Most science labs maintain a temperature far below levels preferred by mice, and it’s taking a toll on their health. New research suggests these chilly mice are skewing science results across a wide range of research areas—and the problem is far worse than anyone realized.
A new paper in Trends in Cancer by researchers from the Roswell Park Cancer Institute in Buffalo shows that environmental factors are impacting the basic biology of mice, in ways that are influencing the outcomes of experiments. The authors also point to serious discrepancies in other research areas, such as cardiovascular disease and obesity. These results may explain why so much irreproducibility exists in mouse studies, and why mice often make for unreliable test subjects.
Most labs maintain a temperature between 68 and 78ºF (20-26ºC), which, if you’re a mouse, is bloody cold. Mice like it considerably warmer, around the 86-90ºF (30-32ºC) mark. It’s not that scientists are being unnecessarily cruel—it’s simply not practical for researchers, who often wear gloves and masks when working with animals‚ to work in such stuffy conditions. It also helps keep the smell down.
According to guidelines by the US National Research Council, mice should be housed within the 68 to 78ºF range and given access to nesting material. Unfortunately, these chilly conditions cause their heart rate and metabolism to change, and they consume more food to compensate.
“Mice are able to survive under a wide variety of temperatures, but they are able to move around and alter their environment for their thermal comfort, such as building elaborate and warm nests,” study co-author Bonnie L. Hylander told Gizmodo. “Also, mice are able to nest in large numbers which assists in conserving warmth. Mice in cages are certainly able to maintain health and body temperature, but it takes more energy for them to do so.”
Hylander’s team, along with others, has found that this extra energy usage is influencing the outcome of experiments. These mice must divert energy towards heat production, weakening their immune systems. That’s a problem if you’re a researcher trying to track a mouse’s ability to fight off a disease.
Hylander’s own cancer research showed that the anti-tumor immune response of the mice, along with their response to chemotherapy and radiation, were all affected by housing temperature. In 2013, these same researchers discovered that mice are better at fighting cancer when they’re cozy and warm.
The researchers decided to investigate the growing body of research on mouse housing temperatures in other fields—and they found similar results.
The team’s review contains not only their own work on cancer, but summarized reviews of the work done by several other investigators in other areas. “Some of the examples in which more significant differences were observed include models of cardiovascular function and obesity, in addition to our work on tumor growth,” Hylander told Gizmodo.
Compounding the problem is that housing temperatures vary between institutes, which may also cause differences in outcomes, and is likely a further source of irreproducibility.
Often, experimental treatments that work fine in one population of mice fail to work in another. These “simplified” models aren’t so simple. Factors that contribute to the irreproducibility problem include food, bedding, exposure to light, and exposure (or lack of exposure) to mice of the opposite sex—even the scientists’ gender.
This is a significant problem given how reliant scientists are on mice for their medical experiments, but the researchers say all’s not lost.
“Right now, it would be important for researchers to be aware of the potential for data skewing and they should report the room temperature at which their mouse experiments were done,” Hylander said. “If it becomes apparent that room temperature is a source of variability in experimental outcomes, then researchers and journals will most likely ask for experiments to be conducted at different temperatures and the outcomes compared.”
Researchers could also keep mice in incubators, and track these results as a unique and separate sample pool. Some labs might even want to just raise the temperature.
“Mouse models are invaluable and irreplaceable in preclinical research,” Hylander said. “But people are always interested in how to improve them.”
[Cell Press/Trends in Cancer (pdf)]