The Na+/I symporter (NIS) mediates electroneutral active transport of the environmental pollutant perchlorate

Abstract

The Na(+)/I(-) symporter (NIS) is a key plasma membrane protein that mediates active I(-) uptake in the thyroid, lactating breast, and other tissues with an electrogenic stoichiometry of 2 Na(+) per I(-). In the thyroid, NIS-mediated I(-) uptake is the first step in the biosynthesis of the iodine-containing thyroid hormones, which are essential early in life for proper CNS development. In the lactating breast, NIS mediates the translocation of I(-) to the milk, thus supplying this essential anion to the nursing newborn. Perchlorate (ClO(4)(-)) is a well known competitive inhibitor of NIS. Exposure to food and water contaminated with ClO(4)(-) is common in the U.S. population, and the public health impact of such exposure is currently being debated. To date, it is still uncertain whether ClO(4)(-) is a NIS blocker or a transported substrate of NIS. Here we show in vitro and in vivo that NIS actively transports ClO(4)(-), including ClO(4)(-) translocation to the milk. A simple mathematical fluxes model accurately predicts the effect of ClO(4)(-) transport on the rate and extent of I(-) accumulation. Strikingly, the Na(+)/ ClO(4)(-) transport stoichiometry is electroneutral, uncovering that NIS translocates different substrates with different stoichiometries. That NIS actively concentrates ClO(4)(-) in maternal milk suggests that exposure of newborns to high levels of ClO(4)(-) may pose a greater health risk than previously acknowledged because ClO(4)(-) would thus directly inhibit the newborns' thyroidal I(-) uptake.

Fig. 1.

NIS-mediated ClO₄⁻ accumulation in maternal milk inhibits thyroidal I⁻ uptake in nursing pups. (A and C) ¹³¹I⁻ imaging of lactating rats treated with ClO₄⁻ (A) or not treated (C). Thirty-four hours after administration of 1 mCi ¹³¹I⁻, ¹³¹I⁻ uptake was barely visible in the thyroid of treated rats (A) but was clearly apparent in the gland (T) of nontreated ones (C). (B and D) Pups from treated rats displayed neither thyroidal ¹³¹I⁻ uptake nor urinary accumulation in the bladder (Bla) (B), whereas pups from nontreated rats displayed both (D). An 18-nCi probe (P) was used as a reference in all images. (E and F) Quantification of thyroidal ¹³¹I⁻ uptake. ¹³¹I⁻ accumulation values were obtained (SI Materials and Methods) from two lactating control rats and nine nursing pups (blue bars) and from two lactating rats treated with ClO₄⁻ and 10 nursing pups (red bars). (G) Milk from ClO₄⁻-treated rats inhibits I⁻ uptake in hNIS-expressing MDCK cells. Steady-state (45-min) I⁻ uptake assays were performed with 20 μM I⁻ and the indicated dilutions of the milk samples (see Material and Methods for details). Data are presented as the reciprocal of the inhibition of I⁻ transport relative to the values obtained with milk dilutions from nontreated dams. The [ClO₄⁻] in the milk was calculated by multiplying the slope of the inverse of I⁻ transport as a function of the milk dilution times the IC₅₀ for ClO₄⁻ (1/0.0058 × 1.25 = 215 μM).

Fig. 2.

NIS actively transports ClO₄⁻. (A) ClO₄⁻ delays I⁻ vectorial transport in polarized Flag-NIS-575-transfected MDCK cells. (Right) Schematic representation of the polarized cell monolayer in a bicameral setup. Ap, apical chamber; BL, basolateral chamber; ML, cell monolayer. White cylinders represent Flag-NIS-575, which is exclusively targeted apically. First, 20 μM I⁻ was added to the Ap chamber, and then its concentration was monitored as a function of time in both the Ap chamber (solid blue line) and the BL chamber (dashed blue line). The Ap concentration of I⁻ started to decrease (and the BL to increase) immediately. When 20 μM ClO₄⁻ was added together with I⁻ to the Ap chamber, there was a delay in the Ap decrease (solid red line) and in the concomitant BL increase (dashed red line) in [I⁻]. The concentration of I⁻ in the Ap chamber of NT-MDCK cells remained constant (dotted black line). (B) Schematic representation of NIS-mediated ClO₄⁻ translocation bioassay. Aliquots from the Ap or BL chamber were taken from the bicameral setup with Flag-NIS-575-transfected or NT-polarized MDCK cells ML, diluted, and added to separate plated, nonpolarized WT-hNIS-expressing MDCK cells for I⁻ uptake assays. (C–E) Effect of aliquot dilutions on I⁻ uptake. First 20 μM ClO₄⁻ alone (without I⁻) was added to the Ap chamber (C), the BL chamber (D), or both (E). After 2.5 h, Ap (open circles) or BL aliquots (filled circles) were added to cells seeded on plastic, and I⁻ uptake assays (initial rates) were carried out. Only ClO₄⁻-containing aliquots inhibited I⁻ uptake. Transport rates as a function of the reciprocal of the aliquot dilutions are shown. Data were fitted by nonlinear least squares with the equation v = V_max·[I⁻]/{K_m,I·(1 + [ClO₄⁻]/K_i,ClO4) + [I⁻]}. The intercept, representing the maximal rate of I⁻ transport at infinite dilution, was fixed to the value measured in the absence of ClO₄⁻. A K_i for ClO₄⁻ (K_i,ClO4) of 1.2 μM and a K_m for I⁻ (K_m,I) of 20 μM were used. When 20 μM ClO₄⁻ was added to the Ap chamber of Flag-NIS-575-transfected MDCK cells, the calculated [ClO₄⁻] in the Ap chamber 2.5 h later was 2.14 ± 0.80 μM; in the BL chamber, where there was no ClO₄⁻ at time 0, [ClO₄⁻] reached 22.95 ± 5.45 μM. In contrast, when 20 μM ClO₄⁻ was added to the Ap chamber of NT-MDCK cells, the BL [ClO₄⁻] was 0.46 ± 0.06 μM (data not shown). When 20 μM ClO₄⁻ was added to the BL chamber of Flag-NIS-575-transfected MDCK cells, the calculated [ClO₄⁻] after incubation was 1.59 ± 0.41 μM in the Ap chamber and 21.88 ± 8.19 μM in the BL chamber (D) and 1.59 ± 0.41 in the Ap chamber when ClO₄⁻ was added to NT cells (data not shown). Finally, when 20 μM ClO₄⁻ was added to both chambers, the resulting [ClO₄⁻] was 1.34 ± 0.35 in the Ap chamber and 44.22 ± 6.78 μM in the BL chamber (E).

Fig. 3.

Kinetic analysis of NIS-mediated ReO₄⁻ transport in MDCK cells. (A–D) Initial rates (2-min time points) of I⁻ (A and C) or ReO₄⁻ (B and D) transport by Flag-WT-NIS (solid lines) or Flag-NIS-575 (dotted lines) were determined at the indicated concentrations of I⁻ (A), ReO₄⁻ (B), or Na⁺ (C and D). For A and B, a constant [Na⁺] of 140 mM was used; for C and D, 20 μM I⁻ and 2 μM ReO₄⁻ were used, respectively. Calculated curves in A and B were generated with the equations v = V_max·[I⁻]/(K_m + [I⁻]) and v = V_max·[ReO₄⁻]/(K_m + [ReO₄⁻]), respectively, by using Gnuplot software. In C and D, isotonicity was maintained constant with choline chloride. Na⁺-dependent data were analyzed with the equation v = V_max·[Na⁺]ⁿ/(K_m + [Na⁺]ⁿ). Data were fitted by nonlinear least squares by using Gnuplot software. Background levels in the kinetic experiments were <2% for I⁻ kinetics, <3% for ReO₄⁻ kinetics, <4% in the Na⁺-dependent I⁻ transport, and <0.5% in the Na⁺-dependent ReO₄⁻ transport. Shown are representative experiments. (E) K_m and V_max values of I⁻ and ReO₄⁻ transport by Flag-WT-NIS and Flag-NIS-575 corrected by percentage of NIS-expressing cells. Representative experiments are shown. Experimental values represent the average of triplicate points ± SD. Kinetic experiments were performed at least three to four times.

Fig. 4.

Transport of I⁻ and ClO₄⁻ from the Ap chamber to the BL chamber and mathematical modeling of vectorial flow. The [I⁻] was measured in both chambers at the indicated times. The experiments being modeled used cells transfected with Flag-NIS-575 that transport I⁻ (and ClO₄⁻) from the chamber facing the Ap surface of the cells to the chamber facing the BL surface. Open symbols, [I⁻] in the BL chamber; filled symbols, [I⁻] in the Ap chamber. Lines represent the flows calculated by using equations 4, 6, 7, and 8. Four adjustable parameters (V_max,I, V_max,ClO4, k_bk,ClO4, and k_bk,I) were used to fit the data. [ClO₄⁻] (light and dark dashed lines) in both chambers were not measured; they were calculated by using equations 8 and 6 after adjusting the parameters to the experimental I⁻ values. The time dependence of the [I⁻] and [ClO₄⁻] in the BL and Ap chambers was modeled with a minimal model based on a limited number of assumptions and parameters. It was assumed that all transport from the Ap chamber to the BL chamber was mediated by NIS with a rate describable by the Michaelis–Menten equation. The dependence of the rate on the [Na⁺] was not included because all experiments were carried out at saturating [Na⁺]. It also was assumed that both I⁻ and ClO₄⁻ use the same NIS-binding site and, therefore, compete with each other for the site in the transporter with inhibition constants equal to their K_m values (K_M,I = 20 μM for I⁻; K_M,C = 1.5 μM for ClO₄⁻). The time dependence of the concentration of both ions in the BL and Ap chambers is obtained by integrating equations 1 and 2. A concentration-dependent backflow is necessary to account for the final steady-state concentrations. The parameter k_bk,x (x = l or C) is the backflow rate constant. The subscript t indicates concentrations at time t and 0 initial concentrations (t = 0). Numerical integration of these equations was carried out by a specially written program, in which the concentrations on the right side of equations 3 and 5 were taken as those at time t, and those on the left side were taken as concentrations at time t + Δt to derive equations 7 and 8. Four constants (V_max,I, V_max,C, k_bk,I, and k_bk,C) were adjusted manually to minimize the sum of the squares of the differences between observed and calculated I⁻ values.