EBDCs are potent fungicides. They are used on many plants, including root and leafy vegetables, fruits, and cereals, both in the field and after harvest. They have been known since the 1960s to cause goiter and to inhibit iodine uptake. The body breaks them down to ethylenethiourea, which has been identified by the EPA as a thyroid carcinogen in rats and mice and a probable human carcinogen. EBDCs can lower levels of thyroid hormone in rats at low doses, and may affect TSH. Ethylenethiourea also inhibits thyroid peroxidase, an enzyme necessary for synthesis of T₃ and T₄.

Perchlorate, used as an oxygen source in missile and rocket fuel, is common in drinking water in the southwestern United States. It inhibits iodine uptake by the thyroid, which can result in hypothyroidism and thyroid tumors. The EPA is in the process of setting a reference dose for perchlorate exposure in humans.

But these toxicants aren't the only chemicals that affect the thyroid. Fluoride, an element common in U.S. drinking water either naturally or added as a dental caries preventive, also suppresses thyroid hormone, although the mechanism is not understood, according to Paul Connett, a professor of chemistry at St. Lawrence University in Canton, New York, and an activist opposing the addition of fluoride to water. But the U.S. Food and Drug Administration emphasizes that fluoride concentrations in drinking water are safe, and most thyroid researchers believe many other chemicals pose a greater threat.

Another culprit, cyanide, occurs naturally in more than 1,000 plants, including cassava, sorghum, and bamboo, all important food sources in many parts of the developing world. At chronic subacute levels of exposure, cyanide can produce goiter and hypothyroidism. According to the Cassava Cyanide Diseases Network, a collaborative group of government officials and academics in Mozambique and Australia, cyanide contributes to goiter when the diet is already iodine-deficient. Traditional methods of processing cassava effectively neutralize cyanide's health effects, but according to WHO regional surveys of iodine deficiency status in Africa, ongoing wars, famines, and resulting mass migrations of African peoples dependent on cassava have interrupted such processing methods in many places, increasing the exposure of these already-stressed populations.

Soy isoflavones, touted by many as a benign substitute for endogenous estrogens for postmenopausal women, can be goitrogenic, although the amounts usually consumed by adults are insufficient to have that effect. Findings by toxicologist Daniel Doerge of the National Center for Toxicological Research and colleague Daniel M. Sheehan, reported in the June 2002 issue of EHP Supplements, suggest that soy isoflavones' ability to disrupt the thyroid depends on other factors such as iodine deficiency, other dietary goitrogens, and underlying thyroid dysfunction. Doergeemphasized the need for further research into the safety of soy isoflavones.

Many other foods contain thyroid-disrupting compounds, such as millet (containing epigenin and luteolin) and the cabbage family (containing goitrin). In parts of Africa where goiter is endemic, such as Sudan and the Republic of Guinea, it probably results from the combination of underlying iodine deficiency and large amounts of millet in the diet.

Radiation: Out, In, High, Low?

The thyroid can be affected by ionizing radiation through the skin by gamma radiation, including X rays; by fission products, such as cesium; or by ingestion or inhalation of iodine-131 (¹³¹I), an isotope present in nuclear fission products. ¹³¹I emits mostly beta radiation, which penetrates surfaces more shallowly than gamma radiation. It has a half-life of only about eight days, but in fallout form it can be inhaled directly or taken up by plants eaten by cows and goats, in whose milk it is then expressed. The thyroid cannot distinguish ¹³¹I from the nonradioactive form.

Ionizing radiation is at present the only known cause of human thyroid cancer. Thyroid tumors, or nodules, rarely result in detectable disruption of thyroid function. About 1% of these tumors are malignant and are treated with surgery, radiation, or both.

Most radiation research has focused on exposures from nuclear bomb tests, wartime bomb drops, and nuclear power plant accidents. Bomb drops and tests occurred from 1945 until aboveground testing was phased out in 1963. Possibly the most famous nuclear events were the Hiroshima and Nagasaki bombings in 1945. Other significant exposures occurred as the result of weapons production and testing in the Marshall Islands from 1946 to 1958, in the American West north and east of the Nevada bomb tests--principally in Utah and Idaho--from 1951 to 1963, and near the Hanford Nuclear Site in eastern Washington State from 1944 to 1957 (lesser releases continued at Hanford until 1972). The 1986 explosion and fire at the Chernobyl nuclear power plant exposed some 5 million people in Belarus, Ukraine, and Russia to radiation.

Far-flung effects. The 1986 explosion and fire at the Chernobyl nuclear power plant exposed some 5 million people in Belarus, Ukraine, and Russia to radioactive fallout. Children are the accident's youngest victims, suffering high rates of thyroid cancers. image credits: Top to bottom: Caroline Penn/Corbis; AFP/Corbis Office

Of the fallout victims studied, the Japanese victims received high doses of mixed external radiation in a very short time, the Marshall Islanders and Utah residents received primarily ¹³¹I exposure periodically over about 12 years, and Hanford downwinders probably received relatively lower doses, primarily of ¹³¹I, from multiple releases over a slightly longer period. The Chernobyl accident produced an acute, rapid exposure, mostly to ¹³¹I but also to a mixture of other fission products. Thus, in none of these events was there a clear-cut relationship among radiation type, duration of exposure, and internal versus external exposure.

High doses of external gamma radiation can damage thyroid gland tissue and lead to hypothyroidism as well as benign or malignant nodules, but the effects of low doses, especially of beta radiation such as ¹³¹I, are less certain. There was some increase in the incidence of both benign and malignant nodules among weapons-exposed populations, and Chernobyl victims, primarily children, suffered a striking spike in thyroid cancer incidence. According to the United Nations Scientific Committee on the Effects of Atomic Radiation's 2000 report, Sources and Effects of Ionizing Radiation, 1,791 children under 18 at the time of the accident were diagnosed with thyroid cancer between 1990 and 1998 in Belarus, Russia, and Ukraine, representing a fourfold increase in incidence. In Belarus, the incidence in children under 15 at diagnosis jumped from 0.2 per 100,000 to 5.6 per 100,000 in 1995, then tapered to 3.9 per 100,000 by 1998. Because some Chernobyl victims were iodine-deficient, their risk of cancer may have been magnified.

Although Chernobyl provided strong evidence that high ¹³¹I exposure increases cancer risk, it did not answer the question of whether low chronic doses do so, but research on Hanford downwinders seems to indicate low risk from low chronic doses. Yet, "nothing's crystal-clear regarding radiation effects," says Scott Davis, chairman of the Department of Epidemiology at the University of Washington School of Public Health and Community Medicine in Seattle.

Davis directed the Hanford Thyroid Disease Study (HTDS), which was conducted in response to a 1988 congressional mandate to the U.S. Centers for Disease Control and Prevention (CDC). The study asked whether exposure to ¹³¹I from Hanford resulted in increased incidence of thyroid disease among 3,441 subjects identified as having lived in the highest-exposure areas during the relevant period. The study found that the risk of thyroid disease did not vary with radiation dose and that Hanford downwinders were at no higher risk of thyroid disease than the general population. The preliminary HTDS results were released by the CDC in its 1999 Hanford Thyroid Disease Study Draft Final Report to a chorus of criticism from downwinders, antinuclear activists, and some scientists. The final CDC report was issued in 2002, but the HTDS results have not yet been published in a peer-reviewed journal; papers based on HTDS data are currently under review by several journals, according to Davis.

The 2000 National Academy of Sciences Review of the Hanford Thyroid Disease Study Draft Final Report supported the study's methodology but noted the report's reliance on highly uncertain dose amounts. In addition, Keith Baverstock, a prominent Chernobyl researcher with the WHO European Center for Environment and Health in Bonn, Germany, noted in a comment posted at the HTDS website (http://www.cdc.gov/nceh/radiation/hanford/htdsweb/index.htm) that the results could indicate an excess of thyroid cancer cases, and that "there are certainly more cases detected than would be expected on the basis of the national rates for invasive thyroid cancer." Follow-up research on the Hanford and Chernobyl victims is ongoing.

Next Steps

At this point in the understanding of environmental effects on the thyroid, there are far more questions than answers, but whether the toxicant under consideration is chemical or radioactive, most of the questions have to do with the effects of low-level exposures rather than acute ones. A further critical issue is whether combinations of toxicants have synergistic effects or sometimes even cancel each other out. Because the effects of low-level exposure may be subtle, Schettler thinks the precautionary principle should be implemented long before unambiguous results accumulate from laborious laboratory and epidemiologic studies.

Xenopus and us. Researchers are working to develop animal models of gene regulation by thyroid hormone including one that studies the metamorphosis from tadpole (above) to frog of the African clawed frog (Xenopus laevis). image credit: Dennis Kunkel

As researchers implement the EPA mandate to study endocrine disruption, they must tackle some difficult procedural issues. For example, the rat thyroid system has been well characterized, but because rats and humans may metabolize certain toxicants differently (and thus, for example, rats may receive noncomparable doses of hydrophobic toxicants such as PCBs via breast milk), it is not certain how confidently researchers can extrapolate rat results to humans.

Kevin Crofton, a behavioral neurotoxicologist at the EPA National Health and Environmental Effects Research Laboratory in Research Triangle Park, North Carolina, points to other problems: "One of the big issues is potential difference in susceptibility of the developing rat brain versus the developing human brain," he says. "For example, epidemiologic studies have shown that a twenty-five percent decrease in maternal T₄ during the first trimester results in decreased IQ in children, [but] there are no rat models that demonstrate this level of sensitivity to T₄ decreases. . . Alternatively, the measures of neurological development in the rat may be crude compared to IQ testing in children."

Many features of brain development are strongly conserved in vertebrates, however. Howdeshell is working to develop a model of the African clawed frog, Xenopus laevis, to shed light on gene regulation by thyroid hormone in human processes including brain development. Xenopus provides a good model for thyroid hormone effects because the metamorphosis from tadpole to frog is heavily mediated by thyroid hormone. "While Xenopus studies cannot be directly extrapolated to human development," says Howdeshell, "the thyroid hormone system is conserved across all vertebrates, including rats and humans, and Xenopus undergo thyroid hormone-directed brain maturation similar to the brain development of more complex vertebrates."

The lack of certainty regarding circulating hormone levels as a measure of toxicant effects is also vexing, and researchers are busy devising better assays and end points. Crofton has identified one such end point that he thinks may be a reliable indicator. In research currently under review, he has correlated postnatal T₄ concentrations with hearing loss in rats. These data demonstrate that, at a minimum, a 50% decrease of thyroid hormone in developing rats is needed to adversely impact hearing function.

A further issue is that endocrine disruption research involves to some extent a culture clash between toxicologists and endocrinologists; the methods and typical questions of interest peculiar to each must be harmonized with the other. For example, toxicologists tend to focus on dose-response determinations, whereas endocrinologists historically haven't considered dose response with respect to hormones at all, says Zoeller.

To help harmonize the disciplines and develop a joint research agenda, the NIEHS sponsored two meetings last year: Thyroid Hormone and Brain Development: Translating Molecular Mechanisms to Population Risk, an international conference that took place 23-25 September 2002 in Research Triangle Park, and Thyroid Toxicants: Assessing Reproductive Health Effects, a workshop held 28-29 April 2003 in Alexandria, Virginia. Many of the most pressing questions were discussed at both, and researchers were encouraged by the interdisciplinary scope of the research program being implemented in response to the EPA's mandate.

The thyroid system is so complex that understanding its normal function is difficult enough, but deciphering environmental disruptions to it is staggeringly convoluted. Yet, the more research reveals, the more pressing the questions become. The thyroid affects nearly every bodily system, and its role in fetal development makes protection of healthy thyroid function imperative. The increasing body of research indicating subtle and possibly additive or synergistic effects of environmental contaminants only adds salience to the issue.

Valerie J. Brown