European Journal of Endocrinology (2001) 145 591±597 ISSN 0804-4643 CLINICAL STUDY Sodium/iodide symporter (NIS) and pendrin are expressed differently in hot and cold nodules of thyroid toxic multinodular goiter Diego Russo, Stefania Bulotta, Rocco Bruno2, Franco Arturi1, Paolo Giannasio2, Michael Derwahl5, Jean-Michel Bidart3, Martin Schlumberger4 and Sebastiano Filetti1 Dipartimento di Scienze Farmacobiologiche and 1Dipartimento di Medicina Sperimentale e Clinica, UniversitaÁ di Catanzaro, 88100 Catanzaro, 2Ospedale Civile di Tinchi-Pisticci, Matera, Italy, 3Departments of Clinical Biology and 4Nuclear Medicine, Institut Gustave-Roussy, 94805 Villejuif, France and 5Department of Medicine, Division of Endocrinology, St Hedwig Hospital and Humboldt University, 10115 Berlin, Germany (Correspondence should be addressed to Sebastiano Filetti, Dipartimento di Medicina Sperimentale e Clinica, UniversitaÁ degli Studi di Catanzaro, Via T. Campanella, 115, 88100 Catanzaro, Italy; Email: ®[email protected]) Abstract Objective: The expression of two iodide transporters, the sodium/iodide symporter (NIS) and pendrin, was analyzed in thyroid tissues of patients with toxic multinodular goiter (TMNG) and non-toxic multinodular goiter (MNG). Methods: The levels of NIS and pendrin proteins were analyzed in total protein extracts from nodular and non-nodular tissues by Western blot. Results: In tissue samples from TMNG, we found an increased expression of NIS (2.5-fold) in the hot nodules, and similar levels between cold nodules and non-nodular tissues. In contrast, the levels of pendrin were slightly increased in both hot and cold nodules from TMNG, and decreased (about twofold) in cold nodules from MNG. We also noticed that there was no relationship between NIS and pendrin expression. Conclusions: Our data demonstrate that hot nodules from TMNG express a higher number of iodide transporters (mainly NIS), whereas cold nodules from TMNG, but not from MNG, show levels of the two proteins comparable with normal tissue, suggesting a role in vivo of TSH in maintaining the expression of NIS and pendrin protein in normal thyroid tissue. Finally, different mechanisms are involved in the regulation of NIS and pendrin expression. European Journal of Endocrinology 145 591±597 Introduction For years, thyroid nodules have been characterized on the basis of their ability to concentrate radioiodine, Iodide uptake is the ®rst step in thyroid hormone thus considered functioning or non-functioning at 131I- production by thyroid follicles. Thyroidal iodide trap- scintiscan and named `hot' or `cold' respectively. Many ping and concentration from blood are achieved by an studies have shown abnormal expression of the NIS active, energy-dependent transport process across the gene in various thyroid disorders (reviewed in 10, 11), baso-lateral plasma membrane of the thyrocytes; iodide especially in thyroid malignancies. In contrast, few data is then transported trans-cellularly to the apical are available on PDS gene expression (8), or on the membrane, where it is organi®ed (1). The way to relationship between NIS and pendrin expression at the clarify the molecular mechanisms underlying iodide protein level (6±9, 12). transport in thyroid cells has been opened up as a result In this study, using Western blot analysis we of the cloning and expression studies of two proteins examined the expression of NIS and pendrin proteins mainly involved in this process, namely the sodium/ in a series of 25 thyroid nodules from ten unrelated iodide symporter (NIS) (2, 3) and pendrin (4, 5). NIS is patients with toxic multinodular goiter (TMNG), located in the baso-lateral plasma membrane of including both hot and cold nodules, in order to clarify thyrocytes (6, 7) and catalyzes active I2 uptake into the role of these two iodide transporters in the the cell, while pendrin, discovered as the defective functional behavior of the nodular lesions. Indeed, the protein in Pendred syndrome (PDS) (4), is a chloride/ simultaneous presence of hot and cold nodules in the iodide transporter expressed in the apical membrane (8, same patient makes this disease a good model for 9), presumably responsible for the I2 accumulation in excluding interference due to genetic and environmen- the colloid. tal factors. For the ®rst time, the protein expression q 2001 Society of the European Journal of Endocrinology Online version via http://www.eje.org Downloaded from Bioscientifica.com at 10/02/2021 04:05:42AM via free access 592 D Russo and others EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 145 levels were compared with the normal, non-nodular another ®ve patients with MNG and non-nodular tissue of each patient. To evaluate the in¯uence of tissues from two patients with Graves' disease were thyroid-stimulating hormone (TSH), which is obviously also examined. Non-nodular, normal tissue from each suppressed in patients with TMNG, ®ve cold nodules patient was also investigated. All patients lived in an from patients with non-toxic multinodular goiter (MNG) area of moderate±severe iodine de®ciency. Informed who had detectable levels of TSH were also investigated. consent was obtained from all patients. A map of the nodules, as they appeared at scintiscan and ultrasound Materials and methods examination, was provided for the surgeon, so that nodules de®ned by scintiscan could be unequivocally Materials identi®ed post-operatively. Tissue specimens, obtained at the time of surgery, were frozen in liquid nitrogen. HEPES, sucrose, dithiothreitol (DTT), EDTA, phenyl- Toxic multinodular goiter was diagnosed on the basis of methylsulfonyl ¯uoride (PMSF), aprotinin, leupeptin clinical thyrotoxicosis, elevated serum levels of free and anti-human beta-actin monoclonal antibody were thyroid hormones, and undetectable TSH levels, asso- obtained from Sigma-Aldrich S.r.l. (Milan, Italy); ciated with a multinodular goiter presenting predomi- nitrocellulose membranes, horseradish peroxidase con- nant 131I uptake in one (or more) nodule(s) as shown jugated anti-rabbit or anti-mouse antibodies, auto- by 131I scanning. The thyroid nodules were classi®ed as radiography ®lm and enhanced chemiluminescence hot or cold according to the scintigraphic examination. (ECL-Plus) Western blotting detection reagents were All nodules were histologically classi®ed as adenomas from Amersham Pharmacia biotech (Milan, Italy); according to the WHO recommendations (13). The protein assay reagent, non-fat dry milk and molecular clinical features of the patients are shown in Table 1. weight markers were from Bio-Rad Laboratories This study was approved by the local ethical committee. (Segrate, Milan, Italy). Protein extraction and Western blot Patients Total proteins were extracted from thyroid tissues as A total of 25 nodules from ten unrelated patients with follows. Frozen tissue was thawed and homogenized in thyroid TMNG were studied. Five nodular tissues from 1 ml of buffer containing 250 mmol/l sucrose, Table 1 Clinical features of the patients with TMNG. Hormonal values were obtained at the time of diagnosis, when 131I scintiscan was performed. The patients were then treated with anti-thyroid drugs and operated on when a state of subclinical hyperthyroidism was reached. Normal values: TSH: 0.35±5.5; FT3 2.3±4.2; FT4: 0.8±1.5. 131 No. Sex/Age TSH (mIU/l) FT3 (pg/ml) FT4 (ng/dl) Nodules Diameter (mm) Ultra-sonography I scintiscan 1 F/51 0.01 4.38 1.5 1a 20 Mixed Cold 1b 25 Mixed Hot 1c 18 Mixed Hot 2 F/65 0.01 4.65 1.76 2a 15 Mixed Cold 2b 18 Mixed Cold 2c 20 Mixed Hot 3 F/60 0.01 4.1 1.2 3a 16 Mixed Cold 3b 18 Mixed Hot 4 F/81 0.1 3.51 1.01 4a 13 Mixed Cold 4b 13 Hypoechoic Hot 5 M/62 0.01 3.92 1.05 5a 70 Hypoechoic Cold 5b 27 Hypoechoic Hot 6 M/71 0.01 3.58 0.8 6a 20 Isoechoic Cold 6b 15 Mixed Cold 6c 30 Hypoechoic Cold 6d 50 Isoechoic Hot 7 F/71 0.38 7.15 1.23 7a 16 Isoechoic Cold 7b 17 Mixed Hot 7c 17 Mixed Hot 8 M/75 1.8* 3.78* 0.8* 8a 50 Mixed Cold 8b 25 Mixed Hot 9 F/54 0.01 5.40 21.2 9a 24 Hypoechoic Cold 9b 56 Hypoechoic Hot 10 F/43 0.07 9.0 3.04 10a 40 Mixed Cold 10b 40 Isoechoic Hot *Values obtained during treatment with anti-thyroid drugs. Almost all patients showed only a T3 thyrotoxicosis, as expected in an area of severe iodine de®ciency. FT3, free tri-iodothyronine; FT4, free thyroxine. www.eje.org Downloaded from Bioscientifica.com at 10/02/2021 04:05:42AM via free access EUROPEAN JOURNAL OF ENDOCRINOLOGY (2001) 145 NIS and pendrin in toxic multinodular goiter 593 10 mmol/l HEPES-KOH (pH 7.5), 1 mmol/l EDTA, 1 mmol/l PMSF, 10 mg/ml leupeptin, 10 mg/ml apro- tinin. The homogenate was centrifuged at 14 000 g (4 8C for 15 min) and the supernatant (which con- tained the whole cell lysate) was quanti®ed spectro- photometrically using the Bradford method. Fifty micrograms of protein were loaded onto a 4±20% gradient SDS polyacrylamide gel and subjected to electrophoresis at a constant voltage (120 V). Figure 1 Western blot analysis under reduced conditions of protein Electroblotting to a Hybond-P ECL nitrocellulose extracts from tissues of two patients with Graves' disease, NIH 3T3 membrane was performed for 2 h at 125 mA using a ®broblasts and normal and nodular tissues from two patients, using Mini Trans blot electroblotting system. Blocking was a polyclonal anti-NIS antibody (upper panel) and a monoclonal done using TTBS/milk (TBS, 1% Tween 20 and 5% anti-human beta-actin antibody (lower panel) (see Materials and methods). A representative of three separate experiments is non-fat dry milk) for 2 h at room temperature. The shown. Lanes 1 and 2=positive control (Graves' tissue); lane membrane was then incubated with a 1/1000 dilution 3=negative control (®broblasts); lanes 4, 5, 6=non-nodular of af®nity puri®ed rabbit anti-NIS (6) or 1/1500 (normal), hypofunctioning (cold) nodule and hyperfunctioning (hot) dilution of anti-pendrin (8) polyclonal antibody or a nodule from a patient with TMNG (no.