The ability to concentrate iodide is a fundamental property of normally functioning thyroid tissue and represents the first step in the production of thyroid hormones. Iodide uptake has been demonstrated in various extrathyroidal tissues, including salivary gland, gastric mucosa, and lactating mammary gland. Recently, cloning and molecular characterization of the human sodium iodide symporter (hNIS) have been reported; however, the patterns of hNIS gene expression in human tissues have remained unidentified. To examine the profiles of human hNIS gene expression in various normal human tissues, we performed high-stringency Northern blot analysis using a 32 P-labeled hNIS-specific complementary DNA (cDNA) probe (nucleotides 1184 -1667). To detect rare hNIS transcripts in small tissue samples, RT-PCR was performed with a pair of hNIS-specific oligonucleotide primers designed to amplify a portion (nucleotides 1184 -1667) of the hNIS gene. hNIS-specific transcripts were confirmed by Southern hybridization using a digoxigenin-labeled internal hNISspecific oligonucleotide probe (nucleotides 1460 -1477). To monitor cDNA integrity and quantity, and to rule out DNA contamination and illegitimate transcription, all samples were coamplified with two pairs of intron-spanning primers designed to amplify fragments of the human ␤-actin and thyroglobulin genes, respectively. Using Northern blot analysis, hNIS transcripts of approximately 4 kb were detected in thyroid gland and parotid gland but not in a broad range of endocrine and nonendocrine tissues. RT-PCR and Southern hybridization revealed hNIS gene expression in thyroid gland, salivary gland, parotid gland, submandibular gland, pituitary gland, pancreas, testis, mammary gland, gastric mucosa, prostate and ovary, adrenal gland, heart, thymus, and lung. By contrast, hNIS transcripts were not detected in normal orbital fibroblasts, colon, and nasopharyngeal mucosa. To further analyze hNIS gene sequences in parotid gland, mammary gland, and gastric mucosa, the EXPAND High Fidelity PCR System and three sets of overlapping NIS oligonucleotide primers were used for amplification and cloning. The resulting PCR products were subcloned into pBluescript-SK II (Ϫ) vector, and at least two independent cDNA clones derived from each tissue were subjected to automated sequencing. The nucleotide sequences of hNIS cDNA derived from parotid gland, mammary gland, and gastric mucosa revealed full identity with the recently published human thyroidderived NIS cDNA sequence. In conclusion, our results demonstrate markedly variable levels of hNIS gene expression in several extrathyroidal tissues. Although the physiological role of hNIS in these tissues awaits further study, our results suggest that the capacity to actively transport iodine may be a feature common to several secretory and endocrine tissues. The diminished capacity to transport and concentrate iodide in extrathyroidal tissues (such as parotid gland, mammary gland, and gastric mucosa), compared with thyroid gland, does not s...

The sodium iodide symporter (NIS), first identified in FRTL-5 cells, plays a critical role in iodide transport in the thyroid gland and in the production of the iodine-containing thyroid hormones. The aim of our study was to examine the regulation of NIS RNA steady-state levels and protein expression as well as functional activity in FRTL-5 cells. FRTL-5 cells cycling in media containing thyrotropin (TSH) were incubated for 48 hours with dexamethasone (10(-8)-10(-5) M), triiodothyronine (T3; 10(-9)-10(-6) M), methimazole (100 microM), propylthiouracil (PTU; 100 microM), perchlorate (10 microM) and potassium iodide (40 microM). In other experiments, cells were treated for 48 hours with various cytokines including interleukin-6 (IL-6) (100 U/mL), interferon-gamma (IFN-gamma) (100 U/mL), tumor necrosis factor-alpha (TNF-alpha) (10 ng/ml), IL-1alpha (100 U/mL), and IL-1beta (100 U/mL). Northern blot analysis using a 32P-labeled rat NIS-specific cDNA probe (nucleotides 1397-1937) revealed NIS mRNA as a single species of approximately 3 kb. When normalized for beta-actin mRNA signal intensities, NIS RNA steady-state levels in viable FRTL-5 cells were suppressed by approximately 80% after incubation with dexamethasone and T3 in a concentration-dependent manner. Iodide accumulation was decreased by up to 40% after incubation with dexamethasone and T3, respectively, in a concentration-dependent manner. Using a rabbit polyclonal rNIS-specific antibody, Western blot analysis of FRTL-5 cell membranes revealed a 60% and 70% suppression of NIS protein expression after treatment with T3 (0.1 microM) and dexamethasone (1 microM), respectively. In additon, NIS RNA steady-state levels were decreased by approximately 50% after treatment of monolayers with methimazole, PTU, and potassium iodide, respectively. Incubation with methimazole and PTU resulted in a 20% and 25% decrease of iodide accumulation, respectively, whereas potassium iodide suppressed iodide accumulation by approximately 50%. Treatment of FRTL-5 cells with IL-6 and IL-1beta resulted in a 30% decrease of NIS RNA steady-state levels. IL-6 did not alter NIS functional activity, but IL-1beta suppressed iodide accumulation by approximately 25%. IFN-gamma and perchlorate failed to alter NIS RNA steady-state levels. In contrast to IFN-gamma that had no effect on iodide accumulation, perchlorate almost completely suppressed iodide accumulation. TNF-alpha and IL-1alpha failed to alter NIS RNA steady-state levels in higher passage numbers of FRTL-5 cells, whereas treatment with TNF-alpha and IL-1alpha of early passages of FRTL-5 cells (<20 cell passages) resulted in a 70% and 40% decrease of NIS RNA steady-state levels, respectively, and in a 20% suppression of NIS functional activity. In conclusion, our data suggest that various agents known to affect iodide transport are capable of differentially altering NIS gene expression and function in cultured thyroid cells. Suppression of NIS gene expression and function by certain cytokines may be responsible, at least in part, for ...

The ability to concentrate iodide, a fundamental property of normally functioning thyroid tissue, is altered in various thyroid diseases. Given the critical role of the Na+/I- symporter (NIS) in controlling iodide access to the thyroid gland, altered expression of NIS may be responsible, at least in part, for an enhanced or diminished capacity to concentrate iodide. In this study, we used Northern blot analysis, a newly established quantitative polymerase chain reaction (PCR) assay and in addition hNIS-directed immunohistochemical analysis to assess the levels of hNIS mRNA and protein expression in various localized and diffuse benign thyroid abnormalities, including Graves' disease (GD), scintigraphically cold solitary benign thyroid nodule (CBTN), nontoxic multinodular goiter (NMNG), solitary autonomously functioning thyroid nodule (AFTN), and mild diffuse iodine deficiency goiter (IDG). In addition, in view of the recent identification of putative binding sites for the transcription factors thyroid transcription factor-1 (TTF-1) and human paired-box-protein-8 (Pax-8) in the human NIS gene promoter, we used reverse transcriptase-polymerase chain reaction (RT-PCR) to assess in these same samples the levels of TTF-1 and Pax-8 gene expression. Northern blot analysis revealed high levels of hNIS gene expression in thyroid specimens derived from patients with GD and AFTN. In contrast, levels of hNIS mRNA expression were moderate in NMNG, low in diffuse IDG, and very low in CBTN. Quantitative RT-PCR analysis of hNIS mRNA transcripts revealed variable but generally low levels of hNIS gene expression in IDG and NMNG, and undetectable or very low levels of hNIS mRNA in all scintigraphically CBTN studied. In contrast, markedly elevated levels of hNIS mRNA transcripts were detected in active GD (up to 17-fold) and AFTN (up to 25-fold). Immunohistochemical analysis revealed abundant hNIS protein expression by thyroid follicular cells in GD, moderate and heterogeneous levels in NMNG, and very low levels in CBTN. hNIS mRNA levels were correlated with TTF-1 and Pax-8 gene expression in GD and, to a lesser degree, in AFTN, NMNG, and IDG, but not in CBTN. In general, hNIS gene expression was more closely correlated with TTF-1 as compared to Pax-8 gene expression. In conclusion, the abundance of hNIS mRNA and protein expression in a broad range of benign thyroid pathologies correlated well with their functional state as assessed by thyroid scintigraphy. In addition to TTF-1 and Pax-8, other transcription factors and enhancer elements may contribute to regulation of NIS gene promoter activity.

Because the thyrotropin receptor (TSHR) has long been considered a thyroid-specific protein, its presence in extrathyroidal tissues has been controversial. In this study, we sought to detect and quantify this potentially low abundance mRNA in various extrathyroidal tissues using liquid hybridization analysis (LHA) and to detect protein with immunohistochemical studies. Strongly positive protected bands, indicating the presence of both intact (2.4 kb) and variant (1.3 kb) TSHr mRNA, were apparent in LHA gel lanes corresponding to normal thyroid, Graves' thyroid, and thymus. Less abundant protected bands of the same sizes were present in lanes corresponding to normal adrenal, and samples from normal kidney were faintly positive. The full-length transcript:variant transcript ratio was approximately 1:1 in all positive tissues. Immunohistochemical analysis of TSHR-like reactivity in paraffin-embedded thymus, adrenal, and kidney revealed specific staining in each of these tissues. No TSHR mRNA or TSHR-like immunoreactivity was detected in samples from several other normal human tissues. We conclude that measurable TSHR mRNA and protein expression is not restricted to the thyroid gland. Further study is warranted to determine whether these extrathyroidal receptors play a role in normal physiology or in disease.

There are several thyroid antigens including human sodium iodide symporter (hNIS), thyrotropin receptor (TSH-R), thyroid peroxidase (TPO), and thyroglobulin (Tg) that have been considered to be thyroid-specific proteins involved in the pathogenesis of autoimmune thyroid diseases. We examined the expression of these thyroid-tolerance related genes in normal human thymus, the lymphoid organ responsible for the induction of central T-cell self. Reverse transcription-polymerase chain reaction (RT-PCR) amplifications were performed with 4 pairs of oligonucleotide primers specific for the hNIS, TSH-R, TPO, and Tg genes, respectively. Gene-specific transcripts were confirmed by Southern hybridization using digoxigenin-labeled internal oligonucleotide probes. To monitor cDNA integrity and quantity, all samples were coamplified with a pair of intron-spanning human beta-actin-specific oligonucleotide primers. Furthermore, using a highly sensitive immunostaining technique and antibodies specific for these 4 antigens, we examined whether NIS-, TSH-R-, TPO-, and Tg-specific immunoreactivity can be detected and localized in normal human thymus. RT-PCR and Southern hybridization revealed expression of each of these 4 thyroid-related genes in normal human thymus. In addition, immunohistochemical analysis of frozen tissue sections derived from normal human thymus showed marked immunoreactivity for NIS, TSH-R, and Tg as well as weaker staining for TPO. Control reactions using isotype matched nonimmune immunoglobulins were consistently negative. Taken together, our results suggest that NIS-, TSH-R-, TPO-, and Tg-RNA are present and actively processed to immunoreactive NIS-, TSH-R-, TPO-, and Tg-like protein in human thymus. These data support the concept that pre-T lymphocytes may be educated to recognize thyroid-related epitopes expressed in thymus, and, thus, to generate self-tolerance against these thyroid-related antigens.