The human solute carrier 26 (SLC26) family of anion transporters consists of ten members that are found in various organs in the body including the stomach, intestine, kidney, thyroid and ear where they transport anions including bicarbonate, chloride and sulfate, typically in an exchange mode. Mutations in these genes cause a plethora of diseases such as diastrophic dysplasia affecting sulfate uptake into chondrocytes (SLC26A2), congenital chloride-losing diarrhoea (SLC26A3) affecting chloride secretion in the intestine and Pendred's syndrome (SLC26A4) resulting in hearing loss. To understand how these mutations affect the structures of the SLC26 membrane proteins and their ability to function properly, 12 human disease-causing mutants from SLC26A2, SLC26A3 and SLC26A4 were introduced into the equivalent sites of the sulfate transporter anti-sigma factor antagonist (STAS) domain of a bacterial homologue SLC26 protein DauA (YchM). Biophysical analyses including size-exclusion chromatography, circular dichroism (CD), differential scanning fluorimetry (DSF) and tryptophan fluorescence revealed that most mutations caused protein instability and aggregation. The mutation A463K, equivalent to N558K in human SLC26A4, which is located within α-helix 1 of the DauA STAS domain, stabilized the protein. CD measurements showed that most disease-related mutants had a mildly reduced helix content, but were more sensitive to thermal denaturation. Fluorescence spectroscopy showed that the mutants had more open structures and were more readily denatured by urea, whereas DSF indicated more labile folds. Overall, we conclude that the disease-associated mutations destabilized the STAS domain resulting in an increased propensity to misfold and aggregate.

INTRODUCTION

The solute carrier 26 (SLC26) proteins are a super-family of anion transporters that is conserved throughout phylogeny in different organisms [13]. The SLC26 membrane proteins transport SO42− in yeast and plants [46], whereas in mammals a wide range of anions are transported including SO42−, HCO3, Cl, I, NO3− and oxalate [716]. Most SLC26 proteins contain an N-terminal membrane domain consist of 14 transmembrane (TM) helices followed by a C-terminal cytosolic STAS (sulfate transporter anti-sigma factor antagonist) domain [1719]. The membrane domain is predicted to have a similar 14 TM fold to that of UraA, a bacterial proton–uracil symporter with a 7+7 TM inverted topology that contains a central substrate-binding site [20]. The crystal structure of a prokaryotic member of the SLC26 family (SLC26Dg) that transports fumarate has confirmed this topology [21]. The STAS domain has been reported to have various functions but is named based on its structural similarity to Bacillus subtilis SpoIIAA, which regulates bacterial sporulation through binding with sigma factor upon phosphorylation. σF is a factor that binds to RNA polymerase to confer fore-spore-specific gene expression, whereas the anti-sigma factor SpoIIAB is a serine/threonine kinase and an antagonist of SpoIIAA [2224]. Some STAS domains are nucleotide-binding phosphoproteins or nucleotidases, whereas others are the interaction or transduction modules [3,24,25]. STAS domains in human SLC26 proteins are important in protein–protein interactions such as interacting with CFTR (cystic fibrosis transmembrane conductance regulator) [2629]. The sole member of the SLC26 (SulP) family found in Escherichia coli is DauA (YchM), annotated in EcoCyc (http://www.ecocyc.org) to be involved in succinate, fumarate and L-aspartate transport. DauA plays a role in the regulation of C4-dicarboxylic acid metabolism at pH 7, via regulation of expression and/or activity of DctA, and it may act as a co-sensor with DcuS [25,30]. The structure of E. coli DauA STAS domain (PDB code 3NY7) adopts a β1-β2-α1-β3-α2-β4-α3-β5-α4 secondary structure that is common to other STAS domains, although the SLC26Dg fumarate transporter is shorter than DauA and lacks the final α4-helix. This core STAS domain of DauA overlays with the rat SLC26A5 (prestin) structure (PDB code 3LLO) in which an intervening loop between α1 and β3 was deleted for structure determination. In addition, the prestin STAS domain contains an extra β-strand (β0) in its N-terminal region [25,31,32]. A structure of a human SLC26 STAS domain is still not available.

In humans, the SLC26 family consists of ten different membrane proteins that are found in various tissues in the body including the stomach, intestine, kidney, thyroid and ear where they typically transport ions including bicarbonate, chloride and sulfate in an exchange mode [3,3335]. Mutations within these genes cause a range of diseases including diastrophic dysplasia in the case of SLC26A2 [3644], intestinal congenital chloride-losing diarrhoea (CCD) in SLC26A3 [8,18,45] and Pendred's syndrome (deafness) in case of SLC26A4 [4649]. Studies of mouse models also generated similar phenotypes of diseases such as distal renal tubular acidosis [50], gastric achlorhydria [50,51], nephrolithiasis [52,53], pancreatic duct defects [54,55] and altered regulation of bicarbonate secretion [56]. Despite their importance in human physiology and disease, solute transporters, including the SLC26 family, remain one of the most poorly studied of all human gene families [57]. Many of these disease-causing mutations are located within the STAS domain. To understand the molecular basis of these diseases, previous studies examined a limited number of mutations within the STAS domain that cause misfolding of SLC26 or trafficking defects in cells [31,58]. A comprehensive study of the effects of the mutations on the behaviour of SLC26 proteins itself is still not available. Therefore an in vitro study of the folding and stability of these mutants is needed in order to give a systematic overview on how mutations give rise to disease at the structural level. Previously we have solved the structure of the E. coli DauA STAS domain, which is readily expressed in bacterial cells with good behaviour in solution [25]. This structure carries a similar fold to other STAS structures. There is only one reported STAS structure from mammals (Rattus norvegicus); however, it was determined by deleting a significant ([505–553]GS[637–718]) [31] intervening loop portion of the protein revealing the core STAS domain. In addition, no tryptophan residues are found in this rat prestin sequence, limiting spectroscopic studies using this natural fluorescent reporter. Based on this, we performed various biophysical analyses on the effect of human SLC26A3, SLC26A4 and SLC26A5 mutations introduced into homologous positions in the E. coli DauA STAS domain as a template to determine the effect of these disease-causing mutations on protein stability and folding.

EXPERIMENTAL

Materials

The pET28b expression vector was supplied by Invitrogen. The Q5® Site-Directed Mutagenesis Kit was from New England BioLabs. Chemicals for bacteria-culturing media and IPTG were from BioShop. Trypsin was obtained from Hampton Research. Urea was from Thermo Scientific. Other chemicals were purchased from Sigma–Aldrich. Mouse anti-His antibody was purchased from Thermo Scientific and the secondary antibody horseradish peroxidase (HRP)-linked anti-mouse (or anti-rabbit) IgG was from Cell Signaling Technology.

Gene cloning and plasmid constructs

The DauA STAS domain (residues Met436–Leu550) construct was amplified by polymerase chain reaction (PCR) from the full-length DauA cDNA and then inserted into a pET28b expression vector fused with an N-terminal His6 tag. Disease-related mutants were created from the wild-type STAS constructs and also cloned into the pET28b expression vector using the Q5® Site-Directed Mutagenesis Kit and the complementary mutagenic primers. All DNA sequences were confirmed by DNA sequencing using GeneSifter at the TCAG sequencing facility at the Hospital for Sick Children, Toronto, Ontario, Canada.

Protein expression assessment

Proteins (wild-type and STAS mutants) were expressed in parallel in 4 ml of BL21(DE3)plysS and cells were grown until the D600 reached 1.0. Induction was performed with the final concentration of 1 mM ITPG at 20°C for 18 h. Cells were centrifuged for 1 min at 15700 g (Eppendorf Centrifuge 5415R, rotor F45-24-11). The pellets were resuspended with 60 μl of 200 mM sodium chloride and 20 mM Tris/HCl, pH 8.0, and then the same volume of SDS sample buffer was added. Protein expression was detected by immunoblotting using a mouse anti-His (and then rabbit anti-LepB for protein loading) antibody followed by HRP-linked mouse (or rabbit) anti-IgG secondary antibody. LepB (Signal peptidase I) is an endogenous protein expressed in bacteria which functions in the cleavage of hydrophobic, N-terminal signal or leader sequences from secreted and periplasmic proteins (http://www.uniprot.org/uniprot/P00803). Here we used rabbit anti-LepB antibody [59].

Protein purification

His-tagged proteins were expressed in E. coli BL21(DE3)plysS cells (Novagen) and cells were cultured in 20 ml of LB medium (20 g/l tryptone, 10 g/l yeast extract and 10 g/l NaCl) with 50 mg/l kanamycin in 125 ml flasks. Large-scale cultivation was performed in 5 litre baffled flasks containing 2 litres of medium. Cells were grown at 37°C at 210 rev./min until reaching a D600 of ∼1.0. Protein expression was induced by adding a 1 mM final concentration of IPTG and cultivation was continued at 18°C at 210 rev./min for 20 h. Cell pellets were harvested by centrifugation at 4000 g for 20 min at 4°C, and resuspended in buffer containing 20 mM Tris/HCl, pH 8.0, 300 mM NaCl and 1% glycerol. After disruption in a French Press (Thermo Spectronic) at 100 MPa, the crude cell extract was separated from cell debris by centrifugation (40000 g, 4°C, 30 min). The protein in the supernatant was purified by nickel-affinity and gel-filtration chromatography as described previously [60]. The purity of the resulting proteins was ascertained by SDS/PAGE to be more than 90% pure. Purified proteins were concentrated by centrifugal ultrafiltration and the protein concentration was determined by absorbance measurements at 280 nm using calculated absorption coefficients. The final protein yield of wild-type STAS domain from 1 litre of cell culture was ∼2 mg.

Mass spectrometry

Samples were analysed on a linear ion trap–Orbitrap hybrid analyser (LTQ-Orbitrap, ThermoFisher) outfitted with a nanospray source and EASY–nLC split-free nano-LC system (ThermoFisher). Freeze-dried peptide mixtures were dissolved in 0.1% formic acid and loaded on to a 75 μm × 15 cm PepMax RSLC EASY-Spray column filled with 2 μM C18 beads (ThermoFisher) at a pressure of 80000 kPa. Peptides were eluted over 60 min at a rate of 250 nl/min using a 0–35% acetonitrile gradient in 0.1% formic acid. Peptides were introduced by nano electrospray into an LTQ–Orbitrap hybrid mass spectrometer (ThermoFisher). The instrument method consisted of one MS full scan (400–1500 m/z) in the Orbitrap mass analyser, an automatic gain control target of 500000 with a maximum ion injection of 200 ms, one microscan and a resolution of 120000. Ten data-dependent MS/MS scans were performed in the linear ion trap using the ten most intense ions at 35% normalized collision energy. The MS and MS/MS scans were obtained in parallel fashion. In MS/MS mode, automatic gain control targets were 10000 with a maximum ion injection time of 100 ms. A minimum ion intensity of 1000 was required to trigger an MS/MS spectrum. The dynamic exclusion was applied using a maximum exclusion list of 500 with one repeat count with a repeat duration of 15 s and exclusion duration of 45 s.

Circular dichroism (CD)

Far-UV CD spectra were collected in a 1-mm-pathlength quartz cell in a Jasco J-810 CD spectropolarimeter. Experiments were performed at room temperature using 0.35 mg/ml protein in 150 mM NaCl and 25 mM sodium phosphate buffer (pH 8.0). Data were recorded from 260 to 185 nm at 1 nm intervals. Data recorded in the 190–240 nm range were analysed using DichroWeb [61,62]. The mean residue ellipticity was measured as a function of wavelength. Thermal stability melting curves were determined by CD with the temperature increasing from 20°C to 80°C with the wavelength set at 222 nm to monitor helical content.

Intrinsic fluorescence emission and urea denaturation assay

Purified protein samples were centrifuged at 15700 g (Eppendorf Centrifuge 5415R, rotor F45-24-11) to remove insoluble material and then mixed with various amounts of a 10 M urea stock solution. The final concentration of the protein was 0.075 mg/ml and the final concentration of urea ranged from 0 to 8 M. Protein samples were equilibrated at room temperature for 20 min prior to fluorescence measurements. Tryptophan fluorescence emission was measured at each urea concentration from 300 to 420 nm upon excitation at 295 nm.

Differential scanning fluorimetry (DSF)

DSF was performed on a CFX96 instrument (Bio-Rad Laboratories). Purified STAS domain or mutant proteins were diluted to a final concentration of 1 mg/ml by mixing with 5× SYPRO Orange (Life Technologies) in 40 μl of solution. To generate the DSF melting curve, fluorescence was measured using the FRET channel of the CFX96 instrument between 20°C and 90°C. The melting temperature was determined from the peak(s) of the first derivatives of the melting curve and calculations were made using the Bio-Rad Laboratories CFX Manager software.

Trypsin digestion assay

For trypsin digestion, the target protein at 1 mg/ml was incubated with various concentrations of trypsin and set for different incubation times at room temperature as indicated in the figures. The reaction was stopped by addition of SDS/PAGE sample buffer.

Immunoblotting

Proteins were resolved by SDS/PAGE using 18% polyacrylamide and transferred on to nitrocellulose for immunoblot analysis. All blots were probed using the mouse monoclonal anti-His antibody (Thermo Scientific) and HRP-linked anti-mouse IgG (Cell Signaling Technology). Expression levels of wild-type STAS domain and mutants were normalized for protein loading by the band intensity of endogenous LepB using a rabbit monoclonal anti-LepB antibody and HRP-linked anti-rabbit IgG (Thermo Scientific). Bands were quantified using NIH ImageJ.

RESULTS

Similarities between E. coli and mammalian SLC26 STAS domain

To study the folding effects of different mutations upon the wild-type STAS domain, we mapped 12 disease-causing mutations in human SLC26 STAS proteins on to equivalent position in the E. coli DauA STAS domain (Figure 1). Although sequence alignment shows that only three mutated residues (Leu459, Arg544 and Ala546) are identical between human SLC26 STAS domains and the E. coli DauA STAS domain, the crystal structures and fold are highly conserved between SLC26A5 STAS (also called prestin) and DauA STAS domain (Supplementary Figure S1A) in containing five or six β-sheets and four α-helix regions (RMSD 2.9). Except for α-helix 1, most of the residues superimpose very well on the structure between E. coli DauA and rat prestin (Supplementary Figure S1B). As opposed to the human proteins, the E. coli STAS domain can be readily expressed in bacteria and purified to homogeneity. In addition, the E. coli STAS domain contains a single tryptophan residue (Trp485 in β3; Figure 1), a useful intrinsic fluorescence probe.

Disease-causing human mutations mapped on to the DauA STAS domain

Figure 1
Disease-causing human mutations mapped on to the DauA STAS domain

(A) Multiple sequence alignment of E. coli DauA STAS domain with human SLC26A2, SLC26A3, SLC26A4 and SLC26A5 STAS domains using Clustal Omega. α-Helices are depicted as blue coils, whereas β-strands are presented as blue arrows. Red amino acids in blue-coloured boxes represent highly conserved residues, whereas residues with a red background refer to complete conservation through all proteins. Dark arrows indicate the disease-related mutations numbered 1–12. (B) The equivalent disease-related mutations of human SLC26A2, SLC26A3 and SLC26A4 are mapped on to the structure of E. coli SLC26 protein DauA STAS domain (PDB code 3NY7). The structure of DauA STAS domain is presented as a ribbon using PyMOL (http://www.pymol.org). Mutations on α-helices 1, 2 and 4, β-strand 3 and Leu459 in one linker region are shown as side-chain sticks with 12 different colours. Trp485 is the fluorescence reporter in the DauA STAS structure that was used for spectroscopic studies.

Figure 1
Disease-causing human mutations mapped on to the DauA STAS domain

(A) Multiple sequence alignment of E. coli DauA STAS domain with human SLC26A2, SLC26A3, SLC26A4 and SLC26A5 STAS domains using Clustal Omega. α-Helices are depicted as blue coils, whereas β-strands are presented as blue arrows. Red amino acids in blue-coloured boxes represent highly conserved residues, whereas residues with a red background refer to complete conservation through all proteins. Dark arrows indicate the disease-related mutations numbered 1–12. (B) The equivalent disease-related mutations of human SLC26A2, SLC26A3 and SLC26A4 are mapped on to the structure of E. coli SLC26 protein DauA STAS domain (PDB code 3NY7). The structure of DauA STAS domain is presented as a ribbon using PyMOL (http://www.pymol.org). Mutations on α-helices 1, 2 and 4, β-strand 3 and Leu459 in one linker region are shown as side-chain sticks with 12 different colours. Trp485 is the fluorescence reporter in the DauA STAS structure that was used for spectroscopic studies.

Protein yield comparison between wild-type and mutant STAS domains

Misfolded proteins are often subjected to degradation or aggregate to form inclusion bodies when overexpressed in E. coli. Most STAS domain mutants expressed as well or better than the wild-type STAS domain in total cell extracts (Figure 2). However, the F499Q mutation that introduces a polar residue within the core of the STAS domain resulted in almost complete loss of expression relative to the wild-type STAS domain. Mutants G457I, K484N, L506S, N542M, A545N and A546V had total expression levels similar to the wild-type STAS domain. Interestingly, some mutants (L459N/A, A463K, T469Q and D470Y) were expressed at higher levels than the wild-type protein. However, mutants G457I, T469Q, D470Y, K484N, L506S, N542M, A545N and A546V had low solubility with most of protein in the insoluble pellet (inclusion bodies) upon cell lysis (results not shown). Specifically, the solubility of G457I was extremely low. Most of these mutations changed the polarity of the side chain, with five of them making the mutant protein more hydrophobic. We concluded that, among all of the mutations, only L459N, L459A and A463K showed significantly higher yields of soluble protein than wild-type STAS domain. Interestingly, these mutations generally introduced more polar residues at these positions.

Immunoblot analysis of the total expression levels of His6-tagged DauA STAS and mutants

Figure 2
Immunoblot analysis of the total expression levels of His6-tagged DauA STAS and mutants

(A) Protein expression in BL21(DE3)plysS cells was detected on immunoblots of whole-cell lysates using a mouse anti-His antibody followed by HRP-linked mouse anti-IgG secondary antibody. Levels of LepB were used as an endogenous protein loading control. (B) Quantification of protein expression levels is based on three separate experiments with the wild-type STAS domain expression level set at 100%. Error bars represent S.D.

Figure 2
Immunoblot analysis of the total expression levels of His6-tagged DauA STAS and mutants

(A) Protein expression in BL21(DE3)plysS cells was detected on immunoblots of whole-cell lysates using a mouse anti-His antibody followed by HRP-linked mouse anti-IgG secondary antibody. Levels of LepB were used as an endogenous protein loading control. (B) Quantification of protein expression levels is based on three separate experiments with the wild-type STAS domain expression level set at 100%. Error bars represent S.D.

Characterization of wild-type and mutant STAS domains by size-exclusion chromatography

We performed size-exclusion chromatography to determine the effect of mutations on the mono-dispersity of the STAS domain protein. The DauA STAS domain migrates at the molecular mass of a monomer (13.8 kDa) by size-exclusion chromatography (Figure 3). Most mutants eluted at a much lower volume suggesting that either they have higher oligomeric states or have begun to unfold and become more asymmetric. According to the migration pattern, the mutants could be classified into three groups (Supplementary Table S1). Group 1, including wild-type STAS and A463K eluted at the predicted molecular weight indicating no or weak oligomerization. In group 2, three mutations have strong effects on the protein's elution pattern where oligomeric peaks could be observed, whereas group 3, containing seven mutants, caused the proteins to elute near the void volume suggesting a higher propensity to self-associate and aggregate compared with the other two groups.

Size-exclusion chromatography profile comparison of DauA STAS and mutants

Figure 3
Size-exclusion chromatography profile comparison of DauA STAS and mutants

All proteins were purified using a nickel-affinity column followed by separation on a Superdex™ 75 10/300 GL (24 ml, GE Healthcare) column in 20 mM Tris/HCl, pH 8.0, and 300 mM NaCl, with 1% glycerol and monitored by absorbance at 280 nm. The void volume of the column was at 8 ml and the STAS monomer eluted at 14 ml.

Figure 3
Size-exclusion chromatography profile comparison of DauA STAS and mutants

All proteins were purified using a nickel-affinity column followed by separation on a Superdex™ 75 10/300 GL (24 ml, GE Healthcare) column in 20 mM Tris/HCl, pH 8.0, and 300 mM NaCl, with 1% glycerol and monitored by absorbance at 280 nm. The void volume of the column was at 8 ml and the STAS monomer eluted at 14 ml.

Effect of mutations on the secondary structure of STAS domain

CD was performed to compare the secondary structures and stability of wild-type STAS domain and the mutants (Figure 4, Supplementary Figure S2). Although minor individual negative ellipticities at 208 nm were observed with mutants, L506S, N542M, A545N and A546V, the CD spectra indicated that mutations had a modest effect on helical content, and that the secondary structural elements were maintained. A difference, however, was noticed in the behaviour of the mutants to thermal denaturation based on loss of helicity measured at 222 nm. Wild-type STAS has a melting temperature of 61°C with a high level of co-operativity. Thermal denaturation assays showed that the mutants L459A, A463K, T469Q, D470Y, K484N, R544A and A546V had a slightly reduced melting temperature, whereas the L459N, L506S, N542M and A545N mutants showed a significant decrease in melting temperature (Figure 4B, Supplementary Table S1). The slopes of the melting curves of the mutants were not as steep as the wild-type protein indicating a loss of co-operative unfolding. Some mutants (e.g. N542M) exhibited a biphasic melting curve, indicating a localized effect of the mutation. Most of the mutations affected the thermal stability of helical elements within the STAS domain.

CD spectra of the DauA STAS domain and mutants

Figure 4
CD spectra of the DauA STAS domain and mutants

(A) CD spectra of DauA STAS domain and mutants. (B) Thermal stability melting curves of the STAS domain and mutants as determined by CD with the temperature ranging from 20°C to 80°C and wavelength set at 222 nm. The y-axis represents normalized mean residue ellipticity at 222 nm.

Figure 4
CD spectra of the DauA STAS domain and mutants

(A) CD spectra of DauA STAS domain and mutants. (B) Thermal stability melting curves of the STAS domain and mutants as determined by CD with the temperature ranging from 20°C to 80°C and wavelength set at 222 nm. The y-axis represents normalized mean residue ellipticity at 222 nm.

STAS stability measured by tryptophan fluorescence emission

To further investigate whether disease-causing mutations changed the local folding environment of the STAS domain, we examined the fluorescence properties of the only tryptophan (Trp485) residue of the E. coli STAS domain located at the beginning of β3 (Figure 1). Compared with wild-type STAS, most mutants showed a dramatic decrease in emission intensity, combined with a red shift of the emission wavelength (Figure 5A). This indicates exposure of the tryptophan to a more polar environment, consistent with an unfolded protein. Only A463K exhibited the same emission profile as wild-type, whereas L459N and T469Q showed a minor change in wavelength and fluorescence intensity, respectively (Figures 5A and 5B).

Intrinsic fluorescence and urea denaturation of DauA STAS and mutants

Figure 5
Intrinsic fluorescence and urea denaturation of DauA STAS and mutants

(A) The intrinsic fluorescence emission of purified STAS and mutants was measured from 300 to 420 nm following excitation at 295 nm. (B) Quantification of the fluorescence intensity of purified STAS and mutants measured at 330 nm. Three independent experiments were carried out for each protein. (C) Urea denaturation of STAS and disease-related mutants. Denaturation of the proteins was performed with increasing concentrations of the urea ranging from 0 to 8 M. Proteins were pre-incubated in 200 mM NaCl and 20 mM Tris/HCl, pH 8.0. Fluorescence emission was measured from 300 to 420 nm following excitation at 295 nm. Wavelength at the maximum fluorescence intensity was recorded for each protein with three individual measurements. An average of these three repeats was plotted as a function of urea concentration. The data represent the means ± S.D. of three repeats.

Figure 5
Intrinsic fluorescence and urea denaturation of DauA STAS and mutants

(A) The intrinsic fluorescence emission of purified STAS and mutants was measured from 300 to 420 nm following excitation at 295 nm. (B) Quantification of the fluorescence intensity of purified STAS and mutants measured at 330 nm. Three independent experiments were carried out for each protein. (C) Urea denaturation of STAS and disease-related mutants. Denaturation of the proteins was performed with increasing concentrations of the urea ranging from 0 to 8 M. Proteins were pre-incubated in 200 mM NaCl and 20 mM Tris/HCl, pH 8.0. Fluorescence emission was measured from 300 to 420 nm following excitation at 295 nm. Wavelength at the maximum fluorescence intensity was recorded for each protein with three individual measurements. An average of these three repeats was plotted as a function of urea concentration. The data represent the means ± S.D. of three repeats.

To further understand the consequence of mutations on the stability of the STAS domain, a urea denaturation assay was performed and tryptophan fluorescence emission was also measured from 300 to 420 nm following excitation at 295 nm. In the absence of urea, the mutants displayed a range of red-shifts indicating exposure of the tryptophan to solvent (Figure 5C); thus the mutants with the greatest red-shift represent more open structures. Only T469Q and R544A showed a slightly lower wavelength at the beginning compared with wild-type STAS (Figure 5C). However, the stabilities of T469Q and R544A fell dramatically when urea was increased to 4 and 2 M, respectively, when compared with wild-type STAS (Figure 5C). In addition, the initial intensity of R544A was very low (Figure 5A). In summary, except for A463K, most mutants were unfolded and had decreased stabilities to urea denaturation compared with wild-type STAS.

STAS stability measured by differential scanning fluorimetry assay

DSF was performed to further measure the thermal stability of the STAS domain and the mutants. DSF, also called thermal shift assay, is a simple and quick method for probing the folding of a protein as it experiences progressive denaturation while the temperature is increased. Fundamentally, the temperature at which the protein unfolds is measured by an increase in the fluorescence of the dye SYPRO Orange that has affinity for hydrophobic parts of the protein [63], which are exposed while the protein unfolds. In this experiment, the melting curves show shifts in temperature, suggestive of significantly decreased stability for the mutants L459N, L459A, L506S, N542M and A546V (Table 1, Supplementary Figure S3 and Supplementary Table S1). These data showed similar results to the CD melting assays. Although no increased stability was observed in any mutants, the stability of the A463K mutant did not change compared with wild-type STAS domain, further indicating that A463K does not affect the stability of the STAS domain.

Table 1
Summary of biophysical properties of wild-type STAS and mutants

Oligomerization: ++++, very strong; +++, strong; ++, moderate; +, weak; − almost no oligomerization.

 Oligomerization (gel filtration) Secondary structure (CD), % α-helix CD midpoint of transition (Tm, °C) DSF midpoint of transition (Tm, °C) 
STAS − 35.0±0.8; n=3 60.7±0.9; n=3 51.0±0.8; n=3 
L459N 27.7±2.4; n=3 49.3±1.9; n=3 45.7±0.5; n=3 
L459A ++ 31.0±0; n=3 56.0±1.6; n=3 43.3±0.5; n=3 
A463K − 37.0±0.8; n=3 58.0±1.7; n=3 51.7±0.5; n=3 
T469Q +++ 28.7±0.9; n=3 56.0±1.6; n=3 50.0±0; n=3 
D470Y ++++ 30.7±0.5; n=3 56.0±1.6; n=3 48.3±0.5; n=3 
K484N ++ 30.0±1.4; n=3 54.0±1.6; n=3 48.0±2.1; n=3 
L506S ++++ 28.0±1.6; n=3 50.0±1.6; n=3 45.0±0.8; n=3 
N542M ++++ 30.3±0.9; n=3 50.7±0.9; n=3 43.0±0; n=3 
R544A +++ 30.7±0.5; n=3 54.7±0.9; n=3 48.0±0.8; n=3 
A545N ++++ 29.7±0.5; n=3 53.3±6.7; n=3 46.0±0; n=3 
A546V ++++ 30.7±0.5; n=3 56.7±1.9; n=3 44.3±2.5; n=3 
 Oligomerization (gel filtration) Secondary structure (CD), % α-helix CD midpoint of transition (Tm, °C) DSF midpoint of transition (Tm, °C) 
STAS − 35.0±0.8; n=3 60.7±0.9; n=3 51.0±0.8; n=3 
L459N 27.7±2.4; n=3 49.3±1.9; n=3 45.7±0.5; n=3 
L459A ++ 31.0±0; n=3 56.0±1.6; n=3 43.3±0.5; n=3 
A463K − 37.0±0.8; n=3 58.0±1.7; n=3 51.7±0.5; n=3 
T469Q +++ 28.7±0.9; n=3 56.0±1.6; n=3 50.0±0; n=3 
D470Y ++++ 30.7±0.5; n=3 56.0±1.6; n=3 48.3±0.5; n=3 
K484N ++ 30.0±1.4; n=3 54.0±1.6; n=3 48.0±2.1; n=3 
L506S ++++ 28.0±1.6; n=3 50.0±1.6; n=3 45.0±0.8; n=3 
N542M ++++ 30.3±0.9; n=3 50.7±0.9; n=3 43.0±0; n=3 
R544A +++ 30.7±0.5; n=3 54.7±0.9; n=3 48.0±0.8; n=3 
A545N ++++ 29.7±0.5; n=3 53.3±6.7; n=3 46.0±0; n=3 
A546V ++++ 30.7±0.5; n=3 56.7±1.9; n=3 44.3±2.5; n=3 

Effect of A463K on sensitivity of the STAS domain to trypsin digestion

A463K seems to be the only mutant to behave similarly to wild-type STAS, based on the above assays. To further confirm whether it is a stabilizing mutant or not, we compared the sensitivity to trypsin digestion of STAS and A463K. Tryptic digestions resulted in similar fragmentations of both proteins into one distinct band just below the intact protein on SDS/PAGE (Supplementary Figure S4). Samples of trypsin-treated wild-type STAS were subjected to mass spectrometric analysis. The LC–MS/MS analysis identified peptides in the lower band corresponding to STAS domain lacking the C-terminal AAMADL sequence with trypsin cutting at Arg544 (Supplementary Figure S5). There was also a loss of the N-terminal His6 tag, confirmed by immunoblotting, indicating trypsin cleavage in this region leaving a stable core structure. Compared with wild-type STAS, A463K seems less sensitive to trypsin digestion, and the degrading band can be detected by anti-His immunoblotting, indicating a possible stabilizing effect of A463K on the N-terminal His6 region.

DISCUSSION

This present study used 12 human SLC26 family disease-related mutations (Table 2) mapped into E. coli DauA STAS domain to systematically determine the effects of these mutations on the folding and stability of the protein. DauA STAS has been shown to be a monomer, which is consistent with most STAS domains in previous studies, except for the STAS dimer observed in YtvA and Sultr1;2 in plants [32,64]. Based on size-exclusion chromatography, most of the disease-related mutations caused the protein to unfold and self-associate into higher oligomers and aggregate, indicating destabilizing effects of these mutations. Expression and purification assays indicated decreased expression and solubility in all mutants except for L459N, L459A and A463K. Tryptophan fluorescence emission, CD and DSF assays showed that most mutants have less compact structure and decreased thermal stability compared with wild-type STAS. The mutations, however, had various effects on protein expression, solubility and stability and could be placed into three different groups.

Table 2
Disease-causing mutations in SLC26 STAS domain

PDS, Pendred's syndrome; DFNB4, deafness, autosomal recessive, 4.

Protein name Anion exchange Expression tissue Mutations DauA (YchM) homologue Human diseases 
SLC26A2 SO42– and possibly HCO3 transport Ubiquitous A715V A546 Diastrophic dysplasia 
SLC26A3 Cl exchange for HCO3, OH, SO42– Intestine, sweat gland, pancreas, kidney I544N L459 CCD 
   R554Q T469  
   D652N K484  
SLC26A4 Cl exchange for HCO3, SO42–, I, formate (kidney) Thyroid, kidney, cochlea S552I G457 PDS 
   N558K A463 DFNB4 
   C565Y D470 PDS 
   L676Q F499 DFNB4 
   F683S L506 DFNB4+PDS 
   T721M N542 DFNB4+PDS 
   H723R R544 DFNB4+PDS 
   D724N A545 PDS 
Protein name Anion exchange Expression tissue Mutations DauA (YchM) homologue Human diseases 
SLC26A2 SO42– and possibly HCO3 transport Ubiquitous A715V A546 Diastrophic dysplasia 
SLC26A3 Cl exchange for HCO3, OH, SO42– Intestine, sweat gland, pancreas, kidney I544N L459 CCD 
   R554Q T469  
   D652N K484  
SLC26A4 Cl exchange for HCO3, SO42–, I, formate (kidney) Thyroid, kidney, cochlea S552I G457 PDS 
   N558K A463 DFNB4 
   C565Y D470 PDS 
   L676Q F499 DFNB4 
   F683S L506 DFNB4+PDS 
   T721M N542 DFNB4+PDS 
   H723R R544 DFNB4+PDS 
   D724N A545 PDS 

Mutations causing no expression or extremely low solubility

Gly457 (equivalent residues are Ala542 in rat prestin and Ser552 in human SLC26A4; Figure 1 and Supplementary Figure S1B) is located on the loop between β2 and α1 of the STAS domain. Glycine is highly important for the flexibility of polypeptides and is commonly found in turns and loops connecting secondary structural elements. We speculate that mutant G457I may decrease the flexibility of the loop and the side-chain orientations inside the STAS domain, leading to the forming of severely misfolded protein. Structure alignment shows that the corresponding residue in rat prestin is Ala542 (Supplementary Figure S1B), a similar small residue sitting in the loop between α1 and β2 (β2 is the third β-strand, since the first β-strand is named β0). However, the corresponding residue in human SLC26A4 is serine (Figure 1A). Serine is also a common residue in loops [65]. Based on the structural similarity between rat prestin and DauA STAS, we speculate that Ser552 in human SLC26A4 may also reside on the loop region. Therefore the disease-causing mutant S552I may affect the flexibility and solubility of the protein since isoleucine is a hydrophobic residue that prefers to be buried within the hydrophobic core of a protein rather than a loop region that generally comprises polar residues.

Similarly to Gly457, mutation of Phe499 (Leu660 in rat prestin and Leu676 in human SLC26A4; Figure 1 and Supplementary Figure S1B) located in α2 shows no or extremely low expression (Figures 1 and 2). Strongly misfolded protein may form due to a disruption of the non-covalent π–π interactions between Phe499 and the aromatic rings of Trp485, Phe468 and Phe502 located inside the protein (Supplementary Figure S7A). The corresponding residue in rat prestin Leu660 (α2, Supplementary Figure S1B) is also located in the hydrophobic core region. Therefore we speculate that the disease-causing mutant L676Q in human SLC26A4 may have a similar effect by interfering the hydrophobic core region of the STAS domain by introduction of a polar residue causing the domain to misfold severely.

Mutations on α1 and the loop connecting α1 and β2

The conserved residue Leu459 (Ile544 in rat prestin and Ile544 in human SLC26A3; Figure 1 and Supplementary Figure S1B) is in the loop connecting α1 and β2 and is close to the hydrophobic Leu491 (Met652 in rat prestin and Leu659 in SLC26A3) and Val488 (Val649 in prestin, Val656 in SLC26A3) (Supplementary Figure S6), implying possible hydrophobic interactions among these residues. A strong trafficking defect of transporter SLC26A3 I544N was observed previously [58]. Therefore, to identify the role of leucine in this region, we mutated the leucine into asparagine and alanine, respectively. The replacement of Leu459 with alanine has a milder effect than mutation into the more polar asparagine. The CD spectrum from L459N showed a larger decrease in helical content compared with L459A. Data from the melting temperature curve also indicates that L459A is more stable than L459N. Unexpectedly, the fluorescence emission assay showed that L459N has a lower intrinsic wavelength, implying a relatively more compact structure than L459A. However, although differences exist between L459N and L459A, these two mutations both changed the natural folding of the STAS domain from different aspects, implying that there is a specific role for leucine (isoleucine) in this loop that contributes to proper folding.

Most residues on α1 of DauA and human STAS domain are poorly conserved (Figure 1A). We speculate that these residues are correlated with the specific IVSs (intervening sequences) that exists in mammalian and plant homologues. Mammalian and plant genes generally encode longer STAS domains compared with bacterial genes [32]. The residues on α1 probably interact with this IVS (∼80 amino acids). Even within mammalian SLC26 family members, some residues are not conserved within all forms (Figure 1A, Mut4 and Mut5), implying specific folding of α1 in different human SLC26 proteins. This IVS is located at the end of α1 in the rat prestin structure but was deleted due to its interference with crystallization of the domain. Deletion of this IVS region in prestin structure makes the analysis of the structure alignment difficult at the region of α1, since we are not sure the tilting away of prestin α1 from DauA α1 (θ1 and θ2 in Supplementary Figure S1A) is due to the deletion of the IVS or to the endogenous structural difference between DauA and prestin. Therefore, the homologous residues on α1 (Ala463, Thr469, Asp470) can only be predicted from a primary sequence alignment at the time.

The A463K mutant in α1 (Asn548 in rat prestin and Asn558 in human SLC26A4; Figure 1) appears to be a stabilizing mutant. The mutation would increase the overall polarity of the domain. In the DauA STAS domain structure, Thr437 and the positively charged His435 and Arg438 are closer to Ala463 than the other residues (Supplementary Figure S7). We speculate that an Ala463 mutation to lysine would cause the helix to tilt away from the positively charged Arg438 and His435 due to the electrostatic repulsion. The mutation also facilitates a possible interaction between Lys463 and Thr437 through hydrogen bonds. All of this would contribute to the decreased flexibility of the N terminal loop, but have no or little impact on the overall fold. This was verified in trypsin digestion assays (Supplementary Figure S4) where wild-type STAS was digested faster than STAS A463K and the degraded band in A463K could be detected by anti-His antibody. However, since the N-terminal region of the STAS domain is likely to contact the TM domain of the SLC26 transporter, the stabilizing effect of A463K is probably an artificial one and may not represent the corresponding disease-causing mutation in vivo.

Both Thr469 (Ser549 in rat prestin, Arg554 in human SLC26A3; Figure 1) and Asp470 (Asp550 in prestin and Cys565 in human SLC26A4; Figure 1) are distributed on the surface of STAS α1 and are exposed to the aqueous environment (Figure 1B). The secondary structures and melting temperatures of T469Q and D470Y showed mild changes (Figure 4 and Supplementary Table S1), but D470Y showed a drastic red-shift of wavelength combined with a decreased intrinsic fluorescence intensity compared with T469Q (Figure 5). This could be explained by the different effects on the size and charge of the residues that the two mutations bring, as threonine to glutamine is a relatively minor change compared with an aspartic acid to tyrosine mutation. However in the human SLC26A3 protein, Arg554 is a strong positively charged residue and mutation to glutamine could strongly affect protein folding.

Mutation of Asp470 to tyrosine would eliminate the negative charge of aspartic acid and increase the volume of the original residue. A change from aspartic acid to tyrosine also decreases the solubility of the wild-type protein, perhaps due to replacement of a charged residue with a hydrophobic residue. D470Y may also contribute to the oligomerization of the protein by promoting surface–surface hydrophobic interactions. This can be seen from the gel-filtration profile in which D470Y has a larger degree of oligomerization compared with T469Q. Therefore, in human SLC26A4, we suspect that mutation of cysteine into tyrosine may also lead to the oligomerization of the natural proteins through non-native surface hydrophobic interactions (π–π interactions). Furthermore, a charge–charge interaction was observed between Asp470 and Arg474 with a distance of 2.9 Å (1 Å=0.1 nm) (Supplementary Figure S7C), although mutation of Asp470 into tyrosine could destroy this interaction and destabilize the loop connecting α1 and β2, which may result in the misfolding of the protein. However, the corresponding residues of Asp470 and Arg474 in human SLC26A4 are Cys565 and Thr569, respectively (Figure 1). It seems that Cys565 and Thr569 could not form a bond as strong as the bond between Asp470 and Arg474. Therefore, this mechanism is not explanatory in human SLC26A4.

Mutations in β3 and α2

Lys484 (Asp645 in rat prestin, Asp652 in human SLC26A3; Figure 1 and Supplementary Figure S1B) is situated on the middle of β3 (tip of β3 in prestin) (Supplementary Figure S8A). Charge repulsion may exist between Lys484 towards Arg454 and Arg543. In addition, there is a strong electrostatic interaction formed between Lys484 and Asp486 (3.2 Å, not shown). We suspect that mutation of Lys484 to asparagine may then increase the attractive force between Asn484 and Arg454 or Arg543 with hydrogen bonds. This may further contribute to a slight tilting of β2 and β3 towards α3 and α4, which in turn would lead to a more opened structure of STAS compared with wild-type protein (model in Supplementary Figure S8B). As shown in the fluorescence emission assay, K484N has a significant red-shift compared with wild-type and most of other mutants. The strong bond between Lys484 and Asp486 in DauA seems have evolved into hydrogen bonds between aspartic acid and threonine/serine (Figure 1, arrow ‘6’) in rat prestin (equalivalent to human SLC26A5) and human SLC26A3. As previously indicated, molecular dynamics showed that linker regions between α1 and α2 in human pendrin tend to be more flexible than in rat or bacteria [66], therefore the unfolding effect of the equivalent mutants between α1 and α2 in human may not be as dramatic as in bacteria and needs further investigation.

Leu506 (Tyr667 in rat prestin, Phe683 in human SLC26A4; Figure 1 and Supplementary Figure S1B) is located at the end of the α2 and probably interacts with Leu512 (valine in rat and human) through hydrophobic interactions (Supplementary Figure S8C). By mutating leucine into serine, the hydrophobic region may be compromised since serine is hydrophilic and tends to face the aqueous environment. The mutations may also contribute to the increased interaction between Ser506 and Arg536 through hydrogen bonds (Supplementary Figure S8C). All of these changes would enable the protein to become less compact. Mutation of the corresponding aromatic residues in human SLC26A4 may have similar effect in eliminating the hydrophobic interactions between Tyr667 and Val674 (rat prestin) (Supplementary Figure S8C).

Mutations in α4

Asn542, Arg544, Ala545 and Ala546 are all distributed on the last α-helix (α4) of the STAS domain (Supplementary Figure S9). Ala545, Ala546 and Ile482 contribute to the formation of a small hydrophobic region (Supplementary Figure S9, orange circle). A salt bridge exists between Asp549 and Arg513, which may function as a lock to secure this hydrophobic region. Mutating Ala546 to valine increased the volume of the residue, whereas mutant A545N increased polarity on the inside of the protein, all of which could interfere the formation of the hydrophobic core. However, the corresponding residue of Ala545 in human SLC26A4 is a hydrophilic Asp724 (Asp708 in rat prestin; Figure 1, Supplementary Figures S1B and S10). Structure comparison with DauA STAS showed that prestin STAS has an additional N-terminal β0 strand followed by a long loop before β1 (Supplementary Figure S10). With these ‘new features’, new interactions formed between His707 (equalivalent to Arg544 in DauA) and Tyr520 through π–π interaction and His707 and Asp518 by salt interaction, which are not observed in DauA STAS. These interactions enable the α4-helix to tilt towards β1, which in turn may facilitate the solvent-facing of site Asp708. This is a likely reason for the corresponding residue of DauA Ala545 in human being a hydrophilic one (Asp724) rather than any hydrophobic residue. Compared with other mutants in α1, Ala546 is the most conserved residue in both DauA and human homologues, and it is constantly buried in the hydrophobic core region. Finally, alanine is a strong α-helix former and mutation to valine (DauA A546V and SLC26A2 A715V), a strong β-strand former, would be expected to destabilize the α4 helix.

The corresponding residues of Asn542 and Arg544 in human SLC26A4 are Thr721 and His723 respectively (Figure 1). These surface residues are exposed to the aqueous solvent environment. Mutations at these two sites (N542M and R544A) to hydrophobic residues would decrease the solubility of the protein and increase the protein's propensity to oligomerize and aggregate through surface hydrophobic interactions. A decreased expression and solubility of these mutants can be seen in the expression and purification assays probably due to formation of insoluble inclusion bodies.

Overall, we conclude that most of these disease-causing mutants changed the natural folding of wild-type protein and decreased the stability of STAS domain resulting in self-association and aggregation in most cases. The present study may provide important information of the STAS working mechanism in regulating SLC26 transporter function. In particular, our studies indicate that, although the mutations caused protein misfolding, it is important to examine the mutations on a case-by-case basis as they vary in the effects on the protein. We are now examining the effect of these STAS mutations on the cell-surface expression of the intact human SLC26 proteins in transfected HEK-293 cells. We hypothesize that the misfolding of the STAS domain will lead to aggregation and retention of the proteins in the endoplasmic reticulum by the quality control system of the cell.

AUTHOR CONTRIBUTION

All experiments were carried out by Xiaoyun Bai, who wrote the manuscript and prepared the figures. Reinhart Reithmeier and Trevor Moraes developed and supervised the research project, and edited the manuscript before submission.

We thank Jing Li for providing the DauA cDNA. Fan Xia and Chloe Rapp are thanked for project discussion. Charles Calmettes, Christine Lai, Andrew Judd, Yogesh Hooda and Anastassia Pogoutse are thanked for help in protein purification and trypsin digestion assay. Nardin Nano, Elisa Leung, Sam and Kaiyu Liu in Dr Walid Houry's laboratory are thanked for technical support in CD and urea denaturation assays. Anti-LepB was received as a gift from Dr Jan-Willem de Gier.

FUNDING

This work was supported by the Canadian Institutes of Health Research [grant number MOP102493 (to R.A.F.R.)]; the Natural Sciences and Engineering Research Council of Canada–Discovery Grants [grant number RGPIN 401975-11 (to T.F.M.)]; and a Postdoctoral Fellowship from ‘CIHR Training Program in Protein Folding and Interaction Dynamics’ (to X.B.).

Abbreviations

     
  • CCD

    congenital chloride-losing diarrhoea

  •  
  • CD

    circular dichroism

  •  
  • DSF

    differential scanning fluorimetry

  •  
  • HEK

    human embryonic kidney

  •  
  • HRP

    horseradish peroxidase

  •  
  • IVS

    intervening sequence

  •  
  • SLC26

    solute carrier 26

  •  
  • STAS

    sulfate transporter anti-sigma factor antagonist

  •  
  • TM

    transmembrane

References

References
1
Sherman
T.
Chernova
M.N.
Clark
J.S.
Jiang
L.
Alper
S.L.
Nehrke
K.
The abts and sulp families of anion transporters from Caenorhabditis elegans
Am. J. Physiol. Cell Physiol
2005
, vol. 
289
 (pg. 
C341
-
C351
)
[PubMed]
2
Felce
J.
Saier
M.H.
Jr
Carbonic anhydrases fused to anion transporters of the SulP family: evidence for a novel type of bicarbonate transporter
J. Mol. Microbiol. Biotechnol.
2004
, vol. 
8
 (pg. 
169
-
176
)
[PubMed]
3
Dorwart
M.R.
Shcheynikov
N.
Yang
D.
Muallem
S.
The solute carrier 26 family of proteins in epithelial ion transport
Physiology
2008
, vol. 
23
 (pg. 
104
-
114
)
[PubMed]
4
Takahashi
H.
Watanabe-Takahashi
A.
Smith
F.W.
Blake-Kalff
M.
Hawkesford
M.J.
Saito
K.
The roles of three functional sulphate transporters involved in uptake and translocation of sulphate in Arabidopsis thaliana
Plant J.
2000
, vol. 
23
 (pg. 
171
-
182
)
[PubMed]
5
Shibagaki
N.
Rose
A.
McDermott
J.P.
Fujiwara
T.
Hayashi
H.
Yoneyama
T.
Davies
J.P.
Selenate-resistant mutants of Arabidopsis thaliana identify Sultr1;2, a sulfate transporter required for efficient transport of sulfate into roots
Plant J.
2002
, vol. 
29
 (pg. 
475
-
486
)
[PubMed]
6
Cherest
H.
Davidian
J.C.
Thomas
D.
Benes
V.
Ansorge
W.
Surdin-Kerjan
Y.
Molecular characterization of two high affinity sulfate transporters in Saccharomyces cerevisiae
Genetics
1997
, vol. 
145
 (pg. 
627
-
635
)
[PubMed]
7
Lohi
H.
Kujala
M.
Makela
S.
Lehtonen
E.
Kestila
M.
Saarialho-Kere
U.
Markovich
D.
Kere
J.
Functional characterization of three novel tissue-specific anion exchangers SLC26A7, -A8, and -A9
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
14246
-
14254
)
[PubMed]
8
Moseley
R.H.
Hoglund
P.
Wu
G.D.
Silberg
D.G.
Haila
S.
de la Chapelle
A.
Holmberg
C.
Kere
J.
Downregulated in adenoma gene encodes a chloride transporter defective in congenital chloride diarrhea
Am. J. Physiol.
1999
, vol. 
276
 (pg. 
G185
-
G192
)
[PubMed]
9
Satoh
H.
Susaki
M.
Shukunami
C.
Iyama
K.
Negoro
T.
Hiraki
Y.
Functional analysis of diastrophic dysplasia sulfate transporter. Its involvement in growth regulation of chondrocytes mediated by sulfated proteoglycans
J. Biol. Chem.
1998
, vol. 
273
 (pg. 
12307
-
12315
)
[PubMed]
10
Soleimani
M.
Greeley
T.
Petrovic
S.
Wang
Z.
Amlal
H.
Kopp
P.
Burnham
C.E.
Pendrin: an apical Cl-/OH-/HCO3- exchanger in the kidney cortex
Am. J. Physiol. Renal. Physiol.
2001
, vol. 
280
 (pg. 
F356
-
F364
)
[PubMed]
11
Xie
Q.
Welch
R.
Mercado
A.
Romero
M.F.
Mount
D.B.
Molecular characterization of the murine Slc26a6 anion exchanger: functional comparison with Slc26a1
Am. J. Physiol. Renal. Physiol.
2002
, vol. 
283
 (pg. 
F826
-
F838
)
[PubMed]
12
Jiang
Z.
Grichtchenko
I.I.
Boron
W.F.
Aronson
P.S.
Specificity of anion exchange mediated by mouse Slc26a6
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
33963
-
33967
)
[PubMed]
13
Karniski
L.P.
Lotscher
M.
Fucentese
M.
Hilfiker
H.
Biber
J.
Murer
H.
Immunolocalization of sat-1 sulfate/oxalate/bicarbonate anion exchanger in the rat kidney
Am. J. Physiol.
1998
, vol. 
275
 (pg. 
F79
-
F87
)
[PubMed]
14
Scott
D.A.
Karniski
L.P.
Human pendrin expressed in Xenopus laevis oocytes mediates chloride/formate exchange
Am. J. Physiol. Cell Physiol.
2000
, vol. 
278
 (pg. 
C207
-
C211
)
[PubMed]
15
Shcheynikov
N.
Wang
Y.
Park
M.
Ko
S.B.
Dorwart
M.
Naruse
S.
Thomas
P.J.
Muallem
S.
Coupling modes and stoichiometry of Cl-/HCO3- exchange by slc26a3 and slc26a6
J. Gen. Physiol.
2006
, vol. 
127
 (pg. 
511
-
524
)
[PubMed]
16
Kim
K.H.
Shcheynikov
N.
Wang
Y.
Muallem
S.
SLC26A7 is a Cl- channel regulated by intracellular pH
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
6463
-
6470
)
[PubMed]
17
Lohi
H.
Lamprecht
G.
Markovich
D.
Heil
A.
Kujala
M.
Seidler
U.
Kere
J.
Isoforms of SLC26A6 mediate anion transport and have functional PDZ interaction domains
Am. J. Physiol. Cell Physiol.
2003
, vol. 
284
 (pg. 
C769
-
C779
)
[PubMed]
18
Makela
S.
Kere
J.
Holmberg
C.
Hoglund
P.
SLC26A3 mutations in congenital chloride diarrhea
Hum. Mutat.
2002
, vol. 
20
 (pg. 
425
-
438
)
[PubMed]
19
Navaratnam
D.
Bai
J.P.
Samaranayake
H.
Santos-Sacchi
J.
N-terminal-mediated homomultimerization of prestin, the outer hair cell motor protein
Biophys. J.
2005
, vol. 
89
 (pg. 
3345
-
3352
)
[PubMed]
20
Lu
F.
Li
S.
Jiang
Y.
Jiang
J.
Fan
H.
Lu
G.
Deng
D.
Dang
S.
Zhang
X.
Wang
J.
Yan
N.
Structure and mechanism of the uracil transporter UraA
Nature
2011
, vol. 
472
 (pg. 
243
-
246
)
[PubMed]
21
Geertsma
E.R.
Chang
Y.N.
Shaik
F.R.
Neldner
Y.
Pardon
E.
Steyaert
J.
Dutzler
R.
Structure of a prokaryotic fumarate transporter reveals the architecture of the SLC26 family
Nat. Struct. Mol. Biol.
2015
, vol. 
22
 (pg. 
803
-
808
)
[PubMed]
22
Min
K.T.
Hilditch
C.M.
Diederich
B.
Errington
J.
Yudkin
M.D.
Sigma F, the first compartment-specific transcription factor of B. subtilis, is regulated by an anti-sigma factor that is also a protein kinase
Cell
1993
, vol. 
74
 (pg. 
735
-
742
)
[PubMed]
23
Losick
R.
Pero
J.
Cascades of Sigma factors
Cell
1981
, vol. 
25
 (pg. 
582
-
584
)
[PubMed]
24
Campbell
E.A.
Westblade
L.F.
Darst
S.A.
Regulation of bacterial RNA polymerase sigma factor activity: a structural perspective
Curr. Opin. Microbiol.
2008
, vol. 
11
 (pg. 
121
-
127
)
[PubMed]
25
Babu
M.
Greenblatt
J.F.
Emili
A.
Strynadka
N.C.
Reithmeier
R.A.
Moraes
T.F.
Structure of a SLC26 anion transporter STAS domain in complex with acyl carrier protein: implications for E
coli YchM in fatty acid metabolism. Structure
2010
, vol. 
18
 (pg. 
1450
-
1462
)
[PubMed]
26
Chang
M.H.
Plata
C.
Sindic
A.
Ranatunga
W.K.
Chen
A.P.
Zandi-Nejad
K.
Chan
K.W.
Thompson
J.
Mount
D.B.
Romero
M.F.
Slc26a9 is inhibited by the R-region of the cystic fibrosis transmembrane conductance regulator via the STAS domain
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
28306
-
28318
)
[PubMed]
27
Homma
K.
Miller
K.K.
Anderson
C.T.
Sengupta
S.
Du
G.G.
Aguinaga
S.
Cheatham
M.
Dallos
P.
Zheng
J.
Interaction between CFTR and prestin (SLC26A5)
Biochim. Biophys. Acta
2010
, vol. 
1798
 (pg. 
1029
-
1040
)
[PubMed]
28
Ko
S.B.
Shcheynikov
N.
Choi
J.Y.
Luo
X.
Ishibashi
K.
Thomas
P.J.
Kim
J.Y.
Kim
K.H.
Lee
M.G.
Naruse
S.
Muallem
S.
A molecular mechanism for aberrant CFTR-dependent HCO(3)(-) transport in cystic fibrosis
EMBO J.
2002
, vol. 
21
 (pg. 
5662
-
5672
)
[PubMed]
29
Shcheynikov
N.
Yang
D.
Wang
Y.
Zeng
W.
Karniski
L.P.
So
I.
Wall
S.M.
Muallem
S.
The Slc26a4 transporter functions as an electroneutral Cl-/I-/HCO3- exchanger: role of Slc26a4 and Slc26a6 in I- and HCO3-secretion and in regulation of CFTR in the parotid duct
J. Physiol.
2008
, vol. 
586
 (pg. 
3813
-
3824
)
[PubMed]
30
Karinou
E.
Compton
E.L.
Morel
M.
Javelle
A.
The Escherichia coli SLC26 homologue YchM (DauA) is a C(4)-dicarboxylic acid transporter
Mol. Microbiol.
2013
, vol. 
87
 (pg. 
623
-
640
)
[PubMed]
31
Pasqualetto
E.
Aiello
R.
Gesiot
L.
Bonetto
G.
Bellanda
M.
Battistutta
R.
Structure of the cytosolic portion of the motor protein prestin and functional role of the STAS domain in SLC26/SulP anion transporters
J. Mol. Biol.
2010
, vol. 
400
 (pg. 
448
-
462
)
[PubMed]
32
Sharma
A.K.
Rigby
A.C.
Alper
S.L.
STAS domain structure and function
Cell. Physiol. Biochem.
2011
, vol. 
28
 (pg. 
407
-
422
)
[PubMed]
33
Mount
D.B.
Romero
M.F.
The SLC26 gene family of multifunctional anion exchangers
Pflugers Arch.
2004
, vol. 
447
 (pg. 
710
-
721
)
34
Sindic
A.
Chang
M.H.
Mount
D.B.
Romero
M.F.
Renal physiology of SLC26 anion exchangers
Curr. Opin. Nephrol. Hypertens.
2007
, vol. 
16
 (pg. 
484
-
490
)
[PubMed]
35
Sterling
D.
Casey
J.R.
Bicarbonate transport proteins
Biochem. Cell Biol.
2002
, vol. 
80
 (pg. 
483
-
497
)
[PubMed]
36
Ayoubi
J.M.
Jouk
P.S.
Pons
J.C.
Diastrophic dwarfism and pregnancy
Lancet
2001
, vol. 
358
 pg. 
1778
 
[PubMed]
37
Bonafe
L.
Hastbacka
J.
de la Chapelle
A.
Campos-Xavier
A.B.
Chiesa
C.
Forlino
A.
Superti-Furga
A.
Rossi
A.
A novel mutation in the sulfate transporter gene SLC26A2 (DTDST) specific to the Finnish population causes de la Chapelle dysplasia
J. Med. Genet.
2008
, vol. 
45
 (pg. 
827
-
831
)
[PubMed]
38
Forlino
A.
Piazza
R.
Tiveron
C.
Della Torre
S.
Tatangelo
L.
Bonafe
L.
Gualeni
B.
Romano
A.
Pecora
F.
Superti-Furga
A.
, et al. 
A diastrophic dysplasia sulfate transporter (SLC26A2) mutant mouse: morphological and biochemical characterization of the resulting chondrodysplasia phenotype
Hum. Mol. Genet.
2005
, vol. 
14
 (pg. 
859
-
871
)
[PubMed]
39
Hall
B.D.
Diastrophic dysplasia: extreme variability within a sibship
Am. J. Med Genet.
1996
, vol. 
63
 (pg. 
28
-
33
)
[PubMed]
40
Hastbacka
J.
de la Chapelle
A.
Mahtani
M.M.
Clines
G.
Reeve-Daly
M.P.
Daly
M.
Hamilton
B.A.
Kusumi
K.
Trivedi
B.
Weaver
A.
, et al. 
The diastrophic dysplasia gene encodes a novel sulfate transporter: positional cloning by fine-structure linkage disequilibrium mapping
Cell
1994
, vol. 
78
 (pg. 
1073
-
1087
)
[PubMed]
41
Hastbacka
J.
Kaitila
I.
Sistonen
P.
de la Chapelle
A.
Diastrophic dysplasia gene maps to the distal long arm of chromosome 5
Proc. Natl. Acad. Sci. U.S.A.
1990
, vol. 
87
 (pg. 
8056
-
8059
)
[PubMed]
42
Megarbane
A.
Farkh
I.
Haddad-Zebauni
S.
How many phenotypes in the DTDST family chondrodysplasias?
Clin. Genet.
2002
, vol. 
62
 (pg. 
189
-
190
)
[PubMed]
43
Panzer
K.M.
Lachman
R.
Modaff
P.
Pauli
R.M.
A phenotype intermediate between Desbuquois dysplasia and diastrophic dysplasia secondary to mutations in DTDST
Am. J. Med. Genet. A
2008
, vol. 
146A
 (pg. 
2920
-
2924
)
[PubMed]
44
Dixon
J.
Edwards
S.J.
Gladwin
A.J.
Dixon
M.J.
Loftus
S.K.
Bonner
C.A.
Koprivnikar
K.
Wasmuth
J.J.
Positional cloning of a gene involved in the pathogenesis of Treacher Collins syndrome
Nat. Genet.
1996
, vol. 
12
 (pg. 
130
-
136
)
[PubMed]
45
Kere
J.
Lohi
H.
Hoglund
P.
Genetic disorders of membrane transport III. Congenital chloride diarrhea
Am. J. Physiol.
1999
, vol. 
276
 (pg. 
G7
-
G13
)
[PubMed]
46
Taylor
J.P.
Metcalfe
R.A.
Watson
P.F.
Weetman
A.P.
Trembath
R.C.
Mutations of the PDS gene, encoding pendrin, are associated with protein mislocalization and loss of iodide efflux: implications for thyroid dysfunction in Pendred syndrome
J. Clin. Endocrinol. Metabol.
2002
, vol. 
87
 (pg. 
1778
-
1784
)
[PubMed]
47
Blons
H.
Feldmann
D.
Duval
V.
Messaz
O.
Denoyelle
F.
Loundon
N.
Sergout-Allaoui
A.
Houang
M.
Duriez
F.
Lacombe
D.
, et al. 
Screening of SLC26A4 (PDS) gene in Pendred's syndrome: a large spectrum of mutations in France and phenotypic heterogeneity
Clin. Genet.
2004
, vol. 
66
 (pg. 
333
-
340
)
[PubMed]
48
Pryor
S.P.
Madeo
A.C.
Reynolds
J.C.
Sarlis
N.J.
Arnos
K.S.
Nance
W.E.
Yang
Y.
Zalewski
C.K.
Brewer
C.C.
Butman
J.A.
Griffith
A.J.
SLC26A4/PDS genotype-phenotype correlation in hearing loss with enlargement of the vestibular aqueduct (EVA): evidence that Pendred syndrome and non-syndromic EVA are distinct clinical and genetic entities
J. Med. Genet.
2005
, vol. 
42
 (pg. 
159
-
165
)
[PubMed]
49
Choi
B.Y.
Stewart
A.K.
Madeo
A.C.
Pryor
S.P.
Lenhard
S.
Kittles
R.
Eisenman
D.
Kim
H.J.
Niparko
J.
Thomsen
J.
, et al. 
Hypo-functional SLC26A4 variants associated with nonsyndromic hearing loss and enlargement of the vestibular aqueduct: genotype-phenotype correlation or coincidental polymorphisms?
Hum. Mutat.
2009
, vol. 
30
 (pg. 
599
-
608
)
[PubMed]
50
Xu
J.
Song
P.
Nakamura
S.
Miller
M.
Barone
S.
Alper
S.L.
Riederer
B.
Bonhagen
J.
Arend
L.J.
Amlal
H.
, et al. 
Deletion of the chloride transporter slc26a7 causes distal renal tubular acidosis and impairs gastric acid secretion
J. Biol. Chem.
2009
, vol. 
284
 (pg. 
29470
-
29479
)
[PubMed]
51
Xu
J.
Song
P.
Miller
M.L.
Borgese
F.
Barone
S.
Riederer
B.
Wang
Z.
Alper
S.L.
Forte
J.G.
Shull
G.E.
, et al. 
Deletion of the chloride transporter Slc26a9 causes loss of tubulovesicles in parietal cells and impairs acid secretion in the stomach
Proc. Natl. Acad. Sci. U.S.A.
2008
, vol. 
105
 (pg. 
17955
-
17960
)
[PubMed]
52
Jiang
Z.
Asplin
J.R.
Evan
A.P.
Rajendran
V.M.
Velazquez
H.
Nottoli
T.P.
Binder
H.J.
Aronson
P.S.
Calcium oxalate urolithiasis in mice lacking anion transporter Slc26a6
Nat. Genet.
2006
, vol. 
38
 (pg. 
474
-
478
)
[PubMed]
53
Dawson
P.A.
Russell
C.S.
Lee
S.
McLeay
S.C.
van Dongen
J.M.
Cowley
D.M.
Clarke
L.A.
Markovich
D.
Urolithiasis and hepatotoxicity are linked to the anion transporter Sat1 in mice
J. Clin. Invest.
2010
, vol. 
120
 (pg. 
706
-
712
)
[PubMed]
54
Wang
Y.
Soyombo
A.A.
Shcheynikov
N.
Zeng
W.
Dorwart
M.
Marino
C.R.
Thomas
P.J.
Muallem
S.
Slc26a6 regulates CFTR activity in vivo to determine pancreatic duct HCO3- secretion: relevance to cystic fibrosis
EMBO J.
2006
, vol. 
25
 (pg. 
5049
-
5057
)
[PubMed]
55
Ishiguro
H.
Namkung
W.
Yamamoto
A.
Wang
Z.
Worrell
R.T.
Xu
J.
Lee
M.G.
Soleimani
M.
Effect of Slc26a6 deletion on apical Cl-/HCO3-exchanger activity and cAMP-stimulated bicarbonate secretion in pancreatic duct
Am. J. Physiol. Gastrointest. Liver Physiol.
2007
, vol. 
292
 (pg. 
G447
-
G455
)
[PubMed]
56
Tuo
B.
Riederer
B.
Wang
Z.
Colledge
W.H.
Soleimani
M.
Seidler
U.
Involvement of the anion exchanger SLC26A6 in prostaglandin E2- but not forskolin-stimulated duodenal HCO3- secretion
Gastroenterology
2006
, vol. 
130
 (pg. 
349
-
358
)
[PubMed]
57
Cesar-Razquin
A.
Snijder
B.
Frappier-Brinton
T.
Isserlin
R.
Gyimesi
G.
Bai
X.
Reithmeier
R.A.
Hepworth
D.
Hediger
M.A.
Edwards
A.M.
Superti-Furga
G.
A call for systematic research on solute carriers
Cell
2015
, vol. 
162
 (pg. 
478
-
487
)
[PubMed]
58
Dorwart
M.R.
Shcheynikov
N.
Baker
J.M.
Forman-Kay
J.D.
Muallem
S.
Thomas
P.J.
Congenital chloride-losing diarrhea causing mutations in the STAS domain result in misfolding and mistrafficking of SLC26A3
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
8711
-
8722
)
[PubMed]
59
de Gier
J.W.
Scotti
P.A.
Saaf
A.
Valent
Q.A.
Kuhn
A.
Luirink
J.
von Heijne
G.
Differential use of the signal recognition particle translocase targeting pathway for inner membrane protein assembly in Escherichia coli
Proc. Natl. Acad. Sci. U.S.A.
1998
, vol. 
95
 (pg. 
14646
-
14651
)
[PubMed]
60
Pang
A.J.
Bustos
S.P.
Reithmeier
R.A.
Structural characterization of the cytosolic domain of kidney chloride/bicarbonate anion exchanger 1 (kAE1)
Biochemistry
2008
, vol. 
47
 (pg. 
4510
-
4517
)
[PubMed]
61
Whitmore
L.
Wallace
B.A.
Protein secondary structure analyses from circular dichroism spectroscopy: methods and reference databases
Biopolymers
2008
, vol. 
89
 (pg. 
392
-
400
)
[PubMed]
62
Whitmore
L.
Wallace
B.A.
DICHROWEB, an online server for protein secondary structure analyses from circular dichroism spectroscopic data
Nucleic Acids Res.
2004
, vol. 
32
 (pg. 
W668
-
W673
)
[PubMed]
63
Lavinder
J.J.
Hari
S.B.
Sullivan
B.J.
Magliery
T.J.
High-throughput thermal scanning: a general, rapid dye-binding thermal shift screen for protein engineering
J. Am. Chem. Soc.
2009
, vol. 
131
 (pg. 
3794
-
3795
)
[PubMed]
64
Jurk
M.
Dorn
M.
Schmieder
P.
Blue flickers of hope: secondary structure, dynamics, and putative dimerization interface of the blue-light receptor YtvA from Bacillus subtilis
Biochemistry
2011
, vol. 
50
 (pg. 
8163
-
8171
)
[PubMed]
65
Betts
M.J.
Russell
R.B.
Barnes
M.R.
Gray
I.C.
Amino acid properties and consequences of substitutions
Bioinformatics for Geneticists
2003
Chichester
John Wiley & Sons
66
Sharma
A.K.
Zelikovic
I.
Alper
S.L.
Molecular dynamics simulations of the STAS domains of rat prestin and human pendrin reveal conformational motions in conserved flexible regions
Cell. Physiol. Biochem.
2014
, vol. 
33
 (pg. 
605
-
620
)
[PubMed]