点击显示 收起
【摘要】 The outer medullary collecting duct (OMCD) plays an important role in bicarbonate reabsorption and acid-base regulation. An apical V-type H + -ATPase and a basolateral exchanger, located in intercalated cells of OMCD, mediate the bicarbonate reabsorption. Here we report the identification of a new basolateral exchanger in OMCD intercalated cells in rat kidney. Northern hybridizations demonstrated the predominant expression of this transporter, also known as SLC26A7, in the outer medulla, with lower expression levels in the inner medulla. SLC26A7 was recognized as a 90-kDa band in the outer medulla by immunoblot analysis and was localized on the basolateral membrane of a subset of OMCD cells by immunocytochemical staining. No labeling was detected in the cortex. Double-immunofluorescence labeling with the aquaporin-2 and SLC26A7 antibodies or anion exchanger-1 and SLC26A7 antibodies identified the SLC26A7-expressing cells as -intercalated cells. Functional studies in oocytes demonstrated that increasing the osmolality of the media (to simulate the physiological milieu in the medulla) increased the exchanger activity mediated via SLC26A7 by about threefold ( P < 0.02 vs. normal condition). We propose that SLC26A7 is a basolateral exchanger in intercalated cells of the OMCD and may play an important role in bicarbonate reabsorption in medullary collecting duct.
【关键词】 anion exchanger basolateral membrane aquaporin
THE COLLECTING DUCT (CD) is composed of three anatomically distinct segments, cortical, outer medullary, and inner medullary collecting duct (CCD, OMCD, and IMCD, respectively), and plays an essential role in acid-base homeostasis. The CCD secretes both acid and bicarbonate by - and -cells, respectively, whereas the OMCD and IMCD are predominantly involved with acid secretion and bicarbonate reabsorption. The intercalated cells in the OMCD are exclusively of the acid-secreting -cell type. Active proton secretion across the apical membrane of the CD is largely mediated via H + -ATPase (and to some extent, H + -K + -ATPase) ( 25, 29 ). The exit across the basolateral membrane of intercalated cells in the CCD and OMCD is mediated via exchanger. The OMCD has the highest rate of H + secretion, and therefore absorption, of the CD segments ( 25 ). As a result, the OMCD has been regarded as a major regulator of acid-base homeostasis.
The apical H + -ATPase in intercalated cells of the OMCD belongs to the family of V-type ATPase and is inhibited by bafilomycin ( 2, 5, 7, 8, 25, 26 ). The apical H + -K + -ATPase in the OMCD is likely a variant of gastric H-K-ATPase ( 29 ). This conclusion is based on the studies that demonstrated the inhibition of net bicarbonate reabsorption by Schering 28080 in OMCD ( 4, 29, 38 ). The exchanger anion exchanger-1 (AE1), a variant of red cell band 3, has been localized on the basolateral membrane of intercalated cells in the OMCD ( 18, 25, 32 - 35 ). However, the possibility that other basolateral exchanger(s) may contribute to net bicarbonate reabsorption in this nephron segment exists.
Recent molecular studies identified a large, highly conserved family of membrane proteins (designated as SLC26A) many of which have been shown to transport anions. Three closely related members of this family are downregulated in adenoma (DRA or SLC26A3), Pendrin (PDS or SLC26A4), and PAT1 (CFEX or SLC26A6) ( 11, 12, 15, 16, 28 ). All three transporters mediate exchange ( 13, 19, 27, 30, 37, 39 ). DRA is expressed on the apical membranes of colonocytes, whereas PAT1 or CFEX is expressed on the apical membranes of kidney proximal tubule and duodenum ( 12, 15, 19, 37 ). Pendrin mRNA expression is detected in proximal tubule and CCD ( 30, 36 ). However, immunocytochemical studies localize pendrin only to the apical membrane of a subpopulation of CCD cells distinct from -intercalated cells, which are thought to be -intercalated cells and non- -, non- -intercalated cells ( 14, 22, 23, 30, 36 ).
A recently cloned member of SLC26A family is SLC26A7, which is shown to be expressed in kidney and testis ( 17 ). Very recent studies from our laboratory demonstrated that in addition to the kidney, SLC26A7 is abundantly expressed in the stomach and mediates exchange ( 20 ). SLC26A7 expression in the stomach is limited to the basolateral membrane of gastric parietal cells ( 20 ). In the current studies, we investigated the renal distribution and regulation of SLC26A7, as little is known about this transporter in the kidney. Our results indicate that SLC26A7 expression in the kidney is predominantly limited to the outer medulla with lower levels in the inner medulla. SLC26A7 is expressed on the basolateral membranes of -intercalated cells in the OMCD, is upregulated by hypertonicity, and colocalizes with AE1. The significance of the results will be discussed.
MATERIALS AND METHODS
Animals. Female Sprague-Dawley rats, weighing 100-150 g, were used for these studies. Animals were allowed free access to water and food. The use of anesthetics (pentobarbital sodium) and the method of euthanasia (pentobarbital sodium overdose) were according to the institutional guidelines and approved protocols.
RT-PCR of SLC26A7 in the kidney. A mouse EST (GenBank accession number BB666404 ), which matched the human SLC26A7 sequence (GenBank accession number AF331521 ), was identified. The following oligonucleotide primers 5'-CTC ACC ACC GAA CCT ATT AC (sense) and 5'-AAC TCG GAT AAG CCC AAC AC (antisense) were designed based on the EST cDNA sequence and used for RT-PCR on RNA isolated from rat stomach and kidney. A 550-bp PCR fragment was purified and sequenced, which corresponded to the human nucleotides 8 to 550. The purified fragment was used as probe for Northern hybridization.
RNA isolation and Northern blot hybridization. Total cellular RNA was extracted from various rat kidney zones (cortex, outer medulla, and inner medulla) according to the established methods ( 9 ), quantitated spectrophotometrically, and stored at -80°C. Total RNA samples (30 µg/lane) were fractionated on a 1.2% agarose-formaldehyde gel, transferred to Magna NT nylon membranes, cross-linked by UV light, and baked. Hybridization was performed according to Church and Gilbert ( 10 ). The membranes were washed, blotted dry, and exposed to a PhosphorImager screen (Molecular Dynamics, Sunnyvale, CA). A 32 P-labeled cDNA fragment corresponding to nucleotides 8 to 550 of SLC26A7 cDNA (see above) was used as a probe for Northern hybridizations.
Antibodies. For SLC26A7, antibodies against human or mouse sequence were used. For mice, a synthetic peptide corresponding to the amino acid residues CGAKRKKRSVLWGKMHTP of mouse SLC26A7 (using the mouse EST with GenBank accession number BB666404 ) was used for polyclonal antibody generation in two rabbits. For humans, SLC26A7-specific antibodies were raised against a synthetic peptide based on human sequence ( 17, 20 ). AE1 antibodies were raised against the NH 2 -terminal end of AE1 and were purchased from Chemicon.
For aquaporin-2 (AQP2), a polyclonal antibody specific to AQP2 water channel was raised against the rat AQP2 peptide CEVRRRQSVELHSPQSLPRGSKA, which corresponds to amino acid residues 250-271 of the COOH-terminal tail of the vasopressin-regulated AQP2 water channel. A cysteine residue was added at the NH 2 -terminal end of the peptide to facilitate its conjugation with a carrier protein for the purpose of purification. Two rabbits were immunized with the conjugate complex. Both rabbits developed ELISA titers greater than 1:100,000. The antiserum was affinity-purified by covalently immobilizing the immunizing peptide on commercially available columns (Sulfo-Link Immobilization kit 2, Pierce, Rockford, IL). This antibody is highly specific and labels the apical membrane of principal cells in the CD ( 3, 22 ).
Immunoblot analysis. These experiments were carried out as previously described ( 30, 37, 40 ). Briefly, the solubilized membrane proteins were size-fractionated on 12% polyacrylamide minigels (Novex, San Diego, CA) and were electrophoretically transferred to nitrocellulose membranes using a Bio-Rad transfer apparatus (Bio-Rad Laboratories, Hercules, CA). The membranes were blocked with 5% milk proteins and then probed with affinity-purified SLC26A7 primary antibodies. The secondary antibody was donkey anti-rabbit IgG conjugated to horseradish peroxidase. The sites of antigen-antibody complexation on the nitrocellulose membranes were visualized using chemiluminescence method (SuperSignal Substrate, Pierce) and captured on light-sensitive imaging film (Kodak). The equity in protein loading in all blots was verified by gel staining using the Coomassie brilliant blue R-250 (Bio-Rad).
Immunofluorescence labeling studies. Mice were euthanized with an overdose of pentobarbital sodium and perfused through the left ventricle with 0.9% saline followed by cold 4% paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.4). Kidneys were removed, cut in tissue blocks, and fixed in formaldehyde solution overnight at 4°C. The tissue was frozen on dry ice, and 6-µm sections were cut with a cryostat and stored at -80°C until used. Single-immunofluorescence labeling was performed as described ( 20 - 22 ) using either Alexa Fluor 488 (green) or Alexa Fluors 568 (red) goat anti-rabbit antibody as secondary antibodies.
For double-immunofluorescence labeling, both AQP2 and SLC26A7 antibodies (or AE1 and SLC26A7) were used at the 1:40 dilution. SLC26A7 antibody was directly labeled by using Zenon Alexa Fluor 568 Rabbit IgG labeling kit and AQP-2 was labeled by Zenon Alexa Fluor 488 Rabbit IgG labeling kit (Molecular Probes, Eugene, OR) according to the manufacturer's instructions. Briefly, SLC26A7 or AQP2 antibody was incubated with a fluorophore-labeled, Fc-specific anti-rabbit IgG Fab fragment. Fab fragment/primary antibody molar ratio was 6:1. The Fab fragment binds to the Fc portion of the rabbit primary antibody, rendering fluorophore-labeled primary antibody ready to be applied to the kidney sections.
Frozen kidney sections were allowed to thaw at room temperature and were subsequently rehydrated in PBS for 15 min and permeabilized in PBS containing 0.3% Triton X-100 (PBT) for 20 min at room temperature. Nonspecific binding was blocked with 1% BSA in PBS for 30 min. Zenon labeling complex, which was freshly prepared as described above, was diluted to a final dilution of 1:40 for both primary antibodies and applied to the sections at room temperature for 2 h in a humidified chamber. Sections were thoroughly washed in PBT for 10 min three times and then in PBS for 5 min two times. Sections were fixed for the second time in 4% formaldehyde in PBS for 15 min at room temperature. The second fixation step prevents dissociation of the Fab fragment from the primary antibody. Sections were then washed and mounted in the antifade mounting medium (ProLong Antifade Kit from Molecular Probes). Sections were examined and images were acquired on a Nikon PCM 2000 laser confocal scanning microscope as 0.5-µm "optical sections" of the labeled cells.
Cloning of mouse SLC26A7. Full-length mouse SLC26A7 cDNA was cloned from mouse stomach by RT-PCR as described ( 20 ). Briefly, the full-length cDNA was amplified from mouse stomach RNA using the following primers: 5'-AGA AGT TGA CTA CTA CAG GAG G(sense) and 5'-AGT TGC CAA GTC ATA TCA TTC (antisense). These primers encode nucleotides 68-2208 of a mouse SLC26A7 cDNA (accession no. BC026928 ). Amplification of the mouse SLC26A7 cDNA by the PCR was performed according to the CLONTECH Advantage 2 PCR kit protocol. After PCR, the product was gel-purified (revealing a single band of 2 kb). Sequence analysis of the PCR product verified the sequence as mouse SLC26A7. The PCR products were ligated into pGEM-T easy vector for expression studies.
Expression of SLC26A7 in Xenopus laevis oocytes. The plasmid containing the full-length mouse cDNA was linearized, and the product was in vitro transcribed to cRNA, as described. X. laevis oocytes were injected with the SLC26A7 cRNA. Briefly, 50 nl cRNA (0.7-1.2 µg/µl) were injected with a Drummond 510 microdispenser via a sterile glass pipette.
Intracellular pH studies. Intracellular pH (pH i ) in oocytes was measured with the pH-sensitive fluorescent probe 2',7'-bis-(3-carboxypropyl)-5-(and-6)-carboxyfluorescein acetoxymethyl ester (BCPCF-AM; Molecular Probes), a close analog of 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein acetoxymethyl ester (BCECF-AM; Molecular Probes), as previously described ( 20, 37, 40 ). Oocytes were loaded with 10 µM BCPCF-AM for 5 min at room temperature, transferred on nylon mesh in a 1-ml perfusion chamber, and perfused at the rate of 3 ml/min with the following Na-free and bicarbonate-free solution (in mM): 63 TMA-Cl, 33 NMDG-gluconate, 2 KCl, 1.8 CaCl 2, 1 MgCl 2, and 5 HEPES at pH 7.5 and gassed with 100% O 2. Fluid was delivered to the chamber by a peristaltic pump via CO 2 -impermeable tubing (Cole Palmer). Ratiometric fluorescence measurements were performed on a Zeiss Axiovert S-100 inverted microscope equipped with Attofluor RatioVision digital imaging system (Attofluor, Rockville, MD). Excitation wavelengths were alternated between 440 and 488 nm, and fluorescence emission intensity was recorded at 520 nm. Data analyses were performed using Attograph and Attoview software packages provided with the imaging system. The ratios were obtained from the submembrane region of the oocytes that were visualized with an achroplan x 40/0.8 water objective with a 3.6-mm working distance. Measured excitation ratios were converted to pH i by using a calibration curve that was constructed with high K + /nigericin method ( 20, 24, 31, 37, 40 ).
The stable baseline pH i was obtained in the above -free solution. To examine the exchanger activity, the perfusate was switched to a Na-free, /CO 2 -containing solution that consisted of (in mM) 63 TMA-Cl, 33 choline-, 2 KCl, 1.8 CaCl 2, 1 MgCl 2, and 5 HEPES, gassed with 95% O 2 -5% CO 2 (pH of 7.5 at room temperature). This maneuver results in cell acidification due to rapid entry of CO 2. The chamber was closed by a lid and constantly superfused with the gas mixture of 5% CO 2 -95% O 2 to prevent CO 2 loss. exchanger was determined as the rate of recovery from acidic pH i due to the exchange of intracellular Cl - with extracellular. Sodium-free solutions were used throughout the experiments to ensure the sodium independence of the observed exchange as well as to keep other possible transport modes (i.e., Na + /H + exchange) inactive. In the experiments in which chloride dependence of the exchanger was examined, experiments were done as described above, except that chloride salts in all the solutions were replaced with equimolar amounts of gluconate. In addition, in the experiments in which the effect of increased osmolarity on the anion exchange was tested, 100 mM mannitol was added to the sodium-free, bicarbonate and chloride-containing solution.
Materials. [ 32 P]dCTP was purchased from New England Nuclear (Boston, MA). Nitrocellulose filters and other chemicals were purchased from Sigma (St. Louis, MO). A RadPrime DNA labeling kit was purchased from GIBCO-BRL. BCECF was from Molecular Probes. A mMESSAGE mMACHINE Kit was purchased from Ambion (Austin, TX). The human multiple tissue blots were purchased from CLONTECH (Palo Alto, CA).
Statistical analyses. Values are expressed as means ± SE. The significance of difference between mean values was examined using Student's t -test or ANOVA. P < 0.05 was considered statistically significant.
RESULTS
SLC26A7 mRNA expression. To examine SLC26A7 mRNA expression levels in kidney, RNA isolated from various rat kidney zones (cortex, outer medulla, inner medulla) was hybridized against an SLC26A7-specific cDNA probe. As indicated, SLC26A7 mRNA was abundantly expressed in the outer medulla, with lower levels in the inner medulla ( Fig. 1 A ). No labeling was detected in the cortex. This pattern of expression is distinct from SLC26A6 (PAT1), which is abundant in cortex but is absent in the outer and inner medulla ( 15, 20a, 37 ). Using specific primers ( MATERIALS AND METHODS ) for RT-PCR procedure on RNA isolated from various kidney zones verified the expression of SLC26A7 in the outer and inner medulla ( Fig. 1 B ). DNA sequencing confirmed the products as rat SLC26A7. Furthermore, comparison of the DNA sequence of the PCR products in kidney and stomach shows perfect identity to one another and to the published mouse sequence.
Fig. 1. A : Northern hybridization of SLC26A7 in rat kidney. Northern hybridization demonstrating SLC26A7 transcript in whole kidney and outer medulla (OM) with lower levels in inner medulla (IM) and no expression in the corterx. SLC26A7 transcript size was 5.2 kb. 28S rRNA levels are shown as constitutive controls. Thirty micrograms of RNA were loaded on each lane. B : expression of SLC26A7 in the kidney by RT-PCR. A representative ethidium bromide staining of agarose gel demonstrates a PCR product of expected size (550 bp) in rat OM and IM.
Immunoblotting of SLC26A7 in rat kidney. We raised antibodies to SLC26A7 and verified their specificity in the stomach ( 17, 20 ). In the stomach, SLC26A7 is detected as a 94-kDa band. The labeling was completely blocked with the preadsorbed immune serum ( 20 ). Similar to the stomach and using the antibody generated against the mouse SLC26A7, a 90-kDa band was detected in the renal outer medulla ( Fig. 2, left ). In addition to the 90-kDa band, a faint band at 80-kDa MW was also detected in the outer medulla. No bands were detected in the cortex. The labeling of the 90- and the faint 80-kDa bands was completely prevented with the preadsorbed immune serum ( Fig. 2 ). The antibody generated against human SLC26A7 also recognized similar bands (data not shown).
Fig. 2. Immunoblotting of SLC26A7 in rat kidney. Left : with the use of the antibody generated against the mouse SLC26A7, a 90-kDa band was detected in the renal OM. No band was detected in the cortex. Right : labeling of the 90-kDa band was completely prevented with the preadsorbed immune serum.
Immunofluorescent labeling of SLC26A7 in rat kidney. To determine the cellular distribution and subcellular localization of SLC26A7, immunofluorescent staining with the purified immune serum was performed in rat kidney. As shown in Fig. 3 A (lower magnification x 40), SLC26A7 is absent in the cortex ( Fig. 3 Aa ) but is abundant in the outer medulla ( Fig. 3 Ab ). To determine the specificity of the SLC26A7 antibody in rat kidney, labeling in the outer medulla was examined with the preadsorbed immune serum. As indicated, preadsorbed serum did not detect any labeling in the outer medulla ( Fig. 3 Ac ). Figure 3 B (higher magnification x 60) demonstrates the localization of SLC26A7 on the basolateral membrane of a subpopulation of cells in the OMCD. There was also some intracellular staining in a number of these cells. These results are consistent with the basolateral membrane (and intracellular localization) of SLC26A7 in certain cells of the OMCD. Our results in mouse kidney demonstrate a similar pattern (data not shown).
Fig. 3. Immunofluorescent labeling with SLC26A7 antibodies. A : lower magnification. a : Cortex ( x 40); b : OM ( x 40); c : labeling with the preadsorbed immune serum in the OM. As indicated, SLC26A7 is only present in the medulla but is absent in the cortex. B : higher magnification ( x 60). SLC26A7 is located on the basolateral membrane of a subset of outer medullary collecting duct (OMCD) cells.
To determine the identity of the SLC26A7-expressing cells, double immunocytochemical staining with antibodies against SLC26A7 and AQP2, which is exclusively expressed on the apical membrane of principal cells, was performed. As shown in Fig. 4 A, SLC26A7 ( left ) and AQP2 ( right ) clearly localized to two distinct cell populations in the OMCD ( middle ). AQP2 localizes on the apical membrane of principal cells, and SLC26A7 localizes on the basolateral membrane of nonprincipal cells. A higher magnification examination of images from Fig. 4 A clearly demonstrates the localization of AQP2 and SLC26A7 in the adjacent cells in the OMCD ( Fig. 4 B ). Taken together, these results demonstrate that SLC26A7 is located on the basolateral membrane of -intercalated cells in the rat OMCD. In the inner medulla, the expression of SLC26A7 was less distinct and showed labeling in segments that were different from the IMCD based on the lack of colocalization with AQP2. Additional studies are underway to determine whether the thin descending and ascending limbs of Henle in the inner medulla express any SLC26A7.
Fig. 4. Immunofluorescent double staining with SLC26A7 [or anion exchanger 1 (AE1)] and aquaporin (AQP)-2 in rat kidney. A : SLC26A7 and AQP2 (lower magnification x 40). Left : immunocytochemical staining in the kidney indicated that SLC26A7 is localized on the basolateral membrane of a subpopulation of OMCD cells. The distribution of AQP2 ( right ) and SLC26A7 ( left ) corresponded to 2 distinct cell subtypes when dual images were acquired ( middle ), indicating that SLC26A7-expressing cells are intercalated cells. B : SLC26A7 and AQP2 (higher magnification). SLC26A7 and AQP2 localize to 2 distinct cell subpopulations in OMCD cells.
Immunofluorescent labeling of SLC26A7 and AE1 in rat kidney. The purpose of the next series of experiments was to determine whether SLC26A7 and AE1 colocalize on the basolateral membrane of -intercalated cells in the OMCD. Accordingly, double immunocytochemical staining with antibodies against AE1 and SLC26A7 was performed. As shown in Fig. 5, SLC26A7 ( A ) and AE1 ( B ) localized on the basolateral membrane of the same cells ( C ). Taken together and coupled to studies in Fig. 4, A and B, these results demonstrate that AE1 and SLC26A7 are located on the basolateral membrane of -intercalated cells in the rat OMCD.
Fig. 5. Immunofluorescent double staining with SLC26A7 and AE1 in rat kidney. SLC26A7 ( A ) and AE1 ( B ) localize on the basolateral membrane of the same cells ( C ) in OMCD. Taken together, these results demonstrate the colocalization of SLC26A7 and AE1 on the basolateral membrane of -intercalated cells in OMCD.
exchanger activity mediated via SLC26A7 is increased by hypertonicity. Recent results from our laboratory demonstrated that SLC26A7 mediates a DIDS-sensitive exchanger in the oocyte expression system ( 20 ). The abundant expression of SLC26A7 in the medulla and its absence in the cortex raised the possibility that the activity of this exchanger may be regulated by hypertonicity. The purpose of the next series of experiments was to examine the effect of hypertonicity on exchanger activity in oocytes that were injected with SLC26A7 cRNA. Oocytes were initially perfused with a /CO 2 -free, Na-free solution. Once baseline pH i was established, the perfusate was switched to the /CO 2 -containing, Na-free solution. This maneuver results in intracellular acidification in both control and SLC26A7-injected oocytes. As shown in Fig. 5 A, the pH i recovered toward baseline level in SLC26A7-injected oocytes but remained acidic in control oocytes ( Fig. 5 B ). To determine whether the recovery from acidosis is indeed dependent on intracellular chloride, oocytes were depleted of intracellular Cl - by incubating them in a Cl - -free perfusate and then exposed to a /CO 2 -containing, Na-free, Cl-free solution. In oocytes injected with SLC26A7 cRNA and depleted of Cl -, the pH i recovery from acidosis was completely abolished ( Fig. 5 C ). The summary of the results is shown in Fig. 6 D. Taken together, these results demonstrate that SLC26A7 is a Na-independent exchanger, which is functionally active at acidic pH i. To determine the effect of hypertonicity on exchange, the rate of pH i recovery from acidosis was examined in the presence of isotonic or hypertonic medium. As shown in Fig. 6 E, switching from an isotonic -free perfusate to /CO 2 -containing isotonic solution resulted in intracellular acidosis. The pH i started to recover in oocytes injected with SLC26A7 cRNA ( Fig. 6 E ). Upon the establishing of the rate of pH i recovery, the solution was switched to a /CO 2 -containing, hypertonic medium (100 mM mannitol was added to the isotonic perfusate). As demonstrated in the representative tracing in Fig. 6 E, the rate of pH i recovery was increased significantly upon the switch to the hypertonic medium. The summary of multiple experiments demonstrated that the rate of pH i recovery was increased by more than threefold in hypertonic medium vs. isotonic perfusate ( P < 0.02, n = 5). In addition, virtually no stimulatory effect of high osmolarity was observed on pH i in control oocytes injected with water ( P < 0.003, n = 5, compared with SLC26A7-injected oocytes). A representative tracing of a control oocyte is shown in Fig. 6 F (the rates of pH i recovery from acidosis in isotonic and hypertonic medium in water-injected oocytes were not different from zero in 3 separate experiments). The exchange mediated via SLC26A7 is sensitive to inhibition by DIDS, with an IC 50 of 125 µM ( 22 ).
Fig. 6. Functional identity of SLC26A7. A : exchange activity (representative tracings) in SLC26A7-expressing oocytes. Representative intracellular pH (pH i ) tracings demonstrating recovery from cell acidosis in oocytes injected with SLC26A7 cRNA. B : no pH i recovery was observed in control oocytes. C : effect of intracellular chloride depletion on pH i recovery in oocytes injected with SLC26A7 cRNA. No pH i recovery from acidosis was observed in Cl-depleted oocytes. D : summary of the results. E : effect of high osmolality on SLC26A7-mediated exchange. Switching to a high-osmolality solution significantly increased the rate of pH i recovery from acidosis in SLC26A7-injected oocytes. Summary of multiple experiments indicated that hypertonicity stimulates the rate of exchanger activity by 3-fold ( P < 0.02, n = 5). F : effect of high osmolality on pH i in control oocytes. No pH i recovery from acidosis was observed upon the switch to isotonic or hypertonic solution.
DISCUSSION
The results of the above experiments indicate that SLC26A7 is predominantly expressed in the kidney outer medulla ( Fig. 1 ). Immunoblot analysis demonstrates the expression of SLC26A7 in the kidney outer medulla ( Fig. 2 ), and immunofluorescent labeling studies localize SLC26A7 on the basolateral membrane of a subset of cells in the OMCD ( Fig. 3 ). Double-labeling studies indicated that the SLC26A7-expressing cells are -intercalated cells ( Fig. 4, A and B ). Immunofluorescent labeling verified the localization of AE1 and SLC26A7 on the basolateral membrane of -intercalated cells ( Fig. 6 D ). Expression studies in oocytes demonstrated that the exchanger mediated via SLC26A7 is sodium independent and significantly enhanced by hypertonicity ( Fig. 5 ). Taken together, these results indicate that SLC26A7 is a basolateral Cl - /base exchanger in the intercalated cells of the OMCD and is regulated by hypertonicity.
The most intriguing aspect of the current studies is the unique localization of SLC26A7 on the basolateral membrane of the OMCD and its absence from the cortex (Figs. 1, 2, 3 ). This contrasts with AE1, which is expressed in both the CCD and OMCD ( 25, 32, 34, 35; Fig. 4 ), and with AE2, which has a wider distribution in kidney tubules ( 6 ). The unique expression pattern of SLC26A7 in the kidney raises the possibility that this exchanger may play a major role in bicarbonate reabsorption in the OMCD. On the basis of immunolocalization studies, it has been proposed and presumed that AE1 is the main basolateral exchanger in OMCD intercalated cells. The presence of SLC26A7 on the basolateral membrane of the same cells indicates that there are at least two distinct basolateral exchangers in OMCD intercalated cells: SLC26A7 and AE1. With respect to epithelial cells, the presence of two transport proteins with similar functional modes on the same membrane domain usually suggests differential regulation or distinct roles related to the function or physiology of that cell or tissue. We therefore suggest that while both AE1 and SLC26A7 are basolateral exchangers in OMCD intercalated cells, they may be essential to apical acid secretion or basolateral reabsorption under dissimilar conditions (i.e., acidosis or hypertonicity). The physiological milieu in the kidney medulla is hypertonic with respect to the cortex and the serum, due, in large part, to the coordinated action of salt and urea transporters in medullary thick ascending limb and medullary collecting duct, respectively. The unique expression of SLC26A7 on the basolateral membrane of the medullary collecting duct, which is exposed to a hypertonic interstitium, and its absence in the cortex, which is facing an isotonic interstitium, raise the possibility that SLC26A7 may function better in a high-osmolality environment. Stimulation of exchanger activity in oocytes by high osmolality confirms this hypothesis ( Fig. 5 ).
A number of disease states are associated with alterations in the tonicity of the medullary interstitium. One such state is water deprivation. Our preliminary data indicate that 3 days of water deprivation, which increases the medullary interstitial osmolality and the osmolyte concentration by several-fold, are associated with enhanced SLC26A7 mRNA expression (by 380%, n = 3, P < 0.02) and protein abundance (by 3-fold, n = 3, P < 0.02) and the downregulation of AE1 mRNA expression (by 46%, n = 3, P < 0.05) and protein abundance (by 51%, n = 3, P < 0.05), indicating differential regulation of these two anion exchangers in chronic hypertonicity (Barone S, Amlal H, Xu J, and Soleimani M, unpublished data). On the basis of these results and the functional studies indicating the activation of SLC26A7 activity by high osmolality, we propose that SLC26A7 may play an important role in bicarbonate reabsorption in OMCD under hypertonic conditions. Whether SLC26A7 and AE1 are differentially regulated in other pathophysiological states such as metabolic acidosis remains unknown. Additional studies are needed to sort out the role of AE1 and SLC26A7 in pathophysiological states in the OMCD.
SLC26A7 is active at both alkaline and acidic pH i (Ref. 20 and Fig. 5 ), whereas the AE family of isoforms (and specifically AE2) are mostly active at neutral and alkaline pH i and remain relatively inactive at acidic pH i ( 1 ). The distinct pH i sensitivity profile of SLC26A7 in in vitro expression systems indicates that this AE can function over a range of pH i in vivo and remains active at both alkaline and acidic pH i. Such a distinct pH i regulatory profile may indicate that bicarbonate reabsorption across the basolateral membrane of OMCD intercalated cells may proceed unimpeded in both acidic and alkaline pH i.
In conclusion, SLC26A7 is expressed in the kidney, with mRNA expression limited to the outer and inner medulla. Immunofluorescent labeling localized SLC26A7 on the basolateral membrane of intercalated cells in the OMCD. Functional studies demonstrated that SLC26A7 functions as a exchanger and is enhanced by hypertonicity. We propose that SLC26A7 is a basolateral Cl - /base exchanger in the kidney OMCD and plays an important role in bicarbonate reabsorption in this nephron segment (specifically in conditions associated with increased tonicity in the medulla).
ACKNOWLEDGMENTS
GRANTS
These studies were supported by a Merit Review grant, a Cystic Fibrosis Foundation grant, National Institutes of Health Grant DK-54430, grants from Dialysis Clinic, Incorporated (to M. Soleimani), and grants by Finska Lakaresallskapet (to M. Kujala and J. Kere).
Address for reprint requests and other correspondence: M. Soleimani, Division of Nephrology and Hypertension, Dept. of Medicine, Univ. of Cincinnati, 231 Albert Sabin Way, MSB G259, Cincinnati, OH 45267-0585 (E-mail: Manoocher.Soleimani{at}uc.edu