When PCBs act like thyroid hormone: mysterious mimicry in the fetal brain.

Children exposed to relatively high levels of polychlorinated biphenyls (PCBs) can have deficits in general intellectual ability, poor short-term memory, and short attention span. Some researchers propose that this may be caused by the ability of PCBs to interfere with thyroid hormone action. Thyroid hormone plays an important role in directing brain development, an action that is largely accomplished by binding to nuclear receptors that either amplify or dampen the expression of genes in specific brain regions at specific developmental times. In this issue, Kelly J. Gauger of the University of Massachusetts Amherst and colleagues report that PCBs can act like thyroid hormone in the fetal brain, but that they don't bind to thyroid hormone receptors; instead, the activity happens through a mechanism that is not yet understood [EHP 112:516-523].

Experimental studies consistently find that PCB exposure decreases levels of the thyroid hormone thyroxine in rats. Therefore, some researchers conclude that PCBs might affect brain development by causing a reduction in thyroid hormone when the developing organ needs it the most. On the other hand, PCBs could act directly on thyroid hormone signaling in the fetal brain--one theory has held that perhaps PCBs bind to thyroid hormone receptors in the brain and either block their action or inappropriately stimulate them. This distinction has important implications for designing a test that can effectively screen for thyroid hormone disruptors.

To study whether PCBs act on the fetal brain through a reduction in maternal thyroid hormone levels or by exerting a direct action on the brain, the team fed the PCB mixture Arochlor 1254 to pregnant rats from gestational days 6 to 16. Six rats received a dose of 1 milligram per kilogram, six received 4 milligrams per kilogram, and six were controls. On gestational day 16, the researchers examined the brain of one fetus from each litter using gene probes and in situ hybridization to obtain a quantitative measurement of gene activity. This was done on gestational day 16 because the fetus cannot make its own thyroid hormone at this time. Therefore, the team could test whether PCB-induced reductions in maternal thyroid hormone would produce effects on the fetus that are similar to those caused by maternal hypothyroidism.

The team found that both doses of Arochlor 1254 reduced thyroid hormone levels in pregnant rats on gestational day 16. Other measures of hypothyroidism, such as maternal weight gain and changes in litter size, were not observed.

The team further found that the Arochlor 1254 exposures caused an increase in the expression of genes normally upregulated by thyroid hormone--strong evidence, the authors write, that PCBs can produce thyroid hormone--like effects in the fetal brain. This and other details of the results suggest that PCBs must be directly activating thyroid hormone receptors, according to the authors. Because they could not identify individual PCBs or their metabolites that exhibited strong binding to thyroid hormone receptors, the mechanism by which PCBs might affect these receptors will likely be a novel finding.