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.