymosan A, which causes pain where it is injected and is often used in pain studies. The scientists showed that the presence of male experimenters caused mice to display 35 percent less pain than when filmed remotely or in the presence of a female experimenter. The men didn’t even have to be present in person. T-shirts that male experiments wore were enough to produce the effect when placed in the same room with the mice. Female experimenters (and their T-shirts) had no effect on the mice.

The scientists hypothesized the mice were experiencing stress-induced analgesia. Further experiments showed that yes, boys stress mice out, significantly increasing levels of the stress hormone corticosterone in the blood. The signs of stress extend to more than pain behaviors. Mice also showed more anxiety.

And the effect wasn’t uniquely human. Males of other mammalian species, including strange male mice, guinea pigs, rats, cats and dogs produced the same effect, while females of those species did not.  The culprit appears to be a cocktail of chemicals that tend to be higher in males than in females, including androstenone, 3-methyl-2-hexanoic acid and androstadienone. The scent of a man, indeed.

The anxiety results and increases in stress hormone could affect other behaviors, many of which use stress and anxiety in their measures. This could mean other behavioral tests such as those for depressive-like behavior or fear could be affected. “There are literally hundreds of labs doing these behavioral tests, and they could all be impacted by whether the experimenter is a male or a female,” says Robert Hallock, a taste and smell researcher at Skidmore College in Saratoga Spring, N.Y.  In the study, the stress effect decreased after 30 to 60 minutes, so Hallock suggests that proper experimental design should include allowing mice to acclimate to the presence of male and female experimenters during experiments.

Hallock also says that the experimenter-gender effect will affect the replication of behavioral studies. “Labs that are having issues replicating a published study might want to take this into account,” he explains. “If they get different results by controlling for whether an experimenter is male or female, those results should be published, too.”

Mogil believes the implications are even more far reaching. “I would predict it’s not just behavioral studies here,” he says. “I predict it extends far beyond that, to cells obtained from the mice. What if a male is performing the sacrifice? The corticosterone levels at sacrifice could make a big difference in the physiology of the cells obtained from those animals.”

But Mogil emphasizes that while the study could change how scientists do their work, it is not a teardown of mouse research. “It’s actually the other way round,” he explains. “Actually animal research works just fine. Biology is just more complicated than we thought. You could get a whole set of different results depending on experimenter sex, but that doesn’t mean the results are wrong.”

And it doesn’t mean that male scientists are doomed to a life outside the animal colony. Instead, this means that the sex of the experimenter is something else to control. After all, scientists already know that light, sound, air and smell matter, so it’s not much of a surprise that characteristics of the experimenter matter, too. David Finn, a pain researcher at the National University of Ireland in Galway, says that in the end, “the paper should prompt scientists to consider the sex of the experimenter as another variable which should be controlled, or at least declared, in behavioral studies, so that others can take it into account.”

Follow me on Twitter: @scicurious

More Humans & Society articles