The Pendred syndrome gene encodes a chloride-iodide transport protein.

Pendred syndrome is the most common form of syndromic deafness and characterized by congenital sensorineural hearing loss and goitre. This disorder was mapped to chromosome 7 and the gene causing Pendred syndrome (PDS) was subsequently identified by positional cloning. PDS encodes a putative transmembrane protein designated pendrin. Pendrin is closely related to a family of sulfate transport proteins that includes the rat sulfate-anion transporter (encoded by Sat-1; 29% amino acid sequence identity), the human diastrophic dysplasia sulfate transporter (encoded by DTD; 32%) and the human sulfate transporter 'downregulated in adenoma' (encoded by DRA; 45%). On the basis of this homology and the presence of a slightly modified sulfate-transporter signature sequence comprising its putative second transmembrane domain, pendrin has been proposed to function as a sulfate transporter. We were unable to detect evidence of sulfate transport following the expression of pendrin in Xenopus laevis oocytes by microinjection of PDS cRNA or in Sf9 cells following infection with PDS-recombinant baculovirus. The rates of transport for iodide and chloride were significantly increased following the expression of pendrin in both cell systems. Our results demonstrate that pendrin functions as a transporter of chloride and iodide, but not sulfate, and may provide insight into thyroid physiology and the pathophysiology of Pendred syndrome.

Effect of formate on volume reabsorption in the rabbit proximal tubule.

Studies on microvillus membrane from rabbit kidney cortex suggest that chloride absorption may occur by chloride/formate exchange with recycling of formic acid by nonionic diffusion. We tested whether this transport mechanism participates in active NaCl reabsorption in the rabbit proximal tubule. In proximal tubule S2 segments perfused with low HCO-3 solutions, the addition of formate (0.25-0.5 mM) to the lumen and the bath increased volume reabsorption (JV) by 60%; the transepithelial potential difference remained unchanged. The effect of formate on JV was completely reversible and was inhibited both by ouabain and by luminal 4,4'-diisothiocyanostilbene-2,2'-disulfonate. Formate (0.5 mM) failed to stimulate JV in early proximal convoluted tubules perfused with high HCO-3 solutions. As measured by miniature glass pH microelectrodes, this lack of formate effect on JV was related to a less extensive acidification of the tubule fluid when high HCO-3 solutions were used as perfusate. These data suggest that chloride/formate exchange with recycling of formic acid by nonionic diffusion represents a mechanism for active, electroneutral NaCl reabsorption in the proximal tubule.

Pendred syndrome, DFNB4, and PDS/SLC26A4 identification of eight novel mutations and possible genotype-phenotype correlations.

Mutations in PDS (SLC26A4) cause both Pendred syndrome and DFNB4, two autosomal recessive disorders that share hearing loss as a common feature. The hearing loss is associated with temporal bone abnormalities, ranging from isolated enlargement of the vestibular aqueduct (dilated vestibular aqueduct, DVA) to Mondini dysplasia, a complex malformation in which the normal cochlear spiral of 2(1/2) turns is replaced by a hypoplastic coil of 1(1/2) turns. In Pendred syndrome, thyromegaly also develops, although affected persons usually remain euthyroid. We identified PDS mutations in the proband of 14 of 47 simplex families (30%) and nine of 11 multiplex families (82%) (P=0.0023). In all cases, mutations segregated with the disease state in multiplex families. Included in the 15 different PDS allele variants we found were eight novel mutations. The two most common mutations, T416P and IVS8+1G>A, were present in 22% and 30% of families, respectively. The finding of PDS mutations in five of six multiplex families with DVA (83%) and four of five multiplex families with Mondini dysplasia (80%) implies that mutations in this gene are the major genetic cause of these temporal anomalies. Comparative analysis of phenotypic and genotypic data supports the hypothesis that the type of temporal bone anomaly may depend on the specific PDS allele variant present.

Activation of Cl-/OH- exchange by parachloromercuribenzoic acid in rabbit renal brush-border membranes.

The effect of the sulfhydryl reagent parachloromercuribenzoic acid (PCMB) on chloride transport was examined in rabbit renal brush-border membrane vesicles (BBMV). PCMB had no effect on the chloride conductive pathway. In the presence of an inside-alkaline pH gradient and a K-/valinomycin voltage clamp, the addition of PCMB stimulated 36Cl uptake and induced a threefold overshoot above the equilibrium value, indicating Cl/OH exchange. The effect of PCMB was reversed by dithiothreitol. Cl/OH exchange was not observed in the absence of PCMB. PCMB-activated Cl/OH exchange persisted even when the membrane potential was made inside-negative relative to the controls, thus, demonstrating that PCMB's effect on 36Cl uptake under pH-gradient conditions is not mediated by parallel Cl- and H+ conductive pathways. PCMB-activated Cl/OH exchange was inhibited by 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) and 5-nitro-2-(3-phenylpropylamino)-benzoic acid (NPPB) with IC50 values of 290 and 80 microM, respectively. These results demonstrate that modification of sulfhydryl groups by PCMB activates Cl/OH exchange in BBMV.

Mutations in the diastrophic dysplasia sulfate transporter (DTDST) gene: correlation between sulfate transport activity and chondrodysplasia phenotype.

The diastrophic dysplasia sulfate transporter (DTDST) gene encodes a transmembrane protein that transports sulfate into chondrocytes to maintain adequate sulfation of proteoglycans. Mutations in this gene are responsible for four recessively inherited chondrodysplasias that include diastrophic dysplasia, multiple epiphyseal dysplasia, atelosteogenesis type 2 and achondrogenesis 1B (ACG-1B). To determine whether the DTDST mutations found in individuals with these chondrodysplasias differ functionally from each other, we compared the sulfate transport activity of 11 reported DTDST mutations. Five mutations, G255E, Delta a1751, L483P, R178X and N425D, had minimal sulfate transport function following expression in Xenopus laevis oocytes. Two mutations, Delta V340 and R279W, transported sulfate at rates of 17 and 32%, respectively, of wild-type DTDST. Four mutations, A715V, C653S, Q454P and G678V, had rates of sulfate transport nearly equal to that of wild-type DTDST. Transport kinetics were not different among the four mutations with near-normal sulfate transport function and wild-type DTDST. When the sulfate transport function of the different DTDST mutations are grouped according to the general phenotypes, individuals with the most severe form, ACG-1B, tend to be homozygous for null mutations, individuals with the moderately severe atelosteogenesis type 2 have at least one allele with a loss-of-function mutation, and individuals with the mildest forms are typically homozygous for mutations with residual sulfate transport function. However, in the X.laevis oocyte expression system, the correlation between residual transport function and the severity of phenotype was not absolute, suggesting that factors in addition to the intrinsic sulfate transport properties of the DTDST protein may influence the phenotype in individuals with DTDST mutations.

Hg2+ and Cu+ are ionophores, mediating Cl-/OH- exchange in liposomes and rabbit renal brush border membranes.

Organometals, including organomercurials, are capable of mediating Cl-/OH- exchange across lipid membranes by forming neutral ion pairs. In this study, the ability of inorganic metals to catalyze Cl-/OH- exchange was examined. In the presence of an inwardly directed chloride gradient, HgCl2 at concentrations as low as 30 nM resulted in quenching of acridine orange fluorescence in liposomes, indicating liposomal acidification. In the presence of the reducing agent, ascorbate, CuSO4 at concentrations as low as 0.6 microM also mediated chloride-dependent liposomal acidification. Copper in the absence of ascorbate, iron (with or without ascorbate), cobalt, cadmium, zinc, nickel, and lead were without an effect. 36Cl efflux from rabbit renal brush border membrane vesicles was also markedly stimulated by micromolar concentrations of mercury or copper plus ascorbate. Vesicle integrity was not altered by the concentrations of mercury or copper employed in these studies. In the absence of ascorbate, CuCl stimulated chloride efflux only under anaerobic conditions, confirming that it is the reduced form of copper (Cu+) that mediates chloride transport across the membrane. In the presence of mercury or reduced copper, an inside alkaline pH gradient stimulated the uphill accumulation of 36Cl and 82Br, respectively, confirming Cl-/OH- exchange. Studies in liposomes and brush border membranes demonstrate that this is an electroneutral process. These results show that Hg2+ and Cu+ are capable of acting as ionophores, mediating electroneutral Cl-/OH- exchange in liposomes and brush border membrane vesicles. This effect could contribute to the toxicity of these two metals.

Effects of sulfate and chloride on three separate oxalate transporters reconstituted from rabbit renal cortex.

Understanding the mechanism of sulfate-dependent, oxalate-stimulated chloride reabsorption in the mammalian proximal tubule is complicated by the presence of multiple oxalate and sulfate transport pathways. Accordingly, we developed a method of reconstituting functional oxalate transport from the rabbit renal cortex so that the individual transporters might be examined. Solubilized microvillus membrane proteins were separated by hydroxyapatite chromatography and then reconstituted into proteoliposomes. Two peaks of oxalate/oxalate exchange activity were observed. Sulfate (10 mM) cis-inhibits oxalate transport in the early peak by 93% and in the later peak by 41%. In contrast, 20 mM chloride inhibits oxalate/oxalate exchange by only 32% in the early peak but inhibits oxalate exchange by 70% in the later peak. Oxalate-stimulated sulfate uptake was observed in the early fractions but not in the later fractions. These data are consistent with the recovery of the sulfate/oxalate exchanger in the early hydroxyapatite fractions and the chloride/oxalate exchanger in the later fractions. The basolateral membrane sulfate/oxalate exchanger was also reconstituted. The reconstituted basolateral and apical membrane sulfate/oxalate exchangers demonstrate nearly identical patterns of substrate specificities. However, 98% of apical sulfate/oxalate exchange activity is lost following exposure to octylglucoside at room temperature, whereas the basolateral sulfate/oxalate exchange activity was reduced 67% (P < 0.05). In conclusion, functional reconstitution of solubilized membrane proteins demonstrates that apical membrane chloride/oxalate exchange and sulfate/oxalate exchange are mediated by different transport proteins. Apical and basolateral sulfate/oxalate exchange may also represent transport on two separate exchangers.

Chloride/formate exchange with formic acid recycling: a mechanism of active chloride transport across epithelial membranes.

The pathways for transport of Cl- and formate in microvillus membrane vesicles isolated from rabbit renal cortex were evaluated. An outward formate gradient stimulated the uptake of Cl-, and an outward Cl- gradient stimulated the uptake of formate, indicating Cl-/formate exchange. In addition, an inside alkaline pH gradient induced the accumulation of formate, consistent with nonionic diffusion of formic acid. Although an inward Na+ gradient also stimulated uphill formate accumulation, suggesting Na+/formate cotransport, this effect was abolished when ionophores were used to prevent the generation of a transmembrane pH gradient, indicating an indirect coupling of formic acid transport to Na+/H+ exchange. An inside alkaline pH gradient only minimally stimulated the uptake of 82Br-, used as tracer for Cl-, confirming the absence of appreciable Cl-/OH- exchange. However, the same pH gradient in the presence of a physiologic formate concentration (0.2 mM) markedly stimulated 82Br- influx. These data suggest that Cl-/formate exchange with recycling of formic acid by nonionic diffusion is a potential mechanism for active Cl- absorption across the luminal membrane in the proximal tubule and perhaps in other epithelia.

Identification and partial purification of a band 3-like protein from rabbit renal brush border membranes.

A monoclonal antibody against the membrane domain of human erythrocyte band 3 was tested for its ability to bind to rabbit renal brush border membranes. A single brush border protein with a molecular mass of 43 kDa was recognized by the band 3 antibody. Using DNase I coupled to an agarose-bead support this 43-kDa protein was partially purified by removing actin and a number of actin-bound proteins from the brush border membranes. The partially purified 43 kDa-band was eluted from sodium dodecyl sulfate-polyacrylamide gels and used to make a highly sensitive and specific guinea pig antiserum. This antiserum, but not serum from control guinea pigs, cross-reacts with purified band 3 from human, rabbit, and bovine erythrocytes confirming the immunologic similarity among these proteins. The 43-kDa protein can be stained by the periodic acid-Schiff base method and binds wheat germ agglutinin and concanavalin A, demonstrating that it is a glycoprotein. Furthermore, in the absence of dithiothreitol, the immunoreactive brush border protein migrates with a molecular mass of 86 kDa on an sodium dodecyl sulfate-polyacrylamide gel suggesting that under nonreducing conditions it exists as a dimer. The 43-kDa protein could be solubilized in octyl glucoside and was further purified using gel filtration chromatography. The amino acid composition of the 43-kDa brush border protein was obtained, and its similarity with erythrocyte band 3 is discussed.

Anion exchange pathways for Cl- transport in rabbit renal microvillus membranes.

We evaluated the mechanisms of chloride transport in microvillus membrane vesicles isolated from the rabbit renal cortex. The presence of Cl-formate exchange was confirmed. Outward gradients of oxaloacetate, HCO3, acetate, lactate, succinate, sulfate, and p-aminohippurate (PAH) stimulated the rate of Cl uptake minimally (less than 25%) or not at all. However, an outward gradient of oxalate stimulated Cl uptake by 70%, and an outward Cl gradient induced uphill oxalate uptake, indicating Cl-oxalate exchange. Moreover, an outward formate gradient induced uphill oxalate uptake, indicating formate-oxalate exchange. Studies of inhibitor and substrate specificity indicated the probable operation of at least two separate anion exchangers in mediating Cl transport. The Cl-formate exchanger accepted Cl and formate as substrates, had little or no affinity for oxalate, was sensitive to inhibition by furosemide, and was less sensitive to inhibition by 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS). The Cl (formate)-oxalate exchanger also accepted Cl and formate as substrates but had high affinity for oxalate, was highly sensitive to inhibition by DIDS, and was less sensitive to inhibition by furosemide. The Cl-formate exchanger was electroneutral, whereas the Cl (formate)-oxalate exchanger was electrogenic. We conclude that at least two separate anion exchangers mediating Cl transport are present on the luminal membrane of the rabbit proximal tubule cell. These exchangers may play important roles in mediating transtubular Cl and oxalate transport in this nephron segment.

Cloning of cDNAs encoding a rabbit renal brush border membrane protein immunologically related to band 3. Sequence similarity with microsomal dipeptidase.

Distinct anion transport processes have been identified in the mammalian renal proximal tubule, but none of the responsible proteins or genes have been isolated. A 43 kDa rabbit microvillus membrane protein that is immunologically related to the erythroid anion exchanger (band 3) was a candidate for a renal anion transporter. To examine the structural relationship with band 3, we cloned cDNAs encoding the 43 kDa protein. The 43 kDa band-3-like protein was purified, and a novel sequence of 24 amino acids was obtained from the N-terminus. Degenerate oligonucleotides were synthesized based on this sequence, and the polymerase chain reaction with single-sided specificity was used to amplify and clone a 1330 bp cDNA from rabbit renal cortex. Additional overlapping 272 bp and 1123 bp cDNAs were obtained by synthesizing and screening a rabbit renal cortical cDNA library. The composite sequence was 1483 bp, terminated with (A)16, and was similar in size to the principal transcript expressed in rabbit renal cortex. The single long open reading frame was predicted to encode a protein composed of 410 amino acids with a molecular mass of 45,193 Da; 15 amino acids predicted to reside at the N-terminus were absent in the mature protein and may constitute a signal peptide. There was only limited sequence similarity with human erythroid band 3. Rather, the sequence was highly similar to microsomal dipeptidase, including the presence of a signal peptide and a consensus sequence for covalent linkage to glycosylphosphatidylinositol. In summary, the 43 kDa protein from rabbit renal cortex that is recognized by a monospecific antibody to erythroid band 3 is most likely a microvillus membrane dipeptidase.

Human pendrin expressed in Xenopus laevis oocytes mediates chloride/formate exchange.

Pendred syndrome, characterized by congenital sensorineural hearing loss and goiter, is one of the most common forms of syndromic deafness. The gene causing Pendred syndrome (PDS) encodes a protein designated pendrin, which is expressed in the thyroid, kidney, and fetal cochlea. Pendrin functions as an iodide and chloride transporter, but its role in the development of hearing loss and goiter is unknown. In this study, we examined the mechanism of pendrin-mediated anion transport in Xenopus laevis oocytes. Unlabeled formate added to the uptake medium inhibited pendrin-mediated (36)Cl uptake in X. laevis oocytes. In addition, the uptake of [(14)C]formate was stimulated in oocytes injected with PDS cRNA compared with water-injected controls. These results indicate that formate is a substrate for pendrin. Furthermore, chloride stimulated the efflux of [(14)C]formate and formate stimulated the efflux of (36)Cl in oocytes expressing pendrin, results consistent with pendrin-mediated chloride/formate exchange. These data demonstrate that pendrin is functionally similar to the renal chloride/formate exchanger, which serves as an important mechanism of chloride transport in the proximal tubule. A similar process could participate in the development of ion gradients within the inner ear.

A 45-kDa protein antigenically related to band 3 is selectively expressed in kidney mitochondria.

Anion exchange similar to that catalyzed by erythrocyte band 3 occurs across many nonerythroid cell membranes. To identify anion-exchange proteins structurally related to band 3, we immunoblotted rabbit kidney medullary membrane fractions with anti-band 3 antibodies. Immunoblots using antibodies to the cytoplasmic domain of band 3 revealed cross-reactive proteins in the plasma membrane fraction only. In contrast, two monoclonal antibodies against band 3 membrane domain labeled a 45-kDa protein; further immunoblotting and immunogold studies of membrane fractions and kidney sections using one of the anti-membrane domain antibodies showed that labeling was strongest in mitochondria of H(+)-secreting collecting duct cells. Tissue-to-tissue expression of the 45-kDa mitochondrial protein was variable: kidney medulla greater than heart greater than kidney cortex much greater than liver. We conclude that a 45-kDa protein with immunological cross-reactivity to the erythrocyte band 3 membrane domain is expressed in mitochondria in a highly cell-specific fashion and speculate that the protein may play a role in mitochondrial anion transport.

Immunolocalization of sat-1 sulfate/oxalate/bicarbonate anion exchanger in the rat kidney.

The rat liver sulfate/bicarbonate/oxalate exchanger (sat-1) transports sulfate across the canalicular membrane in exchange for either bicarbonate or oxalate. Sulfate/oxalate exchange has been detected in the proximal tubule of the kidney, where it is probably involved in the reabsorption of filtered sulfate and the secretion of oxalate and may contribute to oxalate-dependent chloride reabsorption. Screening of a renal cortex cDNA library determined that sat-1 is expressed in the rat kidney. To evaluate this anion exchanger, the sat-1 protein was expressed in Sf9 cells. Sodium-independent sulfate and oxalate uptake was enhanced 7.3-fold and 13.1-fold, respectively, in Sf9 cells expressing the sat-1 protein compared with cells infected with wild-type virus. We determined that sat-1 is glycosylated in the kidney; however, anion exchange via sat-1 is observed despite incomplete glycosylation of sat-1 in Sf9 cells. The sat-1 protein, with an added COOH-terminal 6-histidine tag, was purified on a metal affinity column and used to generate anti-sat-1 monoclonal antibodies. The sat-1 protein was localized to the basolateral membrane, but not the apical membrane, of the proximal tubule by both Western blot analysis and immunohistochemistry. These studies demonstrate that sulfate/oxalate exchange on the apical and basolateral membranes of the proximal tubule represents transport on two different anion exchangers.

Maturation of proximal tubule Na+/H+ antiporter activity in sheep during transition from fetus to newborn.

We have studied maturational changes in the kinetics of the proximal tubule Na+/H+ antiporter. Microvillus membrane vesicles were prepared from renal cortex of fetal and newborn lambs. Amiloride-sensitive uptake of 22Na+ by these vesicles was measured and Woolf-Augustinsson-Hofstee plots were used to determine the Michaelis constant (Km) and rate of maximal uptake (Vmax). Initial studies of fetal lambs at 130-132 days gestation (n = 5; term is 145 days) and 3- to 4-day-old lambs (n = 5) revealed no maturational change in Km (7.27 +/- 1.25 for fetuses and 9.01 +/- 1.03 mM for lambs); however, there was a 242% increase in Vmax (from 1.28 +/- 0.33 in the fetuses to 4.37 +/- 0.85 nmol.s-1.mg protein-1 in the lambs, P = 0.005). Further definition of the developmental change in Na+/H+ antiporter Vmax was obtained when 144-day-gestation fetuses (n = 5) were compared with 24-h-old sibling lambs (n = 5) that had been delivered by cesarean section at 144 days gestation. Again, no significant difference was seen in Na+/H+ antiporter Km (14.9 +/- 6.5 for fetuses and 12.5 +/- 3.4 mM for lambs); however, a significant increase in Na+/H+ antiporter Vmax occurred (from 1.41 +/- 0.51 in the fetuses to 3.32 +/- 0.37 nmol.s-1.mg protein-1 in the lambs, P < 0.01). This study shows that there is a maturational increase in renal cortical Na+/H+ antiporter Vmax during the transition from fetal to newborn life. This increase parallels the increase in renal tubular Na+ reabsorption that occurs at this time.

Role of glucocorticoids in the maturation of renal cortical Na+/H+ exchanger activity during fetal life in sheep.

We have studied the role of glucocorticoids in inducing the maturation in activity of the proximal tubule Na+/H+ exchanger that follows birth. Renal cortical microvillus membrane vesicles were prepared from 132-day gestation sheep fetuses (n = 8) that had received intraperitoneal cortisol (13 micrograms.kg-1.h-1) for the previous 48 h. Membrane vesicles were also obtained from sham-operated twin controls (n = 8). Amiloride-sensitive uptake of 22Na+ by these vesicles was measured, and Woolf-Augustinsson-Hofstee plots were used to determine the Michaelis constant (Km) and maximal velocity (Vmax). There was no significant difference in Km; however, the Vmax was 61% higher in cortisol-treated fetuses. Posttreatment circulating cortisol levels were significantly higher in the treated fetuses. Total RNA was collected from renal cortex of the eight pairs of twins when killed. Renal cortex Na+/H+ exchanger 3 (NHE3) mRNA levels were approximately fourfold higher in cortisol-treated than in control fetuses. Although proximal tubule Na+/H+ exchanger activity and renal cortex NHE3 mRNA levels increased significantly in cortisol-treated fetuses, cortisol infusion did not stimulate renal sodium reabsorption in the fetus but rather produced a natriuresis. These results demonstrate that glucocorticoids can induce an increase in both Na+/H+ exchanger activity and NHE3 mRNA levels during the last trimester of gestation in sheep. However, these changes are not associated with an increased ability of the fetal kidney to reabsorb sodium.

Pendrin, encoded by the Pendred syndrome gene, resides in the apical region of renal intercalated cells and mediates bicarbonate secretion.

Pendrin is an anion transporter encoded by the PDS/Pds gene. In humans, mutations in PDS cause the genetic disorder Pendred syndrome, which is associated with deafness and goiter. Previous studies have shown that this gene has a relatively restricted pattern of expression, with PDS/Pds mRNA detected only in the thyroid, inner ear, and kidney. The present study examined the distribution and function of pendrin in the mammalian kidney. Immunolocalization studies were performed using anti-pendrin polyclonal and monoclonal antibodies. Labeling was detected on the apical surface of a subpopulation of cells within the cortical collecting ducts (CCDs) that also express the H(+)-ATPase but not aquaporin-2, indicating that pendrin is present in intercalated cells of the CCD. Furthermore, pendrin was detected exclusively within the subpopulation of intercalated cells that express the H(+)-ATPase but not the anion exchanger 1 (AE1) and that are thought to mediate bicarbonate secretion. The same distribution of pendrin was observed in mouse, rat, and human kidney. However, pendrin was not detected in kidneys from a Pds-knockout mouse. Perfused CCD tubules isolated from alkali-loaded wild-type mice secreted bicarbonate, whereas tubules from alkali-loaded Pds-knockout mice failed to secrete bicarbonate. Together, these studies indicate that pendrin is an apical anion transporter in intercalated cells of CCDs and has an essential role in renal bicarbonate secretion.

Functional differences of the PDS gene product are associated with phenotypic variation in patients with Pendred syndrome and non-syndromic hearing loss (DFNB4).

The PDS gene encodes a transmembrane protein, known as pendrin, which functions as a transporter of iodide and chloride. Mutations in this gene are responsible for Pendred syndrome and autosomal recessive non-syndromic hearing loss at the DFNB4 locus on chromosome 7q31. A screen of 20 individuals from the midwestern USA with non-syndromic hearing loss and dilated vestibular aqueducts identified three people (15%) with PDS mutations. To determine whether PDS mutations in individuals with Pendred syndrome differ functionally from PDS mutations in individuals with non-syndromic hearing loss, we compared three common Pendred syndrome allele variants (L236P, T416P and E384G), with three PDS mutations reported only in individuals with non-syndromic hearing loss (V480D, V653A and I490L/G497S). The mutations associated with Pendred syndrome have complete loss of pendrin-induced chloride and iodide transport, while alleles unique to people with DFNB4 are able to transport both iodide and chloride, albeit at a much lower level than wild-type pendrin. We hypothesize that this residual level of anion transport is sufficient to eliminate or postpone the onset of goiter in individuals with DFNB4. We propose a model for pendrin function in the thyroid in which pendrin transports iodide across the apical membrane of the thyrocyte into the colloid space.