Mice beat rats: the best model for testing endocrine disruptors. (Science Selections).

Endocrine disruptors, or chemicals that interfere with hormone activity, abound in our environment. They are found in such synthetics as pesticides, preservatives, paints, and plastics, as well as in natural sources such as soy products. To date, approximately 50 chemicals have been identified as endocrine disruptors. However, in the United States alone there are more than 80,000 chemicals now in commercial use that have not yet been tested for such effects. A 1998 U.S. Environmental Protection Agency report called for testing of all environmental chemicals for their estrogenic effects and recommended using an in vivo screen measuring uterine growth in rodents. In this issue, Elizabeth Padilla-Banks, Wendy N. Jefferson, and Retha R. Newbold, all of the NIEHS, compare the sensitivity of mice versus rats as a model for the testing of estrogenic effects of endocrine disruptors [EHP 109:821-826].

In humans, exposure to endocrine disruptors has been proposed to be linked to reproductive and developmental abnormalities, increases in hormone-related cancers, attention deficit/hyperactivity disorder, and behavioral problems. Animals are particularly sensitive to effects from exposure to endocrine disruptors, and as such make good models for testing.

Rats are most commonly used in toxicity testing, but they are more expensive to use than mice--nearly twice as expensive to purchase and three times as expensive to house. Rats can have other disadvantages as well; they are more variable in factors such as body size and uterine response, so more of them must be tested to pick up subtle differences.

In their study, the researchers compared the sensitivity of the immature CD-1 mouse to that of the immature Sprague-Dawley rat. Each species was exposed to varying doses of the female sex hormone 17[beta]-estradiol and compared to an unexposed cohort. After three days of exposure, the animals were sacrificed and their uterine weight:body weight ratios determined. The researchers measured uterine epithelial cell height and number, as well as gland number, all of which increase with estrogenic activity and are therefore useful markers for testing chemicals with unknown estrogenicity. They also measured expression of the estrogen-inducible proteins lactoferrin and complement C3.

In general, the rats and mice proved to be equally well suited for uterotropic bioassay. Both species showed a similar dose-response increase in uterine wet weight as a result of exposure to 17[beta]-estradiol, although mice were more sensitive than rats at all doses tested. (Uterine wet weight, which includes both the tissue and fluid content of the uterus, is a more meaningful measure than dry weight because estrogens increase water absorption and the amount of fluid in the uterine cavity.) Both species showed an increase in uterine epithelial cell height over their respective controls. With respect to epithelial cell number, mice showed a greater increase than rats at any given dose. Further, mice showed an increase in gland number, while rats did not. Both rats and mice showed strong expression of lactoferrin and complement C3 in the uterine epithelial cells following 171[beta]-estradiol treatment.

Over the course of testing, researchers found a significantly greater variation in body weight in the rats versus the mice. Such variability in an experiment can decrease the significance of the results, such that weak estrogens may not be detected. The lower the variability, the fewer animals need to be used for testing. Thus, the ability to use mice might potentially translate into lower animal husbandry costs. Further, the smaller size of the mouse would mean that lesser amounts of chemical compounds would be required for testing. "The bottom line is that we can test more chemicals more efficiently using the mouse model versus the rat model," says Newbold.