Mouse housing temperatures can cook experimental outcomes

Mouse colonies are the bedrock of neuroscience research. In planning their experiments, researchers spend hours reviewing the literature, considering ideal mouse lines and writing detailed protocols to handle the animals humanely. Once the mice arrive, they are housed at institutional facilities that strive to maintain a carefully curated ecosystem to ensure animal well-being.

But sometimes, as researchers, we take for granted just how delicate this ecosystem can be.

Our group has found that the mildly cool temperatures mandated for mouse colonies—typically 22 to 23 degrees Celsius—can influence tumor biology. At that temperature, both the animal’s innate anti-tumor response and its response to various types of treatment are impaired compared with those of mice housed at thermoneutral temperatures between 29 and 30 degrees Celsius. This impairment occurs because the animal needs to generate heat through non-shivering thermogenesis—a process that triggers the release of norepinephrine from sympathetic nerves in brown adipose tissue and beyond, potentially altering numerous cell types.

This temperature effect isn’t limited to cancer models. We have written several reviews summarizing data from the past decade that show how housing temperature and changes in thermoregulation affect experimental outcomes in mouse models of cardiovascular disease, cancer and immunology. And neuroscience is no exception. Housing temperature can also influence the progression of Alzheimer’s disease in the triple-transgenic mouse model 3XTg-ADl, trigger inflammation in the brain and influence slow-wave sleep and amyloid pathology. Even standard housing temperatures can induce brain dysfunction in bacterially challenged mice models.

These and many other examples illustrate that neuroscientists need to pay greater attention to the temperature effect. They also underscore the need for scientists to report housing conditions at animal facilities and test the impact of temperature in pilot experiments.

T

he U.S. National Institutes of Health is increasingly concerned with proper mouse colony temperature and its effect on experiments; a working group dedicated to rigor and reproducibility of animal studies identified it as one of several extrinsic factors—including lighting cycles, food sources, water type and cage-handling frequency—that play a significant role in experiment reproducibility. Given the well-documented effects, why do we house mice at less-than-thermoneutral temperatures?

Animal facilities follow the National Research Council’s guidelines for the ethical and humane housing of laboratory rodents, which standardize nearly every aspect of animal housing in research facilities, including temperature. These guidelines, detailed in the Guide for the Care and Use of Laboratory Animals, 8^th Edition, acknowledge that the standard housing temperature for mice is lower than their preferred thermoneutral temperature but clarifies that this is to avoid heat stress, a potentially fatal condition in which an animal generates more heat than it can dissipate.

The guidelines recommend that instead of turning up the heat, researchers should provide mice with nest-building materials. Unfortunately, this provision may not completely eliminate the effects that low temperature has on experimental outcomes. Studies show that the amount of nesting material needed to avoid non-shivering thermogenesis varies across mouse strains and sex. And different research centers and institutions may add different amounts of nesting materials, as the guide sets no standard amount.

Other housing factors, such as the number of mice in a cage, can affect thermoregulation. A 2019 review article by The Jackson Laboratory, for instance, reported that high-density cages containing many animals are about 3 degrees Celsius warmer than lower-density ones—a temperature closer to the preferred mouse thermoneutral zone—and these differences influenced the animals’ well-being.

How should we solve this problem? Technological advances make it easier to measure environmental parameters in animal facilities. And just as researchers should include these data in their research results, institutions need to work together to standardize essential animal facility assessments, including engineering controls, caging systems, processing methods and standard operating procedures.

We agree with the National Research Council that simply turning up the thermostat is not an optimal solution. But if mice had housing that enabled them to choose their preferred temperature, warmer or cooler, it would help to reduce the risk of thermal stress.

Cage modifications could also help: For instance, a battery-operated warm rack that mildly heats a portion of the cage floor could help mice stay warm without increasing the room temperature beyond a temperature comfortable for staff. Heated mouse cage racks could also help to control the animals’ temperature without affecting the air exchange within the cage or the humidity levels in the animal holding rooms.

We need more research and a detailed understanding of extrinsic factors that affect mice housing. We also need interinstitutional collaborations to work to avoid them. Neuroscientists need to help in this effort not only for the sake of study reproducibility, but also for animal well-being.