compared to the euthyroid state . So, although propylthiouracil is used to treat hyperthyroidism mainly due to its inhibitory action on the enzyme, thyroid peroxidase, propylthiouracil at high doses also inhibits D1 activity .
The production of TH is regulated by the hypothalamo-pituitary-thyroid axis, while the biological activity of TH, i.e. the tissue availability of T3 is mainly regulated by the three deiodinases . The serum concentrations of thyroid function variables are regulated very closely within the individual, while there is a substantial interindividual variation in serum levels of both T4, T3, TSH and thyroglobulin. This was first demonstrated by Feldt-Rasmussen et al.  in 1979, and has later been verified by others [58, 59]. This pattern indicates an important genetic component in the regulation of serum concentrations of thyroid function variables, with an individual set-point for thyroid function. A classical twin study demonstrated results to support this by finding approximately 67% heritability accounting for the variations in plasma concentrations of TSH, and free T3 and T4 , and in a population study of Mexican Americans 26-64% of the interindividual variation was suggested to be due to heredity . Finally, Spencer has in a guideline publication with Baloch as first author  described an individual TSH-free T4 log-linear set point ratio as further support of this concept.
Along these lines polymorphisms have recently been identified in the D1 gene . The T-allele of one of them (D1a) was dose dependently associated with increasing plasma reverse T3 levels and decreasing T3/reverse T3 ratio, while the G-allele of the other (D1b) showed the opposite. Since D1 physiologically plays a key role in production of serum T3 and in the clearance of reverse T3, it might be assumed that the D1a-T variant has a negative effect on tissue D1 activity, while the D1b-G variant could be responsible for a positive effect . Another study performed in a different population showed a dose effect from D1a-T allele on serum T3 concentration and thus supported this hypothesis . In performing such studies it is important to pay attention to the age distribution of the population since a decreased T3 production by D1 may be masked by the production of T3 by skeletal muscle D2 in young subjects [64, 65]. Skeletal muscle size and strength increase during childhood and in young adults, and again gradually declines throughout adult life. D1 activity increases during childhood and adolescence and again decreases during ageing, but the relative contribution of D2 to serum T3 production may be more important in young compared to elderly subjects, resulting in a relatively smaller contribution to T3 production from D1 in young persons.
A polymorphism in the D2 gene did not demonstrate any relationship with plasma concentrations of T3 or reverse T3 , which is possibly explained by the fact that D2 plays the major role in local T3 production in D2-containing tissues. It would therefore not be expected to find polymorphism relations to plasma concentrations, and an effect of the polymorphism on intracellular T3 cannot be excluded. One study described a correlation between the same polymorphism and insulin resistance in obese women . Since there was no concomitant association with their body composition, it was hypothesized that the results might be explained by a linkage to another polymorphism .
No significant association between concentrations of TH and a polymorphism in the D3 gene have been described and there have been no descriptions of deficiencies of deiodinases neither in humans nor in animals . The present conclusion of studies over the recent years have clarified that genetic variation by polymorphisms plays an important role in the serum concentrations of thyroid function variables, and that deiodination of the iodothyronines are crucial players in this unique set-point. In adults it is becoming increasingly clear also, that only minor modifications from this set-point resulting in mild (or subclinical) hypo- or hyperthyroidism, may induce alterations in thyroid hormone bioactivity with consequences for clinical end-points such as bone mineral density, atherosclerosis and heart rate, with increased morbidity and even increased mortality [68, 69]. How frequent such alterations are in children has not been investigated, and therefore it is unknown if, e.g., polymorphisms in the deiodinase genes may have an impact on bone development in children and adolescents.
Because D1 is a selenoprotein, one might expect to find decreased D1 enzyme activity in selenium deficiency, and in rats this was indeed demonstrated for hepatic and renal D1 [70, 71]. There are, however, differences in the organ sensitivity to selenium deficiency, so studies may show difference in results depending on the organs studied. Furthermore, it is difficult to study in humans, because it is difficult to find pure, isolated selenium deficiency. Yet, mildly elevated serum T4 levels have been described in selenium-deficient humans [70-73]. Selenium supplementation in an area with both iodine and selenium deficiency has resulted in an unexpected reduction of serum T4, and in some an increase of serum TSH as indication of worsening of hypothy-roidism [74, 75]. This reaction might be explained by selenium deficiency causing reduced Dl-catalyzed inner ring deiodination of iodothyronines, thereby protecting against hypothyroidism. These results are in contrast to a study by Roti et al. , who examined the effect of selenium supplementation in an area with mild iodine deficiency. The eight female subjects had a positive perchlo-rate discharge test after a previous episode of subacute or postpartum thyroiditis and thus might have been at risk of developing thyroid dysfunction, but they all remained with normal TH concentrations after selenium supplementation.
Nonetheless, it seems that restoration of adequate iodine supply is essential before selenium intake is increased, thereby avoiding selenium-dependent deio-dinative degradation of TH, subsequent urinary loss of iodine and TSH stimulation of an iodine depleted thyroid gland .
The issue of selenium intoxication is still controversial, and intakes of selenium up to 400 |xg/day have not resulted in any adverse effects . Signs of reversible intoxication have been reported by ingesting more than 1,000 |xg/day over a long time . Nevertheless, paramedication, over-the-counter administration, and uncontrolled use of selenium containing preparations with accompanying strong advertisements on the Internet should be monitored and restricted in order to avoid uncontrolled distribution of selenium and its accumulation into body proteins. These commercial preparations are marketed and sold under names such as Thyroid Helper, Daily Energy, Daily Protector, Thyroid Booster and many more.
During critical illness at any age, pronounced alterations in plasma thyroid hormone concentrations occur. It is a whole body response to virtually any serious illness and covers synonyms such as nonthyroidal illness (NTI), low T3 syndrome and euthyroid sick syndrome . The validity of thyroid hormone measurements was initially described as questionable , and although it is generally accepted that a low free T3 perhaps together with low free T4 and TSH at later stages is a hallmark of the disease, the interpretation of serum values of thyroid function variables is still questionable . In fact, an estimate of free TH concentration by total hormone measurement and correction for binding sites on the binding proteins by a TH uptake test is superior to the so-called 'direct' free TH measurements by automated analyses, since the direction of changes of each of the measurements will indicate whether the encountered abnormality is within (i.e. thyroid dysfunction) or outside the thyroid gland (i.e. NTI) .
The typical changes of NTI have initially been described as low T3 (later also T4) together with elevated reverse T3, and studies on the role of deiodi-nases during critical illness focused on D1 and D2, since the reduction of circulating T3 was thought to be due to decreased peripheral deiodination by D1, D2 or both [52, 81, 82]. It is, however, also possible that D3 is induced in the liver and abundant tissues such as skeletal muscle, thereby decreasing the ratio between T3 and reverse T3, a mechanism that might have been underestimated in the previous studies . Cytokines, in particular interleukin-6, may be responsible for part of these changes in NTI, but cannot explain the full effect [83, 84]. Pulsatility of hypothalamic-pituitary hormones including TRH-TSH is almost abolished in this syndrome, and restoration of all the axes by injection of hypothalamic hormones can restore the abnormalities almost completely [85, 86]. Whether this also involves pituitary D1 and D2 is not fully clarified .
Was this article helpful?