Immunolocalization of anion transporter Slc26a7 in mouse kidney

【摘要】  Previous studies have indicated that a major fraction of the filtered Cl - is reabsorbed via apical membrane Cl - /base exchange in the proximal tubule. Recent studies in Slc26a6 null mice have suggested that this transporter mediates only a portion of proximal tubule Cl - /base exchange, raising the possibility that one or more unidentified apical membrane transporters may additionally contribute. Recent studies have identified Slc26a7 as another Cl - /base exchanger expressed in the kidney. We therefore generated Slc26a7-specific polyclonal and monoclonal antibodies to examine cellular and subcellular sites of expression in mouse kidney. The specificity of each antibody was verified by immunoblotting and immunofluorescence of COS-7 cells transiently transfected with mouse Slc26a7. Immunofluorescence microscopy of mouse kidney detected the expression of Slc26a7 subapically in proximal tubule cells, and on the basolateral surface of thick ascending limb cells. Similar staining patterns were demonstrated with two antibodies shown to react with different epitopes on Slc26a7. Immunolocalization of Slc26a7 to proximal tubule and thick ascending limb was also observed in rat kidney. We conclude that Slc26a7 is expressed in the proximal tubule and thick ascending limb of the loop of Henle, and it may therefore contribute to anion transport in these nephron segments.

【关键词】  proximal tubule thick ascending limb anion exchanger

THE MAJORITY OF filtered Na +, Cl -, HCO 3 -, and water are reabsorbed in the proximal tubule (PT) ( 3, 6, 28 ). Earlier studies in brush-border membrane (BBM) vesicles and perfused tubules demonstrated that a major fraction of the filtered Cl - is reabsorbed via different mechanisms of Cl - /base exchange. These mechanisms include Cl - /formate exchange functioning in parallel with Na + /H + exchange and H + /formate cotransport ( 13, 14, 32 - 34 ), and Cl - /oxalate exchange functioning in parallel with SO 4 2- /oxalate exchange and Na + /SO 4 2- cotransport ( 12, 13, 21, 32, 33 ). In addition, Cl - /OH - exchange has been described to mediate apical membrane Cl - /base exchange in the PT ( 5, 29, 36 ).

This laboratory previously identified Slc26a6 as an anion exchanger expressed on the apical brush border of PT cells ( 17 ). Functional characterization in Xenopus laevis oocytes demonstrated that Slc26a6 is capable of operating in multiple exchange modes involving Cl -, formate, oxalate, OH -, HCO 3 -, and SO 4 2- ( 11, 27, 37 ), consistent with the hypothesis that it mediates one or more of the Cl - /base exchange processes described in the PT. Recent studies in Slc26a6 null mice have suggested that this transporter mediates all of the Cl - /oxalate exchange in the PT but only a portion of the Cl - /formate exchange and Cl - /OH - exchange ( 35 ). Thus these studies raise the possibility that one or more unidentified apical membrane transporters may contribute substantially to Cl - /formate exchange and Cl - /OH - exchange across the apical membrane of PT cells. This conclusion is also supported by older studies of transport kinetics in isolated brush-border vesicles indicating that Cl - /base exchange in this nephron segment is likely mediated by two or more distinct anion exchangers ( 13 ).

SLC26A7 was the next member of the SLC26 family to be cloned, characterized, and shown to be expressed in the kidney ( 22, 31 ). Functional characterization of SLC26A7/Slc26a7 has indicated its ability to function as an anion exchanger with affinity for at least Cl -, SO 4 2-, oxalate, and OH - or HCO 3 - ( 22, 25, 26 ). Alternatively, it has been proposed that Slc26a7 principally functions as a Cl - channel regulated by pH ( 16 ). These functional properties raise the possibility that SLC26A7/Slc26a7 might contribute to Cl - transport in the PT.

We therefore initiated studies to immunolocalize Slc26a7 in the mouse kidney. We now report Slc26a7 expression at the apical pole of PT cells detected by immunofluorescence microscopy using two different Slc26a7 antibodies reacting with different epitopes. We also detected Slc26a7 expression on the basolateral membrane of thick ascending limb (TAL) cells.

MATERIALS AND METHODS

Polyclonal Antibody Preparation

A peptide corresponding to amino acids 479-503 of mouse Slc26a7 (CSITDMKEMELKVKTEMHDETSQQIK), including an NH 2 -terminal cysteine, was synthesized by the W. M. Keck Foundation Biotechnology Resource Laboratory at Yale University. The peptide (Slc26a7.479.pep) was coupled to either keyhole limpet hemocyanin or ovalbumin using the Imject Maleimide Activated Immunogen Conjugation Kit (Pierce) and used to immunize three rabbits (named 968, 969, 970) to generate polyclonal antibodies (Pocono Rabbit Farm) according to standard protocols ( http://www.prfal.com/protocolforrabbits.asp ). The specificity and magnitude of the immune response for each animal were examined by ELISA against the immobilized peptide immunogen. Anti-Slc26a7 antibodies were affinity-purified against Slc26a7.479.pep using the SulfoLink Antibody Purification Kit according to the manufacturer's protocol (Pierce). Purified antibody was stored at -20°C. Antibody purified from rabbit 968 produced the best specific signal relative to background based on immunofluorescence and Western blot assays and it was therefore used for the studies in this paper.

Monoclonal Antibody Generation

Preparation of Slc26a7 fusion proteins. A maltose-binding fusion protein (MBP.mSlc26a7.t187) and a glutathione- S -transferase fusion protein (GST.mSlc26a7.t187) containing the hydrophilic COOH-terminal 187 amino acids of mouse Slc26a7 were constructed.

Briefly, two PCR products were amplified from mouse kidney cDNA (Black Swiss) with engineered restriction sites Bam HI/ Xba I or Bam HI/ Eco RI. The PCR products were subcloned into pCR2.1-TOPO (Invitrogen) and sequenced bidirectionally to confirm incorporation of the restriction enzyme sites and to verify the sequence. The vectors with inserts were digested with Bam HI/ Xba I or Bam HI/ Eco RI and the inserts containing the terminal 561 coding nucleotides + stop codon of Slc26a7 were subcloned into pMAL-C2 (New England Biolabs) or pGEX-6P-2 (Amersham Biosciences), respectively. Sequence analysis confirmed the correct orientation and sequence of the inserts. The pMAL-C2/Slc26a7 and pGEX-6P-2/Slc26a7 vectors were transfected into BL21-CodonPlus-competent cells (Stratagene), and the MBP.mSlc26a7.t187 and GST.mSlc26a7.t187 fusion proteins were expressed and purified according to manufacturer protocols.

Immunization of mice and production of hybridomas. Eight-week-old Balb/c mice were immunized with MBP.mSlc26a7.t187 according to standard protocols ( http://info.med.yale.edu/yarc/vcs/immuniza.htm#1 ). A mouse was selected for fusion when its serum showed specific reactivity by immunoblot to Slc26a7 in Slc26a7-transfected COS-7 cells compared with untransfected control cells. Spleen cells were fused to Ag8 mouse myeloma cells as previously described ( 7, 19 ).

Hybridoma selection and monoclonal antibody purification. Hybridomas were selected for the ability of their supernatant to label Slc26a7 by ELISA, immunoblotting, and immunofluorescence. First, 1,700 hybridoma supernatants were screened by ELISA against GST.Slc26a7.t187. Antibodies specific for the Slc26a7 component of the immunogen were identified by using GST.Slc26a7.t187 instead of MBP.Slc26a7.t187. Next, 77 supernatants yielding strong signals were tested for the ability to specifically label Slc26a7 by immunoblot analysis of Slc26a7-transfected COS-7 cells with untransfected COS-7 cells as controls. Nine resultant clones were then screened for the ability to specifically label Slc26a7 by immunofluorescence using PLP-fixed Slc26a7-transfected COS-7 cells vs. PLP-fixed untransfected COS-7 cells plated on coverslips. The two remaining hybridomas that demonstrated specific Slc26a7 labeling by immunofluorescence were cloned and subcloned by limiting dilution.

Two monoclonal antibodies (m3F3, m14H5) were purified from the corresponding hybridoma supernatants by affinity chromatography using Protein-G-Sepharose 4B (Amersham Biosciences) and stored in 50% glycerol/PBS at -20°C.

ELISA. Rabbit sera and hybridoma supernatants were screened for Slc26a7 reactivity as described above. Briefly, MBP.mSlc26a7.t187 (0.2 µg/ml), GST.mSlc26a7.t187 (1 µg/ml), or Slc26a7.479.pep (0.1 mg/ml) were solubilized in PBS and used to coat ELISA plates. The plates were incubated overnight at 4°C. The plates were next washed four times (0.1% Triton X-100/PBS) after which the monoclonal antibody (25 µl hybridoma supernatant diluted in 50 µl of 0.1% Triton X-100, 0.1% BSA/PBS) or polyclonal antibody (rabbit sera diluted in 100 µl of 0.1% Triton X-100, 0.1% BSA/PBS to 1:10 5, 1:10 6, 1:10 7 ) was applied to the corresponding well and incubated 1 h at 25°C. Plates were washed four times followed by the addition of secondary antibody [horseradish peroxidase (HRP)-conjugated goat anti-rabbit or anti-mouse immunoglobulin G, Zymed] and incubated 1 h at 25°C. After a final four washes, HRP substrate was added to each well to identify the wells in which antibody/antigen complexes existed. The optical density, determined at a wavelength of 490 nm, was used as a measurement of antibody binding.

Transient Transfection of Slc26a7 and Slc26a6 in COS-7 Cells

Antibody specificity was verified by immunoblotting COS-7 cells transiently transfected with mouse Slc26a6 ( 17 ) or Slc26a7 cDNA (Black Swiss) that had been subcloned into pcDNA3.1(+) (Invitrogen). The pcDNA 3.1/Slc26a7 construct was prepared by first amplifying the full-length, 1,971-bp Slc26a7 coding sequence from mouse kidney cDNA using gene-specific primers engineered to incorporate Bam HI and Eco RI restriction enzyme sites flanking the coding sequence and to additionally insert a consensus Kozak sequence (CACC) upstream of the start codon. The PCR product was subcloned into pCR2.1-TOPO (Invitrogen) and sequenced bidirectionally to confirm incorporation of the restriction enzyme sites and Kozak consensus sequence and to verify the sequence. The vector with insert was digested with Bam HI/ Eco RI and subcloned into the Bam HI/ Eco RI sites of pcDNA 3.1(+). Sequence analysis confirmed the correct orientation and sequence of the insert.

COS-7 cells were grown in DMEM (Invitrogen/GIBCO) with 10% fetal calf serum, 50 U/ml penicillin, 50 mg/ml streptomycin, and 1 mM Na + -pyruvate at 37°C in 5% CO 2 -95% air. Cells were plated on 24-well tissue culture dishes for 24 h and used at 90-95% confluency. Cells were transfected with a 1:2 ratio of plasmid DNA (pcDNA3.1-a6/a7, 1 µg) to Lipofectamine (2 µl) in 100 µl Opti-MEM I reduced serum medium according to manufacturer's instructions (Invitrogen). Cells were incubated 24-48 h following transfection and then solubilized in SDS sample buffer (10% SDS, 20% glycerol, 2% -mercaptoethanol, 2.9 mM Tris, pH 6.8) for immunoblotting to assay for Slc26a7 and Slc26a6 expression.

Isolation of Mouse BBMs

Renal BBM were isolated from the pooled cortices of seven mice (C57BL/6J) using the magnesium aggregation method and differential centrifugation as described previously ( 2, 15 ). Protease inhibitors [Complete Protease Inhibitor Tablets (Roche), 40 µg/ml PMSF, 0.5 µg/ml leupeptin, and 0.7 µg/ml pepstatin] were included in all buffers. Membranes were resuspended in 200 mM mannitol, 80 mM HEPES titrated to pH 7.5 with KOH. Total protein was assayed by the method of Lowry ( 23, 24 ). Membranes were stored at -70°C until use.

SDS-PAGE and Immunoblotting

COS-7 cells and kidney membranes were solubilized in SDS sample buffer and subjected to SDS-PAGE using 7.5% polyacrylamide gels. Proteins were subsequently transferred to polyvinylidene fluoride (PVDF) microporous membrane (Millipore). Nonspecific binding was blocked by incubating the PVDF membrane at 25°C for 1 h in BLOTTO (5% nonfat dry milk and 0.1% Tween 20 in PBS, pH 7.4) followed by overnight incubation in polyclonal rabbit anti-mouse Slc26a7 [968, (1:200)], monoclonal mouse anti-Slc26a7 [14H5, (1:200)], rabbit anti-mouse Slc26a6 (1:250) ( 17 ) or mouse anti-NHE3 [3H3, (1:1,000)] in BLOTTO at 4°C. The membrane was washed with BLOTTO and incubated for 1 h at 25°C with a 1:2,000 dilution of goat anti-rabbit or anti-mouse HRP-conjugated secondary antibody (Zymed). An enhanced chemiluminescence detection (ECL) system (Amersham Pharmacia) with Kodak Biomax imaging film (Kodak) was used for visualization of the bound antibodies.

To detect Slc26a7 expression in native kidney tissue, a preparation of isolated cortical BBMs was subjected to SDS-PAGE and immunoblotted, same as for COS-7 cells. As presented in RESULTS, detection of Slc26a7 by immunoblotting was extremely weak despite localization by immunofluorescence microscopy. To determine whether Slc26a7 was proteolytically degraded by BBM despite the use of protease inhibitors, BBMs (50 µg) were incubated with a solubilized extract (1% Triton X-100) of Slc26a7-transfected COS-7 cells (one confluent well of 24-well plate) for 1 h at 4°C in the presence of protease inhibitors [Complete Protease Inhibitor Tablets (Roche), 40 µg/ml PMSF, 0.5 µg/ml leupeptin, and 0.7 µg/ml pepstatin]. Additionally, Slc26a7-transfected COS-7 cell lysates were incubated with BBM suspension buffer (K-HEPES) or were left unsupplemented for 1 h at 4°C as controls. To determine whether the proteolysis was specific to Slc26a7, NHE3-transfected COS-7 cells were incubated with BBM (50 µg), same as above, and blotted for NHE3. BBM alone was also blotted for NHE3.

For peptide blocking experiments, 3 µg of either 968, 14H5, or anti-Slc26a6 were incubated with 30 µg of the corresponding immunogen peptide or fusion protein in 500 µl of TBS for 16 h at 4°C. Following this, the antibody/peptide solutions were centrifuged at 18,000 g for 15 min at 4°C. The supernatant was removed and stored at 4°C until use.

Tissue Preparation and Immunofluorescence Labeling of Slc26a7

Adult male C57BL/6J mice or Sprague-Dawley rats were anesthetized with an intraperitoneal injection of pentobarbital sodium followed by perfusion fixation with PLP (2% paraformaldehyde, 750 mM lysine, and 10 mM sodium periodate in phosphate buffer, pH 7.4, 22°C). Kidneys were excised, cut into 1- to 2-mm pieces, and postfixed in PLP for an additional 4 h at 25°C. For cryosections, the tissue was infiltrated with 30% sucrose overnight, cut into 4-µm sections, and slide-mounted. For EMbed-812 sections, following postfixation in PLP, the tissue was rinsed 1 x with TBS and dehydrated in a 70-95-100% ethanol series followed by a final 10-min wash in propylene oxide. Next, the tissue was incubated in 1:1 propylene oxide/EMbed-812 (Electron Microscopy Sciences) for 2 h followed by two incubations in 100% Embed-812 for 1 h each. Finally, the tissue was embedded in EMbed-812 overnight at 60°C. One- to two-micrometer sections were cut and mounted on Superfrost Plus glass slides (Electron Microscopy Sciences). The slides were etched for 5 min in a 1:2 propylene oxide/methanol solution with 2.4 M KOH to remove EMbed-812. Slides were rinsed 2 x in methanol and 1 x in TBS. For antigen retrieval, the slides were next microwave-heated at 40% power for 20 min in 10 mM preheated citrate buffer, pH 6.0. Slides were allowed to cool for 20 min. Rabbit polyclonal anti-Slc26a7 antibody 968 was used with Embed-812 sections and microwave antigen retrieval. Mouse monoclonal anti-Slc26a7 antibody 14H5 was used with cryosections and no microwave antigen retrieval.

All sections were washed in TBS 3 x 5 min, followed by 0.5 M ammonium chloride 1 x 15 min at 20°C in a humidified chamber. This wash was removed and sections were rinsed 1 x in TBS, followed by incubation in 1% SDS in TBS for 5 min. Next, the sections were washed 2 x in TBS and blocked. Sections to be used with anti-Slc26a7 (968, 14H5), anti-Slc26a6, anti- -glutamyl transpeptidase ( -GT), anti-megalin, anti-aquaporin-2 (AQP2), or anti-anion exchanger-1 (AE1, 12B11) were blocked with 10% goat serum/0.1% BSA in TBS for 15 min, whereas sections to be used with anti-Tamm-Horsfall were blocked with 1% BSA-IgG free (Jackson ImmunoResearch) in TBS for 15 min. The sections were next incubated with either rabbit anti-mouse Slc26a7 [968 (1:50)], mouse anti-Slc26a7 [14H5 (1:50)], rabbit anti-mouse Slc26a6 (1:50) ( 17 ), rabbit anti-rat -GT (1:5,000) ( 18 ), rabbit anti-opossum megalin (1:500) ( 38 ), rabbit anti-rat/mouse AQP2 [(1:50) BD Biosciences], mouse anti-AE1 (12B11, 1:100) ( 1 ) or sheep anti-human Tamm-Horsfall (1:2,000; Chemicon) for 16 h at 4°C. All primary antibodies were diluted in TBS containing 10% goat serum/0.1% BSA except for Tamm-Horsfall antibody, which was diluted in 1% BSA-IgG free. The primary antibodies were removed and the sections were washed 3 x 5 min with high-salt TBS (2.5% NaCl) containing 0.1% BSA. The sections were next incubated for 1 h with secondary antibodies (Molecular Probes): Alexa Fluor 488 or 594 goat anti-rabbit or mouse IgG for Slc26a7; Alexa Fluor 594 goat anti-rabbit IgG for Slc26a6, -GT, megalin, and AQP2; Alexa Fluor 594 goat anti-mouse IgG for 12B11; or Alexa Fluor 594 goat anti-sheep IgG for Tamm-Horsfall. All secondary antibodies were diluted 1:200 with the TBS/0.1% BSA/10% goat serum blocking buffer except for the secondary antibody used with anti-Tamm-Horsfall which was diluted 1:200 in 1% BSA-IgG free. The slides were next washed 2 x 10 min with high-salt TBS/0.1% BSA, and 1 x 10 min with standard TBS (0.9% NaCl) and mounted in VectaShield (Vector Laboratories) to sustain the immunofluorescence signal. Slides were visualized on a Zeiss Axiophot phase-contrast microscope or Zeiss LSM 410 confocal microscope.

For peptide blocking experiments, the antibody/peptide conjugates were prepared as described above. The anti-Slc26a7/peptide mixtures were used at a dilution of 1:4.

To perform immunofluorescence on COS-7 cells instead of tissue, the cells were incubated in 0.1% Triton X-100 for 5 min to permeabilize the cells. This was followed by several TBS washes after which the sections were incubated in blocking solution as above.

RESULTS

To examine renal Slc26a7 expression in mouse, we generated polyclonal and monoclonal antibodies against mouse Slc26a7. The polyclonal antibody was raised against a peptide corresponding to amino acids 479-503 of mouse Slc26a7, and the monoclonal antibody was generated against a fusion protein incorporating amino acids 470-656 (COOH terminus) of mouse Slc26a7. To verify the specificity of the antibodies for the intact Slc26a7 protein, immunoblotting was performed using COS-7 cells transiently transfected with either mouse Slc26a7 or Slc26a6 cDNA. As shown in Fig. 1, the polyclonal antibody (968) and monoclonal antibody (14H5) principally labeled two polypeptides in the size range 85-100 kDa in Slc26a7 transfected cells, which were absent in untransfected control cells. The molecular weight of the upper band in the transfected COS-7 cells ( 95 kDa) corresponded to the molecular weight previously reported for mouse Slc26a7 in kidney and stomach by immunoblotting ( 25, 26 ). The lower 85-kDa band detected in the transfected cells may represent an immature, incompletely processed form of Slc26a7. All labeling in the transfected cells was blocked by preincubation of the antibodies with their respective immunogens. No reactivity was observed with 968 or 14H5 toward the related Slc26 gene family member Slc26a6. Additionally, there was no reactivity between an Slc26a6-specific antibody and transfected Slc26a7 protein. Sufficient Slc26a6 expression in the Slc26a6-transfected COS-7 cells was confirmed by immunoblotting with anti-Slc26a6. Taken together, these findings confirm the specificity of both the polyclonal and monoclonal antibodies for Slc26a7.

Fig. 1. Specificity of the anti-Slc26a7 antibodies by immunoblotting. Blotting with rabbit polyclonal anti-mouse Slc26a7 (968), monoclonal anti-mouse Slc26a7 (14H5), and rabbit polyclonal anti-mouse Slc26a6. For 968 and 14H5: Lane 1 : untransfected COS-7 cells. Lane 2 : Slc26a7-transfected COS-7 cells. Lane 3 : Slc26a7-transfected COS-7 cells with preincubation of 968 or 14H5 with corresponding immunogen. Lane 4 : Slc26a6-transfected COS-7 cells. For anti-Slc26a6: Lane 1 : untransfected COS-7 cells. Lane 2 : Slc26a7-transfected COS-7 cells. Lane 3 : Slc26a6-transfected COS-7 cells.

In Fig. 2, 968 and 14H5 labeled a protein 85-100 kDa in Slc26a7-transfected COS-7 cells ( lanes 1 and 3 ), same as above. The signal mediated by polyclonal 968 was blocked by its immunizing peptide (+Slc26a7.479.pep, lane 2 ), whereas the signal mediated by monoclonal 14H5 was not blocked by the same peptide, indicating that its epitope must reside elsewhere in the COOH terminal 187 amino acids of Slc26a7 ( lane 4 ). This blocking study demonstrates that 968 and 14H5 recognize different epitopes within Slc26a7. Because these two antibodies directed against two different epitopes of Slc26a7 labeled the same protein by immunoblotting, the specificity was further confirmed.

Fig. 2. Immunoblots demonstrating polyclonal 968 and monoclonal 14H5 are recognizing different epitopes of Slc26a7. Lane 1 : polyclonal 968 reactivity with Slc26a7-transfected COS-7 cells. Lane 2 : polyclonal 968 reactivity with Slc26a7-transfected COS-7 cells following preincubation with its immunizing peptide (+Slc26a7.479.pep). Lane 3 : 14H5 reactivity with Slc26a7-transfected COS-7 cells. Lane 4 : 14H5 reactivity with Slc26a7-transfected COS-7 cells following preincubation of 14H5 with the 968 immunizing peptide (+Slc26a7.479.pep).

Next, we verified by immunofluorescence microscopy that 968 ( Fig. 3 B ) and 14H5 ( Fig. 3 F ) strongly stained COS-7 cells transiently transfected with Slc26a7 but did not stain untransfected cells ( Fig. 3, A and E ). All labeling in the transfected cells was blocked by preincubation of each antibody with its respective immunogen (+Slc26a7.479.pep or MBP.mSlc26a7.t187; Fig. 3, C and G ). There was no cross-reactivity of the two Slc26a7-directed antibodies for transfected Slc26a6 protein ( Fig. 3, D and H ). As with the aforementioned immunoblotting of Slc26a6-transfected cells ( Fig. 1 ), there was no reactivity between the Slc26a6-specific antibody and transfected Slc26a7 protein ( Fig. 3 J ). Sufficient Slc26a6 expression was confirmed ( Fig. 3 L ) and was blocked by preincubation with the anti-Slc26a6 blocking peptide ( Fig. 3 K ). These findings again confirmed the specificity of the antibodies for Slc26a7 while additionally demonstrating their utility for immunofluorescence studies.

Fig. 3. Specificity of the anti-Slc26a7 antibodies by immunofluorescence. No labeling of untransfected (UT) COS-7 cells by polyclonal 968 ( A ) or monoclonal 14H5 ( E ). Labeling of Slc26a7-transfected (A7-TX) COS-7 cells by 968 ( B ) and 14H5 ( F ). Labeling of Slc26a7-transfected cells by 968 ( C ) and 14H5 ( G ) blocked by respective immunogens. No labeling of Slc26a6-transfected (A6-TX) COS-7 cells by 968 ( D ) or 14H5 ( H ). No labeling of untransfected cells by anti-Slc26a6 ( I ). No labeling of Slc26a7-transfected cells by Anti-Slc26a6 ( J ). Labeling of Slc26a6-transfected (A6-TX) cells by anti-Slc26a6 blocked by its immunogen ( K ). Labeling of Slc26a6-transfected cells by Anti-Slc26a6 ( L ). Magnification: x 1,000.

We next performed studies to characterize the cellular and subcellular localization of Slc26a7 in mouse kidney tissue using indirect immunofluorescence or confocal microscopy. With both 968 ( Fig. 4 A ) and 14H5 ( Fig. 4 D ), abundant expression was seen on the basolateral membrane of distal nephron segments (DS), the apical domain of PTs, and in the glomerulus (G). Confirming the specificity of the staining, the signals from both 968 ( Fig. 4 C ) and 14H5 ( Fig. 4 F ) were completely abolished following preincubation of the antibodies with their respective immunogens. The fact that similar patterns of staining were observed using antibodies reactive with two different epitopes further underscores the specificity of the labeling.

Fig. 4. Immunolocalization of Slc26a7 in mouse renal cortex. A : staining of proximal tubule (PT) apical domain and distal segment (DS) basolateral membranes with polyclonal 968. B : phase-contrast image of A showing distinct proximal tubule and distal segment morphology. C : staining was abolished when 968 was preincubated with the immunizing peptide. D : staining of proximal tubule apical domain and distal segment basolateral membranes with monoclonal 14H5. E : phase-contrast image of D. F : staining was abolished when 14H5 was preincubated with the fusion-protein against which it was generated. Magnification: x 160.

The membrane localization of Slc26a7 is shown in greater detail in the higher-magnification images in Fig. 5. As indicated, 968 ( Fig. 5 A ) and 14H5 ( Fig. 5 B ) clearly localized Slc26a7 to the apical pole of PT cells. In contrast, there was intense staining of the numerous basolateral invaginations of distal nephron segments by both 968 ( Fig. 5 C ) and 14H5 ( Fig. 5 D ).

Fig. 5. Immunolocalization of Slc26a7 in mouse renal cortex at higher magnification. A : Slc26a7 localization to the proximal tubule apical domain using 968. B : same as A but with 14H5. C : Slc26a7 localization to the basolateral membrane of a distal nephron segment using 968. D : same as C but with 14H5. A - D, bottom : corresponding phase-contrast image. Magnification: x 900.

The subcellular localization of Slc26a7 in the PT was examined in greater detail by dual-labeling experiments utilizing antibodies directed against the brush-border enzyme -GT, or against megalin. Figure 6, A and B, shows abundant expression of Slc26a7 and -GT, respectively, at the apical domain of PT cells. However, overlay of the two images ( Fig. 6 C ) clearly demonstrates that Slc26a7 expression is internal to that of -GT, indicating a submicrovillar localization of Slc26a7. The submicrovillar localization of Slc26a7 was confirmed by its colocalization with an anti-megalin antibody ( 38 ) that labels a pool of megalin in the intermicrovillar coated pits and subapical dense tubules but that does not label microvillar megalin ( Fig. 6, D - F ).

Fig. 6. Colocalization of apical Slc26a7 with -glutamyl transpeptidase ( -GT) or megalin. A : Slc26a7 localization to the proximal tubule apical domain using 14H5. B : staining of the same segment for the brush-border marker -GT. C : overlay of A and B demonstrating localization of Slc26a7 to a submicrovillar compartment. D : confocal image showing Slc26a7 localization to the proximal tubule apical domain using 14H5. E : staining of the same segment for megalin. F : overlay of D and E showing colocalization of Slc26a7 and megalin to the same subapical compartment. Magnification: x 900.

The tubule segments exhibiting basolateral staining were characterized by extensive basolateral invaginations and a relatively smooth apical surface suggestive of TAL. We therefore used the TAL marker Tamm-Horsfall glycoprotein ( 4 ) to substantiate this interpretation. Localization of the basolateral membrane staining of Slc26a7 ( Fig. 7 A ) to the same tubule segments and cells as Tamm-Horsfall protein ( Fig. 7 B ) confirmed expression in TAL. Apical expression of Slc26a7 was again noted in PTs.

Fig. 7. Colocalization of basolateral Slc26a7 with Tamm-Horsfall protein. A : localization of Slc26a7 to the PT apical domain and distal segment basolateral membranes using 14H5. B : staining of the same segments with anti-Tamm-Horsfall indicating thick ascending limb (TAL). C : phase-contrast image of A and B. Magnification: x 900.

In addition to characterizing Slc26a7 expression in mouse renal cortex, we examined Slc26a7 localization in outer and inner medulla. Similar to that found in cortex, intense staining of the basolateral membranes of outer medullary TAL was observed ( Fig. 8 A ). Double labeling for Slc26a7 and aquaporin-2 (AQP2) in sections of outer ( Fig. 8 A ) and inner medulla ( Fig. 8 B ) demonstrated a lack of reactivity in collecting duct (CD). Thin limb (TL) and vascular endothelial cells (EC) were also negative.

Fig. 8. Double-staining of mouse outer and inner medulla for Slc26a7 and aquaporin-2. A : differential localization of Slc26a7 to TAL using polyclonal 968 ( left ), and aquaporin-2 to collecting duct (CD; middle ), in outer medulla. Right : phase-contrast. B : lack of reactivity for Slc26a7 ( left ) in CD (stained with anti-aquaporin-2; middle ), thin-limb (TL), and vascular endothelium (EC) in inner medulla. Right : phase-contrast. Magnification: x 300.

To further characterize the renal localization of Slc26a7, expression of the protein was examined in sections of rat renal cortex and medulla. Similar to the above findings in mouse kidney, Fig. 9 A shows localization of Slc26a7 to the apical domain of PTs and basolateral membranes of distal segments (DS). In rat medulla, as in mouse, reactivity was observed on the basolateral membranes of distal segments ( Fig. 9 B ). Figure 9 C shows a double labeling for Slc26a7 and the CD-specific anion exchanger-1 (AE1) in rat. Although this particular experiment failed to recognize Slc26a7 in CD, reactivity with a subpopulation of cells in this nephron segment was inconsistently observed in some experiments. This inconsistent CD staining was observed with polyclonal antibody 968 on Embed-812 embedded rat kidney sections subjected to microwave antigen retrieval and SDS treatment, and with monoclonal antibody 14H5 on cryosections of mouse kidney subjected to SDS treatment. Our findings therefore do not exclude the observed expression of Slc26a7 on the basolateral membrane of intercalated cells as reported in other studies ( 20, 25 ).

Fig. 9. Immunolocalization of Slc26a7 in rat renal cortex and medulla. A : localization of Slc26a7 to cortical PT apical domain and DS basolateral membranes using polyclonal 968. Right : phase-contrast. B : localization of Slc26a7 to medullary distal segment basolateral membranes. Right : phase-contrast. Magnification: x 160. C : differential localization of Slc26a7 to distal segments ( left ) and anion exchanger-1 to CD ( middle ) in outer medulla. Right : phase-contrast. Magnification: x 200.

To confirm the localization of Slc26a7 to the apical domain of PT cells as observed by immunofluorescence microscopy, we next used immunoblotting to detect Slc26a7 expression in a preparation of mouse kidney BBM. The membrane preparation was probed in parallel for Slc26a7 and another membrane-associated transporter, Na + /H + exchanger-3 (NHE3). As shown in Fig. 10, lanes 1 and 5, under the same conditions that revealed NHE3 expression, expression of Slc26a7 was not detected. With 100-fold longer exposure (not shown), a ladder of bands was observed in the Slc26a7 lane, all of which were blocked by preincubation with the immunizing agent, so the specificity of the labeling was uncertain. The same result was obtained with several different membrane preparations of both BBM and renal cortical microsomes, all prepared with a full complement of protease inhibitors (data not shown). Given the abundance of Slc26a7 staining by immunofluorescence microscopy, we then considered the possibility that Slc26a7 was being degraded in kidney membrane preparations, thus preventing detection of the intact protein by immunoblotting. As presented earlier, the presence of Slc26a7 in Slc26a7-transfected COS-7 cells was easily detected by immunoblotting ( Fig. 10, lane 2 ), as was also the case for NHE3 ( lane 6 ). However, on mixing BBM with the Slc26a7-transfected COS-7 cell lysate, the signal obtained for Slc26a7 was dramatically reduced ( lane 3 ) despite the presence of protease inhibitors. In contrast, mixing BBM with NHE3-transfected COS-7 cell lysate caused additivity of detectable NHE3 ( lane 7 ), as would be expected. As an additional control, mixing transfected COS-7 cell lysates with membrane suspension buffer (K-HEPES) caused no loss of Slc26a7 or NHE3 ( lanes 4 and 8 ). These data clearly indicate the presence of substantial proteolytic degradation of Slc26a7 in kidney membrane preparations. This difficulty in the ability to visualize intact Slc26a7 protein precluded any further immunoblotting studies utilizing native kidney membranes.

Fig. 10. Comparison of Slc26a7 and NHE3 expression and degradation in mouse kidney brush-border membranes (BBM). For Slc26a7 using monoclonal 14H5: Lane 1 : immunoblotting Slc26a7 in mouse kidney BBM. Lane 2 : immunoblotting Slc26a7 in Slc26a7-transfected COS-7 cells. Lane 3 : Immunoblotting Slc26a7 in Slc26a7-transfected COS-7 cells to which mouse BBM (50 µg) was added. Lane 4 : immunoblotting Slc26a7 in Slc26a7-transfected COS-7 cells following addition of BBM resuspension buffer (K-HEPES). For NHE3: Lane 5 : immunoblotting NHE3 in mouse kidney BBM. Lane 6 : immunoblotting NHE3 in NHE3-transfected COS-7 cells. Lane 7 : immunoblotting NHE3 in NHE3-transfected COS-7 cells to which mouse BBM (50 µg) was added. Lane 8 : immunoblotting NHE3 in NHE3-transfected COS-7 cells following addition of BBM resuspension buffer (K-HEPES).

DISCUSSION

In the present study, a rabbit polyclonal antibody and a mouse monoclonal antibody directed against mouse Slc26a7 were generated and demonstrated to specifically label Slc26a7 in transfected cells by immunoblotting and immunofluorescence microscopy. The antibodies were then used to characterize Slc26a7 expression patterns in mouse kidney. Immunofluorescence microscopy on sections of mouse renal cortex demonstrated the expression of Slc26a7 at the apical pole but in a submicrovillar compartment in PT cells. In addition to the PT staining, abundant expression was also observed on the basolateral membranes of TAL cells and in the glomerulus. In outer medulla Slc26a7 localization was again observed in TAL basolateral membranes. Similar results were obtained in rat kidney. Importantly, both the PT and loop of Henle localization of Slc26a7 were observed with two antibodies that we demonstrated to be reactive with different epitopes on Slc26a7, confirming the specificity of the findings. It should also be noted that each antibody was used under different conditions of tissue preparation and antigen retrieval, so our findings were not dependent on a single method. Two other studies examining Slc26a7 expression in rat ( 25 ) and human ( 20 ) kidney did not detect Slc26a7 expression in PT or TAL. Rather, both studies detected Slc26a7 expression in -intercalated cells of CD.

The differences in findings between the present study and previous studies could be due to the fact that the antibodies utilized for each study were directed against a different epitope of Slc26a7. An antibody directed at the very NH 2 terminus of Slc26a7 was employed by Petrovic and colleagues ( 25 ), whereas an antibody recognizing the very COOH terminus of SLC26A7 was utilized by Kujala and colleagues ( 20 ). Our polyclonal antibody was raised to a peptide (amino acids 479-503) that corresponds to neither of these sites. Our monoclonal antibody was raised to a large COOH-terminal region (amino acids 470-656) so that it must recognize an epitope different from the NH 2 -terminal antibody of Petrovic et al. ( 25 ) and may react with a different epitope than the antibody of Kujala et al. ( 20 ). Again, we would emphasize that our own antibodies were directed at two different epitopes of Slc26a7 and both detected expression in PT and TAL. It is therefore possible that the extreme NH 2 and COOH termini of Slc26a7 are either proteolytically cleaved or otherwise occluded in PT and TAL cells, accounting for the inability in the previous studies to detect Slc26a7 in these nephron segments. Similarly, our inability to consistently demonstrate labeling of CD as reported in other studies may be due to the fact that the epitopes against which our antibodies were generated are not accessible in this nephron segment. Although two different splice variants of human SLC26A7 have been described ( 31 ), only a single mouse isoform is known to date. Thus it is unlikely that the expression of Slc26a7 at the apical domain in PT and on the basolateral membrane in TAL in mouse is due to differential localization of splice variants.

With regard to function, basolateral Cl - /HCO 3 - exchange activity has been observed in TAL ( 9, 30 ). This is a known mode of operation of Slc26a7 ( 25, 26 ). Thus the functional role of SLC26A7/Slc26a7 in TAL may be to contribute to basolateral Cl - /HCO 3 - exchange activity along with AE2 which is expressed at this site ( 8, 10 ).

As mentioned earlier, recent studies in Slc26a6 null mice have indicated that Slc26a6 mediates only a portion of the Cl - /formate exchange and Cl - /OH - exchange in the PT ( 35 ). Slc26a7 may therefore contribute to mediating these forms of Cl - /base exchange in the PT. If so, the localization of Slc26a7 to a subapical compartment in PT cells suggests that such Cl - /base exchange may be subject to regulation by membrane trafficking. Alternatively, the recent report that Slc26a7 can function as an acid-activated Cl - channel ( 16 ) raises the possibility that Slc26a7 may operate in parallel with H + -ATPase to acidify subapical endosomes in the PT.

In summary, we demonstrated that Slc26a7 is expressed in the PT and TAL of the loop of Henle and it may therefore contribute to anion transport in these nephron segments. Future studies using approaches like knockout mice will be important to determine the functional role of Slc26a7 in the kidney with certainty.

GRANTS

This work was supported by a National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Grant R01-DK-33793 (P. S. Aronson) and a NIDDK Ruth L. Kirschstein Individual National Research Service Award 1-F32 DK-067791 (P. L. Dudas).

【参考文献】
  Alper SL, Stuart-Tilley AK, Biemesderfer D, Shmukler BE, and Brown D. Immunolocalization of AE2 anion exchanger in rat kidney. Am J Physiol Renal Physiol 273: F601-F614, 1997.

Aronson PS. Energy dependence of phlorizin binding to isolated microvillus membranes. J Membr Biol 42: 81-98, 1978.

Aronson PS and Giebisch G. Mechanisms of chloride transport in the proximal tubule. Am J Physiol Renal Physiol 273: F179-F192, 1997.

Bachmann S, Metzger R, and Bunnemann B. Tamm-Horsfall protein-mRNA synthesis is localized to the thick ascending limb of Henle's loop in rat kidney. Histochemistry 95: 517-523, 1990.

Baum M. Evidence that parallel Na + -H + and Cl - -HCO 3 - (OH - ) antiporters transport NaCl in the proximal tubule. Am J Physiol Renal Fluid Electrolyte Physiol 252: F338-F345, 1987.

Berry CA and Rector FC Jr. Mechanism of proximal NaCl reabsorption in the proximal tubule of the mammalian kidney. Semin Nephrol 11: 86-97, 1991.

Biemesderfer D, Rutherford PA, Nagy T, Pizzonia JH, Abu-Alfa AK, and Aronson PS. Monoclonal antibodies for high-resolution localization of NHE3 in adult and neonatal rat kidney. Am J Physiol Renal Physiol 273: F289-F299, 1997.

Castillo JE, Martinez-Anso E, Malumbres R, De Alva E, Garcia C, Medina JF, and Prieto J. In situ localization of anion exchanger-2 in the human kidney. Cell Tissue Res 299: 281-287, 2000.

Eladari D, Blanchard A, Leviel F, Paillard M, Stuart-Tilley AK, Alper SL, and Podevin RA. Functional and molecular characterization of luminal and basolateral Cl - /HCO 3 - exchangers of rat thick limbs. Am J Physiol Renal Physiol 275: F334-F342, 1998.

Frische S, Zolotarev AS, Kim YH, Praetorius J, Alper S, Neilsen S, and Wall SM. AE2 isoforms in rat kidney: immunohistochemical localization and regulation in response to chronic NH 4 Cl loading. Am J Physiol Renal Physiol 286: F1163-F1170, 2004.

Jiang Z, Grichtchenko II, Boron WF, and Aronson PS. Specificity of anion exchange mediated by mouse Slc26a6. J Biol Chem 277: 33963-33967, 2002.

Karniski LP. Effects of sulfate and chloride on three separate oxalate transporters reconstituted from rabbit renal cortex. Am J Physiol Renal Physiol 274: F189-F196, 1998.

Karniski LP and Aronson PS. Anion exchange pathways for Cl - transport in rabbit renal microvillus membranes. Am J Physiol Renal Fluid Electrolyte Physiol 253: F513-F521, 1987.

Karniski LP and Aronson PS. Chloride/formate exchange with formic acid recycling: a mechanism of active chloride transport across epithelial membranes. Proc Natl Acad Sci USA 82: 6362-6365, 1985.

Karniski LP, Wang T, Everett LA, Green ED, Giebisch G, and Aronson PS. Formate-stimulated NaCl absorption in the proximal tubule is independent of the pendrin protein. Am J Physiol Renal Physiol 283: F952-F956, 2002.

Kim KH, Shcheynikov N, Wang Y, and Muallem S. SLC26A7 is a Cl - channel regulated by intracellular pH. J Biol Chem 280: 6463-6470, 2005.

Knauf F, Yang CL, Thomson RB, Mentone S, Giebisch G, and Aronson PS. Identification of a chloride-formate exchanger expressed on the brush-border membrane of renal proximal tubule cells. Proc Natl Acad Sci USA 98: 9425-9430, 2001.

Kocinsky HS, Girardi ACC, Biemesderfer D, Nguyen T, Mentone S, Orlowski J, and Aronson PS. Use of phospho-specific antibodies to determine the phosphorylation of endogenous Na + /H + exchanger NHE3 at PKA consensus sites. Am J Physiol Renal Physiol 289: F249-F258, 2005.

Kohler G and Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256: 495-497, 1975.

Kujala M, Tienari J, Lohi H, Elomaa O, Sariola H, Lehtonen E, and Kere J. SLC26A6 and SLC26A7 anion exchangers have a distinct distribution in human kidney. Nephron Exp Nephrol 101: e50-e58, 2005.

Kuo SM and Aronson PS. Pathways for oxalate transport in rabbit renal microvillus membrane vesicles. J Biol Chem 271: 15491-15497, 1996.

Lohi H, Kujala M, Makela S, Lehtonen E, Kestila M, Saarialho-Kere U, Markovich D, and Kere J. Functional characterization of three novel tissue-specific anion exchangers SLC26A7, -A8, and -A9. J Biol Chem 277: 14246-14254, 2002.

Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265-275, 1951.

Peterson GL. A simplification of the protein assay method of Lowry et al. which is more generally applicable. Anal Biochem 83: 346-356, 1977.

Petrovic S, Barone S, Xu J, Conforti L, Ma L, Kujala M, Kere J, and Soleimani M. SLC26A7: a basolateral Cl - /HCO 3 - exchanger specific to intercalated cells of outer medullary collecting duct. Am J Physiol Renal Physiol 286: F161-F169, 2004.

Petrovic S, Ju X, Barone S, Seidler U, Alper SL, Lohi H, Kere J, and Soleimani M. Identification of a basolateral Cl - /HCO 3 - exchanger specific to gastric parietal cells. Am J Physiol Gastrointest Liver Physiol 284: G1093-G1103, 2003.

Petrovic S, Ma L, Wang Z, and Soleimani M. Identification of an apical Cl - /HCO 3 - exchanger in rat kidney proximal tubule. Am J Physiol Cell Physiol 285: C608-C617, 2003.

Rector FC Jr. Sodium, bicarbonate, and chloride absorption by the proximal tubule. Am J Physiol Renal Fluid Electrolyte Physiol 244: F461-F471, 1983.

Shiuan D and Weinstein SW. Evidence for electroneutral chloride transport in rabbit cortical brush-border membrane vesicles. Am J Physiol Renal Fluid Electrolyte Physiol 247: F837-F847, 1984.

Sun AM. Expression of Cl - /HCO 3 - exchanger in the basolateral membrane of mouse medullary thick ascending limb. Am J Physiol Renal Physiol 274: F358-F364, 1998.

Vincourt JB, Jullien D, Kossida S, Amalric F, and Girard JP. Molecular cloning of SLC26A7, a novel member of the SLC26 sulfate/anion transporter family, from high endothelial venules and kidney. Genomics 79: 249-256, 2002.

Wang T, Egbert J, AL, Abbiati T, Aronson PS, and Giebisch G. Mechanisms of stimulation of proximal tubule chloride transport by formate and oxalate. Am J Physiol Renal Fluid Electrolyte Physiol 271: F446-F450, 1996.

Wang T, Giebisch G, and Aronson PS. Effects of formate and oxalate on volume absorption in rat proximal tubule. Am J Physiol Renal Fluid Electrolyte Physiol 263: F37-F42, 1992.

Wang T, Yang CL, Abbiati T, Shull GE, Giebisch G, and Aronson PS. Essential role of NHE3 in facilitating formate-dependent NaCl absorption in the proximal tubule. Am J Physiol Renal Physiol 281: F288-F292, 2001.

Wang Z, Wang T, Petrovic S, Tuo B, Riederer B, Barone S, Lorenz JN, Seidler U, Aronson PS, and Soleimani M. Renal and intestinal transport defects in Slc26a6-null mice. Am J Physiol Cell Physiol 288: C957-C965, 2005.

Warnock DG and Yee VJ. Chloride uptake by brush border membrane vesicles isolated from rabbit renal cortex. J Clin Invest 67: 103-115, 1981.

Xie Q, Welch R, Mercado A, Romero MF, and Mount DB. Molecular characterization of the murine Slc26a6 anion exchanger, functional comparison to Slc26a1. Am J Physiol Renal Physiol 283: F826-F838, 2002.

Zou Z, Chung B, Nguyen T, Mentone S, Thomson B, and Biemesderfer D. Linking receptor-mediated endocytosis and cell signaling: evidence for regulated intramembrane proteolysis of megalin in proximal tubule. J Biol Chem 279: 34302-34310, 2004.

作者单位：Departments of 1 Internal Medicine and 2 Cellular and Molecular Physiology, Yale University School of Medicine, New Haven, Connecticut