Look Ma, too much soy; hormone in infant food reduces immunity in mice. (This Week).

The health benefits of soybeans are legion, but too much of the food may have a downside. A new study finds that large doses of the estrogenlike hormones that occur naturally in soybeans weaken the immune systems of mice. The finding has yet to be proven relevant to people, but soys widespread use makes the finding worthy of further investigation, some researchers say.

Most U.S. infants either are breastfed or receive cow's-milk-based formulas, but about 25 percent are fed soy-based formulas. Some postmenopausal women take soy-derived supplements. These products contain substantial quantities of plant estrogens, such as genistein. Soy-fed infants may ingest 10 times as much genistein per kilogram of body mass as do adults on high-soy diets.

Genistein has been linked to elevated risks of uterine and breast cancer in animals (SN: 6/16/01, p. 375), but few studies have explored whether soy estrogens hinder the immune system.

To get at that question, Srikanth Yellayi and his colleagues at the University of Illinois at Urbana-Champaign first removed the ovaries of female mice and castrated males to mimic newborn infants in their low concentrations of estrogen and other sex hormones. After injecting or feeding varying doses of genistein to the animals for 1 to 3 weeks, they measured the size of each animal's thymus, a small organ involved in immune-cell development.

The greater an animal's exposure to genistein, the smaller its thymus typically appeared at the end of the experiment. Genistein also reduced the number of immune cells associated with the organ, the researchers report in the May 28 Proceedings of the National Academy of Sciences.

Stephen Barnes of the University of Alabama at Birmingham says it's difficult to extract a practical health message from the mouse study. The researchers relied heavily on injections, which put a larger fraction of biologically active genistein into the blood system than oral consumption does, he notes.

Further, Barnes says, removal of the ovaries causes a mouse's thymus to grow. Since the researchers offer no data about the natural thymus size in mice with intact ovaries, the apparently reduced thymuses of animals given genistein may, in fact, be close to normal in size, he says.

Despite the finding in mice, "soy formulas support normal immunological development in infants," says Christopher T. Cordle of Abbott Laboratories' Ross Products Division, which manufactures soy formula for infants. Cordle and his colleagues reported in the February Journal of Pediatric Gastroenterology and Nutrition that whether infants were fed soy or cow's-milk formula made no difference in their immune systems at 12 months of age.

Paul S. Cooke, a coauthor of the mouse study, acknowledges that the "literature is totally contradictory" and that further research may show that his team's concerns are unfounded. Nevertheless, he says, since soy formulas provide the sole source of nutrition for 750,000 U.S. infants each year, "you want to err on the side of safety."