Abstract

SLC26A6 (PAT1, CFEX) is an anion exchanger that is expressed on the apical membrane of the kidney proximal tubule and the small intestine. Modes of transport mediated by SLC26A6 include Cl/formate exchange, Cl/HCO3 exchange, and Cl/oxalate exchange. To study its role in kidney and intestinal physiology, gene targeting was used to prepare mice lacking Slc26a6. Homozygous mutant Slc26a6−/− mice appeared healthy and exhibited a normal blood pressure, kidney function, and plasma electrolyte profile. In proximal tubules microperfused with a low-HCO3/high-Cl solution, the baseline rate of fluid absorption (Jv), an index of NaCl transport under these conditions, was the same in wild-type and null mice. However, the stimulation of Jv by oxalate observed in wild-type mice was completely abolished in Slc26a6-null mice (P < 0.05). Formate stimulation of Jv was partially reduced in null mice, but the difference from the response in wild-type mice did not reach statistical significance. Apical membrane Cl/base exchange activity, assayed with the pH-sensitive dye BCPCF in microperfused proximal tubules, was decreased by 58% in Slc26a6−/− animals (P < 0.001 vs. wild types). In the duodenum, the baseline rate of HCO3 secretion measured in mucosal tissue mounted in Ussing chambers was decreased by ∼30% (P < 0.03), whereas the forskolin-stimulated component of HCO3 secretion was the same in wild-type and Slc26a6−/− mice. We conclude that Slc26a6 mediates oxalate-stimulated NaCl absorption, contributes to apical membrane Cl/base exchange in the kidney proximal tubule, and also plays an important role in HCO3 secretion in the duodenum.

  • chloride absorption
  • bicarbonate secretion
  • proximal tubule
  • duodenum
  • apical anion exchange

slc26a6 (pat1, cfex) is a member of a large, conserved family of anion exchangers (SLC26) that encompasses at least 10 distinct genes (8, 13, 15, 16, 19, 21, 22, 25, 31, 33, 35, 36, 40, 42). All except SLC26A5 (prestin) function as anion exchangers with versatility with respect to transported anions (8, 13, 15, 16, 19, 21, 22, 25, 31, 33, 35, 36, 40, 42). Slc26a6 was cloned on the basis of homology to downregulated in adenoma (dra; Slc26a3) and pendrin (Slc26a4) (19, 21). In humans, SLC26A6 maps to chromosome 3 and encodes a 738-amino acid protein (21). Immunohistochemical studies in the human pancreas have localized SLC26A6 to the apical membranes of the duct cells (21). In addition to the pancreas, SLC26A6 is expressed on the apical membrane of the kidney proximal tubule (19) and the villi of the duodenum (40). Functional studies in in vitro expression systems have demonstrated that SLC26A6 can mediate multiple anion exchange modes, including Cl/HCO3, Cl/oxalate, Cl/OH, and Cl/formate exchanges (18, 19, 21, 40, 41). Similar anion exchange activities were previously described in apical membranes of the kidney proximal tubule and the small intestine (3, 4, 28). On the basis of the immunolocalization of SLC26A6 in the kidney and its ability to function in multiple Cl/anion exchange modes, it was postulated that SLC26A6 is a major contributor to NaCl absorption in the proximal tubule (18, 19, 28). In the duodenum, the principal form of apical Cl/base exchange activity is Cl/HCO3 exchange, which is responsible for HCO3 secretion in exchange for Cl absorption (1, 9, 17, 30).

To study the role of SLC26A6 in kidney and intestinal physiology, targeted gene disruption was used to prepare mice lacking Slc26a6. The Slc26a6-null mice appear normal, with normal growth, blood pressure, and serum electrolyte profile. Studies in microperfused kidney proximal tubule demonstrate that Slc26a6-null mice have major defects in apical Cl/base exchange and oxalate stimulated NaCl absorption. In the duodenum, Slc26a6-null mice display significant reduction in HCO3 secretion, a pathway essential to protection against acid injury.

EXPERIMENTAL PROCEDURES

Construction of the targeting vector and generation of Slc26a6−/− mice.

To generate the targeting construct, Slc26a6 genomic clones were isolated from a strain of the 129/SvJ mouse phage library and partially characterized by restriction endonuclease mapping, DNA sequencing, and polymerase chain reaction (PCR) analysis. A phosphoglycerate kinase (PGK) Neo cassette was used as the vector. The targeting vector was constructed by site-specific mutagenesis. The short arm is 1.2 kb long, starting 21 bp downstream of exon 3 to the end of the genomic clone. The long arm is 8.7 kb long, from 5 bp upstream of the start codon (ATG) to upstream of the Nsi genomic fragment. By using this strategy, exons 1 and 2 and part of exon 3 were replaced by the Neo gene cassette. Ten micrograms of targeting vector were linearized by NotI and then transfected by electroporation of 129/SvEv iTL1 embryonic stem cells. After selection in G418, surviving colonies were expanded and PCR analysis was performed to identify clones that had undergone homologous recombination. PCR was performed using primer pairs that were designed to identify positive clones. One primer is located outside the short arm, with a sequence of 5′-TAATGGAAGAGGGTGAACCATCTG-3′. The other primer is located in the 5′-promoter region of the Neo gene cassette and has a sequence of 5′-TGCGAGGCCAGAGGCCACTTGTGTAGC-3′. The PCR fragment in colonies expressing the transgene is expected to be 1.4 kb long. Figure 1A is a schematic of the recombinant alleles. More than 400 surviving colonies were screened, and 6 colonies (154, 233, 236, 323, 326, and 381) showed homologous recombination as determined by positive PCR analysis (Fig. 1B). The results were verified by performing Southern blot analysis and PCR fragment sequencing.

Fig. 1.

A: schematic of the recombinant alleles. Immunofluorescent labeling or immunoblotting of Slc26a6 (PAT1, CFEX) in the duodena or the kidneys of wild-type and Slc26a6-null mice is shown. To generate the targeting construct, Slc26a6 genomic clones were isolated from a strain of 129/SvJ mouse phage library and partially characterized by restriction endonuclease mapping, DNA sequencing, and polymerase chain reaction (PCR) analysis. A PGK Neo cassette was used as the vector. The targeting vector was constructed by performing site-specific mutagenesis. The short arm is 1.2 kb long, starting at 21 bp downstream from exon 3 to the end of the genomic clone. The long arm is 8.7 kb long, extending from 5 bp upstream of the start codon (ATG) to upstream of the Nsi genomic fragment. In this strategy, exons 1 and 2 and part of exon 3 were replaced by the Neo gene cassette. Boxes represent exons, and restriction enzyme sites are as indicated. B: PCR analysis of DNA isolated from surviving embryonic stem cells that were electroporated with the targeting construct. At least 6 of 300 colonies showed homologous recombination as determined using positive PCR analysis. Southern blotting confirmed the results. The identity of the 1.4-kb fragment was verified using sequencing analysis. Arrows indicate the expected PCR fragments in transgene-expressing embryonic stem cells. Clones 154, 233, 236, and 381 expressed the recombinant allele. C: mRNA expression of Slc26a6 in the duodena and kidneys of wild-type (+/+), heterozygote (+/−), and Slc26a6-null mice (−/−). Top: short exposure; middle: long exposure; bottom: 28S rRNA. D: immunofluorescent labeling in the duodenum (left and middle) and immunoblotting in the kidney (right) of Slc26a6 in wild-type and Slc26a6-null mice. No labeling was detected in either the duodenum or the kidneys of Slc26a6-null mice. Left: immunofluorescence labeling of Slc26a6 in the duodenum in wild-type (left) and Slc26a6-null (middle) mice; Right: immunoblotting of Slc26a6 in the kidney in Slc26a6-null (left) and wild-type (right) mice.

The embryonic stem cells from one of the positive colonies (clone 154) were microinjected into C57BL/6J blastocysts and implanted into pseudopregnant female mice. The chimeric mice were generated, and, after cross breeding them with wild-type C57BL mice, heterozygous animals were bred.

Animals.

Animals were euthanized with the use of anesthetics (pentobarbital sodium) according to The University of Cincinnati and Yale University institutional guidelines and approved protocols.

RNA isolation and Northern blot hybridization.

Total cellular RNA was extracted from various mouse tissues, including the duodenum and the kidney, according to established methods; quantitated spectrophotometrically; and stored at −80°C. Total RNA samples (30 μg/lane) were fractionated on a 1.2% agarose-formaldehyde gel, transferred to MagnaNT nylon membranes (Micron Separations, Westboro, MA), cross linked using UV light, and baked. Hybridization was performed according to established protocols (11). The membranes were washed, blotted dry, and exposed to a PhosphorImager screen (Molecular Dynamics, Sunnyvale, CA). A 32P-labeled full-length Slc26a6 cDNA fragment was used for Northern blot hybridization.

Immunofluorescent labeling of Slc26a6 in mouse kidney and duodenum.

Slc26a6+/+ and Slc26a6−/− mice were euthanized with a pentobarbital sodium overdose and perfused through the left ventricle with 0.9% saline, followed by cold 4% paraformaldehyde in 0.1 M NaPO4 buffer (pH 7.4). Duodena were removed, cut into tissue blocks, and fixed in formaldehyde solution overnight at 4°C. The tissues were frozen on dry ice, and 6-μm sections were cut with a cryostat and stored at −80°C until used. Single immunofluorescent labeling was performed as described using Slc26a6-specific antibodies (29) and either Alexa Fluor 488 (green) or Alexa Fluor 568 (red) goat anti-rabbit secondary antibodies (29).

Renal hemodynamic measurements.

Baseline renal function was determined in mice maintained on normal chow. Mice were anesthetized with ketamine (50 μg/g body wt) and inactin (100 μg/g body wt) and surgically instrumented for renal and blood pressure measurements with catheters placed in the left femoral artery and vein and in the bladder as described previously (23, 24). Immediately after surgery, a 3 μl/g body wt bolus of 1% FITC-inulin and 3% p-aminohippuric acid (PAH) sodium salt in isotonic saline was administered. This was followed by a maintenance infusion of the same solution at 0.15 μl/min/g body wt. After a 30-min equilibration period, baseline renal function was determined using two 30-min urine samples collected through a catheter in the bladder. At the midpoint of each baseline collection, an arterial blood sample (60 μl) was obtained for determination of plasma FITC-inulin and PAH concentrations, and donor blood was administered to replace the lost volume after each sample was obtained. At the end of the second baseline collection, another blood sample was acquired and plasma electrolyte levels were measured using a pH/blood gas analyzer (Bayer, Medfield, MA). Urinary Na+ and K+ concentrations were determined using a Corning 480 Flame Photometer (Bayer). Glomerular filtration rate (GFR) was calculated from inulin clearance, and effective renal plasma flow (ERPF) was calculated from PAH clearance. Blood pressure was monitored throughout using a fixed-dome transducer and a PowerLab data acquisition system (ADInstruments, Colorado Springs, CO).

In situ microperfusion in kidney proximal tubule.

In situ microperfusion of Slc26a6+/+ and Slc26a6−/− kidney proximal tubules were performed according to established protocols and as described previously (39). Age-matched Slc26a6+/+ and Slc26a6−/− mice weighing 28.1 ± 1.14 g (+/+) and 26.55 ± 1.71 g (−/−), were anesthetized with an intraperitoneal injection of Inactin (100 mg/kg). After surgical preparation, 0.9% saline was infused into the left jugular vein at a rate of 0.15 ml/h. In brief, proximal convoluted tubules were perfused at a rate of 15 nl/min with a solution containing (in mM) 140 NaCl, 5.0 NaHCO3, 4.0 KCl, 2.0 CaCl, 1.0 MgSO4, dibasic 1.0 NaPO4, and 1.0 monobasic NaPO4, pH 6.7. To measure volume absorption, 20 μCi/ml low-Na+ [3H]methoxy-inulin was added to the perfusion solution. One collection was made in each perfused tubule, and two to four collections were taken in the experimental kidney of each animal. The perfused segments were marked with Sudan Black heavy mineral oil, and their lengths were determined after dissection of silicone rubber casts. Calculation of the rate of net fluid absorption (Jv) was based on changes in the concentrations of [3H]inulin, and the rates are expressed per millimeter of tubule length. Data are presented as means ± SE. In Table 1, experimental groups were compared with a control group in either the wild-type or null mice using Dunnett's test. All six groups were compared using ANOVA (see Fig. 3). Differences were considered significant at P < 0.05.

View this table:
Table 1.

Blood pressure, GFR, and serum and urinary electrolytes in wild-type and Slc26a6-null mice

Proximal tubule isolation, in vitro microperfusion, and apical Cl/HCO3 exchanger activity measurement.

Isolation and microperfusion of proximal tubules in Slc26a6+/+ and Slc26a6−/− mice performed according to established protocols and as described previously (2, 6, 10, 28). Intracellular pH (pHi) was measured using 2′,7′-bis(3-carboxypropyl)-5(6)-carboxyfluorescein acetoxymethyl ester (BCPCF-AM) as described previously (2, 6, 10, 28). Excitation wavelengths were recorded at 488 and 440 nm, and emission was measured at 520 nm. Digitized images were analyzed using Attograph software (Atto Bioscience, Rockville, MD). Intracellular calibration was performed at the end of each experiment by using a high-K+-nigericin method (10, 28). The apical Cl/HCO3 exchanger was assayed as the rate of pHi acidification (dpHi/dt) upon switching the luminal perfusate from a Cl-free to Cl-containing solution (28). DIDS (300 μM) was present in bath and ethylisopropyl amiloride (300 μM) was present in the perfusate to block the basolateral Na+-HCO3 cotransporter and apical Na+/H+ exchanger isoform 3 (NHE3), respectively. One tubule per animal was used.

Ussing chamber experiments in duodenum.

To examine HCO3 secretion in the duodenum, mice were anesthetized with 100% CO2 and killed by cervical dislocation. The proximal duodenum was removed, opened along the mesenteric border, stripped of external serosal and muscle layers in Ringer solution with indomethacin (4°C), and mounted in Ussing chambers (0.65 cm2 window area). Experiments were performed under continuous short-circuited conditions. The mucosal solution contained (in mM) 140 Na+, 5.4 K+, 1.2 Ca2+, 1.2 Mg2+, 120 Cl, 25 gluconate, and 10 mannitol. The serosal solution contained (in mM) 140 Na+, 5.4 K+, 1.2 Ca2+, 1.2 Mg2+, 120 Cl, 25 HCO3, 2.4 HPO42−, 2.4 H2PO4, 10 glucose, and 0.001 indomethacin. The osmolality of both solutions was ∼310 mosmol/kg. HCO3 secretion was determined under the automatic control of a pH stat system (PHM290, pH-Stat Controller; Radiometer, Copenhagen, Denmark) (14). After a 30-min measurement of basal parameters, 10 μM forskolin were added to the serosal side of tissue in Ussing chambers for 40 min to measure the changes in duodenal HCO3 secretion and short-circuit current (Isc; reported as μeq/cm2/h). Transepithelial Isc was measured via an automatic voltage clamp (EVC-4000 voltage-current clamp; World Precision Instruments, Berlin, Germany) and calomel electrodes connected to the chamber baths with 4% agar-3 M KCl bridges.

36Cl transport measurement.

The 30-s uptake of 36Cl by luminal membrane vesicle suspensions from duodenum, prepared according to established methods (9), was assayed at room temperature in triplicate using a rapid filtration technique. The reaction was stopped using ice-cold medium. The radioactivity in each filter was assayed using scintillation spectroscopy. Vesicles and all experimental media were continuously gassed with 100% N2 or 5% CO2-95% N2. The uptake of 36Cl was measured under three different conditions: no pH gradient [pHi/extracellular pH (pHo) 7.5/7.5 without CO2/HCO3], outward pH gradient (pHi/pHo 7.5/6.0 without CO2/HCO3), and outward pH and HCO3 gradient (pHi/pHo 7.5/6.0 with CO2/HCO3). The HCO3 concentrations were 25 mM at pH 7.5 and 0 mM at pH 6.0.

Materials.

[32P]dCTP and 36Cl were purchased from New England Nuclear (Boston, MA). Nitrocellulose filters and other chemicals were purchased from Sigma Chemical (St. Louis, MO). The RadPrime DNA labeling kit was purchased from GIBCO-BRL. BCECF was obtained from Molecular Probes (Eugene, OR). The mMESSAGE mMACHINE kit was purchased from Ambion (Austin, TX).

Statistical analyses.

Values are expressed as means ± SE. Statistical analysis was conducted using Student's t-test or ANOVA. P < 0.05 was considered statistically significant.

RESULTS

Germline transmission of the disrupted Slc26a6 gene was accomplished by crossing heterozygous mice. In Fig. 1C (left and right), mRNA expression of Slc26a6 in the duodenum and the kidney of wild-type, heterozygous, and knockout (KO) Slc26a6 mice is compared. Figure 1D (top left and right) depicts immunocytochemical labeling of Slc26a6 in the duodena of wild-type and KO mice and Fig. 1D (bottom left and right) shows Slc26a6 immunoblotting in microsomal membranes isolated from kidneys of wild-type and Slc26a6-null mice. As indicated, there is complete absence of Slc26a6 mRNA and protein in the duodenum and kidneys of KO animals.

Table 2 summarizes the blood pressure, serum electrolyte profile, and parameters of kidney function, such as GFR, urine osmolality, urine volume, and urinary Na+, K+, and Cl in five wild-type and five Slc26a6-null mice. As indicated in the tabular data, Slc26a6-null animals have normal blood pressure and kidney function as measured using GFR. Urinary Na+ was similar, while urinary Cl showed a trend toward increase in Slc26a6-null mice.

View this table:
Table 2.

Effect of formate and oxalate on volume reabsorption in proximal tubules of wild-type and Slc26a6-null mice

To determine the contribution of Slc26a6 to apical Cl/base exchange, isolated proximal tubules (see experimental procedures) were perfused with Cl-free and Na+- and HCO3-containing solutions (140 Na+, 0 Cl, 115 gluconate, and 25 HCO3, pH 7.4) and pHi was measured using ratiometric imaging and the pH-sensitive dye BCPCF (see experimental procedures). Apical Cl/base exchange activity was assayed as the initial rate of pHi acidification in response to the luminal addition of Cl in the presence of CO2/HCO3 using the following solution (in mM): 140 Na+, 115 Cl, and 25 HCO3, pH 7.4. Representative tracings (Fig. 2A) and summarized results (Fig. 2B) are shown. The rate of apical Cl/HCO3 exchanger activity was 0.76 ± 0.05 pH units/min in wild-type animals (n = 21 cells in 6 tubules) and decreased to 0.33 ± 0.04 pH units/min in KO animals (n = 43 cells in 6 tubules), a reduction of 58% (P < 0.001). The magnitude of intracellular acidification caused by apical Cl/HCO3 exchanger was 0.51 ± 0.02 pH units in wild-type animals and decreased to 0.28 ± 0.02 pH units in KO animals (P < 0.001). The pHi returned to baseline upon switching back to the Cl-free perfusate (Fig. 2A). The initial baseline pHi was 7.38 ± 0.02 in Slc26a6-null mice, a value significantly lower than the 7.49 ± 0.01 recorded in wild-type animals (P < 0.001).

Fig. 2.

Apical Cl/HCO3 exchanger activity in isolated microperfused tubules of wild-type and Slc26a6-null mice. A: representative tracings in wild-type (top) and null animals (bottom) are shown. DIDS (300 μM) was present in bath and ethylisopropyl amiloride (300 μM) in the perfusate to block the basolateral Na+-HCO3 cotransporter and apical NHE3, respectively. One tubule per animal was used. B: summary of results. The rate (top) and magnitude (bottom) of acidification caused by apical Cl/HCO3 exchanger activation were decreased significantly in null mice.

To assess the possible role of Slc26a6 in mediating NaCl absorption in the proximal tubule, tubules were microperfused in situ with a late proximal tubule fluid (in mM: 140 Cl and 5.0 HCO3, pH 6.7) and Jv was measured as an index of NaCl absorption. The results are shown in Table 2 and Fig. 3. The baseline rates of Jv measured in the absence of added formate or oxalate were essentially identical in wild-type and null mice (1.51 ± 0.075 and 1.45 ± 0.22 nl·min−1·mm-1, respectively). These results are consistent with previous findings that under the above experimental conditions, DIDS-sensitive Cl/base exchange does not contribute to NaCl absorption in surface proximal tubules in the absence of added formate or oxalate (38).

Fig. 3.

Oxalate and formate stimulated volume reabsorption (Jv) in proximal tubules of wild-type and Slc26a6-null mice. Effects of oxalate and formate were examined by microperfusing the proximal convoluted tubule in vivo in the absence and presence of 1 μM oxalate or 50 μM formate in luminal perfusates. ANOVA indicated that the values for Jv in the presence of oxalate in wild-type and null mice were different (P < 0.05).

Table 2 and Fig. 3 also show that addition of 1 μM oxalate or 50 μM formate to the luminal perfusion solution markedly and significantly stimulated Jv (from 1.51 ± 0.075 to 2.15 ± 0.18 and 2.19 ± 0.17 nl·min−1·mm−1, respectively) in wild-type mice, consistent with previous results in rat and mouse proximal tubules (37, 39). In contrast, stimulation of Jv by oxalate was completely abolished in Slc26a6-null mice (1.42 ± 0.13 nl·min−1·mm−1). This finding demonstrates that the presence of functional Slc26a6 is essential for oxalate stimulation of Jv, consistent with the proposed role of apical membrane Cl/oxalate exchange in this process (3, 4).

Unfortunately, the findings concerning the role of Slc26a6 in mediating formate stimulation of Jv were ambiguous. The value for Jv measured in the presence of formate in null mice (1.70 ± 0.18 nl·min−1·mm−1) was intermediate between the baseline rate and the stimulated rate observed in wild-type mice but was not significantly different from either value. These findings suggest a partial defect in formate-stimulated Jv, but such a defect has not been proved conclusively.

To ascertain the contribution of Slc26a6 in HCO3 secretion in the duodenum, mucosal duodenal tissue was mounted in Ussing chambers and HCO3 secretion was assayed using the pH-stat technique (see experimental procedures). The results (Fig. 4A) demonstrate that at the basal state, HCO3 secretion was decreased by 30.5% (P < 0.03) but that forskolin-stimulated HCO3 secretion remained unchanged in Slc26a6−/− mice. Interestingly, the duodenal Isc, measured in Ussing chambers, was the same in both wild-type and KO animals at basal and stimulated states (Fig. 4B). Forskolin-induced Isc increase in the murine duodenum strongly correlates with CFTR activation (12, 32). Taken together, these studies demonstrate significant reduction in basal HCO3 secretion in the duodena of KO animals, which clearly supports the notion that Slc26a6 is a major apical Cl/HCO3 exchanger in the duodenum. To verify the contribution of Slc26a6 to apical Cl/HCO3 exchanger activity, the 30-s influx of 36Cl into luminal membrane vesicles isolated from the duodenum was assayed in the presence or absence of an outward pH and HCO3 gradient (pHi/pHo 7.5/7.5 vs. 7.5/6.0 ± CO2/HCO3; see experimental procedures). As demonstrated in Fig. 4C, in the presence of an outward pH and HCO3 gradient, the influx of radiolabeled Cl decreased by ∼39% in the apical membrane vesicles from duodenum of Slc26a6-null mice (P < 0.05, n = 3 for each group). The mRNA expression of Slc26a3 (dra) in the small intestine remained unchanged in Slc26a6-null mice (Fig. 4D).

Fig. 4.

HCO3 secretion and short-circuit current (Isc) measurement in the duodena of wild-type and Slc26a6-null mice. A: HCO3 secretion with and without forskolin. HCO3 secretion at the basal state decreased significantly in Slc26a6-null mice, but the magnitude of HCO3 stimulation by cAMP remained comparable in both groups. B: Isc measurement with and without forskolin in the duodena of Slc26a6-null mice. Isc remained comparable wild-type and Slc26a6-null mice at the basal state and in response to cAMP stimulation. C: 36Cl influx in brush-border membrane vesicles from the duodenum. The 30-s uptake of tracer 36Cl was measured as described in experimental procedures according to established techniques (9). The results were normalized per mg of membrane protein and are expressed as %control (no pH gradient in wild-type animals). As indicated, outward pH and HCO3-dependent 36Cl influx (mediated by Cl/HCO3 exchange) was decreased significantly (n = 3 for each group; P < 0.05) in luminal membrane vesicles from the duodena of Slc26a6-null mice. D: mRNA expression of Slc26a3 (dra) in the small intestine (left) and the kidneys (right) of wild-type (+/+), heterozygote (+/−), and Slc26a6-null mice (−/−). As indicated, the expression of dra remained unchanged in the small intestine (left), indicating the lack of compensatory upregulation of Slc26a3 in response to Slc26a6 deletion. The expression of dra remained undetectable in the kidney (right), which is consistent with published reports indicating the absence of this exchanger in kidney nephrons.

DISCUSSION

The majority of filtered Na+, Cl, HCO3, and water are reabsorbed in the proximal tubule (35, 7, 26). Studies performed using membrane vesicles and perfused tubules have led to the concept that a major fraction of the filtered Cl is reabsorbed via different mechanisms of Cl/base exchange. These mechanisms include Cl/formate exchange operating in parallel with Na+/H+ exchange and H+/formate cotransport, and Cl/oxalate exchange operating in parallel with SO42−/oxalate exchange and Na+-SO42− cotransport (3, 4). Formate stimulation of apical Na+/H+ exchanger NHE3 is an additional and/or alternative mechanism to explain formate stimulation of NaCl absorption in the proximal tubule (27). In addition to Cl/formate exchange and Cl/oxalate exchange, Cl/ HCO3 exchange also has been described as a mode of apical membrane Cl/base exchange in the proximal tubule (12, 28). Functional expression studies have indicated that Slc26a6 has the ability to operate in all of these exchange modes (18, 19, 21, 40, 41).

The generation of Slc26a6-null mice has allowed the determination of the exchange modes that are actually mediated by Slc26a6 in the proximal tubule in vivo and the extent to which these exchange modes contribute to transtubular NaCl transport. Our present study (see Fig. 2) has demonstrated that Slc26a6 is the major apical Cl/HCO3 exchanger in the proximal tubule as shown by ∼60% reduction in Cl-dependent HCO3 transport detected using pHi measurements in microperfused tubules of Slc26a6-null animals. The residual Cl-dependent HCO3 transport in Slc26a6-null mice indicates the presence of additional unidentified anion exchangers. Interestingly, the baseline rate of Jv measured in the presence of a high-Cl/low-HCO3 perfusate was essentially identical in wild-type and null mice, suggesting that the apical membrane Cl/HCO3 exchange mediated by Slc26a6 does not contribute to transtubular NaCl absorption. This conclusion is consistent with the previous observation that the baseline rate of Jv in the proximal tubule is not sensitive to concentrations of DIDS that abolish the increments in Jv induced by formate and oxalate (37).

In striking contrast to the normal baseline rate of Jv in Slc26a6-null mice, the increment in Jv induced by oxalate was completely abolished. This finding is consistent with the concept that SLC26A6 represents the Cl/oxalate exchanger proposed to mediate NaCl absorption by operating in parallel with SO42−/oxalate exchange and Na+/SO42− cotransport (3, 4). The increment in Jv induced by formate was blunted in Slc26a6 mice, suggesting a role for Slc26a6 either in mediating Cl/formate exchange coupled to NHE3 and/or in mediating formate entry into the cell to stimulate NHE3. Because the intermediate value for formate-stimulated Jv in null mice was not statistically different from either the baseline or the fully stimulated values, however, no firm conclusion can be drawn concerning this possible role of Slc26a6.

Slc26a6 shows abundant expression in the apical membrane of villi of the duodenum, where it is presumed to mediate apical Cl/HCO3 exchange (40), the main mechanism for basal HCO3 secretion in the small intestine (12, 17, 30, 32). In addition to HCO3 secretion, the apical Cl/HCO3 exchanger functions in parallel with NHE3 and, as a result, is essential for the electroneutral absorption of Na+ and Cl (9, 17, 30). Slc26a6 expression is very low in the colon but high in the small intestine (40), a pattern opposite that of DRA (Slc26a3), which is expressed predominantly in the colon and moderately in the small intestine (25). Our experiments demonstrate a significant reduction in basal HCO3 secretion in the duodenum in Slc26a6-null mice. There was, however, a significant component of basal HCO3 secretion that remained intact in the duodenum of Slc26a6-null mice, suggesting the contribution of other apical anion exchangers, such as Slc26a3 (DRA), to this process. The Cl/HCO3 exchanger was examined more directly in luminal membrane vesicles isolated from duodenum using the 36Cl influx method (Fig. 4C). The results demonstrated significant reduction in Cl/HCO3 exchanger activity, confirming the HCO3 secretion defect that is observed in the duodenum of Slc26a6-null mice (Fig. 4, A and B). The mRNA expression of DRA (Slc26a3) in the small intestine did not change and remained undetectable in the kidney (Fig. 4D), suggesting the absence of compensatory upregulation of Slc26a3 in response to Slc26a6 gene deletion in the duodenum and kidney.

In addition to the Cl/HCO3 exchanger, several studies indicate the presence of a HCO3 conductive pathway that mediates HCO3 secretion into the duodenum (12, 32). Whether the HCO3 conductive pathway is the same as CFTR or is a distinct anion channel that is regulated by CFTR remains to be resolved. The HCO3 conductive pathway is stimulated by cAMP generation as demonstrated by Isc activation and is not affected by luminal Cl removal (34). Our studies demonstrated that forskolin-stimulated HCO3 secretion in the duodenum was comparable in wild-type and Slc26a6-null mice (Fig. 4A). A recent study demonstrated that Slc26a6 stimulation markedly activated CFTR by increasing its overall open probability when both were coexpressed in cultured cells (20). Our current results demonstrate that forskolin-stimulated Isc activation remained the same in wild-type and Slc26a6-null mice, strongly indicating that Slc26a6 does not play an important role in forskolin-activated CFTR activation in the murine duodenum. Taken together, our results demonstrate that Slc26a6 plays an important role in basal HCO3 secretion but does not contribute to forskolin-stimulated HCO3 secretion. Morphometric analysis of kidney structure did not reveal significant abnormalities in epithelial cells of kidney and duodenum in Slc26a6-null animals.

The effects of Slc26a6 deletion on intestine and tubular function are not profound at steady state, with blood pressure, plasma electrolytes, and kidney function remaining comparable to that of wild-type animals. It is worth mentioning that several apical Cl absorbing transporters are present downstream from Slc26a6 in the kidney and intestine (with the furosemide-sensitive Na+-K+-2Cl cotransporter in the thick ascending limb, the thiazide-sensitive Na+-Cl cotransporter in the distal convoluted tubule, pendrin in the cortical collecting duct, and Slc26a3 in the large intestine). It is plausible that one or more of these transporters might show compensatory upregulation in response to Slc26a6 gene deletion, thus blunting the impact of its deficiency on NaCl wasting. Furthermore, it is possible that a deficiency of Slc26a6 may give rise to a more significant phenotype in pathophysiological states such as volume depletion or Na+ or Cl depletion in which proximal tubule NaCl reabsorption plays a more essential role than under baseline conditions.

In conclusion, our studies of Slc26a6-null mice demonstrate that Slc26a6 is a major contributor to apical membrane Cl/base exchange and oxalate-stimulated NaCl absorption in the kidney proximal tubule and also plays an important role in HCO3 secretion in the duodenum.

GRANTS

These studies were supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-62829 (to M. Soleimani), DK-33793 and DK-17433 (to P. S. Aronson), and DK-62289 (to T. Wang), DK-57552 (to J. N. Lorenz), Deutsche Forschungsgemeinschaft Grants Se 460/13-1/2 and Se 460/9-4/5 (to U. Seidler), a National Kidney Foundation grant (to S. Petrovic), and grants from the Department of Veterans Affairs (Merit Review Award) and the Cystic Fibrosis Foundation (to M. Soleimani).

Acknowledgments

The contribution of Dr. Elizabeth Mann is greatly appreciated. We acknowledge the technical assistance of Terry Fettig.

Footnotes

  • * Z. Wang and T. Wang contributed equally to these studies.

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract