Inhibition, Na+/I- symporter (NIS) leads to Thyroidal Iodide, Decreased

NIS is a membrane bound glycoprotein and its main physiological function is to transport one iodide ion along with two sodium ions across the basolateral membrane of thyroid follicular cells. It uses the sodium gradient generated by the Na+/K+ ATPase for the active transport of iodide into the thyrocytes (Eskandari et al., 1997). Extensive studies on NIS protein have identified 14 different mutations and each one of them is related to Iodine Transport Deficiencies (ITD) (reviewed in Spitzweg and Morris, 2010). Most of these mutations have been characterized and it is well known that they even lead to the synthesis of truncated protein (Pohlenz et al., 1997; Pohlenz et al., 1998), partial deletions (Kosugi et al., 2002; Tonacchera et al., 2003; Montanelli et al., 2009) or substitutions of amino acids (Matsuda and Kosugi, 1997; Kosugi et al., 1999; Szinnai et al., 2006) that eventually result in total or partial NIS dysfunction. While most of the NIS mutants have been further investigated and the functional relationship between the NIS dysfunction and ITD is well established (reviewed in Darrouzet et al., 2014; Portulano et al., 2014), the exact structural relationship between mutated NIS and ITD still needs to be elucidated and the molecular modelling of the protein would greatly benefit these studies. At the same time, causative link between  iodide deficiency, thyroid hormones, and neurodevelopment deffects is well documented (Gilbert et al., 2009).

Recent revision of the affinity constant for perchlorate binding to the NIS symporter based on in vitro and human in vivo data, performed by refitting published in vitro data, in which perchlorate-induced inhibition of iodide uptake via the NIS was measured, yielding a Michaelis-Menten kinetic constant (K_m) of 1.5 μm, showed that a 60% lower value for the Km, equal to 0.59 μm. Substituting this value into the PBPK model for an average adult human significantly improved model agreement with the human RAIU data for exposures <100 μg kg^-1 day^-1 (Schlosser PM, 2016).

The effects of maternal hypothyroidism could also contribute to this KER.  During pregnancy TH requirements increase, particularly during the first trimester (Alexander et al. 2004; Leung et al. 2010), due to higher concentrations of thyroxine-binding globulin, placental T4 inner-ring deiodination leading to the inactive reverse T3 (rT3), and transfer of small amounts of T4 to the foetus (during the first trimester foetal thyroid function is absent). Moreover, glomerular filtration rate and clearance of proteins and other molecules are both increased during pregnancy, possibly causing increased renal iodide clearance and a decreased of circulating plasma iodine (Glinoer, 1997). Thus, even though the foetal thyroid can trap iodide by about 12 week of gestation (Fisher and Klein, 1981), high concentrations of maternal perchlorate may potentially decrease thyroidal iodine available to the foetus by inhibiting placental NIS (Leung et al. 2010).

Consequences of TH deficiency depend on the developmental timing of the deficiency (Zoeller and Rovet, 2004). For instance, if the TH deficiency occurs during early pregnancy, offspring show visual attention, visual processing and gross motor skills deficits, while if it occurs later, offspring may show subnormal visual and visuospatial skills, along with slower response speeds and motor deficits. If TH insufficiency occurs after birth, language and memory skills are most predominantly affected (Zoeller and Rovet, 2004).

There are limited data regarding low-level environmental perchlorate exposure and maternal thyroid function during pregnancy. A Chilean study found no increases in TSH or decreases in free thyroxine or urinary iodine concentrations in pregnant women living in three areas (all of which had more than adequate mean urinary iodine levels) with long-term environmental perchlorate exposure (Téllez Téllez et al. 2005). A follow-up analysis of this cohort also confirmed the lack of association between individual urinary iodide or perchlorate concentrations and thyroid function in the pregnant women (Gibbs and Van Landingham, 2008). Studies of large cohorts of first-trimester pregnant women from the U.S., Europe and Argentina found that environmental perchlorate exposure did not affect maternal thyroid function (Pearce et al. 2009).

Many studies have shown inhibition of radioactive iodide uptake by using different cell models and assays. However, there have been identified only few specific NIS inhibitors up to date, while all the others are thought to act through different inhibitory mechanisms. Monovalent anions, others than iodide, are also transported by NIS but Nitrate (NO3-), thiocyanate (SCN-), perchlorate (ClO-4), dysidenin and aryltrifluoroborates are of particular dietary and environmental importance (Jones et al., 1996; Tonacchera et al., 2004; De Groef et al., 2006).

Recent revision of the affinity constant for perchlorate binding to the NIS symporter based on in vitro and human in vivo data, performed by refitting published in vitro data, in which perchlorate-induced inhibition of iodide uptake via the NIS was measured, yielding a Michaelis-Menten kinetic constant (K_m) of 1.5 μm, showed that a 60% lower value for the Km, equal to 0.59 μm. Substituting this value into the PBPK model for an average adult human significantly improved model agreement with the human RAIU data for exposures <100 μg kg^-1 day^-1 (Schlosser PM, 2016).

There are many studies showing the effect of inhibition of NIS on thyroidal iodide uptake:

- Cianchetta et al., 2010 For this study the rat FRTL5 thyroid cell line endogenously expressing NIS, and the monkey kidney fibroblast-like cells (COS-7) transfected with hNIS were used. NIS functionality was assessed with the use of the Yellow Fluorescent Protein (YFP) variant YFP-H148Q/I152L, a genetically encodable biosensor of intracellular perchlorate concentration monitored by real-time fluorescence microscopy. Decrease of YFP-H148Q/I152L fluorescence in FRTL-5 cells occurs as a result of NIS-mediated uptake and binding to the intracellular fluorochrome (Rhoden et al., 2007). The biosensor was used to compare the kinetics of iodide and perchlorate transport by NIS, and to assess the ability of perchlorate to inhibit iodide transport. Additionally, perchlorate was shown to inhibit NIS function (competitive inhibition) by preventing iodide-induced changes in fluorescence of FRTL5 cells. Perchlorate caused a concentration-dependent inhibition of iodide uptake in the initial influx rate (IC50=1.6μM) and in the intracellular concentration of iodide (IC50=1.1μM). Also, both perchlorate and iodide (1–1000 μM) induced concentration-dependent decreases in YFP-H148Q/I152L fluorescence in COS-7 cells expressing hNIS, but had no effect (< 2%) in COS-7 cells lacking hNIS. Additionally, perchlorate induced a significantly smaller decrease in fluorescence (10.6% at 1 mM) than iodide (31.8% at 1 mM iodide). Thus, it was confirmed that the reduction of fluorescence was due to NIS-mediated anion transport into the cells, excluding non-specific effects.

- Tonacchera et al., 2004 Chinese hamster ovary (CHO) cell line had been stably transfected with human NIS and the measurement of iodide uptake was performed with the use of radioactive iodide uptake (RAIU) method. It was shown that the inhibition of iodide uptake was dose-dependent when using the known NIS inhibitors (ClO-4, NO3-, SCN-). Additionally, unlabeled I- (non ¹²⁵I) was used to investigate the inhibition level of radioiodide uptake and to compare it with the potency of the other monoions, which are known NIS inhibitors. The IC₅₀ values for the studied monoions were the following: ClO₄^-: IC₅₀ was 1.22 μΜ; SCN^-: IC₅₀ was 18.7 μΜ; NO₃^-: IC₅₀ was 293 μΜ; I^-: IC₅₀ was 36.6 μΜ. Finally, the present study investigated the joint effects of simultaneous exposure to multiple RAIU inhibitors, by generating multiple dose-response curves in the presence of fixed concentrations of inhibitors. The results of those experiments indicated a competition between the four anions with similar size for access to the binding sites of the NIS. The prediction model developed in this study, actually suggests that thyroidal iodide uptake is approximately proportional to iodide nutrition for any fixed inhibitor concentration, answering to the question whether dietary iodide can modulate the inhibitory effects of the known environmental goitrogens.

- Waltz et al., 2010 Measurement of iodide uptake was performed with a non-radioactive method. By using the rat thyroid low –serum 5 (FRTL5) cells, which endogenously express NIS, a spectrophotometric assay was developed and the iodide accumulation was determined based on the catalytic reduction of yellow cerium to colorless cerium in the presence of arsenious acid (Sandell-Kolthoff reaction). A dose-dependent inhibition of iodide uptake was shown. The IC50 values for the studied compounds were the following: Sodium perchlorate (NaClO4): IC50 was 0.1 μΜ Sodium thiocyanate (NaSCN): IC50 was 12 μΜ Sodium nitrate (NaNO3): IC50 was 800 μΜ Sodium Tetrafluoroborate (NaBF4): IC50 was 1.2 μΜ

- Lecat-Guillet et al., 2007; 2008a A fully automated radioiodide uptake assay was developed and some known NIS inhibitors were tested. A dose-dependent inhibition of iodide uptake was shown. The IC₅₀ values for the studied compounds were the following: Sodium perchlorate (NaClO₄): IC₅₀ was 1 μΜ; Sodium thiocyanate (NaSCN): IC₅₀ was 14 μΜ; Sodium nitrate (NaNO₃): IC₅₀ was 250 μΜ; Sodium Tetrafluoroborate (NaBF₄): IC₅₀ was 0.75 μΜ.  Additionally, a library of 17020 compounds was screened for the identification of new human NIS inhibitors. The identification was based on the magnitude of changes in iodide uptake using Human Embryonic Kidney 293 (HEK293) cells, stably transfected with the hNIS. The same experiments and with similar results were also performed in rat thyroid derived cells (FRTL5), which endogenously express NIS. Compounds that inhibited iodide uptake in a time-dependent manner were considered to act through direct NIS inhibition. In contrast, those compounds that had a delayed effect on iodide uptake were thought to act through a sodium gradient disruption system resulting in indirect inhibition of iodide transport. Perchlorate was used as a positive control in these experiments and, as expected, it blocked iodide uptake immediately and totally throughout the experiment. Dysidenin was also used as a control and the IC50 value identified was 2 μΜ. All the compounds that were used for these experiments were small drug-like molecules that have not been detected in the environment and they were named as ITBs (Iodide Transport Blockers).

- Lecat-Guillet et al., 2008b With the same fully automated radioiodide uptake assay, as described above, new NIS inhibitors were also identified. The organotrifluoroborate (BF3−) was found to inhibit iodide uptake with an IC50 value of 0.4 μM using rat-derived thyroid cells (FRTL5). The biological activity is rationalized by the presence of the ion BF3− as a minimal binding motif for substrate recognition at the iodide binding site.

- Lindenthal et al., 2009 With the use of a patch-clamp technique an analysis of the NIS inhibitors identified by Lecat-Guillet et al., 2008 (named ITB-1 to ITB-10 for "Iodide Transport Blockers") was evaluated in Xenopus oocytes expressing NIS to further assess the inhibitory effect of those molecules specifically on NIS activity. Four of those molecules (ITB-3, ITB-9, ITB-5 and ITB-4) were identified as the most potent, non-competitive NIS inhibitors. The effects of dysidenin were also analyzed with the same technique, as it had been reported to be a specific inhibitor of NIS (Vroye et al., 1998). It was found that dysidenin (50 μM) induced a rapid and reversible inhibition of the iodide (about 40%) of induced current in mNIS-expressing oocytes, but did not evoke any currents in the absence of iodide, suggesting that this effect was due to the inhibition of NIS activity.

- Greer et al., 2002 In human studies, potassium perchlorate was used to predict inhibition of thyroidal iodide uptake by applying the RAIU method. Greer et al., tested body weight adjusted doses of potassium perchlorate and an assessment of RAIU uptake was performed on day 2 and day 14 of treatment and 24 h following treatment termination (on day 15). The NOEL value for inhibition of thyroidal uptake was 0.007 mg/kg-day, while the true NEL value was estimated to be 0.0052 and 0.0064 mg/kg-day. According to the dose-response inhibition of iodide uptake the maximum percentage of iodide inhibition at the doses of 0.0052 and 0.0064 mg/kg-day is 8.3-9.5%, which is physiologically insignificant for a person with dietary sufficient iodine intake.

- Wen et al., 2016 By using human MCF-7 cells, a breast adenocarcinoma cell line, which express inducible NIS in the presence of all-trans retinoic acid (ATRA) it has been shown that inhibition of sterol regulatory element-binding proteins (SREBP) maturation by treatment with 25-hydroxycholesterol (5 µM) for 48 hr reduced ATRA (1 µM)-induced mRNA concentration of NIS and decreased iodide uptake by approximately 20%. This study showed for the first time that the NIS gene and iodide uptake are regulated by SREBP in cultured human mammary epithelial cells.

- Arriagada et al. 2015 This study showed that 2 hr or 5 hr exposure to excess I^- (100 μM) respectively in FRTL-5 cells and in ex-vivo rat thyroid gland (removed after single in vivo i.p. injection of 100 μg of I⁻ in 500 μL of distilled water, and analysis of ¹²⁵I thyroid uptake), induced inhibition of I- uptake through the NIS (~ 30% uptake inhibition after 5 hr in vivo), a process known as the Wolff-Chaikoff effect, which was not associated with a decrease of NIS expression or a change in NIS localization. Incubation of FRTL-5 cells with excess I- for 2 hr increased hydrogen peroxide generation. Also incubation with hydrogen peroxide (100 μM) decreased NIS-mediated I- transport, effect that was reverted by ROS scavengers.

The thyroid system is quite complex and therefore some inconsistent results have been produced by recent studies. For example, it has been observed in healthy volunteers that a 6-month exposure to perchlorate at doses up to 3 mg/d (low doses) had no effect on thyroid function, including inhibition of thyroid iodide uptake as well as serum levels of thyroid hormones, TSH, and Tg (Braverman et al., 2006). However, this study was limited by the small sample size and is obviously underpowered.

The review by Charnley (2008) examines a number of studies where association between perchlorate environmental (low) exposure and thyroid effects were analysed and many inconsistent conclusions have been drawn. For instance, no correlations were found between TH serum levels and urinary iodine concentrations among women exposed to perchlorate participating in the 2000-2001 National Health and Nutrition Examination Survey (NHANES). Available evidence does not support a causal relationship between changes in TH levels and current environmental levels of perchlorate exposure, but does support the conclusion that the US Environmental Protection Agency's reference dose (RfD) for perchlorate is conservatively health-protective. However, potential perchlorate risks are unlikely to be distinguishable from the ubiquitous background of naturally occurring substances present at much higher exposures that can affect the thyroid via the same biological mode of action as perchlorate, such as nitrate and thiocyanate. Therefore, risk management approaches that account for both aggregate and cumulative exposures and that consider the larger public health context in which exposures are occurring are desirable.

Additionally, a more comprehensive study by Pearce et al. (2010) conducted during 2002-2006 on 22,000 women at less than 16-week gestation showed that while low-level perchlorate exposure was ubiquitous in these women (with a median urinary perchlorate concentration of 5 µg/liter in the Turin cohort and 2 µg/liter in the Cardiff cohort), no associations between urine perchlorate concentrations and serum TSH or free T4 in the individual euthyroid or hypothyroid/hypothyroxinemic cohorts were found.

The data assessing the effect of maternal perchlorate exposure in neonates and children and thyroid function remain unclear (Leung et al., 2010).

Decreased iodine intake can decrease TH production, and therefore exposure to perchlorate might be particularly detrimental in iodine-deficient individuals (Leung et al. 2010). Moreover, biologically based dose-response modeling of the relationships among iodide status (e.g., dietary iodine levels), perchlorate dose, and TH production in pregnant women has shown that iodide intake has a profound effect on the likelihood that exposure to goitrogens will produce hypothyroxinemia (Lewandowski et al. 2015).

Consequences of TH deficiency depend on the developmental timing of the deficiency (Zoeller and Rovet, 2004). For instance, if the TH deficiency occurs during early pregnancy, offspring show problems in visual attention, visual processing and gross motor skills, while if it occurs later, offspring may show subnormal visual and visuospatial skills, slower response speeds and motor deficits. If TH insufficiency occurs after birth, language and memory skills are most predominantly affected (Zoeller and Rovet, 2004). Altogether these studies indicate that factors, such as age, gender, developmental stage, and iodide status can affect the impact of perchlorate and other NIS inhibitors. All these variables should be taken into account to explain possible inconsistencies in study findings.