ClC-5 is an integral plasma membrane protein that belongs to the CLC family of voltage-gated chloride channels1. ClC-5 is encoded by the CLCN5 gene, which is located on Xp11.22 and is predominantly expressed in the kidney2. Mutations in CLCN5 are associated with Dent's disease, an X-linked renal tubular disorder characterized by low- molecular-weight proteinuria (LMWP) and renal Fanconi syndrome, associated with hypercalciuria, nephrocalcinosis, and nephrolithiasis3,4. Studies in heterologous expression systems have shown that these mutations abolish or markedly reduce the outwardly rectifying chloride currents generated by ClC-53.
The segmental expression of ClC-5 in the mature human and rat kidney includes the proximal tubule (PT), the thick ascending limb (TAL) of Henle's loop, and the 
-type intercalated cells (IC) of the collecting duct (CD)5,6,7. The codistribution of ClC-5 with the vacuolar H+-ATPase in endosomes suggests that both proteins could interact to regulate endosomal acidification5,7 and apical targeting8 in PT cells. The interaction between ClC-5 and H+-ATPase may also be important for the urinary acidification mediated by -type IC5,6,7. These hypotheses are supported by the major defect in receptor-mediated endocytosis associated with PT dysfunction reported in ClC-5–deficient mice9,10, as well as by the alterations in polarity and expression of H+-ATPase observed in PT cells and IC from patients with Dent's disease8.
The clinical manifestations of Dent's disease, including hematuria, proteinuria, mild polyuria, and nephrolithiasis may often occur during childhood, and LMWP is a consistent feature of the disease4. In particular, the early onset of some cases11,12, in which tubular proteinuria was discovered in the first month of life, suggests that segmental expression of ClC-5 in PT must be acquired before birth. Developmental studies in rats have shown that assembly of brush border components coincides with the onset of glomerular filtration13, but the status of other components of the endocytic pathway in PT cells, such as ClC-5 and H+-ATPase, is not known. Similarly, studies in rat kidney have shown that IC appear during late nephrogenesis and undergo postnatal maturation14. However, unlike rats, nephrogenesis in mouse and humans is essentially achieved at birth, and the pattern of appearance and differentiation of IC in these two species remains unclear.
In this study, we have investigated the renal ontogeny, processing, and distribution of ClC-5 during mouse and human nephrogenesis, in parallel with that of the vacuolar H+-ATPase, which provides the driving force for endosomal acidification in PT cells15, type II carbonic anhydrase (CAII), which is essential for Na+ reabsorption and urinary acidification in PT cells16, and the water channel aquaporin-1 (AQP1) as an ontogeny marker for PT cells17. These studies provide new insights into the maturation of PT and IC during nephrogenesis and help to understand the role of ClC-5 and the early phenotypic variants of Dent's disease.
METHODS
Human and mouse kidney samples
Human fetal kidneys from therapeutic abortion were procured by the International Institute for Advancement of Medicine (Philadelphia, PA, USA), the Anatomic Gift Foundation (Woodbine, GA,USA), and the Necker Hospital (Paris, France). Eighteen human fetal kidneys, ranging from 12 to 25 weeks of gestation (GW), were used for immunoblotting and immunostaining studies. For comparison and additional studies, we also used 3 newborn or infant human kidneys ranging in age from 4 months to 2 years. These normal kidneys were obtained directly at surgery and perfused with ice-cold neutral buffered salt solutions before processing for fixation and/or protein extraction7,17. The use of these samples conformed to local ethical guidelines and were approved by the University of Louvain Ethical Review Board.
Mouse kidney samples were obtained from CD-1 mice (Iffa Credo, Brussels, Belgium). Pregnant mice were sacrified by cervical dislocation, and embryos were dissected under a binocular to isolate kidneys. An average of 12 embryos from 4 different litters were collected every day from embryonic day 13.5 (E13.5) until day 5 after birth (newborn). Comparative studies were performed with 4 adult male kidneys. Kidney samples from wild-type versus ClC-5 knockout (KO) mice10 and adult male Wistar rats were also used.
Antibodies
Affinity-purified polyclonal antibodies (SB499) directed against the N-terminus of human ClC-5 have been characterized previously10. Other antibodies included a monoclonal antibody against the 31 kD E subunit (V1 domain) of the vacuolar H+-ATPase (a gift of Dr. S. Gluck, Washington University, St. Louis, MO, USA), a rabbit polyclonal antibody against the 116 kD a4 subunit (V0 domain) of the vacuolar H+-ATPase18, a monoclonal antibody against 
-actin (Sigma, St. Louis, MO, USA), a rabbit polyclonal antibody against AQP1 (Chemicon, Temecula, CA, USA), and sheep polyclonal antibodies against CAII (Serotec, Oxford, UK) and Tamm-Horsfall protein (Biodesign International, Kennebunk, ME, USA).
Immunoblot analyses and deglycosylation studies
Membrane extracts were prepared as previously described7,10. Briefly, kidney samples were homogenized in ice-cold buffer (300 mmol/L sucrose and 25 mmol/L N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid made to pH 7.0 with 1 mol/L tris (hydroxymethyl)aminomethane [Tris]) containing Complete™ protease inhibitors (Roche, Vilvoorde, Belgium). The homogenate was centrifuged at 1000g for 15 minutes at 4°C, and the resulting supernatant was centrifuged at 100,000g for 120 minutes at 4°C. The pellet ("membrane" fraction) was suspended in ice-cold homogenization buffer before determination of protein concentration and storage at –80°C.
Deglycosylation studies were performed on human and mouse kidney extracts and control glycoproteins (Roche). The samples (15 
g total protein) were incubated for 60 to 90 minutes at 37°C with 12 units of N-glycosidase F, as recommended by the manufacturer.
Immunoblotting was performed as previously described7,10. After blocking, membranes were incubated overnight at 4°C with primary antibodies, washed, incubated for 1 hour at room temperature with appropriate peroxidase-labeled antibodies (Dako, Glostrup, Denmark), washed again, and visualized with enhanced chemiluminescence. Normalization for -actin was obtained after stripping the blots and reprobing with the anti--actin antibody. Specificity of the immunoblot was determined by incubation with: (1) preimmune rabbit serum; (2) nonimmune rabbit or mouse immunoglobulin (Ig)G (Vector Laboratories, Burlingame, CA, USA) or control ascites fluid (Sigma); and (3) anti-ClC5 antibody preadsorbed with a 5-fold (w/w) excess of immunogenic peptide. Determination of the molecular mass was obtained by comparison with different precision standards (MBI Fermentas, Vilnius, Lithuana; Bio-Rad, Hercules, CA, USA). Densitometry analysis of the specific bands was performed with a Hewlett Packard Scanjet model IVC using the National Institutes of Health (NIH) Image V1.60 software, and optical densities normalized to -actin density in the corresponding sample. All immunoblots were at least performed in duplicate.
RT-PCR and semiquantitative RT-PCR
Mouse kidney samples were homogenized in Trizol (Invitrogen, Merelbeke, Belgium) in order to extract total RNA. Total RNA samples (2.7 g) were treated with DNase I (Invitrogen) and reverse-transcribed into cDNA using SuperScript II Rnase H Reverse Transcriptase (Invitrogen). The following primers were designed using Beacon Designer 2.0 (Premier Biosoft International, Palo Alto, CA, USA): ClC-5 (exon 6) sense 5'AAGTGGACCCTTGTCATCAA 3' and antisense 5'ACAAGATGTTCCCACAGCAG 3', AQP1 (exons 1–2) sense 5' GCTGTCATGTATATCATCGCCCAG 3' and antisense 5' AGGTCATTTCGGCCAAGTGAGT 3', GAPDH (exon 1) sense 5' TGCACCACCAACTGCTTAGC 3' and antisense 5' GGATGCAGGGATGATGTTCT 3'. The predicted lengths of the resulting PCR fragments were 115 bp (ClC-5), 107 bp (AQP1), and 176 bp (GAPDH). RT-PCR reactions were performed with 200 nmol/L of both sense and antisense primers in a final volume of 25 
L using 1 unit of Platinum Taq DNA polymerase, 2 mmol/L MgCl2, 400 mol/L dNTP. PCR conditions were 94°C for 5 minutes followed by 35 (ClC-5), 36 (AQP1), and 30 (GAPDH) cycles of 30 seconds at 95°C, 30 seconds at 61°C, and 1 minute at 72°C. The PCR products were size-fractionated on 1.5% agarose gel and then stained with ethidium bromide. Negative controls confirmed that the samples were not cross-contaminated (sterile water instead of RNA) or polluted by genomic DNA (not reverse-transcribed RNA samples). The expression of ClC-5 was also verified using another set of primers (exons 8–9), sense 5' TCCGCACAAACATTGCCTG 3' and antisense 5' AGGCAGCAGCCCCAACCAT 3' (predicted length of the fragment: 519 bp).
Real-time PCR analyses were performed in duplicate with 200 nmol/L of both sense and anti-sense primers in a final volume of 25 L using 1 U of Platinum Taq DNA Polymerase, 2 mmol/L MgSO4, 400 mol/L dNTP, and SYBR Green I (Molecular Probe, Leiden, The Netherlands) diluted to 1/105. PCR mixture contained 10 nmol/L fluorescein for initial well to well fluorescence normalization. PCR conditions were settled as incubation at 94°C for 3 minutes followed by 40 cycles of 30 seconds at 95°C, 30 seconds at 61°C, and 1 minute at 72°C. The melting temperature of PCR product was checked at the end of each PCR by recording SYBR green fluorescence increase upon slowly renaturating DNA. For each assay, standard curves were prepared by serial 4-fold dilutions of mouse adult kidney cDNA. The mRNA levels of ClC-5 and AQP1 were adjusted to GAPDH mRNA level at each stage, and relative changes in mRNA levels during ontogeny were determined by comparison to the adult mRNA level using the following formula: Ratio = (Etarget)
Ct(Adult-Sample)/(EGAPDH)Ct (Adult-Sample)19.
Immunostaining
Kidney samples were fixed in 4% paraformaldehyde (Boehringer Ingelheim, Heidelberg, Germany) in 0.1 mol/L phosphate buffer, pH 7.4, prior to embedding in paraffin as described7,10,17. Six m sections were incubated for 30 minutes with 0.3% hydrogen peroxide to block endogenous peroxidase. For ClC-5 and H+-ATPase immunostaining, antigen retrieval was performed by incubating sections in 0.01 mol/L citrate buffer, pH 5.8, for 75 minutes, in a water bath heated at 97°C, before cooling down and rinsing. Following incubation with 10% normal serum for 20 minutes, sections were incubated for 45 minutes with the primary antibodies diluted in phosphate-buffered saline (PBS) containing 2% bovine serum albumin (BSA). After washing, sections were successively incubated with biotinylated secondary anti-immunoglobulin (Ig)G antibodies, avidin-biotin peroxidase, and aminoethylcarbazole (Vectastain Elite, Vector Laboratories). Sections were viewed under a Leica DMR coupled to a Leica DC300 digital camera (Leica, Heerbrugg, Switzerland). The specificity of immunostaining was tested by incubation: (1) in absence of primary antiserum; (2) with preimmune rabbit serum; (3) with non-immune rabbit serum or control rabbit or mouse IgG (Vector Laboratories); and (4) with the anti-ClC-5 antibody preadsorbed with the immunogenic peptide. The intensity of immunostaining was graded by an observator unaware of the embryonic stage and the antibody used, in comparison with the intensity observed in adult samples.
RESULTS
Comparative ontogeny of ClC-5 in the mouse kidney
ClC-5 was detected as a broad band at 
80 kD in the adult mouse kidney Figure 1a. The band was not detected in kidney extracts from Clcn5 KO mouse (left panel), or when the blot was incubated with anti-ClC-5 antibodies preadsorbed with the immunogenic peptide (right panel). The signal for ClC-5 observed in embryonic samples was different from the adult (right panel); it included two bands at 90 kD and 75 kD, respectively, separated by a smear of diffuse immunoreactive bands. With the exception of the upper band at 90 kD, all the immunoreactive bands were abolished when using preadsorbed antibodies (right panel).
Figure 1.
Comparative ontogeny of ClC-5 in mouse kidney. (A) Characterization of the SB499 anti-ClC-5 antibodies. Left: Kidney extracts (30 g/lane) from wild-type (WT) and knockout (KO) ClCn5 mice were run on 7.5% PAGE and transferred to nitrocellulose. The blot was probed with SB499 anti-ClC-5 antibodies (1:1000). The diffuse band corresponding to ClC-5 at 80 kD is absent in KO mice. The membranes
The expression of ClC-5 in the developing kidney was already detected at E13.5 Figure 1b. At this early stage, the immunoreactive pattern was characterized by the nonspecific upper band at 90 kD, and the diffuse specific bands including the 75 kD band. The signal intensity for ClC-5 peaked early (E14.5), and remained quite stable to E16.5. From E16.5 to the newborn stage, the immunoreactivity pattern for ClC-5 showed a progressive decrease in the higher molecular mass isoforms and apparition of a more focused signal around 75 to 80 kD. Adult samples were characterized by concentrated immunoreactive bands around 80 kD, as described above.
The expression of AQP1 in the kidney is characterized by two isoforms—the core AQP1 (28 kD) and the glycosylated AQP1 (GlyAQP1, 35 to 50 kD). The core AQP1 was detected at E15.5 and it gradually increased during late ontogeny. The GlyAQP1 was detected only at E18.5 and in the newborn kidney, probably reflecting the increased amount of AQP117. The 31 kD band corresponding to the E1 subunit of the H+-ATPase was detected at low level from E13.5 and showed a gradual increase from E16.5 to birth. A progressive induction of the 116 kD a4 subunit of the H+-ATPase from E16.5, in parallel with the E subunit, was observed in developing mouse kidney (data not shown). The CAII band at 30 kD was detected at E14.5, remained stable during all ontogeny, and showed a dramatic increase postnatally Figure 1b. Densitometry analysis Figure 1c confirmed the different expression patterns of ClC-5, which peaked at E14.5 (170% of adult level), remained high (140% to 150% of adult level) during ontogeny, and decreased after birth, AQP1 and H+-ATPase, which showed a gradual increase from E16.5, and CAII, with low expression levels (15% – 30% of adult level) during all ontogeny and a significant postnatal increase.
Quantification of ClC-5 mRNA expression in the developing mouse kidney
Semiquantitative Figure 2a and real-time PCR analyses Figure 2b confirmed the early induction of ClC-5 at E13.5, with a slight decrease of expression during late ontogeny, and a dramatic 2-fold increase in newborns. This expression pattern contrasted with that of the water channel AQP1, which showed a later induction at the end of nephrogenesis to nearly reach the adult level at birth.
Figure 2.
Differential expression of ClC-5 and AQP1 in the developing mouse kidney: semi-quantitative and real-time PCR analysis. (A) Representative semiquantitative RT-PCR analysis of the expression of ClC-5, AQP1, and GAPDH during embryogenesis (E13.5, E15.5, E17.5) and in newborn (NB) and adult (A) mouse kidneys. Negative controls with water (W) and in absence of RT product (N) are shown. (B) Quantitative RT-PCR of ClC-5 and AQP1 mRNA expression during mouse nephrogenesis. The primers for ClC-5, AQP1, and GAPDH were similar to those used above. The ClC-5 and AQP1 mRNA levels were adjusted to GAPDH for each stage. The relative changes in expression during ontogeny were determined by comparison to the adult mRNA level taken as 100%. ClC-5 mRNA is significantly expressed at E13.5 (70% of the adult value), with a subsequent decrease until E17.5 (30%) and a major 2-fold upregulation at birth. This expression pattern contrasts with the low level of AQP1 at E13.5 (4% of the adult value) and its gradual increase from E15.5 (6%) to nearly reach the adult level at birth (73%).
Full figure and legend (25K)Processing of ClC-5 during mouse nephrogenesis: Deglycosylation studies
Sequence analysis predicted that ClC-5 is a glycoprotein with N-glycans linked to Asn at position 4083. Deglycosylation studies with N-glycosidase F were performed in control glycoproteins and developing and adult mouse kidney samples Figure 3. Digestion of transferrin (65 to 60 kD), 1-acid glycoprotein (45 to 22 kD), and ribonuclease B (17 to 15 kD) demonstrated the efficiency of deglycosylation (data not shown). Immunoblot analysis showed that untreated embryonic samples Figure 3a are characterized by diffuse immunoreactive ClC-5 bands (asterisk) located below the nonspecific 90 kD band and a specific, core band at 75 kD (arrow). The latter is above a nonspecific immunoreactive band at 60 kD (arrowhead), which results from incubation in reaction buffer. Untreated adult samples Figure 3b showed the mature ClC-5 band at 80 kD (asterisk) in addition to the nonspecific band at 60 kD (arrowhead), but not the 75 kD core band (arrow). Treatment with N-glycosidase F yielded a shift of the specific ClC-5–positive bands to a lower apparent molecular weight, indicating the existence of Asn-linked glycan chains in both embryonic and adult samples. To note, the molecular weight of the deglycosylated form of ClC-5 was slightly higher in developing than in mature kidney samples.
Figure 3.
Processing and glycosylation of ClC-5 in mouse kidney. The samples [(A) E15.5; (B) adult] were incubated with the reaction buffer containing (+) or not (-) N-glycosidase F, separated (7.5 g protein/lane) on 7.5% PAGE and probed with anti-ClC5 antibodies alone (SB499) or preadsorbed (PreAd) (both diluted 1:1000). Untreated embryonic samples (A) are characterized by diffuse immunoreactive ClC-5 bands (asterisk), located below the nonspecific 90 kD band, and by a 75 kD band (arrow). Incubation in the reaction buffer yielded a nonspecific immunoreactive band at 60 kD (arrowhead) in all samples. Untreated adult samples (B) do not show the 75 kD band (arrow), but are characterized by a focused band at 80 kD (asterisk), in addition to the nonspecific band at 60 kD (arrowhead). The shift of ClC-5-positive bands to a lower molecular weight after N-glycosidase F treatment indicates the existence of Asn-linked glycan chains in both embryonic and adult samples. It must be noted that embryonic samples differ from adult by the existence of a 75 kD, PGNase F-insensitive band, as well as by the higher molecular mass of the PGNase F-sensitive proteins in control and treated conditions.
Ontogeny and glycosylation of ClC-5 in the human kidney
The broad immunoreactive band corresponding to ClC-5 was detected by immunoblotting in the developing human kidney at 12GW Figure 4a. The intensity of ClC-5 expression was sustained during all nephrogenesis, similar to that observed in the infant kidney. Treatment with N-glycosidase F induced a significant shift of ClC-5–positive bands to a lower molecular weight Figure 4b, indicating the existence of Asn-linked glycan chains in developing and mature human kidney samples. As described in mouse, the molecular weight of the deglycosylated form of ClC-5 was higher in fetal than in postnatal samples.
Figure 4.
Ontogeny and glycosylation of ClC-5 in the human kidney. (A) Representative immunoblot for ClC-5 in membrane extracts from rat kidney (positive control, RKC), human fetal kidney (HFK) at 12, 13, 18, and 25 GW, newborn (HNK) and infant (HIK) human kidney. Thirty g of protein were loaded in each lane, and the blot was probed with the anti-ClC-5 antibodies alone (SB499) or preadsorbed (PreAd) (both diluted 1:1000). The broad band corresponding to ClC-5 is detected at 12 GW, and its expression is sustained during nephrogenesis. The ClC-5 signal is abolished when incubation is performed with preadsorbed antibodies. Discrete bands below 60 kD are also detected with nonimmune IgG. (B) Deglycosylation with N-glycosidase F in human kidney samples (25 GW, HFK25; infant, HIK). The samples were incubated with the reaction buffer containing (+) or not (-) N-glycosidase F, separated (7.5 g protein/lane) on 7.5% PAGE and probed with anti-ClC-5 antibodies (SB 499 lanes). The shift of ClC-5 bands to a lower molecular weight confirms the existence of Asn-linked glycan chains. As in mouse, the molecular mass of the PGNase F-sensitive ClC-5 proteins is higher in fetal than in postnatal samples. The pattern observed in HIK samples (more diffuse immunoreactive bands including lower mass isoforms) suggests that they may include residual embryonic isoforms. The signal is abolished when a similar blot is incubated with preadsorbed anti-ClC-5 antibodies (PreAd).
Segmental distribution of ClC-5 during mouse nephrogenesis
ClC-5 was detected in the developing mouse kidney at E13.5 Figure 5a. The signal was located in branching ureteric buds within the inner part of the primitive cortex, whereas the undifferentiated metanephric cap tissue, the condensates, and the inner mesenchyme, were negative. Staining intensity for ClC-5 markedly increased at E14.5, by which dispersed primitive glomeruli were identified Figure 5b. The staining was stronger in the ureteric buds and also appeared in developing PT. Glomeruli, as well as comma- and S-shaped bodies, were all negative. From E15.5 Figure 5c to E18.5 Figure 5d, ClC-5 staining persisted in ureteric buds, became more apparent in developing PT, and gradually appeared in the IC.
Figure 5.
Immunodetection of ClC-5 in the developing mouse kidney. (A) The staining for ClC-5 is first detected in the ureteric buds (arrowheads) located in the inner part of the primitive cortex at E13.5. (B) The staining intensity markedly increases at E14.5, with ClC-5 being located primarily in the ureteric buds (arrowheads) and developing PT (p). (C) ClC-5 staining at E15.5 and (D) E18.5 is observed in ureteric buds (arrowheads), and its intensity increases in developing PT (p); IC are detected from E15.5 (asterisks). Glomeruli, condensates, comma- and S-shaped bodies are negative (Original magnification A-D, 
150).
At higher magnification Figure 6, staining for ClC-5 within the primitive cortex of E14.5 kidneys was identified in the apical area of the tips of ureteric buds Figure 6a. A similar staining was observed within branching ureteric buds in the outer part of the primitive medulla Figure 6b, whereas an apical staining restricted to a subpopulation of ureteric buds cells was observed in the inner medulla Figure 6c. At E15.5, ClC-5 was identified in PT originating from recently formed glomeruli, and scattered CD cells (Figure 6d and inset). The latter were also stained for apical H+-ATPase, indicating that they correspond to -type IC (see below). At E16.5, ClC-5 showed an homogeneous, subapical staining in cells lining developing PT, and a strong apical staining in IC Figure 6e. To note, codistribution of ClC-5 Figure 6f and Tamm-Horsfall protein Figure 6g was observed in some juxtaglomerular tubules at E16.5 and later. Distribution of ClC-5 in E18.5 kidneys included the apical area of PT cells and IC of the cortical CD Figure 6h, exactly like in newborn mouse Figure 6i. Codistribution of ClC-5 Figure 6j and CAII Figure 6k was observed in postnatal IC. The specificity of ClC-5 staining was demonstrated with preadsorbed SB499 antibodies (data not shown). In comparison with ClC-5, CAII was first identified in IC at E15.5 and PT cells at E16.5, whereas AQP1 was detected in PT cells from E16.5 (data not shown). The distribution of ClC-5 and tubular markers during mouse nephrogenesis is summarized in Table 1.
Figure 6.
Segmental distribution of ClC-5 during mouse nephrogenesis. (A, B, C, D, E, F, H, I, J) Immunostaining for ClC-5, (G) Tamm-Horsfall protein, and (K) CAII in the developing mouse kidney at (A–C) E14.5, (D) E15.5, (E–G) E16.5, (H) E18.5 (H), (I–K) and newborn kidney. At E14.5, staining for ClC-5 is restricted to the apical area of the (A, B) ureteric buds, and in a subpopulation of ureteric buds cells in the (C) inner medulla. At E15.5, ClC-5 is located in developing PT and scattered cells within developing CD in the (D) outer cortex. At (E) E16.5, linear apical ClC-5 staining is detected in PT cells, paralleled by a strong apical staining in IC within CD. Staining on serial sections from E16.5 kidneys shows codistribution of (F) ClC-5 and (G) Tamm-Horsfall protein in juxtaglomerular tubules. The distribution of ClC-5 in PT and IC is exactly similar in (H) E18.5 and (I) newborn kidneys. Staining on serial sections shows the codistribution of (J) ClC-5 and (K) CAII in postnatal IC. (Original magnification: A–C, E, H, I, 310; D, F, G, J, K, 410).
Table 1 - Distribution of ClC-5, H+-ATPase, AQP1, and CAII in the developing mouse kidney: Summary.
Full table Segmental distribution of H+-ATPase during mouse nephrogenesis
Staining for H+-ATPase was detected in the developing mouse kidney at E15.5 Figure 7a. The signal was located in the subapical area of developing PT in the inner cortex Figure 7e, whereas ureteric buds, glomeruli and undifferentiated structures were negative. Apical staining for H+-ATPase was also identified in scattered IC Figure 7f. A significant increase in H+-ATPase expression was detected at E16.5 Figure 7b, with staining of developing PT and IC in the cortex and inner medulla (Figure 7g and h). The IC in the medulla were characterized by apical protrusion into the lumen Figure 7h, unlike those observed in the cortex Figure 7g. The distribution of H+-ATPase at E18.5 Figure 7c was similar to that observed in the newborn mouse kidney (Figure 7d and i) and included PT cells (linear, subapical staining) and IC (punctuated, apical staining). H+-ATPase (Figure 7j and l) and ClC-5 (Fig. K and M) colocalized in the developing PT and IC as early as E15.5 (see Table 1 for summary).
Figure 7.
Segmental distribution of H+-ATPase in the developing mouse kidney. (A–L) Immunostaining for H+-ATPase, and (K, M) ClC-5 in the developing mouse kidney at (A, E, F) E15.5 (B, G, H) E16.5 (C, J–M) E18.5, and (D, I) newborn kidney. (A) Staining for H+-ATPase is detected in developing tubules in the inner cortex at E15.5. (B) A progressive increase in staining is then observed at E16.5, (C) E18.5, and further in the (D) newborn, with a distribution that includes developing PT in the cortex and scattered IC within cortical and medullary CD. (E, F) At E15.5, H+-ATPase is located in the subapical area of developing (E) PT cells and scattered IC in (F) cortical CD. Numerous positive IC are also detected in the (G) cortical and (H) inner-medullary CD at E16.5. The H+-ATPase distribution during late ontogeny is similar to that observed in the (I) newborn kidney, including linear, subapical staining in PT cells and focal, apical staining in -type IC. Staining on serial sections confirmed the colocalization of (K, M) ClC-5 with (J, L) H+-ATPase in (J, K) PT and (L, M) -type IC. (Original magnification: A–D, 80; E, F, I–M: 330; G, H, 385).
Segmental distribution of ClC-5 and H+-ATPase in the developing human kidney
Diffuse staining for ClC-5 was identified in 13GW kidneys, located in developing PT from juxtamedullary nephrons (Figure 8a and c). No specific staining was detected in condensates, comma and S-shaped bodies, and glomeruli. At 19 GW, a slightly more intense staining for ClC-5 was detected in maturing PT (Figure 8b and d), whereas no specific staining was detected with preadsorbed SB499 antibodies Figure 8e. The distribution of ClC-5 in developing PT was confirmed in serial sections stained for AQP1; to note, the staining for AQP1 was more polarized than ClC-5 at that stage (Figure 8f and g). In juxtamedullary nephrons, ClC-5 was also located in juxtaglomerular tubule profiles that were positive for Tamm-Horsfall protein (Figure 8h and i). In contrast with CAII (15GW) and H+-ATPase (19GW), ClC-5 could not be clearly identified in IC until the 24 GW (data not shown). Identification of ClC-5 staining in IC was ascertained postnatally by colocalization with CAII (Figure 8j and k).
Figure 8.
Segmental distribution and co-localization of ClC-5 in the developing human kidney. Immunostaining for (A–F, H, J) ClC-5, (G) AQP1, (I) Tamm-Horsfall protein, and (K) CAII in the developing human kidney at (A, C) 13 GW, (E) 17 GW, (B, D) 19 GW, (F–I) 24 GW, and in a (J, K) 4-month infant kidney. Diffuse staining for ClC-5 is identified in developing PT (p) from juxtamedullary nephrons at 13GW, whereas (A, C) glomeruli are negative. A similar pattern is observed at 19 GW, with a slightly more intense staining for ClC-5 in maturing (B, D) PT. No specific staining is observed when using preadsorbed SB499 antibodies on similar sections (E). Staining on serial sections from 24 GW human kidneys demonstrated the colocalization of (F) ClC-5 and (G) AQP1 in PT. ClC-5 is also located in juxtamedullary tubule profiles (H, arrowheads) positive for (I) Tamm-Horsfall protein. Identification of ClC-5 staining in IC (J, arrowhead) is confirmed by colocalization with (K) CAII in serial sections from a 4-month-old kidney. (Original magnification: A, B, 150; C, J, K, 300; D, E, H, I, 320; F, G, 280).
A faint signal for H+-ATPase Figure 9 was detected at 13 GW in the apical area of developing PT in the outer cortex, whereas glomeruli and ureteric buds were negative Figure 9a. From 15 GW to 24 GW Figure 9b, a progressive increase in H+-ATPase expression was detected in the subapical area of PT cells. Isolated cells positive for H+-ATPase Figure 9b were also identified in the CD as early as 19 GW (inset of Figure 9b). The reactivity for H+-ATPase after birth included a strong signal in the apical area of PT cells and IC of the CD Figure 9c.
Figure 9.
Segmental distribution of H+-ATPase during human nephrogenesis. (A) At 13 GW, a faint signal for H+-ATPase is detected in the apical region of the developing PT (p) in the outer cortex. Glomeruli (g) and ureteric buds (u) are negative. (B) At 24 GW, H+-ATPase is located in PT cells (p) and IC (arrowhead). To note, the latter are already positive for H+-ATPase at 19 GW (inset). (C) Typical staining for H+-ATPase in a 2-year-old kidney includes PT cells and -type IC within CD. (Original magnification: A, 225; B–D, 425).
DISCUSSION
Our studies show that ClC-5 undergoes a rapid induction at an early stage of mouse nephrogenesis, followed by a progressive maturation during late nephrogenesis. This ontogeny pattern is distinct from the progressive increase of H+-ATPase and AQP1 expression, and the postnatal induction of CAII. Studies with PGNase F confirm the early N-linked glycosylation of ClC-5, and the progressive maturation of the core protein during nephrogenesis. The distribution of ClC-5 in PT and -type IC, and its colocalization with H+-ATPase in these locations, are already achieved at E15.5 (see Table 1 for summary). In human nephrogenesis, ClC-5 is detected early during the second trimester, with a distribution that includes developing PT cells and later, IC.
Nephrogenesis in mouse and humans is characterized by a repetitive and reciprocal induction between the ureteric bud and the metanephric mesenchyme, resulting in the formation of a mature kidney before birth. In other species, such as rat or rabbit, nephrogenesis continues after birth (20 for review). Glomerular filtration starts at E14 in mouse21, and between the 9th and 12th GW in the human kidney22. The molecular events that take place in renal tubular cells following the onset of glomerular filtration remain partially unknown. The reabsorption of LMW proteins that are freely filtered is confined to PT cells, which are characterized by an intense endocytic activity followed by transport and degradation into the acidic vacuolar-lysosomal system23. Indirect evidence in rats24 and humans25 suggests that the PT endocytic activity is effective during nephrogenesis or immediately after birth. Furthermore, an increase of transporting membrane area and CA activity, as well as the upregulation of AQP1, has been documented in PT cells during nephrogenesis17,22,26. In contrast, the ontogeny and distribution of intracellular transporters such as ClC-5 and H+-ATPase during nephrogenesis have not been characterized to date. The issue is particularly relevant for PT maturation, since several lines of evidence suggest that ClC-5 interacts with the vacuolar H+-ATPase8 and plays an essential role in the megalin receptor-mediated endocytic pathway5,7,9,10.
Immunoblotting and RT-PCR analyses show that ClC-5 undergoes a rapid induction in the developing mouse kidney, followed by progressive maturation during late ontogeny and a marked induction just after birth Figures 1 and 2. Thus, the developmental pattern of ClC-5 expression in mouse clearly differs from that of another glycoprotein such as AQP1, or the co-expressed H+-ATPase. Studies with PGNase F not only confirmed the early N-glycosylation of ClC-530 but also demonstrated biochemical differences between embryonic and adult ClC-5 proteins. These differences include the higher molecular mass of the PGNase F-sensitive proteins in embryonic versus adult samples, as well as the specific expression in mouse embryonic samples of a ClC-5 isoform (75 kD) that is resistant to PGNase F treatment and that may correspond to the "core" ClC-5 protein Figure 3. By analogy with other members of the CLC family, the complex maturation of ClC-5 may result from alternative splicing27,28, post-translational modifications29, or covalent binding to proteins that are expressed during nephrogenesis20. It is tempting to hypothesize that the transient expression of immature ClC-5 isoforms during ontogeny may correspond to the signal observed in ureteric buds, where there is no colocalization with H+-ATPase, whereas a progressive increase in mature ClC-5 isoforms may actually parallel the induction of H+-ATPase during late nephrogenesis. These results, and the postnatal regulation of CAII Figure 1c, confirm recent evidence showing that differentially regulated genes, including transporters, ion-motive ATPases, and enzymes, can be grouped in clusters that participate in different steps of kidney organogenesis31.
As summarized in Table 1, the segmental expression of ClC-5 and H+-ATPase in PT cells is already detected at E15.5 in mouse Figure 5,Figure 6,Figure 7, and essentially achieved during the second trimester of gestation in humans Figure 8. These results complete early observations that the apical expression of brush border components and the distribution of megalin to the apical clathrin-coated membrane domains and endosomes coincide with the onset of glomerular filtration in rats13,32. The coexpression of ClC-5 and H+-ATPase in PT cells immediately after initiation of glomerular filtration could thus indicate the progressive maturation of renal tubular function and explain the decrease in the concentration of LMW proteins in the amniotic fluid during gestation25. These data also help to understand the pathophysiology of early phenotypic variants of Dent's disease, in which LMWP has been identified during the first weeks of life11,12.
In addition to PT cells, ClC-5 and H+-ATPase are codetected from E15.5 in the apical area of isolated cells within the CD. The apical polarity of both markers allow to identify these cells as -type IC, which, in the mature kidney, are primarily involved in urinary acidification. Studies in rat have shown that -type IC simultaneously express H+-ATPase and CAII at the end of the gestation, mature and increase in number during the first weeks of life, and are then partially removed from the CD14,33. Our results provide additional informations on IC maturation in mouse and human nephrogenesis. In mouse nephrogenesis, isolated cells of the medullary ureteric buds, positive for ClC-5, are detected as early as E14.5. IC positive for ClC-5, H+-ATPase, and CAII are observed in the cortical CD at E15.5. In human nephrogenesis, isolated cells positive for CAII are identified within ureteric buds and medullary CD at 15 GW, and rare, isolated cells positive for H+-ATPase are identified at 19 GW. In contrast, ClC-5 could not be identifed in IC before 24 GW. The colocalization of ClC-5 with CAII and/or H+-ATPase in IC was only observed postnatally, confirming data obtained in mature kidney5,7. These results suggest species differences in the maturation of IC14,33 and support the hypothesis that, in humans, the differentiation of IC occurs later than that of the principal cells of the CD17. Finally, ClC-5 is also located in tubules positive for Tamm-Horsfall protein, a marker of maturation of the TAL of Henle's loop34. The location of ClC-5 in maturing TAL confirms observations in mature human and rat kidney6,7. It has been speculated that ClC-5 could play a role in the regulated Ca2+ reabsorption that occurs in TAL35, providing a potential explanation for the hypercalciuria observed both in Dent's disease4 and ClCn5 KO mice10.
CONCLUSION
Our data demonstrate that the segmental distribution of ClC-5 and H+-ATPase is essentially achieved during early nephrogenesis in mouse and the second trimester of gestation in humans. These data provide new insights in the maturation of renal tubules and help to understand the pathophysiology of early phenotypic variants of Dent's disease.
References
| 1. | JENTSCH TJ, STEIN V, WEINREICH F & ZDEBIK AA. Molecular structure and physiological function of chloride channels. Physiol Rev 2002; 82: 503−568. | PubMed | ISI | ChemPort | |
| 2. | FISHER SE, BLACK GC & LLOYD SE et al. Isolation and partial characterization of a chloride channel gene which is expressed in kidney and is a candidate for Dent's disease (an X-linked hereditary nephrolithiasis). Hum Mol Genet 1994; 3: 2053−2059. | PubMed | ISI | ChemPort | |
| 3. | LLOYD SE, PEARCE SH & FISHER SE et al. A common molecular basis for three inherited kidney stone diseases. Nature 1996; 379: 445−449. | Article | PubMed | ISI | ChemPort | |
| 4. | SCHEINMAN SJ. X-linked hypercalciuric nephrolithiasis: Clinical syndromes and chloride channel mutations. Kidney Int 1998; 53: 3−17. | Article | PubMed | ISI | ChemPort | |
| 5. | GÜNTHER W, LÜCHOW A & CLUZEAUD F et al. ClC-5, the chloride channel mutated in Dent's disease, co-localizes with the proton pump in endocytotically active kidney cells. Proc Natl Acad Sci USA 1998; 95: 8075−8080. | Article | PubMed | ISI | ChemPort | |
| 6. | LUYCKX VA, GODA FO & MOUNT DB et al. Intrarenal and subcellular localization of rat ClC-5. Am J Physiol 1998; 275: F761−F769. | PubMed | ISI | ChemPort | |
| 7. | DEVUYST O, CHRISTIE PT & COURTOY PJ et al. Intra-renal and subcellular distribution of the human chloride channel, CLC-5, reveals a pathophysiological basis for Dent's disease. Hum Mol Genet 1999; 8: 247−257. | Article | PubMed | ISI | ChemPort | |
| 8. | MOULIN P, IGARASHI T & VAN DER SMISSEN P et al. Altered polarity and expression of H+-ATPase without ultrastructural changes in kidneys of Dent's disease patients. Kidney Int 2003; 63: 1285−1295. | Article | PubMed | |
| 9. | PIWON N, GÜNTHER W & SCHWAKE M et al. ClC-5 Cl-channel disruption impairs endocytosis in a mouse model for Dent's disease. Nature 2000; 408: 369−373. | Article | PubMed | ISI | ChemPort | |
| 10. | WANG SS, DEVUYST O & COURTOY PJ et al. Mice lacking renal chloride channel, CLC-5, are a model for Dent's disease, a nephrolithiasis disorder associated with defective receptor-mediated endocytosis. Hum Mol Genet 2000; 9: 2937−2945. | Article | PubMed | ISI | ChemPort | |
| 11. | NAKAZATO H, HATTORI S & FURUSE A et al. Mutations in the ClCN5 gene in Japanese patients with familial idiopathic low molecular weight proteinuria. Kidney Int 1997; 52: 895−900. | PubMed | ISI | ChemPort | |
| 12. | BOSIO M, BIANCHI ML, LLOYD SE & THAKKER RV. A familial syndrome due to Arg648Stop mutation in the X-linked renal chloride channel gene. Pediatr Nephrol 1999; 13: 278−283. | Article | PubMed | ISI | ChemPort | |
| 13. | BIEMESDERFER D, DEKAN G, ARONSON P & FARQUHAR MG. Assembly of distinctive coated pit and microvillar microdomains in the renal brush border. Am J Physiol 1992; 262: F55−F67. | PubMed | ISI | ChemPort | |
| 14. | KIM J, TISHER CC & MADSEN KM. Differentiation of intercalated cells in developing rat kidney: An immunohistochemical study. Am J Physiol 1994; 266: F977−F990. | PubMed | |
| 15. | MELLMAN I, FUCHS R & HELENIUS A. Acidification of the endocytic and exocytic pathways. Annu Rev Biochem 1986; 55: 663−700. | Article | PubMed | ISI | ChemPort | |
| 16. | HAMM LL & ALPERN RJ. Cellular mechanisms of renal tubular acidification. inThe Kidney: Physiology and Pathophysiology 2000; edited by Seldin DW, Giebisch G Philadelphia, Lippincott Williams & Wilkins pp 1935−1979. |
| 17. | DEVUYST O, BURROW CR & SMITH BL et al. Expression of aquaporins-1 and -2 during nephrogenesis and in autosomal dominant polycystic kidney disease. Am J Physiol 1996; 271: F169−F183. | PubMed | ISI | ChemPort | |
| 18. | SMITH AN, SKAUG J & CHOATE KA et al. Mutations in ATP6N1B, encoding a new kidney vacuolar proton pump 116-kD subunit, cause recessive distal renal tubular acidosis with preserved hearing. Nat Genet 2000; 26: 71−75. | Article | PubMed | ISI | ChemPort | |
| 19. | PFAFFI MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Research 2001; 29: 2004−2007. |
| 20. | KUURE S, VUOLTEENAHO R & VAINIO S. Kidney morphogenesis: Cellular and molecular regulation. Mech Develop 2000; 92: 31−45. |
| 21. | LOUGHNA S, LANDELS E & WOOLF AS. Growth factor control of developing kidney endothelial cells. Exp Nephrol 1996; 4: 112−118. | PubMed | ISI | ChemPort | |
| 22. | BAUM M & QUIGLEY R. Postnatal renal development. inThe Kidney: Physiology and Pathophysiology 2000; edited by Seldin DW, Giebisch G Philadelphia, Lippincott Williams & Wilkins pp 703−725. |
| 23. | CHRISTENSEN EI & NIELSEN S. Structural and functional features of protein handling in the kidney proximal tubule. Semin Nephrol 1991; 11: 414−439. | PubMed | ISI | ChemPort | |
| 24. | SMAOUI H, SCHAEVERBEKE M, MALLIE JP & SCHAEVERBEKE J. Transplacental effects of gentamicin on endocytosis in rat renal proximal tubule cells. Pediatr Nephrol 1994; 8: 447−450. | PubMed | |
| 25. | MUSSAP M, FANOS V & PICCOLI A et al. Low molecular mass proteins and urinary enzymes in amniotic fluid of healthy pregnant women at progressive stages of gestation. Clin Biochem 1996; 29: 51−56. | PubMed | |
| 26. | LÖNNERHOLM G & WISTRAND PJ. Carbonic anhydrase in the human fetal kidney. Pediatr Res 1983; 17: 390−397. | PubMed | |
| 27. | BORSANI G, RUGARLI EI & TAGLIALATELA M et al. Characterization of a human and murine gene (CLCN3) sharing similarities to voltage-gated chloride channels and to a yeast integral membrane protein. Genomics 1995; 27: 131−145. | Article | PubMed | ISI | ChemPort | |
| 28. | LAMB F, GRAEFF RW & CLAYTON GH et al. Ontogeny of CLCN3 chloride channel gene expression in human pulmonary epithelium. Am J Respir cell Mol Biol 2001; 24: 376−381. | PubMed | |
| 29. | SHAILUBHAI K, SAXENA ES, BALAPURE AK & VIJAY IK. Developmental regulation of glucosidase I, an enzyme involved in the processing of asparagine-linked glycoproteins in rat mammary gland. J Biol Chem 1990; 265: 9701−9706. | PubMed | |
| 30. | ROTH J. Protein n-glycosylation along the secretory pathway: Relationship to organelle topography and function, protein quality control, and cell intercations. Chem Rev 2002; 102: 285−303. | Article | PubMed | ISI | ChemPort | |
| 31. | STUART RO, BUSH KT & NIGAM SK. Changes in global gene expression patterns during development and maturation of the rat kidney. Proc Natl Acad Sci USA 2001; 98: 5649−5654. | Article | PubMed | ChemPort | |
| 32. | SAHALI D, MULLIEZ N & CHATELET F et al. Comparative immunochemistry and ontogeny of two closely related coated pit proteins. The 280-kd taget of teratogenic antibodies and the 330-kd taget of nephritogenic antibodies. Am J Pathol 1993; 142: 1654−1667. | PubMed | ISI | ChemPort | |
| 33. | NARBAITZ R, VANDORPE D & LEVINE DZ. Differentiation of renal intercalated cells in fetal and postnatal rats. Anat Embryol (Berl) 1991; 183: 353−361. | PubMed | |
| 34. | ZIMMERHACKL LB, ROSTASY K & WIEGELE G et al. Tamm-Horsfall protein as a marker of tubular maturation. Pediatr Nephrol 1996; 10: 448−452. | PubMed | |
| 35. | RICCARDI D, LEE WS & LEE K et al. Localization of the extracellular Ca2+-sensing receptor and PTH/PTHrH receptor in rat kidney. Am J Physiol 1996; 271: F951−F956. | PubMed | ISI | ChemPort | |
Acknowledgments
We are grateful to P.J. Courtoy, N.Y. Loh, M-Cl. Gubler, R. Rezsohazy, and F. Wu for helpful discussions and material, and to Ph. Camby, Y. Cnops, H. Debaix, and L. Wenderickx for excellent technical assistance. The Clcn5 KO mice were kindly provided by W.B. Guggino (Dept. of Physiology, Johns Hopkins University Medical School, Baltimore, MD, USA). This work was supported by the Belgian agencies FNRS and FRSM, the Foundation Alphonse et Jean Forton, the Action de Recherches Concertées 00/05-260, and the Wellcome Trust (UK) and MRC (UK). F.J. is a Research Fellow of the FNRS.
MORE ARTICLES LIKE THIS
These links to content published by NPG are automatically generated
RESEARCH
A catalog of gene expression in the developing kidney
Kidney International Letter
A novel renal carbonic anhydrase type III plays a role in proximal tubule dysfunction
Kidney International Original Article
Chloride channels ClC-2 and ICln mRNA expression differs in renal epithelial ontogeny
Kidney International Original Article
Kidney International Original Article
Flow cytometric immunodissection of the human distal tubule and cortical collecting duct system
Kidney International Original Article
