Thyroid hormones and glucose metabolism: the story begins before birth

It has been known for decades that thyroid hormones are important mediators of glucose homeostasis. Although the exact mechanisms involved are still unclear, a number of recent findings have contributed to our understanding of metabolic effects of thyroid hormone in a major way. It is in this context that the paper of Forhead et al. in this issue of Experimental Physiology (Forhead et al. 2009) provides some fascinating results that may boost our understanding of the effects of thyroid hormones on glucose metabolism in preterm infants. Forhead et al. studied the glycogen content in the liver, heart and muscle of sheep fetuses after experimental manipulation of thyroid hormone concentrations in utero by an elegant combination of 3,5,3′-triiodothyronine (T3) infusion and fetal thyroidectomy. They report that thyroid hormones have an important effect on the developmental regulation of glycogen storage in the liver and heart of the ovine fetus during late gestation. Hypothyroidism induced by fetal thyroidectomy prevents both the increment in hepatic glycogen and the fall in cardiac glycogen seen in the normal fetus near term. In addition, administration of T3 earlier in gestation causes a significant increase in glycogen content of the fetal liver and a decrease in the fetal heart. The authors conclude that thyroid hormones are important in the developmental control of hepatic and cardiac glycogen content in the ovine fetus near term, perhaps mediating in part the maturational effects of glucocorticoids on the liver. These findings suggest that hypothyroidism in preterm and growth-retarded infants might result in inadequate hepatic glycogen storage and consequently an inability to maintain euglycaemia in the immediate postnatal period. How could these metabolic effects of thyroid hormone have been brought about? Many genes involved in glucose metabolism are regulated by the active thyroid hormone T3, which exerts its action via binding to the thyroid hormone receptor (TR). These receptors are derived from two separate genes which encode the four major T3-binding isoforms TRα1, TRβ1, TRβ2 and TRβ3. TRα1 and TRβ1 are widely expressed and their relative abundance dictates whether T3-target tissues display either TRα1 or TRβ1 responsiveness, or show no TR isoform specificity at all (Yen, 2001). The TRα1 is thought to be predominantly involved in the metabolic effects of thyroid hormone, while TRβ1 and TRβ2 are thought to be key regulators in balancing the hypothalamic–pituitary–thyroid axis and keeping the euthyroid state. 3,5,3′-Triiodothyronine is derived from the prohormone thyroxine (T4), which can be activated via removing an iodine atom from the phenolic ring by the iodothyronine deiodinases type 1 (D1) and type 2 (D2). Type 3 deiodinase (D3) inactivates thyroid hormone by removing an iodine atom from the tyrosyl ring. The deiodinases are expressed in various tissues, and their expression levels vary enormously during development, regulated (at least in part) by thyroid hormone status. Type 1 deiodinase, predominantly expressed in liver, and D2, expressed in adipose tissue and skeletal muscle, regulate the bioavailability of T3 and thereby the response to insulin; increased T3 concentrations are associated with an increase in glucose turnover and an increase in insulin-mediated glucose disposal in skeletal muscle and adipose tissue via positive regulation of insulin-sensitive glucose transporter-4 (GLUT-4) transcription. Furthermore, molecular scanning of the human D2 gene revealed a novel missense variant (Thr92Ala), which is strongly associated with insulin resistance. Thus, differential expression of thyroid hormone receptor isoforms and deiodinase subtypes result in tissue-specific regulation of thyroid hormone action and target gene transcription patterns. This represents one potential way of modulating carbohydrate metabolism (Chidakel et al. 2005). A recent and fascinating study by Klieverik et al. (2009) in rats has shown that in addition to genomic effects of T3 on hepatic glucose metabolism via the TR expressed in the hepatocyte, stimulation of T3-sensitive neurons in the hypothalamus may also modulate hepatic glucose production via sympathetic projections to the liver, independently of circulating glucoregulatory hormones. This represents a novel central pathway for modulation of hepatic glucose metabolism by thyroid hormone (Klieverik et al. 2009). During the first half of pregnancy, maternal thyroid hormones are important because the fetus is not yet able to produce thyroid hormones by itself. The placenta regulates the amount of thyroid hormones available for the fetus for each stage of development. Early on, fetal plasma T3 concentrations are relatively low as a result of high D3 activity in the placenta and fetal tissues (Darras et al. 1999). Close to term, however, fetal plasma levels of T3 rise, which subsequently results in activation of T3-regulated genes. Many studies have focused on the detrimental effects of a fetal shortage of thyroid hormones on psychomotor development. The critical period for the central nervous system to be dependent on thyroid hormone is from early fetal life until well after birth. Congenital hypothyroidism is, therefore, one of the key targets of neonatal screening programmes that have been instituted by many countries to prevent cerebral damage by early detection and subsequent initiation of thyroid hormone supplementation. The present study of Forhead et al. (2009) shows that a shortage of thyroid hormone during late gestation may not only impair brain development, but also negatively influence carbohydrate metabolism in peripheral organs. By inference, low fetal T3 levels, resulting in suboptimal hepatic glycogen storage, might influence the neonate in its capacity to deal with nutritional challenges that arise at birth. Glycogen content in the cardiac muscle of the fetus is higher than in the newborn and is thought to be important in generating energy for the maintenance of the heart beat during anoxia by converting glycogen to lactic acid. Impaired cardiac glycogen storage is associated with reduced ability to survive in the absence of oxygen. If thyroid hormone availability during development impacts both brain development and metabolism, what more may we expect in the years to come? It is certainly to be hoped that future studies will continue to address permanent effects of changes (perhaps even subtle) in the fetal thyroid hormone milieu. The ultimate aim is to anticipate, and ultimately to prevent, developmental problems in neonates and children as a consequence of fetal thyroid hormone shortage or excess.