TNFα-mediated activation of NF-κB downregulates sodium-iodide symporter expression in thyroid cells

RESEARCH ARTICLE TNFα-mediated activation of NF-κB downregulates sodium-iodide symporter expression in thyroid cells Ma ´rcia Faria 1,2,3☯ , Rita Domingues 1,4☯ , Francisca Paixão 1,4 , Maria João Bugalho 1,5 , Paulo Matos 2,3 , Ana Luı ´sa SilvaID 1,4 * 1Servic¸o de Endocrinologia, Diabetes e Metabolismo do CHULN-Hospital Santa Maria, Lisboa, Portugal, 2BioISI-Biosystems and Integrative Sciences Institute, Faculdade de Ciências da Universidade de Lisboa, Lisboa, Portugal, 3Departamento de Gene ´tica Humana, Instituto Nacional de Sau ´de Doutor Ricardo Jorge, Lisboa, Portugal, 4ISAMB-Instituto de Sau ´de Ambiental, Faculdade de Medicina da Universidade de Lisboa, Lisboa, Portugal, 5Faculdade de Medicina da Universidade de Lisboa, Lisboa, Portugal ☯These authors contributed equally to this work. *[email protected] Abstract The sodium-iodide symporter (NIS) mediates transport of iodide across the basolateral membrane of thyroid cells. NIS expression in thyroid cancer (TC) cells allows the use of radioactive iodine (RAI) as a diagnostic and therapeutic tool, being RAI therapy the systemic treatment of choice for metastatic disease. Still, a significant proportion of patients with advanced TC lose the ability to respond to RAI therapy and no effective alternative therapies are available. Defective NIS expression is the main reason for impaired iodide uptake in TC and NIS downregulation has been associated with several pathways linked to malignant transformation. NF-κB signaling is one of the pathways associated with TC. Interestingly, NIS expression can be negatively regulated by TNF-α, a bona fide activator of NF-κB with a central role in thyroid autoimmunity. This prompted us to clarify NF-kB’s role in this process. We confirmed that TNF-αleads to downregulation of TSH-induced NIS expression in non-neoplastic thyroid follicular cell-derived models. Notably, a similar effect was observed when NF-κB activation was triggered independently of ligand-receptor specificity, using phorbol-myristate-acetate (PMA). TNF-αand PMA downregulation of NIS expression was reverted when NF-κB-dependent transcription was blocked, demonstrating the requirement for NF-kB activity. Additionally, TNF-αand PMA were shown to have a negative impact on TSH-induced iodide uptake, consistent with the observed transcriptional downregulation of NIS. Our data support the involvement of NF-κB-directed transcription in the modulation of NIS expression, where upor down-regulation of NIS depends on the combined output to NF-κB of several converging pathways. A better understanding of the mechanisms underlying NIS expression in the context of normal thyroid physiology may guide the development of pharmacological strategies to increase the efficiency of iodide uptake. Such strategies would be extremely useful in improving the response to RAI therapy in refractory-TC. PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 1 / 11 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Citation: Faria M, Domingues R, Paixão F, Bugalho MJ, Matos P, Silva AL (2020) TNFα-mediated activation of NF-κB downregulates sodium-iodide symporter expression in thyroid cells. PLoS ONE 15(2): e0228794. https://doi.org/10.1371/journal. pone.0228794 Editor: Salvatore Papa, University of Leeds, Faculty of Medicine and Health, UNITED KINGDOM Received: November 6, 2019 Accepted: January 22, 2020 Published: February 12, 2020 Copyright: ©2020 Faria et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files. Funding: This work was supported by the Fundac¸ão para a Ciência e a Tecnologia, grant [PTDC/BIAMOL/31787/2017] from Portugal. MF is recipient of FCT fellowship PD/BD/114388/2016. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction The well-differentiated thyroid carcinomas (DTCs) arise from thyroid follicular cells and represent the most frequent forms of thyroid cancer (TC), including the papillary thyroid cancer (PTC) and follicular thyroid cancer (FTC) subtypes [1]. The majority of DTCs are associated with a favorable prognosis. However, about 30% of patients with advanced forms of DTC become resistant to radioactive iodine (RAI) therapy, the standard treatment for metastatic disease [2]. The lack of efficient therapeutic options alternative to RAI makes the clinical management of these patients challenging, reducing the 10-year survival rate from approximately 90% to 10% [2,3]. The main reason for impaired iodide uptake in refractory-TC is the defective functional expression of the sodium iodide symporter (NIS) [4,5]. NIS belongs to the human solute carrier (SLC) family of transporters, is highly expressed at the basolateral membrane of thyroid follicular cells and is responsible for the active transport of iodide across the plasma membrane into thyroid follicles [6]. The primary regulator of NIS expression in thyroid gland is the thyroid stimulating hormone (TSH) [7,8]. TSH-induced accumulation of cyclic AMP leads to the binding of the PAX8 transcription factor to the NIS upstream enhancer (NUE) element on the NIS gene promoter, a primary requirement for the full activation of NIS expression [9,10]. Despite TSHderived signaling being the key regulator of NIS expression in thyroid tissue, other signaling pathways may have an impact on this process. NIS expression levels and iodine uptake in DTC are reduced when compared to normal tissue [11,12] and this downregulation has been associated with the overactivation of several pathways linked to thyroid malignancy [13]. NF-κB signaling has been implicated in cancer-associated processes of several human malignancies, including thyroid cancer. Increased NF-kB activation has been described in PTC, FTC and anaplastic TC, as being associated with resistance to apoptosis and maintenance of the malignant phenotype [14–16]. Also, in previous studies, we have shown that overexpression of tumor-related RAC1b, a highly activated splice variant of the GTPase RAC1 [17,18], has a significant role in PTC tumorigenesis by inducing resistance to programmed cell death through NF-κB activation [19]. The NF-κB pathway is also responsible for controlling several aspects of cell growth and inflammation [20]. One major pathway responsible for NF-κB activation is the canonical NF-κB pathway, which involves preferentially the heterodimer p65/p50 and is triggered in response to numerous stimuli including pro-inflammatory cytokines such as tumor-necrosisfactor-α(TNF-α) and bacterial lipopolysaccharide (LPS) [20–22]. The understanding of the mechanisms underlying NIS expression in the perspective of normal thyroid physiology may guide the development of strategies to enhance the efficiency of iodide uptake, particularly in the neoplastic context. In fact, in addition to its role in TC, NF-κB has also been implicated in normal thyroid physiology, being necessary for normal thyroid structure development and function, and involved in the expression of several thyroid specific genes, such as PAX8, Tg, TTF-1, TPO and also NIS [23,24]. Indeed, it was shown that the p65 subunit of NF-κB acts in synergy with the transcription factor PAX8 for the promotion of NIS expression, as the result of LPS stimulation [25]. Conversely, despite being widely recognized as a potent activator of the canonical NF-κB pathway [26], TNF-αwas shown to have a negative impact in NIS expression [27]. This is consistent with the observed reduction in RAI uptake in DTCs with concurrent thyroiditis [28], as TNF-αis a central factor in thyroid autoimmunity. This led us to raise the question of whether this negative impact of TNF-αon NIS expression involves activation of NF-κB. TSH-induced NIS expression is hampered by TNF-α-mediated NF-kB activation PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 2 / 11 Competing interests: The authors have declared that no competing interests exist. Materials and methods Cell lines and culture conditions The TSH-responsive cell lines PCCL3 and FRTL5, derived from Fischer rat’s thyroid follicular normal epithelium, were maintained in Coon’s F-12 modified liquid medium (Merck) supplemented with 5% fetal bovine serum (FBS) (Gibco), 10 μg/mL of insulin (SIGMA), 5 μg/mL of Apo-Transferrin (Apo-T) (SIGMA) and 0.1 mU/mL of TSH (SIGMA). Cells were maintained at 37˚C in a 5% humidified CO2 environment and discarded after 20 passages. When appropriate, cells were cultured in starvation medium (F12 Coon’s Modification medium supplemented with 0.2% (v/v) FBS and 5 μg/mL of Apo-T) or TSH-medium (starvation medium supplemented with 1 mU/mL of TSH). Stimulation with TNF-α(10 ng/mL, SIGMA) or phorbol 12-myristate 13-acetate (PMA; 10 ng/mL; SIGMA) was performed (60 minutes to 24 hours) in serum-starved cells for 72h, subsequently subjected to TSH stimulation for 24 hours (TSHmedium). When applicable, cells were treated with the NF-kB selective inhibitor BMS-345541 (10 μM, SIGMA) for 6 hours. RNA extraction, cDNA synthesis and RT-qPCR Total RNA was obtained from cultured cells using the ready-to-use reagent TripleXtractor (GRiSP Research Solutions) following manufacturer’s instructions. cDNA was synthetized from 2 μg of RNA using random primers and RevertAid Reverse Transcriptase (Thermo Scientific), following the manufacturer’s protocol. NIS expression levels were quantified by SYBR Green based quantitative reverse transcription polymerase chain reaction (RT-qPCR) on the Light Cycler 480II (Roche), using Xpert Fast SYBR (Grisp). HPRT1 was used as endogenous control gene. NIS levels were normalized to endogenous control expression level. NIS normalized values were then expressed relative to that of a reference sample (pool of TSH stimulated rat cell lines). Expression values correspond to arbitrary units representing fold differences relative to the reference sample. Primers used (synthesized by NZytech, Portugal) were the following: NIS F (5’-TCCTCACAGGCCGTATC TCA) and NIS R (5’–GAAGGAACCCTGGAGGACAC); HPRT1 F (5’-GCTGAAGATTTGGAA AAGGTG) and HPRT1 R (5’-AATCCAGCAGGTCAGCAAAG). IKBαexpression levels were quantified through the same method as NIS using IKBαthe specific primers IKBαF (5’-TCC TCACAGGCCGTATCTCA) and IKBαR (5’-GACACGTGTGGCCGTTGTAG). HS-YFP–based iodide influx assay Iodide influx was accessed in PCCL3 cells stably expressing the halide-sensitive yellow fluorescent protein (HS-YFP-H148Q/I152L - Y-PCCL3 cells) [29,30]. Cells were seeded in 8-well chamber slides (Ibidi) and serum-starved for 24h. Subsequently, cells were stimulated for 96h with 1 mU/mL TSH and then treated with TNF-α(10 ng/mL, 24h) or PMA (10 ng/mL, 24h). Cells were washed with PBS and then incubated for 15 min at 37˚C with influx-solution (137 mM NaCl, 2.7 mM KCl, 0.7 mM CaCl2, 1.1 mM MgCl2, 1.5 mM KH2PO4, 8.1 mM Na2HPO4, and 10 mM glucose, pH 7.4). Each well was assayed individually for iodide influx by recording fluorescence continuously in a Leica TCS-SPE confocal microscope after addition of isomolar PBS-NaI solution (1 mM final NaI concentration). Decay of YFP fluorescence was followed for 600 seconds, acquiring an image every 10s. To confirm that the iodine influx observed was specifically mediated by NIS, additional assays were performed in the presence of ClO4- (a competitive inhibitor of iodide uptake by NIS), in which cells were incubated with 1mM ClO4-, for 10min, before PBS-NaI solution was added. Images were stacked and analyzed with Image J, defining whole field regions of interest (ROIs) for pixel intensity measurements that excluded TSH-induced NIS expression is hampered by TNF-α-mediated NF-kB activation PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 3 / 11 saturated cells. Cell fluorescence recordings were normalized for the initial average value measured before addition of I−. Total protein lysates and western blot. Y-PCCl3 cells were serum-starved for 24h, stimulated for 96h with 1 mU/mL TSH and then treated with TNF-α(10 ng/mL, 24h) or PMA (10 ng/mL, 24h). Protein extracts were prepared in Laemmli sample buffer (supplemented with 1 U/μl Benzonase, Sigma-Aldrich) and resolved, according to standard protocols, in 12% SDS-PAGE and transferred to PVDF membranes (Bio-Rad). The primary antibody rabbit polyclonal anti-NIS (Proteintech) was used in Western blot at 1:1000. Antibody mouse monoclonal anti-PCNA (Merck) was used at 1:10000. Detection was carried out using secondary peroxidaseconjugated anti-rabbit or anti-mouse IgG (Bio-Rad) antibodies followed by chemiluminescence. Data and statistical analysis The data used to support the findings of this study are included within the article. Statistical analysis was carried out using GraphPad Prism statistical software (San Diego, CA). Quantitative results are shown as means ±SD of at least three independent observations. Statistical comparisons of data sets were made using unpaired two-tailed Student’s t-test and statistical significance was accepted when pvalues <0.05. In the halide-sensitive YFP-based functional assay, the signal decay caused by YFP fluorescence quenching was fitted to an exponential decay function to derive the maximal slope that corresponds to initial influx of I − into the cells. Maximal slopes were converted into rates of variation of the intracellular I − concentration (in nmol/min) using the equation d[I − ]/dt = K d [d(F/F 0 )/dt], where K d is the affinity constant of YFP for I − , and F/F 0 is the ratio of the cell fluorescence at a given time versus the initial fluorescence [31,32]. Results and discussion TNF-αis a pro-inflammatory cytokine widely defined as a potent activator of the canonical NF-κB pathway. TNF-αhas been previously reported to induce a negative impact on NIS expression by downregulating NIS mRNA levels up to 70% in FRTL5 rat thyroid cells [27]. However, whether this effect was dependent of NF-κB activation remained elusive. NIS expression levels are typically reduced in malignant thyroid tissue relative to normal tissue [11,12]. This results from multiple mechanisms elicited by several signaling pathways involved in thyroid tumorigenesis [33]. Consistently, most of the DTC-derived cell lines available fail to express thyroid-specific genes, including NIS and the TSH receptor [34], which hampers the study of TSH-mediated regulation on NIS expression in these neoplastic cellular models (S1 Fig). Nevertheless, the study of the mechanisms underlying NIS functional regulation using non-neoplastic, TSH-responsive, thyroid cell models have been shown to allow the identification of key regulatory events that can be further translated into the malignant context [35–38]. In this study, we started by assessing TNF-αimpact on TSH-induced NIS expression using the PCCL3 cell line as an additional cellular model. This, like the FRTL5 cell line, is also a non-neoplastic rat thyroid follicular cell-derived model representative of TSH-responsive system for NIS expression and iodide uptake [8]. Serum-starved cells were subjected to TSH stimulation and then treated with TNF-αfor 1h or 24h. The effects on NIS mRNA levels were accessed by qRTPCR (Fig 1A). We confirmed that TNF-αled to downregulation of TSH-induced NIS expression in FRLT5. This effect was also observed in PCCL3 cells, reaching a statistically significant decrease with a 24h treatment (Fig 1A). Next, we asked whether NIS downregulation was a specific outcome of TNF-αstimulation or if it may result from stimulation with other NF-κB activators. In fact, in contrast to that observed with TNF-αtreatment, NIS levels were shown to increase upon stimulation of FRTL5 cells with LPS, another well-defined activator of the canonical NF-κB TSH-induced NIS expression is hampered by TNF-α-mediated NF-kB activation PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 4 / 11 pathway. This increase was actually shown to be a result of the synergistic action of PAX8 and the p65-NF-κB subunit in the NUE regulatory region [25]. This suggests that different triggers for NF-κB activation may result in opposite NIS expression outcomes. These may be a consequence of the concerted action between the signaling cues generated by ligand-receptor interaction and the NF-κB active dimers produced. Thus, we then tested the effect of NF-κB activation on NIS expression when this is triggered independently of ligand-receptor specificity by stimulating the cells with phorbol-myristate-acetate (PMA). This has been shown to induce canonical NF-κBdependent transcription by acting through the direct activation of protein-kinase-C [39]. We observed that, similarly to TNF-α, PMA downregulates NIS mRNA expression in both PCCL3 and FRTL5 cells (Fig 1B). It should also be noted that, albeit PMA has been shown to affect NIS protein acting as a downregulator of NIS function and iodine uptake in human breast cells [39,40], no effects at NIS transcriptional control level have been reported so far. Our next step was to clarify whether this negative impact on NIS expression was in fact dependent on the NF-κB-mediated transcriptional activity. Thus, we assessed the effect of TNF-αand PMA in the presence of BMS-345541, an allosteric inhibitor of the IkB Kinase (IKK) complex [41] whose activity is mandatory for canonical NF-κB activation [20]. We Fig 1. Effect of TNF-αand PMA on NIS transcriptional expression in non-neoplastic, TSH-responsive thyroid follicular cell lines. NIS mRNA levels were assessed by RT-qPCR and correspond to arbitrary units representing fold differences relative to a reference sample, corrected to HPRT levels used as endogenous control gene. Impact on NIS transcript levels of TNFα(A) and PMA (B) stimuli. FRTL5 and PCCL3 were subjected to a 96h starvation period followed by stimulation with TSH (1 mU/mL for 24h), in the presence of either TNFα(10 ng/mL; 1h and 24h) or PMA (10 ng/mL; 1h and 24h). As controls, non-stimulated cells and cells stimulated with TSH for 24h in the absence of TNFαand PMA were included. Plotted values are the mean ±SD (error bars) of three independent assays. Comparisons of TNF-αand PMA stimulation to non-stimulated conditions were made using two-tailed Student’s t-tests (�p�0.05; ��p�0.01). https://doi.org/10.1371/journal.pone.0228794.g001 TSH-induced NIS expression is hampered by TNF-α-mediated NF-kB activation PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 5 / 11 observed that, in the presence of BMS-345541, the decrease in NIS transcript levels induced by both TNF-αand PMA was reverted, with NIS mRNA reaching levels similar to those of TSH stimulation alone (Fig 2A). This observation is consistent with NF-κB activation being required for NIS downregulation. NF-κB activation by TNF-αand PMA was further monitored using IκBαmRNA levels as a readout of NF-κB transcriptional activity. In fact, one of the early events resulting from canonical NF-κB activation is the increase of IκB-αmRNA levels [20] and it has been previously shown that IκBαmRNA levels correlate robustly with the activation of NF-κB in several cellular models, making it a broadly effective readout of NF-κB activation [42]. A significant increase in IκB-α mRNA levels was observed upon stimulation of both PCCL3 and FRTL5 cells with TNF-αand PMA. Moreover, this increase was strongly inhibited by BMS-345541, indicating a specific response from the NF-κB canonical pathway (Fig 2B). Finally, we asked whether TNF-αand PMA would also have a negative impact on iodide uptake. To address this question, we established an iodide influx assay in PCCL3 cells, modified to stably express the halide-sensitive yellow fluorescent protein (YFP) F46L/H148Q/I152L mutant (HS-YFP) [17,31,43] which has been validated for NIS activity assessment [29,30]. The HS-YFP inside these modified PCCL3 cells (Y-PCCL3) is quenched by iodide, enabling changes in iodide intracellular concentration to be monitored in thyroid cells by live cell imaging. Y-PCCL3 cells were thus subjected to TSH stimulation in the presence of TNF-αand PMA. A nine-fold faster decay rate of HS-YFP fluorescence was observed upon TSH stimulation, compared to that detected in the absence of TSH (Fig 3A and 3B), consistent with TSH Fig 2. NF-κB impact on TNF-αand PMA-mediated downregulation of NIS transcriptional expression. NIS (A) and IKBα(B) mRNA levels were quantified by RT-qPCR. TSH-stimulated cells were treated for 24h with TNF-αand PMA either in the absence (vehicle) or presence of the NF-κB inhibitor BMS-345541 (10 μM; 6h). Plotted values are the mean ±SD (error bars) of three independent assays, compared with the group treated with only TSH. Comparisons were made using two-tailed Student’s t-tests (�p�0.05; ��p�0.01; ��� p�0.001). https://doi.org/10.1371/journal.pone.0228794.g002 TSH-induced NIS expression is hampered by TNF-α-mediated NF-kB activation PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 6 / 11 Fig 3. Effect of TNF-αand PMA on TSH-induced iodide uptake and NIS protein levels in Y-PCCL3 cells. Cells PCCL3 cells stably expressing the YFP-halide sensor were subjected to a 24h starvation period and then treated with TSH for 96h (TSH ctrl), in the presence or absence of either TNF-αor PMA Additional iodide influx assays were performed in the presence of both (24h treatment), and subjected to iodide influx assays and Western blot. TSH and ClO4- (1 mM, 10 min), a competitive inhibitor of iodide uptake by NIS. Representative HS-YFP fluorescence decay TSH-induced NIS expression is hampered by TNF-α-mediated NF-kB activation PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 7 / 11 enhancement of NIS expression that ultimately promotes NIS-mediated iodide uptake. We confirmed that the observed TSH-induced iodide influx was mediated by NIS since it was completely abolished in the presence of perchlorate (ClO4-) (Fig 3A and 3B), a competitive inhibitor of iodide uptake by NIS [44]. In addition, both TNF-αand PMA stimuli induced a meaningful decrease in TSH-induced iodide influx in Y-PCCL3 cells (Fig 3A and 3B). Consistent with the iodide influx rates and the observed downregulation of NIS expression at transcriptional level, a considerable reduction in NIS protein levels was also noted in the presence of both TNF-αand PMA (Fig 3C). Transcriptional and post-transcriptional regulation of NIS expression involves different molecular players and signaling pathways that are also implicated in tumor progression and aggressiveness. NF-κB family of transcription factors is responsible for regulating many cellular processes such as inflammatory responses, cellular growth and cell death, and has been also implicated in TC development. We have previously reported NF-kB activation as a key mechanism through which the tumor-related RAC1b exerts its oncogenic action in the context of PTC [19]. Additionally, we have also recently shown an inverse correlation between RAC1b overexpression and NIS levels in DTCs [45]. This prompted us to explore, in the present study, the impact of the NF-κB signaling on NIS expression. Our data support that the canonical NFκB pathway may also be involved in the downregulation of NIS. The interplay between the NF-κB activation and NIS expression, however, requires further clarification. This is particularly important when we take into account that the effects on NIS transcription induced by TNF-αand PMA are opposite to those described for LPS [25]. One possible explanation for these apparent antagonistic effects of NF-κB activation in NIS transcription may be the activation and interplay of additional pathways that concertedly modulate the affinity of NF-κB transcription factors for different promoters. Thus, NF-κB-driven transcription may lead to different results depending on the overall signaling involved in its activation and regulation. Understanding the mechanisms underlying physiological NIS functional regulation may allow the identification of regulatory events that can be pharmacological targeted to enhance the iodide uptake efficiency and improve the response to RAI therapy in RAI-refractory thyroid cancers. Supporting information S1 Fig. Effect of TSH on NIS transcriptional expression in thyroid follicular cell-derived cell lines. NIS mRNA levels were assessed by RT-qPCR and correspond to arbitrary units representing fold differences relative to a reference sample, corrected to endogenous control expression levels. HPRT1 and TBP were used as endogenous control genes for rat and human, respectively. RT-qPCR preformed as previously described [45]. The different cell lines were subjected to a 96h starvation period followed by stimulation with TSH (1 mU/mL for 48h). Plotted values are the mean ±SD (error bars) of three independent assays. (TIF) traces (A) recorded continuously for 600 seconds, acquiring an image every 10 s, after exposure to 1mM NaI (as described in [30]). Fluorescence (F) was plotted over time as percentage of fluorescence at time 0 (F0). Iodide influx rates (B) calculated by fitting the curves to the exponential decay function to derive the maximal slope that corresponds to initial influx of I − into the cells. Endogenous NIS protein expression upon TSH, TNF-αor PMA treatment was monitored Western Blot using anti-NIS primary antibody. Endogenous PCNA expression was used as loading control. Data are means ±SEM of three independent assays. Comparisons were made using one-tailed Student’s t-tests (�p�0.05; ��p�0.01; ���p�0.001). https://doi.org/10.1371/journal.pone.0228794.g003 TSH-induced NIS expression is hampered by TNF-α-mediated NF-kB activation PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 8 / 11 Author Contributions Conceptualization: Ana Luı ´sa Silva. Formal analysis: Ma ´rcia Faria, Ana Luı ´sa Silva. Funding acquisition: Paulo Matos, Ana Luı ´sa Silva. Investigation: Ma ´rcia Faria, Rita Domingues, Francisca Paixão. Methodology: Ma ´rcia Faria, Rita Domingues, Paulo Matos. Project administration: Ana Luı ´sa Silva. Resources: Maria João Bugalho, Paulo Matos, Ana Luı ´sa Silva. Supervision: Ana Luı ´sa Silva. Validation: Ma ´rcia Faria, Rita Domingues. Visualization: Ana Luı ´sa Silva. Writing – original draft: Ana Luı ´sa Silva. Writing – review & editing: Maria João Bugalho, Paulo Matos. References 1. Dralle H, Machens A, Basa J, Fatourechi V, Franceschi S, Hay ID, et al. Follicular cell-derived thyroid cancer. Nat Rev Dis Primer. 2015; 1: 15077. 2. Cooper DS, Doherty GM, Haugen BR, Kloos RT, Lee SL, Mandel SJ, et al. Management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid Off J Am Thyroid Assoc. 2006; 16: 109–142. 3. Schmidt A, Iglesias L, Klain M, Pitoia F, Schlumberger MJ. Radioactive iodine-refractory differentiated thyroid cancer: an uncommon but challenging situation. Arch Endocrinol Metab. 2017; 61: 81–89. https://doi.org/10.1590/2359-3997000000245 PMID: 28225999 4. Spitzweg C, Bible KC, Hofbauer LC, Morris JC. Advanced radioiodine-refractory differentiated thyroid cancer: the sodium iodide symporter and other emerging therapeutic targets. Lancet Diabetes Endocrinol. 2014; 2: 830–842. https://doi.org/10.1016/S2213-8587(14)70051-8 PMID: 24898835 5. Durante C, Haddy N, Baudin E, Leboulleux S, Hartl D, Travagli JP, et al. Long-term outcome of 444 patients with distant metastases from papillary and follicular thyroid carcinoma: benefits and limits of radioiodine therapy. J Clin Endocrinol Metab. 2006; 91: 2892–2899. https://doi.org/10.1210/jc.20052838 PMID: 16684830 6. Doha ´n O, Carrasco N. Advances in Na+/I−symporter (NIS) research in the thyroid and beyond. Mol Cell Endocrinol. 2003; 213: 59–70. https://doi.org/10.1016/j.mce.2003.10.059 PMID: 15062574 7. Ravera S, Reyna-Neyra A, Ferrandino G, Amzel LM, Carrasco N. The Sodium/Iodide Symporter (NIS): Molecular Physiology and Preclinical and Clinical Applications. Annu Rev Physiol. 2017; 79: 261–289. https://doi.org/10.1146/annurev-physiol-022516-034125 PMID: 28192058 8. Kogai T, Endo T, Saito T, Miyazaki A, Kawaguchi A, Onaya T. Regulation by thyroid-stimulating hormone of sodium/iodide symporter gene expression and protein levels in FRTL-5 cells. Endocrinology. 1997; 138: 2227–2232. https://doi.org/10.1210/endo.138.6.5189 PMID: 9165005 9. Kogai T, Brent GA. The sodium iodide symporter (NIS): regulation and approaches to targeting for cancer therapeutics. Pharmacol Ther. 2012; 135: 355–370. https://doi.org/10.1016/j.pharmthera.2012.06. 007 PMID: 22750642 10. Taki K, Kogai T, Kanamoto Y, Hershman JM, Brent GA. A thyroid-specific far-upstream enhancer in the human sodium/iodide symporter gene requires Pax-8 binding and cyclic adenosine 3’,5’-monophosphate response element-like sequence binding proteins for full activity and is differentially regulated in normal and thyroid cancer cells. Mol Endocrinol Baltim Md. 2002; 16: 2266–2282. https://doi.org/10. 1210/me.2002-0109 PMID: 12351692 11. Lazar V, Bidart JM, Caillou B, Mahe ´C, Lacroix L, Filetti S, et al. Expression of the Na+/Isymporter gene in human thyroid tumors: a comparison study with other thyroid-specific genes. J Clin Endocrinol Metab. 1999; 84: 3228–3234. https://doi.org/10.1210/jcem.84.9.5996 PMID: 10487692 TSH-induced NIS expression is hampered by TNF-α-mediated NF-kB activation PLOS ONE | https://doi.org/10.1371/journal.pone.0228794 February 12, 2020 9 / 11