Enolases (2-phospho-D-glycerate hydrolases) have been characterized as glycolytic enzymes that catalyse the interconversion of 2-phosphoglycerate to phosphoenolpyruvate and may be involved in development and differentiation. Various tumors are known to display increased enolase expression and enolase seems to be involved in tumor progression. A single human ENO1 transcript encodes two proteins of 48 kDa and 37 kDa (MBP-1), both are discussed to participate in the down-regulation of c-myc expression. In this study we investigated the influence of retinoic acid (RA) on (a) the expression of both ENO1 gene products and (b) the proteome pattern in the follicular thyroid carcinoma cell line FTC-133. Reverse transcriptase polymerase chain reaction (RT-PCR) revealed an initial increase in ENO1 expression at 24h and 48h followed by a subsequent decrease at 72h following RA stimulation. ENO1 proteins were downregulated by RA at all incubation times tested. The downregulation of ENO1 protein correlated with RA-induced reduction in ATP levels and phosphorylation status in FTC-133. Moreover, RA treatment downregulated c-myc protein expression and decreased proliferation of FTC-133 cells. Two dimensional gel analyses revealed a significant RA-induced alteration in proteome profile in FTC-133. In summary, our data show that RA downregulates both ENO1 gene products at 48 kDa and 37 kDa in FTC-133. Thus, RA may potentially impair the metastatic potential of FTC-133 cells by different mechanisms including decrease in metabolism and proliferation and these effects could in part be mediated by a downregulation of c-Myc.

The biosynthesis of thyroid hormone is a complex process implying various regulated steps. In particular H2O2 is indispensable for iodination of tyrosyl residues and subsequent coupling of iodotyrosyls of thyroglobulin (Tg), but on the other hand it may have strong cytotoxic effects. Therefore, the thyrocyte is dependent on antioxidative defence mechanisms against this lifelong exposure to high H2O2 concentrations. We hypothesize this protective function could be performed by GPx3, a selenoenzyme highly expressed in the epithelial cells of the thyroid gland. GPx3 might be released into the lumen of the follicles and catalyze the detoxification of excess H2O2. To test this hypothesis, we purified human Tg from post mortem thyroid tissues. Isolated Tg globules were incubated in PBS with 0.5% SDS to extract proteins loosely adherent to the surface of the globules. After centrifugation, the pellet was incubated in PBS with 50 mM DTT to solubilize proteins bound by covalent disulfide crosslinks. By Western blot analysis we detected GPx3 in the SDS soluble fraction while no GPx3 was observed in the DTT fraction. Contamination of Tg with cellular proteins was excluded as incubation with an antibody against the intracellular membrane protein 5'deiodinase type I did not produce any signal in neither of the two fractions. Our data demonstrate that GPx3 is indeed present in the follicular lumen and attached to the surface of Tg globules. Here, GPx3 might (1) influence the rate of thyroid hormone production by controlling the availability of H2O2, (2) participate in crosslinking of Tg, by using excess H2O2 released into the follicular lumen or (3) might even be incorporated itself into the highly polymerized storage form of Tg. So far nothing is known about the activity state of a GPx3 linked to Tg, but possibly this is a further mechanism to fastly provide an enzyme which can interfere with the local H2O2 availability.

Endocrine disrupting compounds (EDC) of natural or synthetic origin were originally found to impair development and endocrine regulation of the reproductive system. Here, we assessed whether they also exert effects on the thyroid hormone axis. Female ovariectomized rats were treated for 12 weeks with two substances used as UV absorbers in sunscreens: Octyl-methoxycinnamate (OMC; 66 and 310mg/day) and 4-methylbenzylidene-camphor (4-MBC; 0.4 and 1.74mg/day) alone or in combination with soy-containing food. These substances are known to be resorbed by the skin, circulating in the blood (Janjua et al. 2004J Invest Dermatol 123:57–61) and suspected to have endocrine disrupting effects. The soy isoflavone genistein has estrogenic and anti-estrogenic effects and is known to inhibit thyroid hormone biosynthesis. As type I 5'-deiodinase (5'DI) is a well known T3-regulated endpoint, 5'DI activity was measured by deiodination of 125I--reverse T3, and 5'DI gene expression was determined by real time RT-PCR. Treatment with OMC caused a significant reduction of 5'DI activity from a basal level of 6.6 to 3.0 and 3.1 pmol iodide released x mg–1 x min–1 in the liver in the absence and presence of soy in the diet, respectively, while 4MBC did not show any effects. A similar decrease was also observed for 5'DI mRNA levels. In the kidney OMC as well as 4MBC led to a marked decline of 5'DI activity from 8.4 to 3.0 pmol iodide released x mg–1 x min–1, independent of the diet. Additionally, higher 5'DI activities were found in the kidney of animals consuming soy containing diet, while no such effect was obtained in liver. This data indicates that 4MBC and OMC affect thyroid hormone metabolism via tissue-specific inhibition of 5'DI activity and gene expression. Thus, 4MBC and OMC do not only interfere with steroid-hormone dependent regulation, but may also disturb endocrine regulation via the thyroid hormone axis.

Hürthle cell carcinomas (HTC) are characterized by mitochondrial amplification and enhanced oxygen metabolism. To clarify if defects in enzymes scavenging reactive oxygen species are involved in the pathogenesis of HTC, we analyzed selenium (Se)-dependent expression of various detoxifying selenoproteins in the HTC cell line XTC.UC1. Glutathione peroxidase and thioredoxin reductase activity was found both in cell lysates and conditioned media of XTC.UC1 cells and was increased by Na(2)SeO(3). Western blot analysis demonstrated the presence of thioredoxin reductase both in cell lysates and conditioned media and of glutathione peroxidase 3 in conditioned media. Type I 5'-deiodinase, another selenoprotein that catalyzes thyroid hormone metabolism, was detectable only in cell lysates by enzyme assay and Western blot, and responded to stimulation by both Na(2)SeO(3) and retinoic acid. A selenoprotein P signal was detected in conditioned media by Western blot, but was not enhanced by Na(2)SeO(3) treatment. In situ hybridization revealed glutathione peroxidase mRNAs in HTC specimen; glutathione peroxidase 3 mRNA levels were reduced. These data suggest adequate expression and Se-dependent regulation of a couple of selenoproteins involved in antioxidant defense and thyroid hormone metabolism in XTC.UC1 cells, so far giving no evidence of a role of these proteins in the pathogenesis of HTCs.

Enolases (2-phospho-D-glycerate hydrolases) are glycolytic enzymes that catalyse the interconversion of 2-phosphoglycerate to phosphoenolpyruvate, may also be involved in cellular differentiation. In this study we investigated the influence of retinoic acid (RA) on the proteome, the 48 kDa human α-enolase (ENO1) and its post-translational modifications in the follicular thyroid carcinoma cell line FTC-133.

Xanthohumol (XN) is the major prenylated flavonoid of the female inflorescences (cones) of the hop plant (Humulus lupulus). It is also a constituent of beer, a major dietary source of prenylated flavonoids, known to interfere with the sex steroid axis. In the present study, we analyzed in vitro and in vivo effects of XN on thyroid hormone transport and metabolism.

OBJECTIVE:Proinflammatory cytokines are involved in the pathogenesis of non-thyroidal illness (NTI), as shown by studies with IL-6-/- and IL-12-/- mice. Interleukin (IL)-6 changes peripheral thyroid hormone metabolism, and IL-12 seems to be involved in the regulation of the central part of the hypothalamic-pituitary-thyroid (HPT) axis during illness. IL-18 is a proinflammatory cytokine which shares important biological properties with IL-12, such as interferon (IFN)-gamma-inducing activity.DESIGN:By studying the changes in the HPT-axis during bacterial lipopolysaccharide (LPS)-induced illness in IL-18-/-, IFNgammaR-/- and wild-type (WT) mice, we wanted to unravel the putative role of IL-18 and IFNgamma in the pathogenesis of NTI.RESULTS:LPS induced a decrease in pituitary type 1 deiodinase (D1) activity (P<0.05, ANOVA) in WT mice, but not in IL-18-/- mice, while the decrease in D2 activity was similar in both strains. LPS decreased serum thyroid hormone levels and liver D1 mRNA within 24 h similarly in IL-18-/-, and WT mice. The expression of IL-1, IL-6 and IFNgamma mRNA expression was significantly lower in IL-18-/- mice than in WT, while IL-12 mRNA expression was similar. IFNgammaR-/- mice had higher basal D1 activity in the pituitary than WT mice (P<0.05); LPS induced a decrease of D2, but not of D1, activity in the pituitary which was similar in both strains. In the liver, the LPS-induced increase in cytokine expression was not different between IFNgammaR-/- mice and WT mice, and the decrease in serum T3 and T4 levels and hepatic D1 mRNA was also similar.CONCLUSIONS:The relative decrease in serum T3 and T4 and liver D1 mRNA in response to LPS is similar in IL-18-/-, IFNgammaR-/- and WT mice despite significant changes in hepatic cytokine induction. However, the LPS-induced decrease in D1 activity in the pituitary of WT mice is absent in IL-18-/- mice; in contrast, LPS did not decrease pituitary D1 activity in the IFNgammaR-/- mice or their WT, which might be due to the genetic background of the mice. Our results suggest that IL-18 is also involved in the regulation of the central part of the HPT axis during illness.

Type I 5’-deiodinase (5’DI) is involved in the metabolism of thyroid hormones, both activating the „prohormone“ T4 to the biologically active T3 and inactivating thyroid hormones and their metabolites. High 5’DI activity is mainly found in the liver, the kidney and the euthyroid pituitary. In the thyroid gland, 5’DI is a sensitive differentiation marker, as its expression and activity are high in the normal thyroid gland, but reduced or even undetectable in thyroid carcinomas in correlation to their degree of de-differentiation. Here, we used in situ hybridization of frozen sections and immunocytochemistry of paraffin sections to study the expression of 5’DI in thyroid tissues. By in situ hybridization, 5’DI signals were detected in the thyrocytes of rat thyroids. An antiserum was raised against a C-terminal peptide of feline 5’DI and used in Western blots. 27 kDa bands, corresponding to the substrate binding catalytic subunit of 5’DI, were detected in tissue samples and cell lines deriving from the liver or the thyroid gland of rat, mouse, pig or human origin. The antiserum also stained tubular regions and parenchymal cells in human kidney and liver sections, respectively. In human goiter samples, staining for 5’DI was solely confined to the intracellular compartment of the thyrocytes. Signals were of variable intensity differing from follicle to follicle, and within follicles, signal intensity varied between individual thyrocytes. Stromal cells, blood vessels and lymphocytic infiltrates of pathological thyroid tissues were free of staining. A tissue microarray was used to study expression levels of 5’DI protein in 11 thyroid carcinomas and matched normal controls (1 undifferentiated, 3 follicular and 7 papillary thyroid carcinomas). Down-regulation of 5’DI protein levels was seen in all of the 11 matched thyroid normal/cancer pairs examined. In summary, our results suggest a localization of 5’DI exclusively in the epithelial cells of the thyroid gland and a reduced expression in thyroid carcinomas.

To assess interference with endocrine regulation of the thyroid axis, rats (female, ovariectomised) were treated for 12 weeks with the suspected endocrine active compounds (EAC) or endocrine disrupters (ED) 4-nonylphenol (NP), octyl-methoxycinnamate (OMC) and 4-methylbenzylidene-camphor (4-MBC) as well as 17β-estradiol (E2) and 5α-androstane-3β,17β-diol (Adiol) on the background of a soy-free or soy-containing diet, and endpoints relevant for regulation via the thyroid axis were measured. Thyrotropin (TSH) and thyroid hormone (T4, T3) serum levels were altered, but not in a way consistent with known mechanisms of feedback regulation of the thyroid axis. In the liver, malic enzyme (ME) activity was significantly increased by E2 and Adiol, slightly by OMC and MBC and decreased by soy, whereas type I 5′-deiodinase (5′DI) was decreased by all treatments. This may be due rather to the estrogenic effect of the ED, as there is no obvious correlation with T4 or T3 serum levels. None of the substances inhibited thyroid peroxidase (TPO) in vitro, except for NP. In general, several endocrine active compounds disrupt the endocrine feedback regulation of the thyroid axis. However, there was no uniform, obvious pattern in the effects of those ED tested, but each compound elicited its own spectrum of alterations, arguing for multiple targets of interference with the complex network of thyroid hormone action and metabolism.

The most potent natural and synthetic plant derivatives affecting thyroid function and secretion are flavonoids. In the normal adult rat, thyroid hormones are transported to target tissues primarily bound to transthyretin (TTR). This is in contrast to human serum, where thyroxine-binding globulin (TBG) binds T4 with highest affinity. Flavonoids are strongly and preferentially bound to TTR in most species including man but show no or only minor competition for thyroid hormone bound to TBG, or to serum albumin. Xanthohumol is the major prenylated flavonoid of the female inflorescences (cones) of the hop plant (Humulus lupulus). It is also a constituent of beer, our major dietary source of prenylated flavonoids. In the present study, we analysed the effects of Xanthohumol on the thyroid hormone axis, in particular the binding of T4 to its transport and distributor proteins in human as well as in rat serum.

OBJECTIVE The sick euthyroid syndrome in critically ill patients without primary disease of the thyroid gland is characterised by low serum total triiodothyronine (T3), normal to elevated thyroxine (T4), elevated reverse T3 (rT3) and normal TSH levels. The aim of this work was to clarify if impaired T4 and rT3 5'-deiodination is an underlying mechanism. DESIGN AND METHODS We analysed the effect of the human recombinant proinflammatory cytokines interleukin (IL)-6 and IL-1beta, tumour necrosis factor-alpha (TNF-alpha) and interferon-gamma (IFN-gamma) on human type I 5'-iodothyronine deiodinase (5'DI) enzyme activity in the human hepatocarcinoma cell line HepG2, i.e. in a homologous human system. Furthermore, we analysed transcriptional effects of the cytokines by transient transfection assays using the luciferase or chloramphenicol acetyltransferase (CAT) reporter genes under the control of 1480 nucleotides of the human 5'DI promoter. RESULTS IL-6 at 500 pg/ml and TNF-alpha at 25 ng/ml had no significant effect, whereas 100 ng/ml IFN-gamma or 10 ng/ml IL-1beta reduced 5'DI enzyme activity to 77.9 and 59.5% of control values. IFN-gamma did not alter, IL-6 and TNF-alpha moderately decreased (in the case of IL-6 only in the CAT system), and IL-1beta (0.01-10 ng/ml) dose-dependently inhibited 5'DI promoter activity to a minimum of 38.1%. CONCLUSION IL-1beta inhibited both 5'DI enzyme and promoter activity and, thus, may exert its effect on thyroid hormone metabolism at least partially through direct inhibition of hepatic 5'DI gene transcription.