|
Press Release
VA Study Finds Environment Can Alter Genetic Mouse Behavior
January 1, 2000
Sounding a caution for scientists trying to sort out how genes affect behavior, researchers report in the June 4 issue of Science that genetically identical mice behaved differently in seemingly identical tests at three laboratories.
Inbred mice have become mainstays in research to determine how particular genes may influence behavior such as drinking alcohol or abusing drugs, said first author John Crabbe, Ph.D., director of the Portland Alcohol Research Center at the Portland VA Medical Center and Oregon Health Sciences University. "Of course, the underlying assumption has been that behavior measured by one person will show the same genetic differences as behavior measured by somebody else." Techniques used to study genetically identical mice, including the so-called "knockout mouse" in which researchers delete (or knock out) a suspect gene to determine the effect, depend on consistent behavior in mice of the same strain that haven't been genetically altered, Crabbe pointed out.
Because no two labs are alike, Crabbe and his colleagues Douglas Wahlsten, Ph.D., professor of psychology at the University of Alberta in Edmonton, and Bruce Dudek, Ph.D., professor of psychology at the State University of New York at Albany, set out to determine whether lab environment influences genetic behavior and thus may skew results when scientists attempt to determine the role particular genes play in behavior. "We chose to test environment by trying to make it go away," Crabbe said. "We made the three labs as identical as we could."
For example, the researchers provided the same food, bedding and other care, used mice born the same day, and used identical techniques to evaluate six behaviors, such as general activity, alcohol drinking, sensitivity to cocaine, and simple learning. The team could not duplicate certain factors, Crabbe noted, such as the people performing tests at different sites. The researchers tested several strains of genetically identical mice and one knockout strain at their laboratories in Portland, New York, and Edmonton. "What we found is both good news and bad news," Crabbe said. As expected, the team found large differences in behavior between different strains at each of the labs. Although scientists previously thought shipped mice may be stressed and behave differently, the researchers also found that animals shipped to the labs several weeks before testing behaved the same as mice of the same strain that were born there.
Some behaviors, such as how much alcohol mice consumed, were extremely consistent among the same strains tested at all three labs. Other behaviors, however, differed markedly. For example, some mouse strains showed much more sensitivity to cocaine at the Edmonton lab than at the other two sites, while other strains showed similar sensitivity at all three labs. For some measures, all labs obtained different results when comparing the knockout mice and identical "wild type" mice that retained the knockout gene. In Edmonton, for example, no difference was observed between knockout and wild type mice in the activity test, whereas the knockout mice were more active at the other two sites. In another test, the knockouts were considerably more active than wild type mice only in Portland.
"Clearly, if you drew conclusions from these tests in one lab," Crabbe said, "you might not be able to replicate your results somewhere else." The study's findings call for caution in interpreting the effects of manipulating or deleting genes, he emphasized. Especially when resultant behavioral differences are slight, he said, genetic effects should be replicated before drawing conclusions.
|
