Triumphs of the thyroid despite lesser conversion.

In a seemingly wasteful fashion, the thyroid gland focuses its effort into secreting an inactive version of thyroid hormone T4. In fact, a hormone, from the Greek word for impulse, is a biologically active molecule, so in the case of thyroid hormone, the actual active hormone is that of the triply iodinated derivative, T3. To successfully produce T3, the human thyroid processes about 1000 nmol iodide daily, packaging the iodide into a very large molecule [thyroglobulin (Tg)] that is subsequently hydrolyzed. The resulting iodinated products are sequentially unleashed in a controlled manner so that, eventually, only 10 nmol of the highly potent T3 are secreted. The bulk of T3 production (40 nmol), however, occurs outside of the thyroid parenchyma through the conversion of T4 to T3 by the activating deiodinases, a group of selenoenzymes that can selectively remove iodine moieties from T4 (reviewed in Ref 1). Given that only 10 nmol or less of T3 are directly secreted from the thyroid, one would assume that an absence of the deiodinases should impair the overall daily production of T3, leading to systemic hypothyroidism. In this issue of Endocrinology, Galton et al. (2) challenge the conventional paradigm of thyroid hormone homeostasis by showing that mice lacking both of the activating deiodinases do just fine, because they are still capable of maintaining normal amounts of T3 in serum and do not suffer from systemic hypothyroidism. The low abundance of T3 in the thyroid secretion of humans and rodents is explained by the high T4/T3 ratio in their Tg, with ratios of about 15:1 in humans and 8:1 in rats (1). Even considering that some T4 to T3 conversion occurs in the thyroid parenchyma after Tg hydrolysis, of which very little is known (3), T4 still predominates over T3 in the thyroid secretion. Notably, this varies among species, where the human thyroid produces less than 20% of the body’s T3, whereas the rodent’s thyroid contributes about 40% (1). Still, this leaves the majority of T3 production to extrathyroidal tissues in both species. Conversion of T4 to T3 is catalyzed by the type 1 or type 2 iodothyronine deiodinases (D1 or D2), which activate thyroid hormone through removal of an outer-ring iodine on T4 (known as 5 -deiodination). D2 is the main deiodinase to catalyze local activation of thyroid hormone, being most notable for its expression in pituitary gland, brain, and brown adipose tissue (4). D1, on the other hand, is kinetically inefficient but is expressed at such high levels in kidney and liver that it, too, plays a sizable role in extrathyroidal T3 production, particularly in rodents (5). In fact, it is commonly accepted that D1 and D2 contribute equally to extrathyroidal T3 production in mice and rats. This is unlike the situation in humans, where D2 is argued to play a greater role in the generation of plasma T3 (1). Despite these differences in T3 production between humans and rodents, the generation of mice lacking specific deiodinase activity has allowed for in-depth study of the physiological roles played by D1 and D2 in thyroid hormone homeostasis (6). Remarkably, mice with targeted disruption of Dio1 [D1 knockout (D1KO)] or Dio2 (D2KO) have been found to exhibit a normal serum T3 concentration (7, 8). Even the C3H/D2KO mouse that lacks D2 and expresses only residual D1 activity maintains euthyroid serum T3 levels (9). All three of these mouse models have increased concentrations of serum T4, and both the D2KO and C3H/D2KO have elevated TSH. These adjustments in T4 and TSH indicate that keeping serum T3 concentration within the normal range is the default hypothalamic primary directive for the TRH-TSH-thyroid axis, despite marked elevation in serum T4 concentrations. This innate ability to avoid catastrophic hypothyroidism is fascinating and has only recently been appreciated as a result of these deiodinase-deficient mouse models. To better appreciate the plasticity and adaptive capacity of the T3-generating system, and to rule out the effect of possible compensation by D1 or D2 in the aforementioned mouse models, the present study by Galton et al. (2) evaluates mice lacking both D1 and D2 activity. As with the other mouse models, D1/D2KO mice sustain a normal serum concentration of T3 at the expense of markedly altered thyroidal status. There is a 2-fold greater serum T4 concentration and a 2.6-times higher TSH serum level when compared with wild-type mice. The elevated level of T4 is higher than that seen in either the D1KO or D2KO, and the elevated TSH level is comparable to that measured in D2KO