THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 286, NO. 1, pp. 131–137, January 7, 2011 © 2011 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

Surprising Substrate Versatility in SLC5A6

Naⴙ-COUPLED Iⴚ TRANSPORT BY THE HUMAN Naⴙ/MULTIVITAMIN TRANSPORTER (hSMVT) Received for publication, July 21, 2010, and in revised form, October 26, 2010 Published, JBC Papers in Press, October 27, 2010, DOI 10.1074/jbc.M110.167197

Fernanda Delmondes de Carvalho‡ and Matthias Quick‡§¶1 From the ‡Center for Molecular Recognition and §Department of Psychiatry, Columbia University College of Physicians and Surgeons and the ¶Division of Molecular Therapeutics, New York State Psychiatric Institute, New York, New York 10032 Iodide (Iⴚ) is an essential constituent of the thyroid hormones triiodothyronine and thyroxine, which are required for the development of the central nervous system in the fetus and newborn. Iⴚ uptake in the thyroid is mediated by the Naⴙ/Iⴚ symporter (NIS). NIS has gained particular medical interest due to its sensitivity to the environmental pollutant perchlorate (ClO4ⴚ) and its implication in radioiodide cancer treatment. Recently, others have shown that Iⴚ absorption in the intestine is mediated by NIS (Nicola, J. P., Basquin, C., Portulano, C., Reyna-Neyra, A., Paroder, M., and Carrasco, N. (2009) Am. J. Physiol. Cell Physiol. 296, C654 – 662). However, their data suggest the participation of other systems in the homeostasis of Iⴚ, in particular because in vivo uptake studies revealed a ClO4ⴚ-insensitive transport component. Here, we describe Naⴙ-coupled Iⴚ uptake by the human Naⴙ/multivitamin transporter (hSMVT), a related protein isolated from the placenta, where it was suggested to supply the fetus with the water-soluble vitamins biotin and pantothenic acid, and ␣-lipoic acid. hSMVT-mediated Naⴙ/Iⴚ symport is inhibited by the other three organic hSMVT substrates but not by NIS substrates; notably, hSMVT is insensitive to ClO4ⴚ. Because hSMVT is found in the intestine and in many other tissues, we propose that hSMVT may play an important role in the homeostasis of Iⴚ in the body.

The Na⫹/multivitamin transporter (SMVT)2 is a member of the SLC5 family (1) of Na⫹/solute symporters (SSS) (2) and mediates the Na⫹-dependent uptake of the structurally diverse water-soluble vitamins pantothenic acid (PA) and biotin (3, 4). ␣-Lipoic acid (LA), another vitamin-like substance with strong antioxidant properties, inhibits uptake of biotin and PA, and its addition to the bath solution elicits inward currents in oocytes expressing the rat SMVT (5) and human SMVT (hSMVT) (4). Although direct evidence for the uptake of LA has been missing, these results imply that LA is the third substrate transported by SMVT in a Na⫹-coupled symport reaction.

Among the members of the SSS family, hSMVT (SLC5A6) shares the highest sequence identity and similarity with the Na⫹/iodide symporter (NIS; SLC5A5) (6) (40 and 64%, respectively) and SLC5A8, a recently identified organic cation transporter (41 and 57%, respectively). NIS is directly implicated in the physiology of the thyroid hormones thyroxine and triiodothyronine by transporting I⫺ in Na⫹-coupled manner from the blood across the basolateral membrane of thyroid cells (7). Due to its high sequence homology with NIS (46% identity and 70% similarity), SLC5A8 was initially proposed to function as an I⫺ transport system in the apical membrane of the thyrocyte, where it was suggested to mediate the transport of I⫺ into the colloidal lumen, where the thyroid hormones are synthesized (8). However, uptake of I⫺ by SCL5A8 could not be unequivocally confirmed, and Coady et al. (9) and Miyauchi et al. (10) described this protein as Na⫹-coupled monocarboxylate transporter. In a similar vein, the high sequence homology between NIS and hSMVT gave rise to the hypothesis that hSMVT may exhibit I⫺ transport as well. This notion was further supported by the observation that uptake of PA by hSMVT was inhibited when NaCl was replaced with NaI (4). The authors questioned whether this inhibition could be attributed to the competition of I⫺ with PA transport, i.e. Na⫹/I⫺ symport, or to the lack of Cl⫺ in a putative Na⫹/Cl⫺-dependent substrate transport reaction. To address this question, we analyzed hSMVT-mediated substrate transport and substrate-elicited currents in Xenopus laevis oocytes. We show that the activity of hSMVT is not dependent on Cl⫺ but that instead, I⫺ inhibits the Na⫹-dependent uptake of PA, biotin, and LA in a concentrationdependent manner. Remarkably, 125I⫺ uptake studies in hSMVT-expressing oocytes unequivocally revealed that hSMVT catalyzes Na⫹-dependent transport of I⫺ that is not sensitive to the NIS-specific I⫺ uptake blocker ClO4⫺ (11). In addition, the other substrates of hSMVT, namely, PA, biotin, and LA, specifically inhibit I⫺ transport, suggesting a common uptake mechanism that is shared among these structurally diverse substrates. As hSMVT is universally expressed in all tissues of the human body (4), it is feasible to speculate that hSMVT contributes to the homeostasis of I⫺ in the body.

1

To whom correspondence should be addressed: 650 W. 168th St., BB1119, New York, NY 10032. Tel.: 646-797-2902; Fax: 212-305-5594; E-mail: [email protected]. 2 The abbreviations used are: SMVT, Na⫹/multivitamin transporter; hSMVT, human SMVT; SSS, Na⫹/solute symporters; LA, ␣-lipoic acid; PA, pantothenic acid; NIS, Na⫹/I⫺ symporter; Tris, 2-amino-2-hydroxymethyl-propane-1,3-diol; Mes, 2-(N-morpholino) ethanesulfonic acid; SGLT, Na⫹/ glucose transporter.

JANUARY 7, 2011 • VOLUME 286 • NUMBER 1

EXPERIMENTAL PROCEDURES Molecular Biology—The hSMVT gene was amplified by PCR using vector pSPORT-hSMVT (4) as template and flanking primers that contained unique EcoRV and HindIII sites at the 5⬘ and 3⬘ ends, respectively. The EcoRV/HindIII-cut PCR JOURNAL OF BIOLOGICAL CHEMISTRY

131

Iodide Transport by hSMVT product was ligated into similarly treated vector pGhS-4 (12), replacing the entire hSGLT1 gene. The integrity of the hSMVT gene in the resulting vector, pMV1, was confirmed by DNA sequencing and found to be identical to that in the template plasmid pSPORT-hSMVT. The mRNA transcript for hSMVT was produced in vitro with the Ambion MegaScript kit, using the T3 RNA polymerase and an RNA Cap analog, following NotI linearization of pMV1 (12). X. laevis oocytes were injected with 50 ng of mRNA transcript for hSMVT and subsequently incubated in Barth’s medium containing gentamicin (5 mg/ml) at 18 °C for 3– 6 days. Transport Assays and Electrophysiological Techniques— Transport measurements and electrophysiological experiments were performed in assay buffers composed of 10 mM HEPES-Tris, pH 7.4, 2 mM KCl, 1 mM MgCl2, 1 mM CaCl2, and 100 mM NaCl, LiCl, or choline chloride. Cl⫺-free assay buffers were prepared by replacing all Cl⫺ salts with gluconate salts, whereas choline chloride was replaced with N-methyl-D-glucosamine. For Na⫹ activation experiments at Cl⫺-free conditions, sodium gluconate was varied between 0 and 100 mM by equimolar replacement with N-methyl-D-glucosamine to give a final concentration of 100 mM. The effect of H⫹ on hSMVT activity was tested by lowering the pH of assay buffer containing 100 mM choline chloride to 5.5, by titration with Mes. Electrophysiology—Electrophysiological measurements were performed using the two-electrode voltage clamp technique (13). In experiments using Cl⫺-free conditions, the reference electrode was connected to the experimental oocyte chamber via an agar bridge (3% agar in 3 M KCl). Currentvoltage (I/V) relations were obtained with a pulse protocol (pCLAMP or Clampex, Axon Instruments, Inc., Foster City, CA) that consisted of 100-ms voltage steps, from a holding potential (Vh) of ⫺50 mV to a series of test voltages (Vt), from ⫹50 to ⫺150 mV, in 20-mV increments. Currents were low pass-filtered at 500 Hz and sampled at 10 kHz. Data Analysis—Substrate-evoked steady-state currents were fitted to Equation 1, ⫺

⫺

II ⫽ ⫺

I Imax ⫻ 关I⫺兴n ⫺

I 共K0.5 兲n ⫹ 关I⫺兴n

⫺

(Eq. 1)

I where II and Imax represent substrate-induced current and I⫺ maximal substrate-induced current, respectively. K0.5 equals ⫺ I [I⫺] at 50% Imax , and n represents the Hill coefficient. Figs. 3 and 4 are based on data obtained from representative experiments, and data in individual panels originate from parallel determinations using the same batch of oocytes. Data fitting was performed using non-linear regression algorithms in Sigma Plot and errors represent the S.E. of the fit. Radiotracer Uptake Studies—Uptake of RS-[3H]␣-lipoic acid (12 Ci/mmol; American Radiochemicals), D-[14C]pantothenic acid (51.5 mCi/mmol; Perkin Elmer), D-[14C]biotin (54 mCi/mmol; GE Healthcare), or Na125I (2280.6 Ci/mmol; Perkin Elmer; diluted to a specific activity of 0.1 Ci/mmol with KI, yielding a final Na⫹ concentration of 1.9 nM) was performed in hSMVT-expressing oocytes or control oocytes, in

132 JOURNAL OF BIOLOGICAL CHEMISTRY

parallel. To eliminate Na⫹ or Cl⫺ contaminations from Barth’s medium, oocytes were subjected to a wash in 100 mM choline chloride or N-methyl-D-glucosamine assay buffer, respectively, prior to assessing their uptake activity at given experimental conditions. The uptake reaction was stopped by removing the oocytes from the radiotracer-containing assay buffer and subsequently washing the oocytes in ice-cold assay buffer without radiotracer. The accumulated radioactivity was determined by scintillation counting after solubilization of individual oocytes with 100 ␮l of 5% SDS, followed by addition of 5 ml of scintillation mixture (Econo-Safe, Research Products International Corp.). All uptake experiments were performed at least in duplicate with oocytes from different donor frogs. Data are expressed as mean ⫾ S.E. of ⱖ 7 oocytes, and kinetic constants were obtained by nonlinear regression fitting of the data, with errors indicating the S.E. of the fit. Immunological Detection of hSMVT in Oocytes— hSMVTexpressing oocytes were solubilized with 100 mM Tris-Cl, pH 7.5, 100 mM NaCl, 1% Triton X-100 (w/v). The soluble fraction of three oocytes was separated from insoluble cell debris by centrifugation and subjected to 10% SDS-PAGE analysis followed by electrotransfer of the protein to PVDF membranes (14). For Western blot analysis, the membrane was blocked with 5% nonfat dry milk in PBS containing 0.1% Tween 20, followed by incubation of the blot with a rabbit polyclonal antibody (SMVT-11A; Alpha Diagnostic, San Antonio, TX). Goat anti-rabbit IgG secondary antibodies conjugated to horseradish peroxidase were used in conjunction with the ECL method (SuperSignal威 West Pico kit, Thermo Scientific), and immunoreactions were detected with the AlphaImager HP system (Cell Biosciences).

RESULTS AND DISCUSSION Iodide Inhibits hSMVT-mediated Substrate Transport—To standardize the preparation of hSMVT mRNA to the protocol that has been successfully applied to hSGLT1 (15) and allows direct comparison between hSMVT and hSGLT1, the hSMVT cDNA was inserted in the hSGLT1 expression plasmid pGhS-4 (12) under the control of the T3 promoter, yielding plasmid pMV1. To detect the presence of hSMVT in oocytes, we performed Western blot analysis of oocyte protein from oocytes injected with the hSMVT mRNA originating from pMV1 or from control (water-injected) oocytes using a polyclonal anti-SMVT antibody raised against the LYHACRGWGRHTVGELLMADRK peptide. This peptide corresponds to amino acids 44 – 65 of hSMVT and has been successfully used for the detection of the human (16) and rat (17) isoforms of SMVT. Two distinct bands (at an apparent molecular mass of ⬃55 and ⬃69 kDa that may correspond to the core-glycosylated and the fully glycosylated protein, respectively (14)) were detectable in hSMVT-expressing oocytes but not in control oocytes (Fig. 1). Oocytes expressing hSMVT displayed robust uptake activity of [14C]PA (data not shown) and [14C]biotin (Fig. 2A). Uptake of [14C]biotin featured the same characteristics previously attributed to hSMVT in terms of Na⫹ dependence and sensitivity to PA and racemic RS-LA (4). Wang et al. (4) have shown that reVOLUME 286 • NUMBER 1 • JANUARY 7, 2011

Iodide Transport by hSMVT placement of NaCl with NaI resulted in a marked reduction of the uptake activity of hSMVT. Because the authors also observed a reduction in the uptake activity when NaCl was replaced with sodium gluconate, it was speculated whether hSMVT catalyzes Na⫹- and Cl⫺-dependent substrate cotransport, as featured by most mammalian members of the neurotransmitter:sodium symporter family (18) or whether hSMVT activity appears to be blocked by I⫺ because I⫺ may also be transported by hSMVT in a Na⫹-dependent manner. The latter option seemed feasible since hSMVT, as mentioned above, shares high sequence homology with NIS, the Na⫹/I⫺ symporter. Fig. 2A shows that removal of Cl⫺ from the Na⫹containing assay buffer (by equimolar replacement of NaCl with sodium gluconate) had no significant effect on the Na⫹dependent transport of [14C]biotin by hSMVT. Remarkably,

FIGURE 1. Western blot analysis of oocytes expressing hSMVT. Immunological detection of hSMVT was performed using a polyclonal antibody raised against a 22-amino acid peptide corresponding to positions 44 – 65 in hSMVT as described under “Experimental Procedures.” Protein originating from three oocytes (co, water-injected control oocytes; hSMVT, hSMVT-expressing oocytes) was subjected to 10% SDS-PAGE and electroblotted onto a PVDF transfer membrane. The membrane was incubated with rabbit antihSMVT polyclonal IgG, followed by incubation with goat anti-(rabbit IgG) horseradish peroxidase conjugate. Immunoreactions were visualized with the enhanced chemiluminescence method. Positions of the protein standards are indicated (kDa).

the addition of 250 ␮M I⫺ resulted in a reduction of accumulated radiolabeled biotin by ⬃52% (Fig. 2A). Likewise, uptake of 50 ␮M racemic RS-[3H]LA (Fig. 2B) in oocytes expressing hSMVT was Na⫹-dependent and Cl⫺-independent (as demonstrated by replacement of NaCl with sodium gluconate). The hSMVT-specific Na⫹-dependent uptake of RS-[3H]LA was inhibited by 100 ␮M biotin, PA, or I⫺ by ⬃63, ⬃32, and ⬃29%, respectively. This result demonstrates radiotracer flux measurements of [3H]LA for the first time, providing direct evidence that LA is transported by hSMVT in a Na⫹-dependent manner, as suggested previously based on electrophysiological data (4). The fact that the other established substrates of hSMVT inhibit the transport of LA strongly suggests that these substrates share a common pathway through the transporter. Consistent with this notion, increasing [I⫺] abolished the Na⫹-dependent uptake of RS-[3H]LA in concentrationdependent manner, exhibiting an apparent inhibition conI⫺ stant (IC50 , corresponding to [I⫺] that yielded 50% inhibition of LA uptake) of 258.8 ⫾ 56.5 ␮M (Fig. 2C). Iodide Elicits Electrical Currents in hSMVT—The electrogenicity of hSMVT-mediated substrate transport is shown in Fig. 3. Oocytes injected with hSMVT mRNA were superfused in 100 mM NaCl-containing buffer under voltage-clamped conditions (Vh ⫽ ⫺50 mV), and compounds were added as indicated. Notably, 25 ␮M PA, biotin, LA, or I⫺ elicited the same magnitude of inward currents, which were reversible upon the removal of each compound from the bath solution. In striking contrast to several members of the SLC5 family, hSMVT exhibited no detectable Na⫹ leak currents under our experimental conditions (compare Fig. 4A). For example, these currents have been observed for NIS (19) and SGLT1 (20) when 100 mM choline chloride was replaced with 100 mM NaCl in assay buffer at pH 7.4. For NIS and SGLT1, the Na⫹ leak currents reached ⬃35 and ⬃10% of the maximum Na⫹/ substrate cotransport currents, respectively, and they are attributed to the uncoupled flux of Na⫹ in the absence of substrate (12, 21–23). However, the molecular basis of the leak currents has recently been challenged by Longpre et al. (24). The authors show that the transporter-associated cation leak current is directly proportional to the SGLT1-mediated passive water permeability, and they suggest that water and cations share a common pathway through the transporter. In

FIGURE 2. Transport activity of hSMVT in Xenopus oocytes. A, uptake of 25 ␮M D-[14C]biotin by oocytes injected with hSMVT mRNA. Transport of radiolabeled biotin was assayed for 15 min at 25 °C in the presence of 100 mM choline chloride, pH 7.4 (CHO), 100 mM NaCl (Na⫹) plus 250 ␮M of the indicated compounds: PA, RS-␣-lipoic acid (LA), or NaI (I⫺). B, uptake of 50 ␮M RS-[3H]␣-lipoic acid by hSMVT-expressing (solid bars) and control (open bars) oocytes. Assays were performed as described in the legend to A by including Cl⫺-free conditions; 100 mM NaCl were equimolarly replaced with sodium gluconate (NaG). Compounds, biotin (BIO), PA, or I⫺ were added at a final concentration of 100 ␮M. C, the inhibition of LA uptake is dependent on the concentration of I⫺. Uptake of RS-[3H]␣-lipoic acid was performed in 100 mM NaCl in the presence of increasing [KI]. Data were plotted as function of I⫺ concentration and fitted according to a modified Equation 1 (n set to 1; see text for details). Data are from representative experiments.

JANUARY 7, 2011 • VOLUME 286 • NUMBER 1

JOURNAL OF BIOLOGICAL CHEMISTRY

133

Iodide Transport by hSMVT

FIGURE 3. Substrate-elicited inward currents of hSMVT. Currents were recorded from a representative oocyte injected with hSMVT mRNA at Vh ⫽ ⫺50 mV in 100 mM NaCl-containing buffer. 25 ␮M biotin (BIO), PA, RS-␣lipoic acid (LA), or NaI (I⫺) were added as indicated by the bar.

contrast to the conclusions raised in this study, Loo et al. (25) have previously shown that passive water and Na⫹ transport through cotransporters depend on saturable uncoupled flux of Na⫹, whereas water permeation occurs through a low conductance water channel. To date, no data are available that address the water transport function of hSMVT. The fact that hSMVT does not exhibit Na⫹ leak currents could be a useful tool to dissect the molecular bases of water and Na⫹ flux through transporters. To test the effect of the membrane potential on the hSMVT-elicited electrical currents, we performed electrophysiological measurements by stepping the membrane potential from a holding potential (⫺50 mV) to the test potential (from ⫺150 to ⫹50 mV in 20-mV decrements). Fig. 4A shows current/voltage (I/V) relations of a representative hSMVT-expressing oocyte in the presence of different ionic conditions. Electrical currents observed in assay buffer at pH 7.4 containing either 100 mM NaCl, 100 mM sodium gluconate, or 100 mM choline chloride were not appreciably different. Larger currents were observed when NaCl was equimolarly replaced with LiCl or when the pH of 100 mM choline chloride-containing buffer was lowered to 5.5 (e.g. ⫺1310 nA in choline chloride, pH 5.5, versus ⫺825 nA in 100 mM NaCl, pH 7.4, when compared at a Vt of ⫺150 mV). Although the magnitude of the hSMVT-elicited currents in LiCl and choline chloride at pH 5.5 was markedly different to those observed in NaCl (or sodium gluconate) and chlonine chloride at pH 7.4, substrate-induced currents for hSMVT were only observed in the presence of Na⫹ (Fig. 4, B and C). In contrast, rat and human SGLT1 have been shown to couple glucose symport to the flux of Na⫹, Li⫹, and/or H⫹ (12, 20).3 Consistent with uptake data of radiolabeled biotin and LA (see Fig. 2), substitution of NaCl with sodium gluconate (Cl⫺free conditions) had no significant effect on hSMVT-elicited currents in the absence (Fig. 4A) or in the presence of substrate (biotin, PA, LA, or I⫺; data not shown). These results emphasize that Na⫹ is strictly required as coupling ion and Cl⫺ is dispensable for substrate transport by hSMVT. Moreover, these observations also point out that the individual members of the SLC5 family may have distinct ionic requirements for cation-coupled substrate symport. Fig. 4B shows the substrate-induced currents (Isubstrate ⫽ total I ⫺ Ino substrate) evoked by 25 ␮M PA, biotin, or LA in the presence of 100 mM NaCl. These currents were not observed in control (water-injected) oocytes. The magnitude of the substrate-elicited currents in NaCl increased with hypopolar3

M. Quick and E. Wright, unpublished results.

134 JOURNAL OF BIOLOGICAL CHEMISTRY

ization in supralinear fashion and did not saturate at ⫺150 mV, the most negative potential tested. To test whether the I⫺-elicited currents observed in hSMVT-expressing oocytes (Fig. 3) resulted from an unspecific interaction of I⫺ with hSMVT, we performed electrophysiological measurements by applying the voltage-step protocol in the presence of high concentrations of I⫺ as well as substrates that have been shown to elicit currents in other members of the SCL5 family. Monovalent anions that are transported by NIS, such as NO3-, isethionate, Br⫺ (19), and perchlorate (ClO4⫺) (11), as well as the SGLT1 substrate ␣-methyl-D-glucose, failed to elicit currents in oocytes injected with hSMVT mRNA in the presence (Fig. 4C) or absence of NaCl (data not shown). Also, I⫺-induced currents in hSMVT-expressing oocytes were not inhibited by the potent NIS inhibitor ClO4⫺ when tested at a concentration of 1 mM. ⫺ Note that the apparent inhibitory constant of ClO4⫺ (KiClO4 ) on I⫺ transport by NIS was determined to be 1.8 ␮M (19). Fig. 4D shows current-voltage (I/V) relations of I⫺-elicited currents obtained at representative [I⫺], at Vt ranging from ⫺150 to ⫹50 mV. At [I⫺] ⬍ 1 mM, currents were reversed, and small net outward currents were observed at positive potentials. The magnitude of the I⫺-induced inward currents increased and saturated with hyperpolarization. The I⫺-induced inward current at each Vt was concentration-dependent and saturable. Plotting the current as a function of [I⫺] and fitting to Equation 1, from Vt between ⫺150 mV and ⫹10 I⫺ mV, yielded half-maximum concentration constants (K0.5 ) that were relatively independent of Vt between ⫺150 mV and ⫺70 mV (⬃150 ␮M) and increased to 1.15 ⫾ 0.1 mM at ⫹10 I⫺ mV (Fig. 4E). The maximum I⫺-induced current (Imax ) was dependent on Vt, saturated with hypopolarizing Vt at about ⫺200 nA but did not saturate at ⫺150 mV, the most hyperpolarizing Vt tested (Fig. 4F). It is noteworthy to mention that in some oocytes injected with hSMVT mRNA (but not in control oocytes), the addition of [I⫺] ⱖ 5 mM induced large inward steady-state currents at all Vt tested that increased in superlinear fashion with hyperpolarization. Iodide Transport by hSMVT—Although the above experiments revealed either the effect of I⫺ on the uptake of organic solutes (biotin and LA; Fig. 2) or on electrical currents elicited by hSMVT (Figs. 3 and 4), these results do not necessarily validate per se the hypothesis of hSMVT-mediated uptake of I⫺. To rule out the possibility that I⫺ merely binds to hSMVT, i.e. acting as blocker of substrate symport and inducing electrical signals in response to binding to the protein without being transported, we performed direct 125I⫺ uptake studies. Fig. 5A reveals that hSMVT-expressing oocytes exhibit Na⫹dependent uptake of I⫺ that is ⬃10-fold higher than that observed in control oocytes. Again, replacement of Cl⫺ with gluconate in the presence of Na⫹ had virtually no effect on the uptake activity. Replacement of Na⫹ with Li⫹ or choline (Cl⫺ salts), or decreasing the pH from 7.4 to 5.5 (of the 100 mM choline chloride assay buffer) resulted in hSMVT-mediated uptake activity that was indistinguishable from that observed in control oocytes. The time course of Na⫹-dependent I⫺ transport (Fig. 5B) showed that the steady-state level of I⫺ accumulation (101 ⫾ 6.7 pmol per oocyte) was reached after VOLUME 286 • NUMBER 1 • JANUARY 7, 2011

Iodide Transport by hSMVT

FIGURE 4. hSMVT-elicited electrical currents. A, current-voltage (I/V) relationship of steady-state currents elicited in hSMVT-expressing oocytes in the presence of different ions upon stepping the membrane potential from Vh of ⫺50 mV to a series of test voltages (Vt) from ⫹50 to ⫺150 mV, in 20-mV increments. The relevant ionic condition of the assay buffer is indicated: 100 mM choline chloride, pH 7.4 or 5.5 (CHO, pH 7.4 or CHO, pH 5.5, respectively); 100 mM NaCl or sodium gluconate (NaG), or LiCl. B, substrate-induced currents. 25 ␮M PA, biotin, or LA were added to assay buffer containing 100 mM NaCl, and currents were recorded as described in the legend to A. The substrate-induced currents were calculated by subtracting the currents observed in 100 mM NaCl from the total current in NaCl plus the indicated substrate. C, effect of representative potential substrates (1 mM isethionate, 1 mM perchlorate [ClO4⫺], 100 mM ␣-methyl-D-glucose (␣-MDG) on hSMVT-mediated steady-state currents in 100 mM NaCl. I/V relations of the net currents (Icompound ⫽ Itotal ⫺ Ino compound) are shown. D, I/V relation of I⫺-induced steady-state currents in 100 mM NaCl and increasing concentrations of [I⫺]. I⫺-induced currents were plotted as a I⫺ I⫺ function of [I⫺] for each Vt from ⫺150 to ⫹10 mV and fitted to Equation 1 with n ⫽ 1, yielding K0.5 (E) and Imax (F) at each indicated Vt. ⫺

FIGURE 5. Iodide transport by hSMVT. Uptake of 500 ␮M 125I⫺ was measured in oocytes injected with hSMVT mRNA or in control oocytes. A, cation dependence of 125I⫺ uptake. Transport was measured for 5 min in the presence of the indicated assay buffer (100 mM choline chloride, pH 7.4 (⫺), or pH 5.5 (⌬␮H⫹), 100 mM LiCl ((⌬␮Li⫹), 100 mM NaCl (⌬␮Na⫹/⌬␮Cl⫺) or 100 mM sodium gluconate (⌬␮Na⫹). B, time course of Na⫹-dependent I⫺ uptake in hSMVT-expressing (f) and control (E) oocytes. C, inhibition of 125I⫺ transport by hSMVT substrates. D, isotopic dilution of 500 ␮M 125I⫺ with nonraI⫺ dioactive I⫺ yielding an IC50 of 267.1 ⫾ 39.5 ␮M. E, Na⫹ dependence of the initial rates of 125I⫺ transport by hSMVT. Transport of 500 ␮M 125I⫺ was measured at increasing concentrations of sodium gluconate for 5 min.

⬃30 min. Uptake of 125I⫺ was inhibited by the other hSMVT substrates, biotin, PA, and LA, but ClO4⫺ failed to inhibit hSMVT-mediated I⫺ transport (Fig. 5C), paralleling the electrophysiological observations. Isotopic replacement of 500 ␮M 125 ⫺ I with increasing concentrations of nonradioactive I⫺ JANUARY 7, 2011 • VOLUME 286 • NUMBER 1

I yielded a 50% reduction of 125I⫺ uptake (IC50 ) at a concentration of 267.1 ⫾ 39.5 ␮M (Fig. 5D). This number is in perfect I⫺ agreement with the K0.5 obtained with electrophysiological measurements (Fig. 4E), considering that the membrane potential of non-clamped hSMVT-expressing oocytes was ⫺37 ⫾ 4 mV (n ⫽ 17). The initial rates of 125I⫺ transport by hSMVT were dependent on the Na⫹ concentration in the assay buffer and reached a maximum at ⬃50 mM (Fig. 5E). Fitting the curve according to the Hill equation (compare ⫺ Equation 1 in which II and [I⫺] were replaced with the initial 125 ⫺ rate of I transport and [Na⫹], respectively) yielded a concentration of Na⫹ at 50% of the maximum 125I⫺ uptake velocNa⫹ ity (apparent K0.5 ) of 10.6 ⫾ 0.6 mM, with a Hill (n) coefficient of 1.7 ⫾ 0.1 mM. Although the Hill (n) coefficient does not provide an absolute measure for the stoichiometry of Na⫹:I⫺ co-transport, a Hill coefficient of ⬃2 is consistent with an apparent stoichiometry of 2 Na⫹:1 I⫺ for coupled uptake. This is in line with much functional data in other SLC5 family members, such as a Na⫹-to-substrate stoichiometry of 2:1 for Na⫹/glucose and Na⫹/I⫺ cotransport by SGLT1 (26) and NIS (19), respectively. In addition, the identification of two bound Na⫹ ions in the crystal structure of LeuT (27) correlates well with a Hill coefficient of ⬃2 for Na⫹-coupled Leu binding and direct 22Na⫹ binding studies (28). Taken together, our results unequivocally show that hSMVT, in addition to transporting biotin and pantothenic acid (4), catalyzes Na⫹/I⫺ and Na⫹/␣-lipoic acid co-transport (symport) with the electrochemical Na⫹ gradient as the immediate driving force. With the exception of SGLT3 (SLC5A3), which has been shown to serve as glucose-sensing Na⫹ channel (29), this feature is shared among all characterized members of the SSS family (30). However, in contrast to rat and human SGLT (12, 20) and PutP (31), which can use

JOURNAL OF BIOLOGICAL CHEMISTRY

135

Iodide Transport by hSMVT Na⫹, Li⫹, and/or H⫹ as coupling cation for substrate transport, hSMVT-mediated substrate transport is strictly dependent on Na⫹. In this regard, hSMVT shares a common feature with LeuT, which cannot catalyze substrate transport when Na⫹ is replaced with H⫹ or Li⫹.4 Note that in the crystal structures of LeuT, the carboxyl group of the substrate serves as coordinating partner for a Na⫹ (in the Na1 site) (27, 32, 33). This observation may provide a general explanation, at the molecular level, of how the energy stored in the electrochemical cation concentration gradient is transduced into the transport of a specific substrate against its concentration gradient. From recent crystallographic data (27, 34, 35), it appears feasible to assume that overlapping binding sites for substrate and co-substrate (i.e. cations) may represent a common principle of cation-coupled co-transport. The coordinated interaction of all binding partners, namely substrate, co-substrate, protein backbone, and amino acid side chains, seems to determine the specificity of a particular transport protein for its substrate(s) and co-substrate(s). In this regard, it is interesting to note that some members of the SSS family possess a strict or slightly overlapping substrate specificity for structurally related substrates (e.g. PutP exhibits an exclusive specificity for proline, SGLT transports the sugars glucose and galactose, and NIS mediates transport of monovalent inorganic and small organic anions, such as I⫺, Br⫺, SCN⫺ (19), or ClO4⫺ (11)), whereas hSMVT mediates the Na⫹-coupled transport of relatively large organic compounds (biotin, PA, and LA) and I⫺, an inorganic anion. This broad substrate spectrum positions hSMVT in a rather unique niche among the members of the SSS family, and, although the organic substrates exist in anionic form at physiological pH (36), this substrate versatility is surprising from a structural perspective. What is the physiological relevance of hSMVT-mediated Na⫹/I⫺ cotransport? NIS has been shown to be the primary I⫺-translocating protein in the thyroid and in the lactating mammary gland, where it mediates active I⫺ transport into the thyroid follicular cells, the first step in thyroid hormone biosynthesis, and into the milk for thyroid hormone biosynthesis by the nursing newborn, respectively (37). Recently, Nicola et al. (38) demonstrated NIS-mediated Na⫹-dependent uptake of I⫺ in the intestine. However, early reports (39, 40) indicate a rather complex I⫺ homeostasis in the gastrointestinal tract that may involve other proteins in addition to NIS. This notion is supported by the fact that the accumulation of pertechnetate (TcO4⫺), a commonly used NIS substrate, in the blood of rats that were administered 99m TcO4⫺ via a duodenal catheter, was only partially inhibited by the potent NIS blocker ClO4⫺ (38). The isolation of hSMVT from the placenta points to the central role of hSMVT in the supply of the developing fetus with the essential nutrients biotin, PA, and LA, as well as I⫺. Because hSMVT is expressed in virtually every tissue in the body (4, 41), it might contribute to a hitherto unknown homeostasis of I⫺.

4

M. Quick and J. Javitch, unpublished observations.

136 JOURNAL OF BIOLOGICAL CHEMISTRY

Acknowledgments—We are indebted to Dr. Puttur Prasad for kindly providing the pSPORT-hSMVT-expression vector. We thank Drs. Ernest Wright, Don Loo, and Bruce Hirayama for help and advice at the initial phase of the project and Drs. Ming Zhou and Jonathan Javitch for support and critical discussions. REFERENCES 1. Hediger, M. A., Romero, M. F., Peng, J. B., Rolfs, A., Takanaga, H., and Bruford, E. A. (2004) Pflugers Arch 447, 465– 468 2. Saier, M. H., Jr. (2000) Microbiol. Mol. Biol. Rev. 64, 354 – 411 3. Prasad, P. D., Wang, H., Kekuda, R., Fujita, T., Fei, Y. J., Devoe, L. D., Leibach, F. H., and Ganapathy, V. (1998) J. Biol. Chem. 273, 7501–7506 4. Wang, H., Huang, W., Fei, Y. J., Xia, H., Yang-Feng, T. L., Leibach, F. H., Devoe, L. D., Ganapathy, V., and Prasad, P. D. (1999) J. Biol. Chem. 274, 14875–14883 5. Prasad, P. D., Wang, H., Huang, W., Fei, Y. J., Leibach, F. H., Devoe, L. D., and Ganapathy, V. (1999) Archives of biochemistry and biophysics 366, 95–106 6. Dai, G., Levy, O., and Carrasco, N. (1996) Nature 379, 458 – 460 7. Doha´n, O., and Carrasco, N. (2003) Mol. Cell. Endocrinol. 213, 59 –70 8. Rodriguez, A. M., Perron, B., Lacroix, L., Caillou, B., Leblanc, G., Schlumberger, M., Bidart, J. M., and Pourcher, T. (2002) J. Clin. Endocrinol. Metab. 87, 3500 –3503 9. Coady, M. J., Chang, M. H., Charron, F. M., Plata, C., Wallendorff, B., Sah, J. F., Markowitz, S. D., Romero, M. F., and Lapointe, J. Y. (2004) J. Physiol. 557, 719 –731 10. Miyauchi, S., Gopal, E., Fei, Y. J., and Ganapathy, V. (2004) J. Biol. Chem. 279, 13293–13296 11. Doha´n, O., Portulano, C., Basquin, C., Reyna-Neyra, A., Amzel, L. M., and Carrasco, N. (2007) Proc. Natl. Acad. Sci. U.S.A. 104, 20250 –20255 12. Quick, M., Loo, D. D., and Wright, E. M. (2001) J. Biol. Chem. 276, 1728 –1734 13. Loo, D. D., Hazama, A., Supplisson, S., Turk, E., and Wright, E. M. (1993) Proc. Natl. Acad. Sci. U.S.A. 90, 5767–5771 14. Hirayama, B. A., and Wright, E. M. (1992) Biochim. Biophys. Acta 1103, 37– 44 15. Hediger, M. A., Coady, M. J., Ikeda, T. S., and Wright, E. M. (1987) Nature 330, 379 –381 16. Balamurugan, K., Ortiz, A., and Said, H. M. (2003) Am. J. Physiol. Gastrointest. Liver Physiol. 285, G73–77 17. Nabokina, S. M., Subramanian, V. S., and Said, H. M. (2003) Am. J. Physiol. Renal Physiol 284, F737–742 18. Zomot, E., Bendahan, A., Quick, M., Zhao, Y., Javitch, J. A., and Kanner, B. I. (2007) Nature 449, 726 –730 19. Eskandari, S., Loo, D. D., Dai, G., Levy, O., Wright, E. M., and Carrasco, N. (1997) J. Biol. Chem. 272, 27230 –27238 20. Hirayama, B. A., Loo, D. D., and Wright, E. M. (1997) J. Biol. Chem. 272, 2110 –2115 21. Vandenberg, R. J., Arriza, J. L., Amara, S. G., and Kavanaugh, M. P. (1995) J. Biol. Chem. 270, 17668 –17671 22. Panayotova-Heiermann, M., Loo, D. D., and Wright, E. M. (1995) J. Biol. Chem. 270, 27099 –27105 23. Wright, E. M., Loo, D. D., Turk, E., and Hirayama, B. A. (1996) Curr. Opin. Cell Biol. 8, 468 – 473 24. Longpre´, J. P., Gagnon, D. G., Coady, M. J., and Lapointe, J. Y. (2010) Biophys. J. 98, 231–239 25. Loo, D. D., Hirayama, B. A., Meinild, A. K., Chandy, G., Zeuthen, T., and Wright, E. M. (1999) J. Physiol. 518, 195–202 26. Mackenzie, B., Loo, D. D., and Wright, E. M. (1998) J. Membr. Biol. 162, 101–106 27. Yamashita, A., Singh, S. K., Kawate, T., Jin, Y., and Gouaux, E. (2005) Nature 437, 215–223 28. Shi, L., Quick, M., Zhao, Y., Weinstein, H., and Javitch, J. A. (2008) Mol. Cell 30, 667– 677 29. Diez-Sampedro, A., Hirayama, B. A., Osswald, C., Gorboulev, V., Baum-

VOLUME 286 • NUMBER 1 • JANUARY 7, 2011

Iodide Transport by hSMVT

30. 31. 32. 33. 34. 35.

garten, K., Volk, C., Wright, E. M., and Koepsell, H. (2003) Proc. Natl. Acad. Sci. U.S.A. 100, 11753–11758 Wright, E. M., and Turk, E. (2004) Pflugers Arch. 447, 510 –518 Quick, M., and Jung, H. (1998) Biochemistry 37, 13800 –13806 Singh, S. K., Yamashita, A., and Gouaux, E. (2007) Nature 448, 952–956 Singh, S. K., Piscitelli, C. L., Yamashita, A., and Gouaux, E. (2008) Science 322, 1655–1661 Abramson, J., Smirnova, I., Kasho, V., Verner, G., Kaback, H. R., and Iwata, S. (2003) Science 301, 610 – 615 Boudker, O., Ryan, R. M., Yernool, D., Shimamoto, K., and Gouaux, E.

JANUARY 7, 2011 • VOLUME 286 • NUMBER 1

(2007) Nature 445, 387–393 36. Ball, G. F. (2005) Vitamins in Foods: Analysis, Bioavailability, and Stability, Saint Lucie Press/Taylor & Francis G, Boca Raton, FL 37. Doha´n, O., De la Vieja, A., Paroder, V., Riedel, C., Artani, M., Reed, M., Ginter, C. S., and Carrasco, N. (2003) Endocr. Rev. 24, 48 –77 38. Nicola, J. P., Basquin, C., Portulano, C., Reyna-Neyra, A., Paroder, M., and Carrasco, N. (2009) Am. J. Physiol. Cell Physiol. 296, C654 – 662 39. Acland, J. D., and Illman, O. (1959) J. Physiol. 147, 260 –268 40. Pastan, I. (1957) Endocrinology 61, 93–97 41. Said, H. M. (1999) J. Nutr. 129, 490S– 493S

JOURNAL OF BIOLOGICAL CHEMISTRY

137