Click here for free access to our latest coronavirus/COVID-19 research, commentary, and news.

Support nonprofit science journalism

Science’s extensive COVID-19 coverage is free to all readers. To support our nonprofit science journalism, please make a tax-deductible gift today.

Brown mouse (left) and lab mouse (right): a tale of two immune systems.

Brown mouse (left) and lab mouse (right): a tale of two immune systems.


Can ‘dirty mice’ save animal research?

When it comes to animal models of the immune system, laboratory mice are like the clean version of a profane song; they’re still pretty good, but they may not capture the grit of real life. Unlike wild rodents, lab mice are bred, raised, and housed in essentially sterile environments. That makes experiments with them easier to reproduce and control, but it has also raised serious concerns about whether or not they’re realistic models for humans. Now, in two new studies, researchers have proposed a way to “dirty up” these mice, though some worry that the work could compromise future studies.

“I think these are very, very important studies,” says immunologist John Wherry at the University of Pennsylvania, who was not part of the work. “They have a lot of implications for how we think about chronic infection and global health.”

In the first of the new studies, published today in Nature, immunologist David Masopust at the University of Minnesota, Twin Cities, in Minneapolis and his team set out to create lab mice whose immune systems looked more like their wild counterparts. They did so by procuring mice from pet stores, which possess none of the sterile precautions of the facilities reserved for lab mice, and placing them in a cage with laboratory mice. Before the cohabitation, lab mice displayed a low number and diversity of memory T cells, which are specialized for fighting a variety of foreign invaders. Their “immune profile” closely resembled that of a newborn baby, likely because they rarely have to deal with infections. “[Lab mice] have had a very privileged existence,” Masopust says. “We don’t live like that.”

Things changed drastically after 2 months with the pet store mice. Twenty-two percent of the lab mice died, but those that survived showed profound changes in their immune systems. Exposure to pathogens increased a variety of immune system components in their blood—particularly subsets of specialized T cells—shifting the immune profile closer toward that of an adult human. Tests for common mouse pathogens showed that the lab mice had been exposed to viruses, bacteria, and worm parasites. The cohousing also changed how the mice responded to new infections: After mingling with the dirty mice, the lab mice showed 10,000-fold increase in their immune response against a bacteria known as Listeria monocytogenes—a reaction on par with pet store mice themselves.

Masopust believes the experiment produced mice with more realistic immune systems—the kind of animals you would need, he says, to more accurately test the power and safety of new drugs and vaccines. He says that in the future, researchers could expose  a lab mouse to a mix of pathogens akin to what humans might be experiencing to get a better sense of what that repertoire does to immune systems in general. In a perfect world, researchers could tailor that repertoire to match the pathogens of particular regions of the world as well.

And indeed, scientists are investigating exactly this idea. In the second study, published today in Cell Host & Microbe, viral immunologist Skip Virgin of Washington University in St. Louis in Missouri and his team infected laboratory mice with herpes and flu viruses, as well as with intestinal worms in order to chart how the rodents responded to a vaccine for yellow fever. “We’re interested in seeing whether we could change and, at least in theory, improve the mouse model by making its exposure to environmental agents more normal,” Virgin says. These pathogens were chosen because they represent common infections in many parts of the world; if their presence changes how animals respond to vaccine it could help explain why some vaccines appear to work better in some populations than others.

Initially, immune responses to the yellow fever vaccine were similar between the ordinary lab mice and the group coinfected with other pathogens, but after 34 days the coinfected group showed signs of a reduced immune response to the vaccine, as evidenced by a decrease in the amount of antibodies against the vaccine components found in the rodents’ blood. Analyses of the DNA in the mice’s blood also showed that gene expression was different between the two groups as well. The scientists observed changes in the genetics underlying T cell activation, cell death, and the response to immune signaling molecules, suggesting that coinfection with other common pathogens changes the immune system at the genetic level.

“[Masopust] showed that the natural state of the animals is very different in nature or a pet store than it is in a [lab],” Virgin says. “We showed that under controlled conditions you can make substantially similar changes by taking that same clean mouse and giving it a series of infections.”

Harnessing this knowledge to create mouse populations that closely mimic human immune systems is still a long way off, however. One of the largest challenges will be to do so in a way that doesn’t ruin the many advantages of the lab mouse. Imperfect as they may be, laboratory mice have proven an excellent model for human in countless scientific experiments. Their sterile environment and blank slate immune system ensures that all scientists are working with the same models. Introducing changes to the immune system will need to be done in a very careful and calculated manner if it works at all, Virgin says.

Wherry is also cautious, pointing out that, aside from the reproducibility issue, it will also be expensive and time consuming to create and house mouse populations with the desired immune profiles. “I think we’ll see this used selectively in certain areas,” he says. “It would be very difficult to do it systematically and apply it across the board.” In spite of the difficulties, Wherry says he can envision a future in which scientists can choose from a library of mice with defined immune profiles that reflect different human populations.

“That’s the dream,” Virgin says. 

*Correction, 21 April, 9:03 a.m.: The original version of this article mentioned both mice and rats. The experiments were only done on mice.