Claudin-2 is selectively expressed in proximal nephron in mouse kidney Alissa H. Enck, Urs V. Berger and Alan S. L. Yu Am J Physiol Renal Physiol 281:F966-F974, 2001. First published 15 August 2001; doi:10.1152/ajprenal.0021.2001 You might find this additional info useful... This article cites 26 articles, 15 of which can be accessed free at: http://ajprenal.physiology.org/content/281/5/F966.full.html#ref-list-1 This article has been cited by 26 other HighWire hosted articles, the first 5 are: Secretory Carrier Membrane Protein 2 Regulates Exocytic Insertion of NKCC2 into the Cell Membrane Nancy Zaarour, Nadia Defontaine, Sylvie Demaretz, Anie Azroyan, Lydie Cheval and Kamel Laghmani J. Biol. Chem., March 18, 2011; 286 (11): 9489-9502. [Abstract] [Full Text] [PDF] Calcium Inhibits Paracellular Sodium Conductance through Claudin-2 by Competitive Binding Alan S. L. Yu, Mary H. Cheng and Rob D. Coalson J. Biol. Chem., November 19, 2010; 285 (47): 37060-37069. [Abstract] [Full Text] [PDF] Differential expression of claudin tight junction proteins in the human cortical nephron Adam Kirk, Sara Campbell, Paul Bass, Juan Mason and Jane Collins Nephrol. Dial. Transplant., July , 2010; 25 (7): 2107-2119. [Abstract] [Full Text] [PDF] Claudin-2−deficient mice are defective in the leaky and cation-selective paracellular permeability properties of renal proximal tubules Shigeaki Muto, Masaki Hata, Junichi Taniguchi, Shuichi Tsuruoka, Kazumasa Moriwaki, Mitinori Saitou, Kyoko Furuse, Hiroyuki Sasaki, Akio Fujimura, Masashi Imai, Eiji Kusano, Shoichiro Tsukita and Mikio Furuse PNAS, April 27, 2010; 107 (17): 8011-8016. [Abstract] [Full Text] [PDF] Updated information and services including high resolution figures, can be found at: http://ajprenal.physiology.org/content/281/5/F966.full.html Additional material and information about AJP - Renal Physiology can be found at: http://www.the-aps.org/publications/ajprenal This information is current as of March 27, 2012. AJP - Renal Physiology publishes original manuscripts on a broad range of subjects relating to the kidney, urinary tract, and their respective cells and vasculature, as well as to the control of body fluid volume and composition. It is published 12 times a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2001 by the American Physiological Society. ISSN: 1931-857X, ESSN: 1522-1466. Visit our website at http://www.the-aps.org/. Downloaded from ajprenal.physiology.org on March 27, 2012 Cytoplasmic Accumulation of Incompletely Glycosylated SHBG Enhances Androgen Action in Proximal Tubule Epithelial Cells Eui-Ju Hong, Biswajyoti Sahu, Olli A. Jänne and Geoffrey L. Hammond Molecular Endocrinology, February , 2011; 25 (2): 269-281. [Abstract] [Full Text] [PDF] Am J Physiol Renal Physiol 281: F966–F974, 2001. First published August 15, 2001; 10.1152/ajprenal.00021.2001. Claudin-2 is selectively expressed in proximal nephron in mouse kidney ALISSA H. ENCK, URS V. BERGER AND ALAN S. L. YU Renal Division and Membrane Biology Program, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts 02115 Received 26 January 2001; accepted in final form 29 June 2001 tight junction; paracellular transport; renal tubule; gene expression a key role in the reabsorption of solutes by the renal tubule, particularly in “leaky” nephron segments. In the proximal tubule, approximately one-third of NaCl reabsorption is thought to be passive and occur via paracellular transport (17). Active transcellular reabsorption of Na⫹ and HCO3⫺ in the initial proximal tubule generates a high luminal Cl⫺ concentration that then drives diffusive paracellular Cl⫺ flux. This process is critically dependent on the preferential permeability of the paracellular pathway of the mid- and late proximal tubule for Cl⫺ over HCO3⫺ (10). The resultant lumen-positive electrical potential drives concomitant paracellular Na⫹ reabsorption. The slight transtubular osmotic gradient that this creates is sufficient to drive water reabsorption, most of which likely occurs transcellularly and is mediated by aquaporin-1 (26). The structural and moTHE PARACELLULAR PATHWAY PLAYS Address for reprint requests and other correspondence: A. S. L. Yu, Renal Division, Brigham and Women’s Hospital, 77 Ave. Louis Pasteur, Boston, MA 02115 (E-mail: [email protected]). F966 lecular basis for the unique permeability properties of the paracellular pathway of the proximal tubule is presently unknown. Recent studies have demonstrated that the paracellular permeability barrier in all cells is constituted by the tight junction, which is a complex of multiple structural proteins (6). One class of tight junction proteins that has recently been identified, the claudin family, is of particular interest (25). Claudins are likely to be important determinants of paracellular epithelial permeability for five reasons. First, claudins are transmembrane proteins and therefore contribute peptide domains that physically protrude into the intercellular space. Second, overexpression of a claudin in cultured epithelia has been shown to reduce paracellular permeability (12). Third, one isoform, claudin-16 (paracellin-1), is localized to the thick ascending limb and distal convoluted tubule of the kidney and, when mutated, causes autosomal-recessive hypomagnesemia with hypercalciuria, an inherited disorder that is likely due to defective paracellular divalent cation transport (24). Fourth, at least 20 mammalian isoforms of claudin have now been identified, and all of those that have been studied have been found to be highly expressed in some epithelial tissues, including the kidney (7, 18). Finally, different isoforms of claudin can form heteropolymeric tight junction strands in a single cell, and heterophilic paired interactions between cells, and therefore have the potential to generate considerable combinatorial diversity (9). We (29) and others (24, 25, 27) have therefore proposed that the combination of paracellular permeability properties to solutes and/or water that is unique to each epithelial tissue, and to each nephron segment within the renal tubule, is specified by the unique claudin isoform, or combination of isoforms, that is expressed. To map the claudin isoforms expressed in each segment of the renal tubule, we have used in situ hybridization and immunofluorescent staining of mouse kidney cryosections. We now report that claudin-2, an isoform that is restricted to the kidney and liver (7), is highly and selectively expressed in the proximal renal tubule and early thin descending limb. 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. 0363-6127/01 $5.00 Copyright © 2001 the American Physiological Society http://www.ajprenal.org Downloaded from ajprenal.physiology.org on March 27, 2012 Enck, Alissa H., Urs V. Berger and Alan S. L. Yu. Claudin-2 is selectively expressed in proximal nephron in mouse kidney. Am J Physiol Renal Physiol 281: F966–F974, 2001. First published August 15, 2001; 10.1152/ajprenal. 00021.2001.—The proximal nephron possesses a leaky epithelium with unique paracellular permeability properties that underlie its high rate of passive NaCl and water reabsorption, but the molecular basis is unknown. The claudins are a large family of transmembrane proteins that are part of the tight junction complex and likely form structural components of a paracellular pore. To localize claudin-2 in the mouse kidney, we performed in situ hybridization using an isoform-specific riboprobe and immunohistochemistry using a polyclonal antibody directed against a COOH-terminal peptide. Claudin-2 mRNA and protein were found throughout the proximal tubule and in the contiguous early segment of the thin descending limb of long-looped nephrons. The level of expression demonstrated an axial increase from proximal to distal segments. In confocal images, the subcellular localization of claudin-2 protein coincided with that of the tight junction protein ZO-1. Our findings suggest that claudin-2 is a component of the paracellular pathway of the most proximal segments of the nephron and that it may be responsible for their uniquely leaky permeability properties. CLAUDIN-2 EXPRESSION IN KIDNEY METHODS AJP-Renal Physiol • VOL Immunostaining was then performed, as described below, in the presence of 0.3% Triton X-100. In vitro translation and immunoblotting. The two templates that worked optimally for translation were plasmid constructs containing the coding region and in-frame COOHterminal FLAG epitope tag of claudin-2 in pcDNA3 (see above) and claudin-7 in the pOX(⫹) vector, which has a -globin 5⬘-untranslated sequence and a poly-A⫹ tail (4). In vitro translation was performed with the TNT-coupled reticulocyte lysate system in the presence of Transcend biotinylated lysine-tRNA (both from Promega) and detected by streptavidin blotting, according to the manufacturer’s instructions. Translation of firefly luciferase (predicted molecular mass, 61 kDa) was used as a positive control, and omission of the template as a negative control. To isolate proteins from native tissue, mouse kidney cortex was dissected and homogenized, and a crude membrane preparation and soluble fraction were separated by differential centrifugation at 47,000 g. Immunoblots were performed as described previously (4). Rabbit polyclonal antiserum raised against a synthetic peptide derived from the COOH terminal of human claudin-2 and epitope affinity purified (Zymed) was used at a concentration of 1 g/ml. Immunohistochemistry. Immunohistochemistry was performed on 5-m mouse kidney cryosections essentially as described previously (15), except that perfusion-fixation with paraformaldehyde was not performed because we found that it obscured detection of the claudin-2 antigen. Instead, sections were postfixed for 15 min in 4% paraformaldehyde at room temperature, then washed in PBS before immunostaining. The claudin-2 primary antibody was used directly on unamplified sections at a concentration of 3 g/ml; rabbit polyclonal aquaporin-1 antibody (kind gift of Dr. Dennis Brown) was used at 1:200, rat monoclonal ZO-1 antibody (kind gift of Dr. Bradley Denker) at 1:2, and M2 anti-FLAG monoclonal antibody (Sigma) at 1:100. In double-label colocalization experiments, a series of rigorous control studies were performed to exclude cross-reactivity between the two primary antibodies, as we have described in detail previously (15). Specifically for double-labeling with rabbit antibodies against claudin-2 and aquaporin-1, the sections were stained first with claudin-2 antibody at a low concentration (0.12 g/ml, undetectable without amplification), after which the signal was amplified with the tyramide amplification system (TSA, NEN Life Sciences) and detected with a Cy3 fluorophore. The sections were then incubated with the aquaporin-1 antibody and detected without amplification using a fluorescein-conjugated anti-rabbit IgG secondary antibody. To counterstain for nuclear DNA, sections were mounted in Vectashield mounting medium containing propidium iodide (Vector Laboratories). Slides were visualized with a Bio-Rad MRC-1024 confocal krypton-argon laser scanning microscope. For double-labeled slides, images were acquired sequentially for each fluorophore in single-label mode to minimize “bleed-through” between channels. Each pair of images was then imported into Adobe Photoshop 3.0, where false color was added, and the pair was merged to generate dualcolor images. Peptide-blocking studies. The COOH-terminal 47 residues (amino acids 184–230) of mouse claudin-2 were amplified by PCR and cloned downstream of, and in-frame with, the Escherichia coli glutathione-S-transferase (GST) coding sequence under the control of the lacZ␣ promoter, using the bacterial expression vector pGEX-4T (Amersham-Pharmacia Biotech). Transformants were grown in the presence of isopropyl--Dthiogalactopyranoside to induce synthesis of the GSTCLDN-2 fusion protein, harvested, lysed by sonication, and 281 • NOVEMBER 2001 • www.ajprenal.org Downloaded from ajprenal.physiology.org on March 27, 2012 Preparation of riboprobes. The full 693-bp coding region of mouse claudin-2 cDNA (GenBank accession no. AI789490) was amplified by PCR, cloned into pBluescript II KS(⫹), and verified by DNA sequencing. Digoxigenin-labeled sense and antisense cRNA probes were prepared by in vitro transcription and alkali hydrolyzed to a length of 200 nucleotides, as described previously (15). Unlabeled sense cRNA transcripts of claudin-2 to be used as targets in dot-blot studies were generated similarly, except that digoxigenin was omitted. Labeled and unlabeled transcripts of claudin-7 and claudin-14 (GenBank accession nos. AA521724 and AA261472, respectively), the homologous claudin isoforms, were generated similarly. In situ and blot hybridization. In situ hybridization was performed on 5-m sections of unperfused frozen mouse kidney, essentially as described previously (21). In brief, the cryosections were fixed with 4% paraformaldehyde at room temperature for 15 min and acetylated with acetic anhydride. They were then immersed in plastic slide mailers containing 200–350 ng/ml riboprobe in a buffer of 50% formamide, 5⫻ standard sodium citrate (SSC), 2% blocking reagent (Boehringer Mannheim), 0.02% SDS, 0.1% N-laurylsarcosine at pH 7, and hybridized at 68°C for 17 h. The sections were then rinsed in three changes of 2⫻ SSC followed by two highstringency washes in 0.1⫻ SSC for 30 min at 68°C. Sections were visualized using alkaline phosphatase-conjugated antidigoxigenin Fab fragments (Boehringer Mannheim) and 5-bromo-4-chloro-3-indolylphosphate/nitroblue tetrazolium, according to the manufacturer’s instructions. For doublelabeling studies, the sections were washed three times in 1⫻ PBS after color development, then directly incubated with the following tubule-specific markers: FITC-conjugated Lotus tetragonobulus agglutinin (LTA), 10 g/ml (Sigma); polyclonal antibody to the thiazide-sensitive NaCl cotransporter, 1:500, and to the bumetanide-sensitive Na-K-2Cl cotransporter (BSC-1), 1:200 (kind gifts of Dr. Steven C. Hebert); and the E11 monoclonal antibody to the vacuolar H⫹-ATPase, undiluted (kind gift of Dr. Stephen L. Gluck). Antibody labeling was detected with the appropriate FITC-conjugated secondary antibody. For dot-blots, 1-l aliquots containing 0.01–1 ng of unlabeled sense cRNA were spotted directly onto nylon membrane (Duralon, Stratagene) and immobilized by ultraviolet cross-linking. Hybridization and washing of the blots were performed under the same conditions as for in situ hybridization, after which they were processed for detection using a chemiluminescent reagent (CPD-Star, Boehringer Mannheim), according to the manufacturer’s instructions. Tissue culture and transient expression studies. The full coding region of mouse claudin-2, except the translational termination codon, was amplified by PCR, using a sense primer with an upstream KpnI restriction site and Kozak consensus sequence and an antisense primer with a downstream BamHI site, and cloned into the KpnI-BamHI sites of a pcDNA3 “shell” vector with a downstream in-frame FLAG octapeptide epitope tag followed by a stop codon (4) to generate a COOH-terminal epitope-tagged mammalian expression construct. HEK-293 cells, cultured to 70–80% confluence on glass coverslips in Dulbecco’s modified Eagle’s medium with 5% fetal bovine serum at 37°C in 5% CO2, were transiently transfected with the claudin-2 expression construct, or pcDNA3 alone as a vector control, using LipofectAMINE Plus Reagent (Life Technologies). After 24 h, the cells were washed in PBS and fixed in methanol at ⫺80°C for 10 min. F967 F968 CLAUDIN-2 EXPRESSION IN KIDNEY Fig. 1. The claudin-2 (CLDN-2) antisense riboprobe is highly isoform specific. A: ethidium bromide-stained agarose gel demonstrating sense cRNA transcripts generated by in vitro transcription of claudin-2 and its paralogs to be used as isoform-specific control targets. B: autoradiograph of dot-blots spotted with 0.01–1 ng of each sense cRNA control target and hybridized to the indicated digoxigenin-labeled antisense probes. RESULTS To determine the intrarenal localization of claudin-2 mRNA, we first generated an antisense riboprobe to the coding region of the claudin-2 sequence for in situ hybridization. Claudin-2 is a member of a large multigene family. Claudin-7 and -14, two of its closest homologs, share 44 and 54% nucleotide identity within the coding region, respectively. To assess the specificity of our claudin-2 riboprobe, we generated dot-blots spotted with serial dilutions of sense cRNA for claudin-2, -7, and -14 and hybridized them at high stringency to our antisense riboprobes (Fig. 1). The claudin-2 antisense riboprobe was able to detect 0.01 ng of claudin-2 without demonstrating any cross-reaction with up to 1 ng of claudin-7 or -14, indicating at least 100-fold isoform specificity. Using this probe, we performed in situ hybridization of mouse kidney cryosections at high stringency. Labeling with the antisense probe was detected in the majority of tubules in the cortex and outer stripe of outer medulla, and in a few tubules in the inner stripe, but not in the inner medulla (Fig. 2). No labeling was observed in negative control sections that were hybridized with a sense probe. To determine the identity of the claudin-2-expressing tubules, we developed a double-label technique in which sections were first hybridized to the claudin-2 RNA probe, processed for colorimetric detection, and then labeled with a fluorescent lectin conjugate or antibody marker. Claudin-2 mRNA localized to tubules stained with the lectin from LTA (Fig. 3, D and E), a marker found on the apical and, to a lesser extent, basolateral membrane of the proximal tubule (23). Claudin-2 was absent from tubules that stained with antibodies to the apical bumetanide-sensitive Na-K-2Cl cotransporter (Fig. 3, F and G), the thiazide-sensitive NaCl cotransporter (Fig. 3, H and I), and the vacuolar proton pump (Fig. 3, J and K), indicating that it was not expressed in thick ascending limb, distal convoluted tubule, or collecting tubule, respectively. Within the proximal tubule, claudin-2 mRNA was expressed in all of segments S1–S3 and coincided precisely with LTA staining in every tubule. Claudin-2 mRNA labeling was observable but weak in the S1 Fig. 2. In situ hybridization of claudin-2 sense (left) and antisense riboprobes (right) to mouse kidney sections and colorimetric detection. Arrowheads, position of the capsule of the kidney. CX, cortex; OS, outer stripe of outer medulla; IS, inner stripe of outer medulla. Scale bar, 500 m. AJP-Renal Physiol • VOL 281 • NOVEMBER 2001 • www.ajprenal.org Downloaded from ajprenal.physiology.org on March 27, 2012 isolated by affinity purification with glutathione-sepharose. Bacteria transformed with pGEX-4T alone were used to generate control GST protein. In peptide-blocked immunofluorescence studies, GST or GST-CLDN-2 was mixed in a 3:1 molar ratio with the claudin-2 antibody, incubated at room temperature for 30 min, and then applied to kidney sections. CLAUDIN-2 EXPRESSION IN KIDNEY F969 segments, identified as tubules emerging from the urinary pole of Bowman’s capsule and contiguous with the parietal epithelium (Fig. 3A). Claudin-2 labeling was stronger in presumptive S2 convoluted segments in the cortical labyrinth and S3 segments, identified morphologically as straight, axially oriented tubules in the medullary rays of the cortex, and the outer stripe of the outer medulla (Fig. 3B). A subset of tubules in the inner stripe near the border with the outer stripe also demonstrated labeling, suggestive of expression in the early part of the thin descending limb of Henle’s loop (Fig. 3C). In situ hybridization with probes to three other claudin isoforms showed completely different patterns of labeling (Enck AH and and Yu ASL, unpublished observations), suggesting the absence of AJP-Renal Physiol • VOL cross-hybridization to homologous mRNA species at this stringency. Because discrepancies between mRNA and protein localization have been described for other genes, we confirmed our data on claudin-2 expression by immunofluorescent localization of the protein, using a commercially available polyclonal antibody raised against a COOH-terminal peptide. The COOH-terminal region diverges markedly between all known claudin isoforms (maximum of 25% amino acid identity), so antibodies raised against this region are likely to be quite isoformspecific. By immunoblotting, we found that the claudin-2 antibody detected a band of approximately the expected size of 24.4 kDa in mouse kidney cortex membranes (Fig. 4C). To confirm that the antibody was 281 • NOVEMBER 2001 • www.ajprenal.org Downloaded from ajprenal.physiology.org on March 27, 2012 Fig. 3. Nephron segment localization of claudin-2 mRNA. A: claudin-2 mRNA is weakly expressed in the S1 segment of proximal tubule, seen emerging from the urinary pole of a glomerular tuft (G), as well as the contiguous parietal epithelium (arrowhead), in contrast to its strong expression in nearby S2 segments. B: strong expression in long, straight, axially oriented tubules in the medullary rays, representing S3 segments. C: expression in occasional tubules in the inner stripe near the border with the outer stripe, likely representing early thin descending limbs (arrowheads). D–K: paired images of kidney cortex sections double-labeled with claudin-2 cRNA probe, by in situ hybridization, and with a fluorescent nephron segment-specific marker. Claudin-2 is expressed in tubules that costain with Lotus tetragonobulus (LTA), a marker of proximal tubule apical membranes (D and E). It is absent from tubules (asterisks) that stain with the bumetanide-sensitive Na-K-2Cl cotransporter (BSC-1) antibody, a marker of the thick ascending limb (F and G), the thiazide-sensitive NaCl cotransporter (TSC) antibody, a marker of the distal convoluted tubule (H and I), and the H⫹-ATPase antibody, a marker of intercalated cells of the cortical collecting duct (J and K). Scale bars: 200 m (C); in all other panels, 50 m. F970 CLAUDIN-2 EXPRESSION IN KIDNEY indeed recognizing claudin-2, we generated a bacterial fusion protein of GST with the COOH-terminal peptide of claudin-2 (Fig. 4A) and an in vitro translated claudin-2 protein (Fig. 4B). Both were positive by immunoblotting with the claudin-2 antibody, whereas their respective negative controls, GST alone and an in vitro translation product in the absence of the template, were clearly negative (Fig. 4C). Although claudin-7 was also efficiently translated (Fig. 4B), its protein was not stained by the claudin-2 antibody (Fig. 4C), indicating that our antibody is isoform specific. Furthermore, in cultured epithelial cells transiently overexpressing a FLAG epitope-tagged claudin-2 construct, our claudin-2 antibody stained plasma membranes, and particularly the intercellular junctions, in a pattern identical to that observed using an antibody against the FLAG epitope (Fig. 5A). By immunofluorescence of mouse kidney sections using this antibody, moderate levels of claudin-2 staining were found in the majority of tubules throughout the cortex and the outer stripe of outer medulla, strong staining in a subset of tubules in the inner stripe of outer medulla, and absence of staining in the inner medulla (Fig. 6). There was no staining with nonimmune serum (not shown), and staining with the claudin-2 antibody was blocked by preincubation with the fusion protein of GST with its cognate COOH-terminal peptide but not when preincubated with GST alone (Fig. 5B). Claudin-2 protein expression in the cortex and outer stripe colocalized to tubules that stained with LTA, indicating that these are proximal tubules (Fig. 7A). Claudin-2 expression extended distally from the S3 segment of proximal tubules and became markedly more intense in the upper segments of the thin descending limb, where it colocalized to the same tubules that strongly express aquaporin-1 (Figs. 6 and 7C). These tubule segments were distributed diffusely in the region between the vascular AJP-Renal Physiol • VOL bundles, indicating that they are likely to be early segments of the descending limbs of long-looped nephrons (TDL1) (11, 13). Claudin-2 expression ended abruptly near the border between outer and inner medulla and Fig. 5. Specificity of claudin-2 antibody as determined by immunohistochemistry. A: immunofluorescence staining of HEK cells transiently transfected with FLAG epitope-tagged claudin-2 using antibodies to claudin-2 or to the FLAG epitope. A: en face views. Bottom left: vertical section reconstructed from 0.2-m confocal series in the plane (arrow). Indicated are the approximate level of the apical (A) and basolateral (B) membrane and strong staining at the intercellular junction (arrowhead). B: immunofluorescence staining of kidney sections with claudin-2 antibody blocked by preincubation with GST-CLDN-2 fusion protein or with control GST peptide. 281 • NOVEMBER 2001 • www.ajprenal.org Downloaded from ajprenal.physiology.org on March 27, 2012 Fig. 4. Specificity of the claudin-2 antibody as determined by immunoblotting. A: Coomassie blue-stained denaturing polyacrylamide gel of the purified bacterial expression proteins glutathione-S-transferase (GST; 28 kDa) and a fusion of GST and the COOH-terminal peptide of claudin-2 (GST-CLDN2; 31 kDa). B: in vitro translated products from FLAG-tagged claudin-2 (CLDN2; 25.6 kDa) and claudin-7 (CLDN7; 23.5 kDa), luciferase (positive control, 61 kDa), or no template (Neg ctrl) detected by incorporation of biotinylated lysine-tRNA followed by streptavidin blotting. C: Western blot of the indicated in vitro translated proteins (IVT; 1 l/lane), bacterial expression proteins (10 ng/lane) and crude membrane (P47) and soluble (S47) protein fractions of mouse kidney cortex (20 g/lane), using the claudin-2 antibody. Molecular mass (in kDa) of the markers in the ladder are indicated on the left of A–C. Note that in vitro translated claudin-2 is slightly larger than the native protein in mouse kidney cortex (predicted molecular mass, 24.5 kDa), due to the presence of the epitope tag. CLAUDIN-2 EXPRESSION IN KIDNEY F971 Downloaded from ajprenal.physiology.org on March 27, 2012 Fig. 6. Low-power overview of claudin-2 protein expression in mouse kidney. Each row represents 1 of 4 overlapping views, progressing from the capsule (arrow) to the medulla, of a single section double-stained with antibodies to claudin-2 and aquaporin-1 (AQP1). Colocalization is observed in proximal convoluted tubules (PCT) in the cortex (CX), S3 segments of proximal tubule in the outer stripe of outer medulla (OS), and the upper segments of thin descending limbs (TDL1) in the inner stripe of inner medulla (IS), but not in vascular bundles (VB) or in the inner medulla (IM). AJP-Renal Physiol • VOL 281 • NOVEMBER 2001 • www.ajprenal.org F972 CLAUDIN-2 EXPRESSION IN KIDNEY was absent from late thin descending limbs (TDL2) (Figs. 6 and 7D). Within the tubules, claudin-2 was clearly localized to the lateral intercellular membrane (Fig. 7, A and B). In both proximal tubules and thin descending limbs, it appeared to colocalize with the ubiquitously expressed tight junction protein ZO-1 (Fig. 8), indicating that claudin-2 is located at or very close to the zona occludens. The presence of tubules (presumably distal in origin) that were positive for ZO-1 but completely negative for apparent claudin-2 staining confirmed the absence of artifacts due to antibody cross-reactivity or bleed-through of fluorescence. In proximal tubules, moderate intracellular claudin-2 staining was also observed. DISCUSSION We have used in situ hybridization and immunohistochemistry to localize expression of a single claudin isoform, claudin-2, within the mouse kidney. Because of the high degree of homology between claudin family members, it was important to ensure that our techniques were sufficiently specific to detect a single isoform. Our results AJP-Renal Physiol • VOL by in situ hybridization are likely to be specific because we used the highest possible stringency (hybridization at 68°C in the presence of 50% formamide and washes at 68°C in 0.1⫻ SSC) and also demonstrated that the claudin-2 antisense riboprobe showed no cross-reaction to its two closest homologs, claudin-14 and -7, by dot-blot hybridization under the same conditions. Our results by immunohistochemistry are likely to be specific because we used an antibody directed against the COOH terminal of claudin-2, which is poorly conserved between homologous family members. We confirmed that the antibody recognizes claudin-2 by immunoblot of mouse kidney cortex membranes, in vitro translated protein, and a bacterially expressed GST fusion protein with the claudin-2 COOH-terminal peptide, and by immunofluorescence staining of cultured cells transfected with claudin-2. Specific immunorecognition of claudin-2 was absent with nonimmune serum and was blocked specifically by the GST-CLDN2 fusion protein. Finally, we demonstrated that the antibody was isoform specific because it did not recognize in vitro translated protein of the closely homologous isoform, claudin-7. 281 • NOVEMBER 2001 • www.ajprenal.org Downloaded from ajprenal.physiology.org on March 27, 2012 Fig. 7. Nephron segment localization of claudin-2 protein by double-labeling studies. A: claudin-2 colocalizes to the lateral membrane (L) of proximal tubules, identified by costaining of their apical (A) and basolateral (B) membranes with the lectin LTA. B: propidium iodide highlights the cell nuclei, showing that claudin-2 staining is intercellular. C: claudin-2 staining is stronger in the early thin descending limb (TDL1) than in the proximal tubule. Arrowhead, transition between S3 segment and thin descending limb, which also coincides with increased expression of AQP1. D: claudin-2 expression is absent from the late thin descending limb (TDL2). Arrowhead, transition between the more proximal segment of the thin descending limb that expresses both claudin-2 and AQP1 and the more distal segment that expresses only AQP1. VR, vasa recta. CLAUDIN-2 EXPRESSION IN KIDNEY F973 The results we obtained by in situ hybridization and by immunohistochemistry were very similar and therefore nicely corroborate each other. We found that claudin-2 is expressed throughout the proximal tubule and in upper segments of the thin descending limb near the border of the inner and outer stripe of outer medulla. We believe that the latter areas represent the early segments of thin descending limbs of long-looped nephrons because they also stain very strongly for aquaporin-1, which is reported to be predominantly expressed in long loops (20), and because they are distributed diffusely in the interbundle region, whereas the descending limbs of shortlooped nephrons are known to be closely associated with the vascular bundles in the mouse (11, 13). Both techniques demonstrated axial variation in claudin-2 expression, increasing from proximal to distal portions of this part of the tubule. By immunohistochemistry, claudin-2 was localized to the tight junction, as expected. A modest amount of staining was also found intracellularly in the proximal tubule and may represent an immature pool in the synthetic pathway, a reserve pool to be recruited to the tight junction only when needed, or simply crossreactivity of the antibody to another intracellular epitope. An important role for the paracellular pathway in passive NaCl reabsorption by the proximal nephron was proposed 35 years ago by Rector and colleagues (22). In the early proximal tubule, Na⫹ is reabsorbed transcellularly, primarily with HCO3⫺; isosmotic water reabsorpAJP-Renal Physiol • VOL tion follows, likely via the transcellular route through aquaporin-1 water channels (26). Na⫹-coupled reabsorption of phosphate, glucose, and amino acids and secretion of organic acids and bases also occur. Thus the tubular fluid-to-plasma concentration ratio in the late proximal tubule is high (⬃1.4) for Cl⫺ and low (⬃0.4) for HCO3⫺. The paracellular pathway is leaky and exhibits a preferential permeability for Cl⫺ over HCO3⫺ (10), thus favoring paracellular reabsorption of Cl⫺ and the development of a lumen-positive electrical potential. This in turn drives further reabsorption of Na⫹, passively via the paracellular route, and therefore further osmotic water reabsorption (19). Thus approximately one-third of all Na⫹ and water reabsorption in the proximal tubule is dependent on a leaky but anion-selective paracellular pathway. At the same time, significant transtubular gradients for glucose, amino acids, and organic anions and cations accumulate by the end of the proximal tubule, and so the paracellular pathway must maintain a tight barrier against their backleak. By contrast, the thin descending limb is widely thought to be a segment that lacks active, transcellular NaCl reabsorption and is impermeable to passive, paracellular NaCl transport but highly permeable to water, thus allowing it to play a key role in the generation of the medullary concentrating gradient according to the Kokko-Rector-Stephenson model (16). However, it is in fact a heterogeneous nephron segment in 281 • NOVEMBER 2001 • www.ajprenal.org Downloaded from ajprenal.physiology.org on March 27, 2012 Fig. 8. Subcellular colocalization of claudin-2 with ZO-1. Arrowhead, intercellular junction of a proximal tubule (top right) and an early thin descending limb (bottom); arrow, adjacent tubules that are positive for ZO-1 but not for claudin-2. F974 CLAUDIN-2 EXPRESSION IN KIDNEY We thank Dr. Dennis Brown for helpful discussions. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. REFERENCES 1. Amemiya M, Loffing J, Lotscher M, Kaissling B, Alpern RJ, and Moe OW. Expression of NHE-3 in the apical membrane of rat renal proximal tubule and thick ascending limb. Kidney Int 48: 1206–1512, 1995. 2. Dieterich HJ, Barrett JM, Kriz W, and Bulhoff JP. The ultrastructure of the thin loop limbs of the mouse kidney. Anat Embryol (Berl) 147: 1–18, 1975. 3. Dobyan DC, Magill LS, Friedman PA, Hebert SC, and Bulger RE. Carbonic anhydrase histochemistry in rabbit and mouse kidneys. Anat Rec 204: 185–197, 1982. 4. Dowland LK, Luyckx VA, Enck AH, Leclercq B, and Yu ASL. Molecular cloning and characterization of an intracellular chloride channel in the proximal tubule cell line, LLC-PK1. J Biol Chem 275: 37765–37773, 2000. 5. Ernst SA and Schreiber JH. Ultrastructural localization of Na⫹,K⫹-ATPase in rat and rabbit kidney medulla. J Cell Biol 91: 803–813, 1981. 6. Fanning AS, Mitic LL, and Anderson JM. Transmembrane proteins in the tight junction barrier. J Am Soc Nephrol 10: 1337–1345, 1999. 7. Furuse M, Fujita K, Hiiragi T, Fujimoto K, and Tsukita S. Claudin-1 and -2: novel integral membrane proteins localizing at AJP-Renal Physiol • VOL 23. 24. 25. 26. 27. 28. 29. tight junctions with no sequence similarity to occludin. J Cell Biol 141: 1539–1550, 1998. Furuse M, Furuse K, Sasaki H, and Tsukita S. Conversion of zonulae occludentes from tight to leaky strand type by introducing claudin-2 into Madin-Darby canine kidney I cells. J Cell Biol 153: 263–272, 2001. Furuse M, Sasaki H, and Tsukita S. Manner of interaction of heterogeneous claudin species within and between tight junction strands. J Cell Biol 147: 891–903, 1999. Green R and Giebisch G. Reflection coefficients and water permeability in rat proximal tubule. Am J Physiol Renal Fluid Electrolyte Physiol 257: F658–F668, 1989. Imai M and Yoshitomi K. Heterogeneity of the descending thin limb of Henle’s loop. Kidney Int 38: 687–694, 1990. Inai T, Kobayashi J, and Shibata Y. Claudin-1 contributes to the epithelial barrier function in MDCK cells. Eur J Cell Biol 78: 849–855, 1999. Kriz W and Koepsell H. The structural organization of the mouse kidney. Z Anat Entwicklungsgesch 144: 137–163, 1974. Kurtz I. Apical and basolateral Na⫹/H⫹ exchange in the rabbit outer medullary thin descending limb of Henle: role in intracellular pH regulation. J Membr Biol 106: 253–260, 1988. Luyckx VA, Goda FO, Mount DB, Nishio T, Hall A, Hebert SC, Hammond TG, and Yu ASL. Intrarenal and subcellular localization of rat CLC5. Am J Physiol Renal Physiol 275: F761– F769, 1998. Masilamani S, Knepper MA, and Burg MB. Urine concentration and dilution. In: Brenner and Rector’s The Kidney (6th ed.), edited by Brenner BM. Philadelphia, PA: Saunders, 2000, p. 595–635. Moe OW, Berry CA, and Rector FC Jr. Renal transport of glucose, amino acids, sodium, chloride, and water. In: Brenner and Rector’s The Kidney (6th ed.), edited by Brenner BM. Philadelphia, PA: Saunders, 2000, p. 375–416. Morita K, Furuse M, Fujimoto K, and Tsukita S. Claudin multigene family encoding four-transmembrane domain protein components of tight junction strands. Proc Natl Acad Sci USA 96: 511–516, 1999. Neumann KH and Rector FC Jr. Mechanism of NaCl and water reabsorption in the proximal convoluted tubule of rat kidney. J Clin Invest 58: 1110–1111, 1976. Nielsen S, Pallone T, Smith BL, Christensen EI, Agre P, and Maunsbach AB. Aquaporin-1 water channels in short and long loop descending thin limbs and in descending vasa recta in rat kidney. Am J Physiol Renal Fluid Electrolyte Physiol 268: F1023–F1037, 1995. Peng JB, Chen XZ, Berger UV, Vassilev PM, Brown EM, and Hediger MA. A rat kidney-specific calcium transporter in the distal nephron. J Biol Chem 275: 28186–28194, 2000. Rector, FC Jr, Martinez-Maldonado M, Brunner FP, and Seldin DW. Evidence for passive reabsorption of NaCl in proximal tubule of rat kidney. J Clin Invest 45: 1060–1070, 1966. Schulte BA and Spicer SS. Histochemical evaluation of mouse and rat kidneys with lectin-horseradish peroxidase conjugates. Am J Anat 168: 345–362, 1983. Simon DB, Lu Y, Choate KA, Velazquez H, Al-Sabban E, Praga M, Casari G, Bettinelli A, Colussi G, RodriguezSoriano J, McCredie D, Milford D, Sanjad S, and Lifton RP. Paracellin-1, a renal tight junction protein required for paracellular Mg2⫹ resorption. Science 285: 103–106, 1999. Tsukita S and Furuse M. Occludin and claudins in tightjunction strands: leading or supporting players? Trends Cell Biol 9: 268–273, 1999. Vallon V, Verkman AS, and Schnermann J. Luminal hypotonicity in proximal tubules of aquaporin-1-knockout mice. Am J Physiol Renal Physiol 278: F1030–F1043, 2000. Wong V and Goodenough DA. Paracellular channels (Abstract). Science 285: 62, 1999. Yoshitomi K and Imai M. Electrophysiological characterization of upper portion of descending limb of long-looped nephron. Am J Physiol Renal Fluid Electrolyte Physiol 260: F311–F316, 1991. Yu AS. Paracellular solute transport: more than just a leak? Curr Opin Nephrol Hypertens 9: 513–515, 2000. 281 • NOVEMBER 2001 • www.ajprenal.org Downloaded from ajprenal.physiology.org on March 27, 2012 all species except the rabbit (11) and is composed of three ultrastructurally distinct cell types, type I in descending limbs of short-looped nephrons (SDL), type II in the early segments of the descending limb of long-looped nephrons largely in the inner stripe of outer medulla (TDL1), and type III in the late segments of the descending limb of long-looped nephrons (TDL2) confined to the inner medulla (2). Type II cells are quite different from types I and III and more closely resemble proximal tubule cells in that they are tall and have well-developed apical microvilli, abundant mitochondria, and a single intercellular tight junctional strand, all suggestive of a cell that is active in transcellular solute and water transport, while having a leaky paracellular pathway. Consistent with this, the TDL1 is unique among thin descending limb segments in expressing basolateral Na-K-ATPase (5, 28), luminal carbonic anhydrase (3), and both functional (14) and molecular (1) evidence of apical Na⫹-H⫹ exchange. Furthermore, the TDL1 has a very low transepithelial resistance (28) and a 10-fold higher passive Na⫹ permeability than either TDL2 or SDL (11). It is therefore likely that the TDL1, very much like the late proximal tubule, mediates both active transcellular and passive paracellular salt reabsorption concomitant with osmotic water reabsorption. Interestingly, claudin-2 has been found to be expressed in a strain of the canine kidney cell line, Madin-Darby canine kidney (MDCK), that has a low transepithelial resistance but is absent from the highresistance strain of MDCK. Transfection of claudin-2 into the high-resistance cells induced a dramatic reduction in transepithelial resistance (8), suggesting the intriguing possibility that claudin-2 may be able to reconstitute a relatively leaky paracellular pore. We therefore speculate that the role of claudin-2 in proximal tubule and the TDL1 may be to create a lowresistance paracellular shunt that is critical for passive NaCl reabsorption in these segments.
© Copyright 2024 Paperzz