Anoctamin-1/TMEM16A is the major apical iodide channel of the thyrocyte

L. Twyffels,1,2 A. Strickaert,3 M. Virreira,4 C. Massart,3 J. Van Sande,3 C. Wauquier,1 R. Beauwens,4

J. E. Dumont,3 L. J. Galietta,6 A. Boom,4,5 and V. Kruys1,2

1Laboratoire de Biologie Moléculaire du Gène, Faculté des Sciences, Université libre de Bruxelles (ULB), Brussels, Belgium;2Center for Microscopy and Molecular Imaging, Université libre de Bruxelles (ULB), Brussels, Belgium; 3Institut deRecherche Interdisciplinaire en Biologie Humaine et Moléculaire, Université libre de Bruxelles (ULB), Brussels, Belgium;4Laboratoire de Physiologie Moléculaire et Cellulaire, Faculté de Médecine, Université libre de Bruxelles (ULB), Brussels,Belgium; 5Laboratoire d’Histologie, Histopathologie et Neuroanatomie, Faculté de Médecine, Université libre de Bruxelles(ULB), Brussels, Belgium; and 6Laboratory of Molecular Genetics, Istituto Giannina Gaslini, Genoa, Italy

Submitted 29 April 2014; accepted in final form 24 September 2014

Twyffels L, Strickaert A, Virreira M, Massart C, Van Sande J,Wauquier C, Beauwens R, Dumont JE, Galietta LJ, Boom A,Kruys V. Anoctamin-1/TMEM16A is the major apical iodide channelof the thyrocyte. Am J Physiol Cell Physiol 307: C1102–C1112, 2014.First published October 8, 2014; doi:10.1152/ajpcell.00126.2014.—Iodide is captured by thyrocytes through the Na�/I� symporter (NIS)before being released into the follicular lumen, where it is oxidizedand incorporated into thyroglobulin for the production of thyroidhormones. Several reports point to pendrin as a candidate protein foriodide export from thyroid cells into the follicular lumen. Here, weshow that a recently discovered Ca2�-activated anion channel,TMEM16A or anoctamin-1 (ANO1), also exports iodide from ratthyroid cell lines and from HEK 293T cells expressing human NIS andANO1. The Ano1 mRNA is expressed in PCCl3 and FRTL-5 ratthyroid cell lines, and this expression is stimulated by thyrotropin(TSH) in rat in vivo, leading to the accumulation of the ANO1 proteinat the apical membrane of thyroid follicles. Moreover, ANO1 prop-erties, i.e., activation by intracellular calcium (i.e., by ionomycin or byATP), low but positive affinity for pertechnetate, and nonrequirementfor chloride, better fit with the iodide release characteristics of PCCl3and FRTL-5 rat thyroid cell lines than the dissimilar properties ofpendrin. Most importantly, iodide release by PCCl3 and FRTL-5 cellsis efficiently blocked by T16Ainh-A01, an ANO1-specific inhibitor,and upon ANO1 knockdown by RNA interference. Finally, we showthat the T16Ainh-A01 inhibitor efficiently blocks ATP-induced iodideefflux from in vitro-cultured human thyrocytes. In conclusion, ourdata strongly suggest that ANO1 is responsible for most of the iodideefflux across the apical membrane of thyroid cells.

anoctamin-1; iodide release; pendrin; thyrocyte; TMEM16A

THE FUNCTION OF THE THYROCYTE is to take up iodide, oxidize it,and bind it to tyrosine residues of thyroglobulin to finallycouple these iodotyrosines into iodothyronines, the thyroidhormones (T3 and T4). The steps involved in this biosynthesisare the trapping of iodide in thyrocytes from the extrafollicularmedium by the Na�/I� symporter (NIS), its transport from thethyrocyte to the follicular lumen, its oxidation there intoiodotyrosine, and the coupling of iodotyrosines in thyroglob-ulin by thyroperoxidase. Pendrin has been proposed as theprotein mediating the step of iodide export from the cell intothe follicular lumen, i.e., of iodide crossing the thyrocyte’sapical membrane (3). However, several arguments indicate that

pendrin is not critical for this transport (38, 44) as micedeficient for pendrin are euthyroid (4, 6, 42) and patientsaffected by the Pendred syndrome, i.e., homozygotic pendrininactivation, may develop goiters in iodine-deficient conditionsbut usually not before the second decade of life (3). Anotherpathway for iodide transport across the apical membrane musttherefore be considered, and the definition of its identity andhormonal responsiveness constitutes the aim of the presentstudy. Most cells are equipped with ion channels permeable toanions. Among those, Ca2�-activated chloride channels havebeen identified in the 1980s on the basis of their Ca2� sensi-tivity and biophysical properties as defined by patch-clampexperiments. These channels play a large variety of rolesdepending on the cell type in which they are expressed, frommembrane excitability, olfactory transduction, photoreception,and transepithelial secretion, to preventing polyspermy (seeRef. 11 for review). However, their molecular identity re-mained elusive until 2008, when three groups independentlycloned the gene encoding the transmembrane protein calledTMEM16A or anoctamin-1 (ANO1) (5, 33, 46). This proteinwas observed to function as a Ca2�-activated chloride channel(26, 37) and to be widely expressed at the plasma membrane ofmany cells in liver, skeletal muscle, heart, pancreas, placenta,sensory neurons, kidney, airways epithelial cells, lung alveoli,and mammary and salivary glands where they concentrate atthe apical membrane (8). The functional properties of ANO1were found to be regulated by alternative splicing and somevariants were observed to be expressed in the thyroid (9). Asmost variants transport iodide better than chloride (i.e., with ahigher affinity; Ref. 8), we decided to investigate whetherANO1 might play a role in iodide export in thyroid. Thepresent observations demonstrate that in the rat and humaniodide release into the colloid space is carried out mainly byANO1.

MATERIALS AND METHODS

Materials

Enzymes were purchased from Invitrogen (Carlsbad, CA) andRoche (Penzberg, Germany). Oligonucleotides were purchased fromSigma-Aldrich (St. Louis, MO). Cell culture media were purchasedfrom Invitrogen. The ANO1 inhibitor T16Ainh-A01 was from Calbio-chem (cat 613551; San Diego, CA) (21). 125I as NaI (�17 Ci/mg; 629GBq) was purchased from Perkin Elmer (Zaventem, Belgium).

Cell Line Culture and Transfection

Human embryonic kidney 293T (HEK 293T) cells were grown inDulbecco’s modified Eagle’s medium Glutamax (Invitrogen) supple-

A. Boom and V. Kruys contributed equally to this work.Address for reprint requests and other correspondence: J. E. Dumont, Institut

de Recherche Interdisciplinaire en Biologie Humaine et Moléculaire(IRIBHM), Université libre de Bruxelles, Campus Erasme, 808 Route deLennik, Building C, Brussels, Belgium (e-mail: jedumont@ulb.ac.be).

Am J Physiol Cell Physiol 307: C1102–C1112, 2014.First published October 8, 2014; doi:10.1152/ajpcell.00126.2014.

0363-6143/14 Copyright © 2014 the American Physiological Society http://www.ajpcell.orgC1102

mented with 10% fetal bovine serum, penicillin (50 U/ml), andstreptomycin (50 �g/ml). Rat PCCl3 (10) and FRTL-5 (1) cells weregrown in Coon’s modified Ham’s F-12 medium supplemented with5% decomplemented fetal bovine serum (Invitrogen), bovine thyro-tropin (TSH; 1 mU/ml), transferrin (5 �g/ml), insulin (1 �g/ml),penicillin (100 U/ml), streptomycin (100 �g/ml), and fungizone (2.5�g/ml).

For iodide uptake/release experiments and Western blot analyses,cells were plated at a density of 105 cells/well (HEK 293T cells) or5 � 105 cells/well (PCCl3 and FRTL-5 cells) in six-well plates. HEK293T cells were transfected the next day using 2 �g of plasmid DNAmixed with 6 �l of Fugene 6 (Roche) according to the manufacturer’sinstructions. Typically, cells were transfected with 1 �g of NIS-pCDNA3 and 1 �g of a plasmid encoding either GFP (pEGFP-C2;Clontech, Mountain View, CA) or ANO1. For pendrin expression, theplasmid DNA amount was lowered to 100 ng and supplemented with900 ng of the pEGFP-C2 plasmid.

Human Primary Thyrocyte Culture

Human thyroid tissue was obtained from patients undergoingthyroidectomy for solitary cold nodules or multinodular pathology.The protocol was approved by the ethics committee of the Erasmehospital. Only the healthy normal looking, nonnodular tissue was usedwithin 30–45 min of surgical removal. The tissue was digested intofollicles and grown in primary culture as described previously (29)except that forskolin (5 �M) was added to the culture medium the last3 days to support differentiation.

RT-PCR Analysis of Ano1 Expression in Tissues

RNA extraction from tissue and subsequent classical PCR analysiswere performed as described in Best et al. (2) except that reversetranscription was performed using iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA). For the comparison of Ano1 expression in controland methimazole-treated rats, cDNA was prepared from thyroid tissueas described above, and a quantitative real-time PCR was performedon a CFX 96 Real-Time System C1000 Thermal Cycler (Bio-Rad)using iQ SYBR Green Supermix (Bio-Rad). The reference gene usedwas Hprt. Primer sequences are available upon request.

Analysis of Rat Ano1 Isoforms

Alternatively spliced Ano1 transcripts expressed in PCCl3 cellsand rat thyroid were detected by RT-PCR. Conditions for PCRwere the following: 95°C for 5 min for the initial denaturation;95°C for 15 s, 58°C for 30 s, and 72°C for 45 s for 38 cycles; and72°C for 10 min for the final extension. We used the followingprimers: for exon 6b, forward: 5=GAACAACGTGCACCAAG-GCC3= and reverse: 5=GCTCCAAGCCAGGCAAAGTAC3=; forexon 13, forward: 5=GAAACGGAAGCAGATGAGAC3= and re-verse: 5=GGCTTCATACTCTGCTCTGG3=; and for exon 15, for-ward: 5=GTATGAAGCCAGAGTCTTAG3= and reverse: 5=ATA-AACGTAGTCACCAGGCCG3=. The amplified products were re-solved on 2% agarose gels. Expected sizes of PCR productsamplified from 6b� and 6b� transcripts: 310 and 244 bp, respec-tively; 13� and 13� PCR products: 100 and 88 bp, respectively;and 15� and 15� PCR products: 553 and 475 bp, respectively.

Protein Extraction

HEK 293T cells were collected, washed twice in ice-cold PBS, andlysed in EBC buffer (50 mM Tris·HCl, pH 8.0, 100 mM NaCl, 0.2%NP40, and cOmplete EDTA-free protease inhibitor cocktail; Roche)for 30 min at 4°C. Thyroid tissues were lysed in RIPA buffer (50 mMTris·HCl pH 7.4, 150 mM NaCl, 01% Nonidet P-40, and 0.25%sodium deoxycholate) containing phosphatase and proteases inhibi-tors. The lysates were centrifuged at 10,000 rpm for 10 min, and thesupernatants were saved at �80°C.

Western Blotting

Ten micrograms of total protein extract from HEK 293T cells or 40�g from thyroid tissue were electrophoresed on a 7.5% polyacryl-amide gel and then transferred onto a nitrocellulose membrane forWestern blot analysis. The membrane was blocked with PBST con-taining 2–5% nonfat dry milk, incubated overnight at 4°C with theanti-ANO1 antibody sc-69343 (1/500–1/100 dilution), rinsed exten-sively with PBST, and incubated for 1 h at room temperature with thedonkey anti-goat IgG-HRP sc-2020 (1/25000–1/2,000 dilution; SantaCruz Biotechnology). The signals were detected by chemilumines-cence using either the SuperSignal West Pico detection kit (ThermoFisher Scientific, Waltham, MA) or the Western Lightning Plus-ECLdetection kit (Perkin Elmer, Waltham, MA).

Detection of ANO1 by Immunofluorescence Staining of Rat ThyroidSections

Ethics statement. Adult male Wistar rats were housed and usedfollowing the rules of the Belgian Regulations for Animal Care, withapproval of the Ethics Committee of the School of Medicine of theUniversité Libre de Bruxelles.

Sample preparation. Fourteen-week-old male Wistar rats receivedor not methimazole for 2 wk (200 mg/l in drinking water). They werethen anesthetized with ketamine hydrochloride and xylazine, and theywere perfused intracardiacally with a fresh solution of 4% (wt/vol)paraformaldehyde in 0.1 M phosphate buffer at pH 7.4. Thyroids werequickly dissected and further fixed by overnight immersion in 4%(wt/vol) paraformaldehyde in 0.1 M phosphate buffer at pH 7.4. Thetissue was then transferred to successive graded sucrose solutions (10,20, and 30%, overnight each) and finally embedded in Tissue-TekOCT compound, snap-frozen in cold 2-methylbutane, and stored at atemperature of �80°C. Cryosections (10 �m) were cut on a cryostat(Leitz), mounted on slides coated with 0.1% poly-L-lysine (Sigma),and stored at �20°C until use. Hematoxylin and eosin staining wasroutinely used to evaluate the sections before immunostaining.

Immunofluorescence. Immunofluorescence was carried out at roomtemperature. The slides were dropped 30 min in methanol containing0.3% hydrogen peroxide to block endogenous peroxidase activity.They were then washed in water, preincubated 10 min in normalrabbit serum, and incubated overnight at room temperature withanti-ANO1 sc-69343 antibody in Tris-buffered saline (TBS) contain-ing 1% of normal rabbit serum. Thereafter, the slides were washed inTBS and incubated for 30 min with a secondary goat anti-rabbitbiotinylated antibody (1/300 dilution; Vector), rinsed in TBS, andincubated for 1 h with an avidin-biotin complex conjugated withperoxidase. The slides were washed again in TBS and incubated for10 min with Tyramide Alexa350 (1/100 dilution; Molecular Probes).After being rinsed twice in TBS, the slides were mounted in Gelvatol/DABCO aqueous medium. Sections were examined with a ZeissAxioplan microscope, and images were acquired using an AxioCamHRc camera. The ability of the sc-69343 antibody to recognize ANO1in immunofluorescence experiments was validated by showing acolocalization of anti-Flag and anti-ANO1 signals at the membrane ofHEK 293T cells transfected to express Flag-ANO1 (data not shown).

DNA Constructs

The NIS-pcDNA3 was previously described (40a). The pcDNA3DNA construct encoding mouse pendrin was kindly provided by D.Eladari (Centre de Recherche Cardio-vasculaire, Inserm U970,Équipe 123, Paris, France). PcDNA3.1 constructs encoding humanANO1 ac and abc isoforms were previously described (9). Weexpressed the abc isoform unless stated otherwise. Flag-tagged ANO1(abc) construct was generated by PCR amplification of Ano1 (abc)open reading frame (ORF) using a reverse primer encoding Flagepitope to generate the ORF with an in-frame Flag sequence at its

C1103ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

3=-end. The resulting PCR product was cloned in the XhoI site of thepcDNA3 plasmid. The DNA construct was subsequently sequenced.

Iodide Uptake and Release from Cells

Seventy-two hours after transfection, HEK 293T cells were rinsedwith Krebs-Ringer HEPES buffer (KRH) and incubated for 45 min in1 ml of KRH supplemented with 125I� (Perkin Elmer; 1 �Ci/ml,10�7M KI) at 37°C. The wells were then rinsed twice with KRH at22°C and further incubated at 22°C in 1 ml of KRH supplementedwith 10�7 M KI, 1 mM ClO4

� (to suppress iodide reuptake), and theagent (e.g., ionomycin) and/or the condition (e.g., chloride-free me-dium) under study. The six-well plates were incubated in an open airincubator with mild shaking. Fifty microliters of medium were with-drawn after 1, 2, 3, 5, 10, and 30 min. The cells were then rinsed twicewith 1 ml of KRH at 0°C and lysed in 1 ml of 1 M NaOH. Themedium aliquots and the cells were counted in a �-counter (PerkinElmer-Wallac Wizard 1470 auto � counter). The release at each timewas expressed as percentage of total uptake, i.e., 125I released in themedium plus the amount remaining in the cells. The uptake wasalways checked in parallel wells in which cells were incubated with125I�, 10�7 M KI, plus or minus 1 mM ClO4

�, for the duration of cellloading, i.e., 45 min. Only preparations with a cell-to-medium ratioover 6 were used. The cells were then rinsed twice in KRH at 0°C andcounted. In chloride deprived medium, KCl, NaCl, and CaCl2 werereplaced by gluconate salts of K, Na, and Ca.

The test agent (e.g., ionomycin) or test medium (i.e., chloridedeprived) was present during the release incubation only, except forT16Ainh-A01, the ANO-1 inhibitor, which was added during the 125Iloading and release incubations. Ionomycin and T16Ainh-A01 weredissolved in DMSO (final DMSO concentration in cell culture me-dium: 0.1%). Control wells analyzed in parallel with ionomycin- orT16Ainh-A01-treated wells were supplemented with 0.1% DMSO tobe fully comparable. The experimental protocol was exactly the samefor PCCl3 and FRTL-5 cells except that the cells were not transfectedand that the experiments were performed 2 days after plating.

Generation of PCCl3 Stably Expressing shRNA

The DNA oligos encoding shRNA sequences were designed andcloned into the expression vector pLV-H1-EF1�-puro using the singleoligonucleotide RNAi technology developed by Biosettia (San Diego,CA). A shRNA targeting an irrelevant gene (-galactosidase) was used asa negative control (control shRNA). The oligonucleotide sequences werethe following: for control shRNA, AAAA-GCAGTTATCTGGAAGAT-CAGG-TTGGATCCAA-CCTGATCTTCCAGATAACTGC; for pen-drin shRNA, AAAA-AGGAATTAAACGATCGATTTA-TTGGAT-CCAA-TAAATCGATCGTTTAATTCCT; and for Ano1 shRNA,AAAA-AATCCATGGTGTCGGGTTTG-TTGGATCCAA-CAAAC-CCGACACCATGGATT. All plasmids were verified by DNA se-

quencing. The production of the lentiviruses was performed in HEK293T cells by cotransfection of the lentiviral constructs with psPAX2and pMD2.G plasmids (Addgene) according to the manufacturer’sinstructions. PCCl3 lentiviral transduction was performed accordingto the following procedure. Fifty thousand cells were seeded insix-well/plates on day 1. On day 2, the culture medium was removedand cells were incubated with 1 ml of viral suspension complementedby 1 ml of PCCl3 culture medium and polybrene (20 �g/ml, finalconcentration) for 8 h. The medium was then replaced with completePCCl3 culture medium. From day 4, transduced cells were selected byincubation with puromycin (1 �g/ml) for 2 wk. Resistant cells werepooled and knockdown efficiency was determined by quantitativeRT-PCR as described below.

Quantitative RT-PCR Analysis of shRNA-Mediated ExpressionKnockdown in PCCl3 Cells

Total RNA was isolated from shRNA-expressing PCCl3 cells usingTri reagent (Thermo Fisher Scientific), and reverse transcriptions andquantitative PCR analyses were performed as described by Twyffelset al. (39). The reference gene used was GAPDH. Primer sequencesare available upon request.

RESULTS

Expression of ANO1 in Rat Tissues and in PCCl3 Cells

Expression studies in murine tissues indicated that Ano1 ispredominantly expressed in epithelia like trachea, pancreas,colon, salivary gland, prostate tissue, and the thyroid (32).Ano1 primary transcript undergoes alternative splicing, thusresulting in the generation of multiple isoforms. Three alter-native exons, 6b, 13, and 15, coding for segments of 22(segment b), 4 (segment c), and 26 (segment c) amino acids,respectively, are differently spliced in human organs (9, 40).Functional studies on ANO1 isoforms indicated that the inclu-sion/exclusion of specific exons influenced ANO1 channelproperties (9). In human thyroid, the predominantly expressedANO1 transcript contains exons 6b and 13 but lacks exon 15 (9,40). Here, we analyzed the expression of Ano1 in rat tissues aswell as in the rat thyroid PCCl3 cell line by RT-PCR. Asshown in Fig. 1A, Ano1 is expressed in several tissues includ-ing the thyroid, the pancreas, and the parotid and submaxillaryglands. These results are in accordance with a previous expres-sion profiling of Ano1 in mouse (32). In PCCl3 cells and ratthyroid tissue, the Ano1 gene is expressed as alternativelyspliced mRNAs containing or not exon 6b in various ratios(6b� 6b� in PCCl3, 6b� � 6b� in rat thyroid). All

Fig. 1. Expression of anoctamin-1 (Ano1) inrat tissues and in the rat thyroid PCCl3 cellline. A: detection of Ano1 transcripts in rattissues by RT-PCR. B: comparative analysisof Ano1 alternatively spliced forms in PCCl3cells and in rat thyroid tissue by RT-PCR.

C1104 ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

transcripts contained exon 13 but lacked exon 15 (Fig. 1B). Thecloning and sequencing of Ano1 cDNA ORF obtained fromPCCl3 and rat thyroid confirmed the alternative inclusion ofexon 6b and the presence and absence of exons 13 and 15,respectively (data not shown). Of note, the peptide sequenceencoded by rat Ano1 6b exon was identical to the one encodedby human ANO1 6b exon, contrarily to the one deduced fromthe current National Center for Biotechnology Informationreference sequence for rat Ano1 mRNA (accession no.:NM_001107564).

Expression and Localization of ANO1 in Rat ThyroidFollicles

To evaluate the interaction between thyroid metabolism andANO1 localization and expression, we treated rats with theantithyroid drug methimazole, which inhibits thyroid hormonesynthesis and consequently increases serum TSH. We observedthat this classical stimulating treatment increased Ano1 expres-sion at the mRNA (Fig. 2A) and protein (Fig. 2B) levels.

We then immunostained thyroid tissues from rats treated ornot with methimazole for ANO1. In the thyroids from controlrats, the signal was generally low (Fig. 3A). However, adiscrete but clear-cut labeling was observed in some follicles;in a few of them, ANO1 was detected at the apical membrane

of rat thyrocytes (Fig. 3B). As expected from our previousresults, the ANO1 signal was considerably stronger in thethyroid from methimazole-treated rats. Moreover, it was lo-cated at the apical side of the thyrocytes in almost all follicles(Fig. 3, C and D). This correlates with increased serum TSHlevel (20). ANO1 localization in rat thyroid is therefore com-patible with a role in iodide export from thyrocytes to thefollicle lumen.

Release of Iodide Is Induced by Ectopic Expression of ANO1and Pendrin in HEK 293T Cells

In polarized thyroid cells, iodide release takes place at theapical but not at the basal membrane, as shown in porcinethyroid cells in primary culture (23). In unpolarized cells,influx and release take place at the same membrane sites, thusestablishing a steady state. We tested the ability of ANO1 tomediate iodide release and compared it to that of pendrin in aheterologous expression setup. HEK 293T cells were trans-fected with plasmid DNA constructs encoding the sodium/iodide symporter NIS in combination with ANO1 (abc) iso-form or pendrin or, as a control, with GFP. We then loadedthese cells with radioiodide and measured the subsequentiodide release from these cells. As shown in Fig. 4, HEK 293Tcells expressing GFP displayed a low ability to release iodide.

Fig. 2. The classical stimulating drug methimazoleincreases Ano1 expression in rat thyroid. A: quan-titative real-time PCR showing the effect of amethimazole treatment on the expression of theAno1 mRNA in rat thyroid tissue. Bars representthe average value � SE over 3 independent ex-periments. B: Western blot showing the effect of amethimazole treatment on the expression of theANO1 protein in rat thyroid tissue.

Fig. 3. Expression and localization of ANO1in rat thyroid follicles. A–D: rat thyroid sec-tions were immunostained for ANO1 (blue)and cell nuclei were counterstained with pro-pidium iodide (red). A merged view of the 2signals is presented except for B. A and B:representative pictures of ANO1 immunode-tection in thyroid follicles of untreated rat. Adiscrete labeling is observed in some folli-cles (A); only a few follicles express ANO1at the level of the apical side of thyrocytes(B, arrows). C and D: representative picturesof ANO1 immunodetection in thyroid of ratstreated with methimazole (200 mg/l) indrinking water during 2 wk. Strong fluores-cence intensity is detected at the apical sideof almost all thyroid follicles. No significantsignal was observed when the anti-ANO1antibody was omitted during sample prepa-ration (data not shown) suggesting that theweak signal observed in the interstitium isnot artifactual, probably reflecting vascularANO1.

C1105ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

By contrast, expression of ANO1 (abc) or pendrin stronglyincreased iodide release from HEK 293T cells. PCCl3 cellsalso showed a significant iodide release.

Comparative analysis of ANO1 isoforms (abc and ac) inthe same experimental setup (Fig. 5A) did not reveal qual-itative differences in the ability of these isoforms to mediateiodide release (Fig. 5B). In conclusion, both ANO1 andpendrin allow the release of iodide at physiological concen-trations of I� and Cl�.

Iodide Release Is Stimulated by Increased Intracellular Ca2�

Concentration in PCCl3 Cells and in ANO1-ExpressingHEK 293T Cells but not in Pendrin-Expressing HEK 293TCells

Ca2� ionophore enhances the release of iodide from dogthyroid cells (28) as well as in kidney proximal tubular cellsunder the control of purinergic stimulation (7). We thereforestudied iodide release by pendrin- and ANO1-expressing HEK293T cells in the presence or the absence of the Ca2� iono-phore ionomycin. While ionomycin increased iodide releasemediated by ANO1 isoforms, no change was observed inpendrin-expressing HEK 293T cells (Figs. 5 and 6). The iodidechannel expressed in PCCl3 cells shared ANO1 activationproperties, i.e., the Ca2� ionophore stimulated iodide releasefrom these cells, as previously reported for dog thyroid cells(28). Moreover, this enhanced iodide efflux is reproduced upon

purinergic stimulation (Fig. 6). Thus, while increase of intra-cellular calcium concentration does not affect pendrin-medi-ated iodide efflux, it greatly stimulates iodide release fromANO1-expressing HEK 293T cells, PCCl3 cells, and dogthyrocytes. Of note, Pesce et al. (27) showed weak activationof iodide loss by TSH but not by forskolin in PCCl3 cells,which suggests a TSH effect through the weakly respondingphospholipase C-PIP2-Ca2� cascade in this model.

Ionomycin Stimulates 99mTcO4� Release from

ANO1-Expressing HEK 293T Cells and from PCCl3 Cells

99mTcO4� is often used as a substitute for radioiodide in

uptake experiments, and 99mTcO4� is transported into thyroid

cells by the iodide transporter NIS (22, 40a). We testedwhether this absence of specificity also applied to the iodidechannels. In NIS/GFP-expressing HEK 293T cells as well as inPCCl3 cells, 99mTcO4

� was better concentrated and releasedthan radioiodide. On the other hand, there was no increase inbasal release from HEK 293T cells also expressing ANO1 orpendrin. In NIS/GFP and in NIS/pendrin-expressing HEK293T cells, this release was not enhanced by ionomycin.However, ionomycin induced an increased release of 99mTcO4

�

from NIS/ANO1 HEK 293T and from PCCl3 cells (Fig. 7).Thus 99mTcO4

� is not transported by ANO1, except uponstimulation of the cells with Ca2� ionophore. Moreover, thesimilar capacity of PCCl3 cells to transport 99mTcO4

� uponincreased Ca2� intracellular concentration supports the pres-ence of active ANO1 channels in these cells.

The Release of Iodide Occurs in the Absence of Chloride inANO1-Expressing HEK 293T Cells and in FRTL-5 andPCCl3 Cells but not in Pendrin-Expressing HEK 293T Cells

As pendrin exchanges iodide (or bicarbonate) for chloride,pendrin-mediated release of iodide is expected to be drasticallydecreased in the absence of chloride in the medium. To assessANO1 and pendrin iodide transport activity in the presence orthe absence of Cl� ions, transfected HEK 293T cells wereincubated in normal or chloride-free medium and iodide re-lease was measured after prior loading incubation in normalmedium. We observed that chloride deprivation markedlydecreased iodide release from pendrin-expressing HEK 293Tcells but did not affect iodide release from ANO1-expressingHEK 293T cells (Fig. 8). The same conditions were thenapplied to PCCl3 and FRTL-5 cells. As shown in Fig. 8,

Fig. 4. Comparison of the release of radioiodide from 125I-preloaded PCCl3cells (Œ) or HEK 293T cells transfected to express Na�/I� symporter (NIS) incombination with GFP (�) or ANO1 (Œ) or pendrin (�) The data are repre-sentative of more than 5 independent experiments.

Fig. 5. Comparison of iodide release by ANO1 abcand ac isoforms. A: ectopic expression of ANO1 acand abc isoforms in transiently transfected HEK293T cells. Detection of ANO1 protein isoformswas performed by Western blot (WB)using 10 �g ofprotein extract and anti-ANO1 antibody. Negativecontrol corresponds to untransfected HEK 293Tcells. Gel loading was controlled by unspecific la-beling of a HEK 293T endogenous protein. B: com-parison of the release of radioiodide from 125I-preloaded HEK 293T cell transfected to express NISin combination with ANO1 (ac) (Œ) or ANO1 (abc)(�) and incubated in the presence (�, �) or absence(Œ, �) of ionomycin (1 �M). ac And abc refer to theamino acid segments a, b, or c. The data are repre-sentative of at least 3 independent experiments.

C1106 ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

bottom right, chloride deprivation did not significantly modifyiodide release from either cell line. These results indicate thatANO1, in contrast to pendrin, does not require chloride anionsto promote iodide release and that the channel mediating basaliodide release from rat thyroid cells shares this characteristicwith ANO1.

The T16Ainh-A01 Inhibitor Reduces Iodide Release fromPCCl3 and FRTL-5 Cells

Novel ANO1 inhibitors have been recently developed, the mostpotent one being the aminophenylthiazole T16Ainh-A01 (21). Wefirst tested the specificity of T16Ainh-A01 to block iodide release

Fig. 6. Stimulation of 125I release from preloadedHEK 293T cells transfected with NIS in combinationwith GFP or pendrin or ANO1 or from PCCl3 cells.Ionomycin concentration was 0.5 �M for HEK 293Tcells and 0.2 �M for PCCl3 cells. ATP concentra-tion was 5 �M. Control values (Œ), ionomycin-treated cells (�), ATP-treated cells (�) are shown.The data are representative of at least 3 independentexperiments.

Fig. 7. Release of 125I (�, Œ) or 99mTc (�, �) frompreloaded HEK 293T cells transfected with NIS incombination with GFP or pendrin or ANO1 or fromPCCl3 cells. Cells were incubated in the presence ofionomycin (�, �) or not (Œ, �). Ionomycin concen-tration was 1 �M for HEK 293T cells and 0.5 �Mfor PCCl3 cells. The data are representative of atleast 3 independent experiments.

C1107ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

from ANO1-expressing HEK 293T cells compared with GFP- andpendrin- expressing HEK 293T cells. As shown in Fig. 9,T16Ainh-A01 strongly inhibited iodide release from ANO1-ex-pressing HEK 293T cells but not from pendrin-expressing HEK293T cells, thereby confirming T16Ainh-A01 specificity towards

ANO1 transport activity. We then tested T16Ainh-A01 inhibitoryactivity on iodide release by PCCl3 and FRTL-5 cells and showedan inhibition of basal iodide release and a complete inhibition ofCa2� ionophore-stimulated release. The IC50 of the inhibitor is ofthe order of 30 �M.

Fig. 8. Effect of chloride on 125I release from pre-loaded HEK 293T cells transfected with NIS incombination with GFP, pendrin or ANO1 or fromPCCl3 cells and FRTL-5 cells. Normal medium: Œ,�, o; chloride-deprived medium: �, �, Œ; HEK293T: �, Œ; FRTL-5 cells: �, �; PCCl3 cells: Œ, o.The data are representative of at least 3 independentexperiments.

Fig. 9. Effect of the ANO1 inhibitor T16Ainh-A01on 125I release from HEK 293T cells transfectedwith NIS in combination with pendrin or Ano1 orfrom PCCl3 cells or from FRTL-5 cells. �, �:Ionomycin-stimulated cells (1 �M for HEK 293Tcells; 0.2 �M for PCCl3 and FRTL-5 cells); �, �:cells incubated with T16Ainh-A01 (100 �M). Thedata are representative of at least 3 independentexperiments.

C1108 ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

Knockdown of ANO1 Expression Downmodulates IodideEfflux in Ionomycin- and ATP-Exposed PCCl3 Cells

To confirm that ANO1 is the Ca2�-dependent iodide channelin thyroid cells, we knocked down ANO1 synthesis in PCCl3cells by stable expression of interfering shRNA. In parallel, weestablished PCCl3 cell lines expressing a control shRNA or ashRNA targeting pendrin. As shown in Fig. 10A, shRNAstargeting ANO1 and pendrin led to a three- and sixfold de-crease in ANO1 and pendrin mRNA, respectively, comparedwith their levels in PCCl3 expressing control shRNA. WhileANO1 knockdown did not modify iodide efflux in basal con-ditions (Fig. 10B, left), it markedly reduced iodide efflux inionomycin- and ATP-exposed cells (Fig. 10B, middle andright). In contrast, downmodulation of pendrin expression didnot affect iodide efflux in the three conditions. All together,these results confirm the primary role of ANO1 in iodide effluxfrom stimulated thyroid cells.

The T16Ainh-A01 Inhibitor Reduces Iodide Release fromHuman Thyrocytes

We then examined the role of ANO1 in iodide efflux byhuman thyrocytes. First, we validated that Ano1 transcript andprotein are expressed in human thyroid (Fig. 11, A and B).Second, we measured iodide efflux from primary human thy-rocytes exposed or not to ATP in the presence or the absenceof the T16Ainh-A01 inhibitor. As shown in Fig. 11C, ATPstrongly increased iodide efflux as in PCCl3 cells and thisincrease was suppressed by T16Ainh-A01, thereby confirmingthe conserved role of ANO1 in human thyroid.

DISCUSSION

To be oxidized and bound to the tyrosines of thyroglobulinand for the following synthesis of thyroid hormones, iodidemust be first transported into the lumen of the thyroidfollicle (34, 44). This uptake requires a first active transportfrom the medium to the cytosol of the follicular cell,mediated by the basal membrane Na�/I� symporter NIS,and then a passive release to the follicular lumen through theapical membrane driven by a favorable electrochemical gra-

dient. The latter has been ascribed to the apical Cl�/HCO3�(I�)

exchanger pendrin. However, there are several argumentsagainst an exclusive role of pendrin: 1) in human congenitalhomozygotic defects of pendrin (the Pendred syndrome), hy-pothyroidism only appears in iodine-deficient areas (18, 25);and 2) the inactivation of pendrin gene in mouse does notinduce a thyroid-deficient phenotype (6). The question there-fore arises whether other membrane proteins fulfill this role.ANO1 is expressed within the apical membrane of severalchloride secreting epithelia often in parallel with another chlo-ride channel, the cystic fibrosis transmembrane conductanceregulator (CFTR); they both constitute major rate-limitingsteps for secretions controlled by increased cytosolic Ca2� forANO1 and by the cAMP cascade for the CFTR (10, 34). Bothchannels can also mediate HCO3

� transport; yet, only ANO1exhibits a greater selectivity for iodide than chloride and it hasrecently been suggested as a potential candidate mediatingiodide exit into the follicular space (41). Expression of ANO1gene is decreased in some papillary thyroid carcinoma and in11 tested anaplastic carcinomas (15). It is slightly increased infive thyroids of nonimmune hyperthyroidism and is markedlyupregulated in autonomous adenomas (16). Expression of pen-drin/Slc26a4 gene is decreased in most papillary and in 11anaplastic carcinomas investigated (15). It is slightly increasedin autonomous adenomas (16). These expression patterns areconsistent with a role of both proteins in functional differen-tiation.

In this work, we show that Ano1, which has been recentlycloned and characterized, is expressed in human and rat thyroidprimary cells and in PCCl3 and FRTL-5 rat thyroid cell lines.Among the alternatively spliced forms, both the abc and the acforms are expressed. The ac protein form, which is moresensitive to calcium than the abc form (8, 9), is equally if notmore expressed than the abc form in FRTL-5 and PCCl3 cells.In the thyroid of methimazole-treated rats in which serum TSHis elevated, Ano1 expression is increased and the ANO1protein is detected at the apical membrane of most follicles.The higher permeability of ANO1 for iodide vs. chloride isconsistent with the release observed here even though intracel-lular chloride is present at much higher concentration than

Fig. 10. Effect of Ano1 and pendrin knockdown on the release of 125I from PCCl3 cells. A: validation of Ano1 and pendrin knockdown in PCCl3 cells stablytransfected to express shRNAs targeting Ano1 or pendrin mRNAs. Total RNA was extracted and the levels of pendrin, and Ano1 mRNAs were measured byquantitative RT-PCR. Bars show the average value � SD. over 3 independent experiments. B: comparison of 125I release from PCCl3 expressing control shRNA(Œ), Ano1 shRNA (�), or pendrin shRNA (�). Left: control values (untreated cells). Middle: cells incubated with ionomycin (0.1 �M). Right: cells incubatedwith ATP (10 �M). The data are representative of at least 3 independent experiments.

C1109ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

iodide (9). Both pendrin and ANO1 confer to NIS-expressingHEK 293T cells the property of iodide release. They bothdiffer in specificity from the NIS transporter, which takesalmost exclusively I� and 99mTcO4

�, while they transport manyother anions including Cl� and HCO3

�. The two mainly ex-pressed isoforms of ANO1 (ac and abc isoforms) transportiodide. Moreover, our work suggests that iodide release by thethyroid PCCl3 and FRTL-5 cells reproduces the characteristicsof ANO1 but not those of pendrin. Whereas release of iodideby pendrin-expressing cells is inhibited in a chloride-freemedium, as expected for a chloride-anion exchanger, it is notdownmodulated in ANO1-expressing cells nor is it in thyroidFRTL-5 and PCCl3 cells. On the other hand, iodide release isactivated by increased intracellular Ca2� concentration inANO1-expressing HEK 293T and in PCCl3 cells but not inpendrin-expressing HEK 293T cells. Moreover, the more phys-iological purinergic stimulation of PCCl3 cells reproduces themarked increase in iodide efflux observed upon ionomycinexposure. It is worth noting that this property is one of themajor characteristics of dog thyroid cells in primary culture(28). The increased release of 99mTcO4

� from the Ca2� iono-phore-stimulated ANO1-expressing HEK 293T and PCCl3cells is consistent with the very low affinity but sizeable effectof ClO4

�, the chemical analog of TcO4�, on the high affinity

iodide channel from bovine thyroid membrane vesicles (13,14). A stimulatory effect of Ca2� ionophore on iodide effluxhad already been observed in the FRTL-5 rat thyroid cell line(43). Most importantly, T16Ainh-A01 abolished nearly com-pletely iodide release from ANO1-expressing HEK 293T,

FRTL-5, and PCCl3 cells but not from pendrin-expressingHEK 293T cells. Moreover, shRNA-mediated knockdown ofANO1 expression strongly decreased ionomycin and ATP-stimulated iodide efflux in PCCl3 cells. Several reports furthersupport the role of ANO1 as a thyroid iodide channel. Indeed,both the iodide channels of bovine thyroid membranes andANO1 are very sensitive to SCN� inhibition (6, 12). SCN�

was also shown to inhibit iodide transport from cell to lumenin rat thyroid in vivo (45). Thus our experimental evidence andthe literature strongly support a role for ANO1, but not forpendrin, in PCCl3 and FRTL-5 cell iodide release and suggestsuch a role in rat thyroid cells. The location of ANO1 proteinat the apical border of rat thyroid cells in vivo further supportsthis conclusion.

The pathway of iodide exit into the follicular space of humanthyroid has not been solved so far. Here, we show that ANO1is also expressed in human thyrocytes and is involved in iodideefflux from these cells under conditions of increased intracel-lular Ca2� (Fig. 11), thereby highlighting the conserved role ofANO1 in human. We confirm that pendrin is also able to exportiodide from the cells as previously reported (12, 18, 36).Moreover, pendrin is present at the apex of thyroid cells andiodide release from a thyroid cell line derived from a Pendredpatient is decreased (25). However, neither knockdown ofpendrin expression nor chloride deprivation in the cell culturemedium modified iodide efflux both under basal and activatedconditions in PCCl3 cells. Yet, these experiments do notcompletely exclude a possible role of pendrin or of otherchannels in iodide efflux under basal conditions in these and

Fig. 11. Human Ano1: expression and functionalityin iodide efflux from thyrocytes. A: detection ofAno1 transcripts in human tissues including thy-roid by RT-PCR. B: detection of ANO1 protein inhuman thyroid and in thyroid from methimazole(MMI)-treated rats by Western blot. C: effect ofthe ANO1 inhibitor T16Ainh-A01 on 125I effluxfrom resting and ATP-stimulated human thyro-cytes in primary culture. Resting cells: Œ, �; ATP(1 mM)-stimulated cells: �, �. Left: control values(without T16Ainh-A01). Right: cells incubated withT16Ainh-A01 (100 �M).

C1110 ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

other species. As a matter of fact, the relative role of thesemembrane proteins may vary between species as does thesecond messenger at play during TSH stimulation: cAMPincreases iodide release from pig thyroid cells in primarycultures (24, 28) while increase in intracellular Ca2� plays theprominent role in PCCl3 and FRTL-5 rat thyroid cells asshown here and in dog and human thyroid. Many such bio-chemical interspecies differences have been observed amongmammalian thyroids (19, 35).

Taken together, the present data support the conclusion thatTmem16A/Ano1 is expressed at the apex of the thyrocyte andthat ANO1 constitutes the thyrocyte apical iodide channelaccounting for stimulated iodide release under conditions ofincreased intracellular Ca2�.

NOTE ADDED IN PROOF

While this work was under revision, an article suggesting a role ofANO1 in iodide release from rat thyroid cells, based on other evidence, waspublished (17).

GRANTS

This work was supported by the Fund for Medical Scientific Research(Belgium, Grant 2.4.511.00.F), the “Actions de Recherches Concertées”(Grant AV.06/11-345), the Fonds Brachet, the Fonds Van Buuren, the FondsJ. P. Naets of the King Baudouin Foundation, and the Fonds Alphonse et JeanForton, Brussels, Belgium. The Center for Microscopy and Molecular Imagingis supported by the European Regional Development Fund (FEDER) and theWalloon Region.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: L.T., J.V.S., R.B., J.D., and V.K. conception anddesign of research; L.T., A.S., M.V., C.M., J.V.S., C.W., A.W.B., and V.K.performed experiments; L.T., M.V., C.M., J.V.S., R.B., J.D., A.W.B., and V.K.analyzed data; L.T., A.S., M.V., J.V.S., R.B., J.D., A.W.B., and V.K. inter-preted results of experiments; L.T., M.V., C.M., J.V.S., and A.W.B. preparedfigures; L.T., J.V.S., R.B., J.D., and V.K. drafted manuscript; L.T., J.V.S.,R.B., J.D., L.J.V.G., and V.K. edited and revised manuscript; L.T., A.S., M.V.,C.M., J.V.S., C.W., R.B., J.D., L.J.V.G., A.W.B., and V.K. approved finalversion of manuscript.

REFERENCES

1. Ambesi-Impiombato FS, Parks LA, Coon HG. Culture of hormone-dependent functional epithelial cells from rat thyroids. Proc Natl Acad SciUSA 77: 3455–3459, 1980.

2. Best L, Brown PD, Yates AP, Perret J, Virreira M, Beauwens R,Malaisse WJ, Sener A, Delporte C. Contrasting effects of glycerol andurea transport on rat pancreatic beta-cell function. Cell Physiol Biochem23: 255–264, 2009.

3. Bizhanova A, Kopp P. Controversies concerning the role of pendrin as anapical iodide transporter in thyroid follicular cells. Cell Physiol Biochem28: 485–490, 2011.

4. Calebiro D, Porazzi P, Bonomi M, Lisi S, Grindati A, De Nittis D,Fugazzola L, Marinò M, Botta G, Persani L. Absence of primaryhypothyroidism and goiter in Slc26a4 (-/-) mice fed on a low iodine diet.J Endocrinol Invest 34: 593–598, 2011.

5. Caputo A, Caci E, Ferrera L, Pedemonte N, Barsanti C, Sondo E,Pfeffer U, Ravazzolo R, Zegarra-Moran O, Galietta LJ. TMEM16A, amembrane protein associated with calcium-dependent chloride channelactivity. Science 322: 590–594, 2008.

6. Everett LA. Targeted disruption of mouse Pds provides insight about theinner-ear defects encountered in Pendred syndrome. Hum Mol Genet 10:153–161, 2001.

7. Faria D, Rock JR, Romao AM, Schweda F, Bandulik S, Witzgall R,Schlatter E, Heitzmann D, Pavenstädt H, Herrmann E, KunzelmannK, Schreiber R. The calcium-activated chloride channel Anoctamin 1

contributes to the regulation of renal function. Kidney Int 85: 1369–1381,2014.

8. Ferrera L, Caputo A, Galietta LJV. TMEM16A protein: a new identityfor Ca2�-dependent Cl� channels. Physiology (Bethesda) 25: 357–363,2010.

9. Ferrera L, Caputo A, Ubby I, Bussani E, Zegarra-Moran O, Ravaz-zolo R, Pagani F, Galietta LJV. Regulation of TMEM16A chloridechannel properties by alternative splicing. J Biol Chem 284: 33360–33368, 2009.

10. Fusco A, Berlingieri MT, Di Fiore PP, Portella G, Grieco M, VecchioG. One- and two-step transformations of rat thyroid epithelial cells byretroviral oncogenes. Mol Cell Biol 7: 3365–3370, 1987.

11. Galietta LJV. The TMEM16 protein family: a new class of chloridechannels? Biophys J 97: 3047–3053, 2009.

12. Gillam MP, Sidhaye AR, Lee EJ, Rutishauser J, Stephan CW, KoppP. Functional characterization of pendrin in a polarized cell system.Evidence for pendrin-mediated apical iodide efflux. J Biol Chem 279:13004–13010, 2004.

13. Golstein P, Abramow M, Dumont JE, Beauwens R. The iodide channelof the thyroid: a plasma membrane vesicle study. Am J Physiol CellPhysiol 263: C590–C597, 1992.

14. Golstein PE, Sener A, Beauwens R. The iodide channel of the thyroid.II. Selective iodide conductance inserted into liposomes. Am J Physiol CellPhysiol 268: C111–C118, 1995.

15. Hébrant A, Dom G, Dewaele M, Andry G, Trésallet C, Leteurtre E,Dumont JE, Maenhaut C. mRNA expression in papillary and anaplasticthyroid carcinoma: molecular anatomy of a killing switch. PLoS One 7:e37807, 2012.

16. Hébrant A, Van Sande J, Roger PP, Patey M, Klein M, Bournaud C,Savagner F, Leclère J, Dumont JE, van Staveren WC, Maenhaut C.Thyroid gene expression in familial nonautoimmune hyperthyroidismshows common characteristics with hyperfunctioning autonomous adeno-mas. J Clin Endocrinol Metab 94: 2602–2609, 2009.

17. Iosco C, Cosentino C, Sirna L, Romano R, Cursano S, Mongia A,Pompeo G, di Bernardo J, Ceccarelli C, Tallini G, Rhoden KJ.Anoctamin 1 is apically expressed on thyroid follicular cells and contrib-utes to ATP- and calcium-activated iodide efflux. Cell Physiol Biochem34: 966–980, 2014.

18. Kopp P, Pesce L, Solis-S JC. Pendred syndrome and iodide transport inthe thyroid. Trends Endocrinol Metab 19: 260–268, 2008.

19. Massart C, Hoste C, Virion A, Ruf J, Dumont JE, Van Sande J. Cellbiology of H2O2 generation in the thyroid: investigation of the control ofdual oxidases (DUOX) activity in intact ex vivo thyroid tissue and celllines. Mol Cell Endocrinol 343: 32–44, 2011.

20. Milenkovic M, De Deken X, Jin L, De Felice M, Di Lauro R, DumontJE, Corvilain B, Miot F. Duox expression and related H2O2 measure-ment in mouse thyroid: onset in embryonic development and regulation byTSH in adult. J Endocrinol 192: 615–626, 2007.

21. Namkung W, Phuan PW, Verkman AS. TMEM16A inhibitors revealTMEM16A as a minor component of calcium-activated chloride channelconductance in airway and intestinal epithelial cells. J Biol Chem 286:2365–2374, 2011.

22. Nicola JP, Basquin C, Portulano C, Reyna-Neyra A, Paroder M,Carrasco N. The Na�/I� symporter mediates active iodide uptake in theintestine. Am J Physiol Cell Physiol 296: C654–C662, 2009.

23. Nilsson M, Björkman U, Ekholm R, Ericson LE. Iodide transport inprimary cultured thyroid follicle cells: evidence of a TSH-regulatedchannel mediating iodide efflux selectively across the apical domain of theplasma membrane. Eur J Cell Biol 52: 270–281, 1990.

24. Nilsson M, Björkman U, Ekholm R, Ericson LE. Polarized efflux ofiodide in porcine thyrocytes occurs via a cAMP-regulated iodide channelin the apical plasma membrane. Acta Endocrinol (Copenh) 126: 67–74,1992.

25. Palos F, García-Rendueles MER, Araujo-Vilar D, Obregon MJ, CalvoRM, Cameselle-Teijeiro J, Bravo SB, Perez-Guerra O, Loidi L, Czar-nocka B, Alvarez P, Refetoff S, Dominguez-Gerpe L, Alvarez CV,Lado-Abeal J. Pendred syndrome in two Galician families: insights intoclinical phenotypes through cellular, genetic, and molecular studies. J ClinEndocrinol Metab 93: 267–277, 2008.

26. Pedemonte N, Galietta LJ. Structure and function of TMEM16 proteins(Anoctamins). Physiol Rev 94: 419–459, 2014.

27. Pesce L, Bizhanova A, Caraballo JC, Westphal W, Butti ML, Comel-las A, Kopp P. TSH regulates pendrin membrane abundance and enhancesiodide efflux in thyroid cells. Endocrinology 153: 512–521, 2012.

C1111ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org

28. Raspé E, Dumont JE. Control of the dog thyrocyte plasma membraneiodide permeability by the Ca(2�)-phosphatidylinositol and adenosine3=,5=-monophosphate cascades. Endocrinology 135: 986–995, 1994.

29. Roger P, Taton M, Van Sande J, Dumont JE. Mitogenic effects ofthyrotropin and adenosine 3=,5=-monophosphate in differentiated normalhuman thyroid cells in vitro. J Clin Endocrinol Metab 66: 1158–1165, 1988.

32. Schreiber R, Uliyakina I, Kongsuphol P, Warth R, Mirza M, MartinsJR, Kunzelmann K. Expression and function of epithelial anoctamins. JBiol Chem 285: 7838–7845, 2010.

33. Schroeder BC, Cheng T, Jan YN, Jan LY. Expression cloning ofTMEM16A as a calcium-activated chloride channel subunit. Cell 134:1019–1029, 2008.

34. Song Y, Driessens N, Costa M, De Deken X, Detours V, Corvilain B,Maenhaut C, Miot F, Van Sande J, Many MC, Dumont JE. Roles ofhydrogen peroxide in thyroid physiology and disease. J Clin EndocrinolMetab 92: 3764–3773, 2007.

35. Song Y, Massart C, Chico-Galdo V, Jin L, De Maertelaer V, DecosterC, Dumont JE, Van Sande J. Species specific thyroid signal transduc-tion: conserved physiology, divergent mechanisms. Mol Cell Endocrinol319: 56–62, 2010.

36. Taylor JP, Metcalfe RA, Watson PF, Weetman AP, Trembath RC.Mutations of the PDS gene, encoding pendrin, are associated with proteinmislocalization and loss of iodide efflux: implications for thyroid dysfunc-tion in Pendred syndrome. J Clin Endocrinol Metab 87: 1778–1184, 2002.

37. Tian Y, Schreiber R, Kunzelmann K. Anoctamins are a family ofCa2�-activated Cl� channels. J Cell Sci 125: 4991–4998, 2012.

38. Twyffels L, Massart C, Golstein PE, Raspe E, Van Sande J, DumontJE, Beauwens R, Kruys V. Pendrin: the thyrocyte apical membraneiodide transporter? Cell Physiol Biochem 28: 491–496, 2011.

39. Twyffels L, Wauquier C, Soin R, Decaestecker C, Gueydan C, KruysV. A masked PY-NLS in Drosophila TIS11 and its mammalian homologTristetraprolin. PLoS One 8: e71686, 2013.

40. Ubby I, Bussani E, Colonna A, Stacul G, Locatelli M, Scudieri P,Galietta L, Pagani F. TMEM16A alternative splicing coordination inbreast cancer. Mol Cancer 12: 75, 2013.

40a.Van Sande J, Massart C, Beauwens R, Schoutens A, Costagliola S,Dumont JE, Wolff J. Anion selectivity by the sodium iodide symporter.Endocrinology 144: 247–252, 2003.

41. Viitanen TM, Sukumaran P, Löf C, Törnquist K. Functional couplingof TRPC2 cation channels and the calcium-activated anion channels in ratthyroid cells: implications for iodide homeostasis. J Cell Physiol 228:814–823, 2013.

42. Wangemann P, Kim HM, Billings S, Nakaya K, Li X, Singh R, SharlinDS, Forrest D, Marcus DC, Fong P. Developmental delays consistentwith cochlear hypothyroidism contribute to failure to develop hearing inmice lacking Slc26a4/pendrin expression. Am J Physiol Renal Physiol297: F1435–F1447, 2009.

43. Weiss SJ, Philp NJ, Grollman EF. Iodide transport in a continuousline of cultured cells from rat thyroid. Endocrinology 114: 1090–1098,1984.

44. Wolff J. What is the role of pendrin? Thyroid 15: 346–348, 2005.45. Wollman SH, Reed FE. Transport of radioiodide between thyroid gland

and blood in mice and rats. Am J Physiol 196: 113–120, 1959.46. Yang YD, Cho H, Koo JY, Tak MH, Cho Y, Shim WS, Park SP, Lee

J, Lee B, Kim BM, Raouf R, Shin YK, Oh U. TMEM16A confersreceptor-activated calcium-dependent chloride conductance. Nature 455:1210–125, 2008.

C1112 ANO1 IS THE MAJOR APICAL IODIDE CHANNEL OF THE THYROCYTE

AJP-Cell Physiol • doi:10.1152/ajpcell.00126.2014 • www.ajpcell.org