The Na+/I- symporter (NIS): mechanism and medical impact

Abstract

The Na(+)/I(-) symporter (NIS) is the plasma membrane glycoprotein that mediates active I(-) transport in the thyroid and other tissues, such as salivary glands, stomach, lactating breast, and small intestine. In the thyroid, NIS-mediated I(-) uptake plays a key role as the first step in the biosynthesis of the thyroid hormones, of which iodine is an essential constituent. These hormones are crucial for the development of the central nervous system and the lungs in the fetus and the newborn and for intermediary metabolism at all ages. Since the cloning of NIS in 1996, NIS research has become a major field of inquiry, with considerable impact on many basic and translational areas. In this article, we review the most recent findings on NIS, I(-) homeostasis, and related topics and place them in historical context. Among many other issues, we discuss the current outlook on iodide deficiency disorders, the present stage of understanding of the structure/function properties of NIS, information gleaned from the characterization of I(-) transport deficiency-causing NIS mutations, insights derived from the newly reported crystal structures of prokaryotic transporters and 3-dimensional homology modeling, and the novel discovery that NIS transports different substrates with different stoichiometries. A review of NIS regulatory mechanisms is provided, including a newly discovered one involving a K(+) channel that is required for NIS function in the thyroid. We also cover current and potential clinical applications of NIS, such as its central role in the treatment of thyroid cancer, its promising use as a reporter gene in imaging and diagnostic procedures, and the latest studies on NIS gene transfer aimed at extending radioiodide treatment to extrathyroidal cancers, including those involving specially engineered NIS molecules.

Figure 1.

Schematic representation of TH biogenesis in polarized thyroid epithelial cells. The thyroid gland (upper left) is composed of follicles (upper right), each of which consists of a single layer of epithelial cells (main panel). NIS (red circle), located on the basolateral surface of the thyrocyte, mediates Na⁺/I⁻ symport with a 2:1 stoichiometry. The driving force for this process is the Na⁺ electrochemical gradient generated by the basolateral Na⁺/K⁺ ATPase (brown triangle). I⁻ crosses the apical membrane via an unidentified carrier or carriers (purple structure with question mark) and is then oxidized and incorporated into Tg (green wire in the colloid) via an enzymatic oxidation carried out by TPO (pink), with the H₂O₂ generated by Duox2 (also pink). Duox2A, an endoplasmic reticulum chaperone required for Duox2 maturation and function, is represented as a blue structure in the endoplasmic reticulum. Iodinated Tg is endocytosed and proteolyzed in the endocytic/lysosomal compartment (green intracellular vesicles), and THs are released into the bloodstream. I⁻ incorporated into MIT and DIT is recycled through a deiodination reaction by the iodotyrosine dehalogenase Dehal1.

Figure 2.

Secondary structure of NIS. The secondary structure model for NIS was constructed on the basis of biochemical data and the protein's structural homology with transporters whose 3-dimensional structures are known (for details, see Section VI). NIS is a 13-TMS protein, with the amino terminus facing the extracellular milieu and an intracellular carboxy terminus. TMSs I and XIII, and the loops between TMSs I and II and XII and XIII, are represented by dotted lines; their position relative to the other TMSs in the secondary structure could not be defined by homology modeling. Ten helical TMSs form an internal inverted topology repeat consisting of 5 transmembrane helical bundles (repeat 1, comprising TMSs II–VI, and repeat 2, comprising TMSs VII–XI), related to each other by a 2-fold pseudosymmetry axis running parallel to the membrane plane through the center. There are 3 glycomoieties (black trees), 1 between TMSs VI and VII and 2 between TMSs XII and XIII. Based on the crystal structures of several prokaryotic transporters (–136, 178), unwound regions in the helices are predicted to provide flexibility and expose side- and main-chain contacts critical for substrate coordination. TMSs II and VII are proposed to define the substrate cavity accommodating both the anion and Na1, and TMS IX contributes to coordination of Na2 (as discussed in Sections VI.D. and VI.B. respectively).

Figure 3.

Evolution of THs: from unicellular algae to mammals. A, In the phylum Chordata, TH biosynthesis is required to signal for metamorphosis (in Protochordata and amphibians) and for development and homeostasis (in vertebrates). In the subphylum Protochordata, which includes Tunicata and Amphioxus, THs are produced in the endostyle, an exocrine gland associated with the pharynx and used for feeding, and end up in the esophagus together with food. In the parasitic lamprey, the endostyle is present during the larval stage, but after metamorphosis, it is replaced by a thyroid. Control of thyroid function by TSH has never been demonstrated in jawless fish, and pituitary modulation of TH synthesis is a characteristic of higher vertebrates. B, Unicellular algae, which use I⁻/I₂ as an antioxidant system, synthesize iodothyronines from marine components. Echinoid larvae feeding on phytoplankton (green box) evolved to sense algal T₄ as a cue to the amount of energy available for their development, in a process called cross-kingdom cross-talk. Lecithotrophic larvae, which evolved from echinoids with planktotrophic larvae (purple box), are not exposed to environmental iodothyronines; however, TH signaling became essential for their metamorphosis earlier in evolution, and rather than losing THs as a development-inducing signal, these larvae evolved the ability to synthesize their own iodothyronines. Facultative feeding larvae (blue box) acquire T₄ from the plankton they feed on but also synthesize endogenous T₄, which in some cases can be enough to trigger metamorphosis in the absence of food. In larvae that depend on food to undergo metamorphosis, exogenous administration of T₄ rescues the effect of food withdrawal (ie, inhibition of metamorphosis), whereas the TH synthesis inhibitor thiourea does not, demonstrating that T₄ is both necessary and sufficient for metamorphosis and indicating that endogenous TH biosynthesis does not occur in these organisms. In contrast, thiourea inhibits metamorphosis in nonfeeding larvae, and these larvae also incorporate ¹²⁵I⁻ into THs. This indicates that endogenous TH biosynthesis occurs in nonfeeding larvae and that it triggers a developmental step in these organisms.

Figure 4.

THs in human development and effects of ID and maternal hypothyroidism on fetal development. A, Maternal dietary I⁻ and T₄ are critical over the entire course of gestation. The fetus's thyroid starts producing THs at ∼20 weeks, but the mother's T₄ contribution is still crucial. After birth, maternal I⁻, which is transferred by NIS into the milk, is key for the newborn's development. Light blue arrows indicate the time of development of specific brain areas and body features that depend on adequate T₄ levels and can be affected by low T₄ at the indicated time of gestation. Neurological or myxedematous cretinism features occur depending on the onset and levels of maternal hypothyroxinemia or hypothyroidism. Congenital hypothyroidism due to fetal mutations in TH synthesis genes have a negative impact after birth, when the supply of maternal T₄ is interrupted. B, Effects of maternal thyroid conditions on fetal development and preventive measures.

Figure 5.

Comparison of hormone levels in hypothyroxinemia and hypothyroidism. Red line indicates normal TSH range. ID is the most common cause of maternal hypothyroxinemia. Maternal serum T₄ is correlated to fetal development. In contrast to hypothyroidism, hypothyroxinemia does not produce significant changes in TSH or T₃ concentrations. Therefore, TSH screenings are not appropriate for the identification of pregnant mothers whose T₄ levels need corrective measures.

Figure 6.

In vitro and in vivo bioassays reveal that perchlorate is a NIS substrate actively transported against its concentration gradient. A, A polarized monolayer of MDCK cells stably transfected with a NIS mutant that is targeted exclusively to the apical surface was exposed to ClO₄⁻ to allow for vectorial transepithelial transport. After a 2-hour incubation, an aliquot was taken from each compartment to assess its effect on ¹²⁵I⁻ transport in NIS-expressing cells grown on a plastic dish. The degree of I⁻ transport inhibition in cells reflects the amount of ClO₄⁻ present in the aliquot, and can be quantified by serial dilutions and by measuring the inhibition by ClO₄⁻ of I⁻ transport in kinetic experiments (129). B, ClO₄⁻ is actively concentrated in maternal milk. To assess whether NIS transports ClO₄⁻ into maternal milk in vivo, lactating dams were injected ip with ClO₄⁻ or saline solution. Pups were separated from their mothers, and lactation was stimulated with oxytocin. Milk collected from the dams was diluted to investigate its effect on ¹²⁵I⁻ transport in NIS-expressing cells grown on a plastic dish. The degree of inhibition of I⁻ transport is directly correlated to the amount of ClO₄⁻ present in the sample, and ClO₄⁻ concentration can be quantified by serial dilutions and inhibition kinetics. Based on the distribution volume of ClO₄⁻ and its IC₅₀, it was determined that ClO₄⁻ is actively concentrated in maternal milk (129).

Figure 7.

Experimentally tested NIS secondary structure model. Gray cylinders represent TMS; gray lines, extracellular segments; black lines, intracellular segments; and branches, N-linked glycosylation sites (N225, 485, and 497). NIS mutations identified in patients with ITD are indicated using the single-letter amino acid code. Those for which the letters are outlined in red (V59E, R124H, Q276E, T354P, G395R, G543E, and Δ339–443) have been studied at the molecular level, as has the effect of different amino acid substitutions at these positions. Δ indicates deletions, and X terminations.

Figure 8.

NIS homology model. A, An NIS 3-dimensional homology model was generated by sequence and structural alignment with vSGLT (27% identity and 57% similarity). The homology model was built on manually corrected sequence alignment. The resulting model was then manually adjusted to improve side-chain interactions and energy minimized. The color code used for the TMSs is the same one used for the secondary structure model (inset, and Figure 2). Helices are represented by cylinders. The model is oriented with the extracellular and intracellular milieux on the top and bottom, respectively. The amino and carboxy termini, along with TMSs I and XIII, are not present in the reference 3-dimensional structures and therefore cannot be modeled. B, Helical model depicting the bundle (TMSs II–III from repeat 1 in red, and VII–VIII from repeat 2 in orange), and the scaffold (TMSs III–V from repeat 1 in blue, and VIII–X from repeat 2 in cyan). The remaining helices and loops are omitted for clarity. The 2 Na⁺ and the I⁻ are shown at positions inferred from experimental data and molecular dynamics calculations.

Figure 9.

A single amino acid substitution converts NIS-mediated ClO₄⁻ and ReO₄⁻ transport from electroneutral to electrogenic. The stoichiometry of WT NIS is 2 Na⁺ per I⁻. Thus, I⁻ transport is electrogenic, as 1 net positive charge is translocated, and it elicits inward positive currents in 2-electrode voltage-clamp experiments in X. laevis oocytes (upper left current trace). In contrast, ClO₄⁻ and ReO₄⁻ do not elicit any current (traces in green box). However, the experiments described in Figure 6 showed that ClO₄⁻ and ReO₄⁻ are transported by NIS, with a V_max that is only approximately one-third that of I⁻, which would have produced currents of recordable magnitude. Therefore, it is concluded that ClO₄⁻ (and ReO₄⁻) transport occurs with a different stoichiometry (1 Na⁺ to 1 ClO₄⁻/ReO₄⁻) than that of I⁻ transport and that it is electroneutral. Interestingly, it is possible to modify the Na⁺/anion transport stoichiometry by replacing a single amino acid in TMS III (Figures 2 and 7); G93N/T/D/Q/E NIS, for example, transports ClO₄⁻ and ReO₄⁻ in an electrogenic fashion (yellow rectangle).

Figure 10.

I⁻ and SCN⁻ transport and oxidation play an antimicrobial role in airway epithelium. A new function for NIS? A, In thyrocytes, I⁻ is translocated across the basolateral membrane by NIS, using the Na⁺ electrochemical gradient generated by the Na⁺/K⁺ ATPase. I⁻ exits the apical surface via an unknown carrier or carriers; pendrin, an anion exchanger, is a candidate for this function. Stimulation of TSH receptor increases the concentration of cAMP and intracellular Ca²⁺, which activates Duox2-catalyzed H₂O₂ synthesis for TPO-catalyzed oxidation of I⁻ to iodine. B, Airway epithelial cells share certain features with thyrocytes; SCN⁻ is actively transported into these cells across the basolateral membrane, likely by NIS. SCN⁻ is secreted in the ASL at concentrations 50 times those found in blood. CFTR is involved in the apical exit of SCN⁻ upon intracellular cAMP increase under basal conditions (342). Ca²⁺-activated Cl⁻ channels (CaCC), and pendrin, which is also expressed in airway epithelial cells, are also responsible for SCN⁻ translocation, in particular under conditions that increase their expression (for instance, under the stimulation of IL-4, an anti-inflammatory cytokine) (342). Duox1, which is expressed at the apical surface in airway epithelium, generates H₂O₂, a substrate for the oxidation by LPO of SCN⁻ to OSCN⁻, a species that has antibacterial activity. I⁻ at the cell air/liquid interface has also recently been found to have an analogous and complementary function; it reduces viral infection in the airway epithelia in vitro (336).

Figure 11.

Wolff-Chaikoff effect and the escape therefrom. I⁻ itself is a major regulator of I⁻ accumulation in the thyroid. A, High plasma I⁻ levels decrease TH biosynthesis; this mechanism is known as the Wolff-Chaikoff effect. B, Thyroid cells adapt to persistent high I⁻ by downregulating NIS and restoring TH production; this process is called the escape from the Wolff-Chaikoff effect.