Origins of obesity.

Fifty-nine million Americans are considered obese, and 300,000 die each year from related causes, making this disorder the second-leading U.S. cause of death after smoking. The ever-expanding reach of obesity is leading some scientists to look beyond traditional explanations that implicate genes, diet, and physical activity. At a 20 February 2004 symposium titled Obesity: Developmental Origins and Environmental Influences, cosponsored by the NIEHS and the Duke University Integrated Toxicology Program, presenters discussed data that support the hypothesis that in utero or neonatal exposures to environmental chemicals, notably endocrine disruptors, play a role in the etiology of obesity.

It has long been known that estrogens regulate the size of adipocytes (fat cells) in adult humans and animals. The same phenomenon may occur in children, too. "Humans with childhood obesity as well as infants born to mothers with gestational diabetes have adipocyte hyperplasia, which predisposes them to adult obesity and the metabolic syndrome [characterized by metabolic risk factors such as excessive abdominal fat, blood fat disorders, and insulin resistance]," says Paul Cooke, a professor of reproductive biology at the University of Illinois at Urbana-Champaign. "Estrogens seem to influence the process of adipocyte hyperplasia as well as the consequences of adipocyte hyperplasia in terms of producing metabolic alterations and inducing obesity."

Retha Newbold, a developmental biologist with the NIEHS Environmental Toxicology Program, says there is compelling evidence that exposure to endocrine disruptors during critical phases of cell differentiation may have long-lasting consequences. "These exposures likely alter mechanisms involved in weight homeostasis," she says. "We're still trying to determine if it's a direct effect on the adipose cells and how they differentiate or proliferate, or whether it's a disruption of the endocrine feedback loops."

Newbold's own research in this area has focused primarily on prenatal exposures to diethylstilbestrol (DES), a synthetic estrogen used in as many as 8 million pregnancies from the 1940s through the 1970s to prevent miscarriage. In animal studies, Newbold and others have shown that high-dose DES exposures during pregnancy produce small to normal-size offspring that tend to stay small as adults, whereas low-dose exposures produce normal-size offspring that tend to fatten as they age. This dose-dependent link with adult weight gain has been seen with prenatal exposure to other xenoestrogens and applies to neonatal exposure as well, says Newbold.

Dosage thresholds may be paramount, says Frederick vom Saal, a professor in the Division of Biological Sciences at the University of Missouri--even extremely low-level intrauterine exposure to the ubiquitous xenoestrogen bisphenol A, a constituent in food and beverage containers, can have far-teaching effects. "Low-dose exposure to bisphenol A increases differentiation of fat cells and also increases glucose transport, both of which may affect later development of obesity," he says. In research published in the August 1999 issue of EHP Supplements, vom Saal and colleagues found that a brief period of low-dose bisphenol A exposure during gestation in mice--an exposure resulting in blood levels much lower than those typically found in pregnant women and their fetuses--resulted in an accelerated rate of postnatal growth for both male and female mouse pups, as well as early puberty in females. He emphasizes that the bisphenol A connection in humans is still only a hypothesis.

Obesity later in life also is associated with intrauterine exposure to other environmental exposures including cigarette smoke. "With nicotine exposure in utero we see generalized underresponsivity of the autonomic nervous system posnatally," says Edward Levin, an associate professor of environmental sciences and policy at Duke. Nicotine exposure could alter appetite-regulating neural systems in the brain as well as the peripheral noradrenergic system, which is involved in metabolic control over adipose tissue.

Nutritional stresses during gestation are other possible contributors to future obesity. There are strong positive correlations between gestational diabetes, birthweight, and adolescent obesity; the mechanism may involve transfer of maternal glucose to the fetus, thereby stimulating the production of fetal insulin, which acts as a fetal growth hormone. Given the known associations, it is reasonable to speculate that simultaneous exposure to environmental estrogens might compound the weight-enhancing effects of insulin resistance during pregnancy, says Sheila Collins, who heads the Endocrine Biology Program at CIIT Centers for Health Research in Research Triangle Park, North Carolina.

"At this point, all of the data must be considered preliminary," says Jerrold Heindel, a scientific program administrator for the NIEHS and one of the conference organizers "Most of the data are from animal models, and there is still the possibility that the phenomena observed in animals may not apply to humans. Nonetheless, some of the findings presented at the meeting were rather striking." The meeting presentations are available online at http://www.niehs.nih.gov/ multimedia/qt/dert/obesity/agenda.htm.