The elemental importance of sufficient iodine intake: a trace is not enough.

Thyroid hormone is an important regulator of energy metabolism and crucial for the development of different tissues, in particular the brain (1, 2). Iodine is an essential trace element for the synthesis of thyroid hormone; as their names indicate, the prohormone T4 contains four iodine atoms, and the principal bioactive hormone T3 contains three iodine atoms. If iodine intake is sufficient, the normal human thyroid gland secretes predominantly T4 and only about 20% of daily T3 production (3). Most T3 is generated outside the thyroid gland by enzymatic outer ring deiodination (ORD) of T4 in different tissues. T4 and T3 are converted by inner ring deiodination to the receptor-inactive metabolites rT3 and 3,3 -diiodothyronine. The peripheral metabolism of thyroid hormone involves three deiodinases (3). Two deiodinases (D1 and D2) have ORD activity and are thus capable of producing T3 from T4. Conversion of T4 to T3 is the most efficient reaction catalyzed by D2, but this is far from true for D1, which is much more effective in the ORD of rT3 (4). In addition to expression of D1 in the thyroid gland, the enzyme is particularly abundant in liver and kidney (3). Hepatic and renal D1 are important sources of circulating T3. D2 is expressed in human brain, pituitary, thyroid, and skeletal muscle (3). In particular in the brain and also in the anterior pituitary, D2 is extremely important for local T3 production (3). Although modest levels of D2 are found in the normal human skeletal muscle, they appear sufficient to contribute a major part of peripheral T3 production in view of the large size of this tissue (5). The relative importance of skeletal muscle D2 for plasma T3 production increases in hypothyroidism because this is associated with a decrease in D1 expression in liver and kidney and an increase in D2 expression in skeletal muscle and other tissues (6). The opposite is true for hyperthyroidism. D1 expression is stimulated by its product T3 at the transcriptional level (3). Conversely, D2 activity is under negative control of its substrates T4 and rT3, which induce the ubiquitination and degradation of the enzyme (3). Although D1 also has inner ring deiodination activity, a third deiodinase (D3) is the major player in the degradation of thyroid hormone, showing preference for T3 over T4 as the substrate (3). In adults, the brain may be the major site of D3 expression, although normal skin also contains substantial D3 activity (7, 8). Even higher D3 levels are expressed in fetal tissues, in particular brain and liver (9, 10). The role of D3 in fetal development is intriguing. In addition to different fetal tissues, very high D3 activities are expressed in the placenta and the pregnant uterus (11–13). Despite this high D3 expression, placental transfer of maternal T4 represents the only source of fetal plasma T4 in the first half of gestation (1, 2). In the second half of gestation, the fetal thyroid gland becomes an ever more important source of circulating T4 (1, 2). The high D3 activities expressed in the fetoplacental unit are probably important to prevent premature exposure of growing tissues to bioactive T3, which induces cellular differentiation. Nevertheless, sufficient fetal plasma T4 accumulates to supply the brain with substrate for local T3 production (10). A recent extensive study of fetal and neonatal human brain development has demonstrated region-specific temporary profiles of tissue T4, T3, and rT3 levels, and D2 and D3 activities (10). These findings support the view that normal brain development requires the coordinated expression of D2 and D3 to secure intracellular T3 levels that are optimal for the particular brain region and stage of development. From recent work in different laboratories, a picture has emerged emphasizing the interaction between astrocytes and neurons in the local regulation of T3 levels in the brain (14, 15). Neurons are thought to be the major target cells for T3 in the developing brain, and this T3 is supplied by neighboring astrocytes (14, 15). Several steps are required to use plasma T4 and make it available as bioactive T3 to central neurons (Fig. 1). In addition to transfer of T4 at the choroid plexus from plasma to CSF and subsequently to periventricular cells, T4 supply to the brain requires its transport across the blood-brain barrier, but little is known about this process (16–18). Recently, one member of the organic anion transporting polypeptide family, OATP1C1, was shown to be highly specific for T4 and expressed almost exclusively in brain capillaries, suggesting that it is important for T4 transport across the blood-brain barrier (19, 20). Also the transporters involved in T4 uptake in astrocytes and T3 release from these cells have not been identified. However, recent findings suggest that a member of the monocarboxylate transporter family, MCT8, is extremely important for neuronal T3 uptake (15, 21). The MCT8 gene is located on the X chromosome, and males with an inactivating MCT8 mutation show a distinct phenotype of severe psychomotor retardation in combination with high serum T3 levels (22–24). This syndrome is explained by the impaired neuronal T3 uptake if MCT8 is inactivated, preventing T3 access to its nuclear receptor as well as to D3 also present in these cells, with a consequent defect in neuronal T3 action and metabolism (25). The clinical features of patients with MCT8 mutations dramatically underscore the crucial role of thyroid hormone in brain development. This has been known for a long time from the severe neurological deficits in subjects that have Abbreviations: D1–D3, Deiodinases 1–3; ORD, outer ring deiodination.