Endosomal Chloride-Proton Exchange Rather Than ...

Endosomal Chloride-Proton Exchange Rather Than Chloride Conductance Is Crucial for Renal Endocytosis Gaia Novarino, et al. Science 328, 1398 (2010); DOI: 10.1126/science.1188070 This copy is for your personal, non-commercial use only.

If you wish to distribute this article to others, you can order high-quality copies for your colleagues, clients, or customers by clicking here. Permission to republish or repurpose articles or portions of articles can be obtained by following the guidelines here.

Updated information and services, including high-resolution figures, can be found in the online version of this article at: http://www.sciencemag.org/cgi/content/full/328/5984/1398 Supporting Online Material can be found at: http://www.sciencemag.org/cgi/content/full/science.1188070/DC1 A list of selected additional articles on the Science Web sites related to this article can be found at: http://www.sciencemag.org/cgi/content/full/328/5984/1398#related-content This article cites 25 articles, 12 of which can be accessed for free: http://www.sciencemag.org/cgi/content/full/328/5984/1398#otherarticles This article has been cited by 2 articles hosted by HighWire Press; see: http://www.sciencemag.org/cgi/content/full/328/5984/1398#otherarticles This article appears in the following subject collections: Cell Biology http://www.sciencemag.org/cgi/collection/cell_biol

Science (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. Copyright 2010 by the American Association for the Advancement of Science; all rights reserved. The title Science is a registered trademark of AAAS.

Downloaded from www.sciencemag.org on August 11, 2010

The following resources related to this article are available online at www.sciencemag.org (this information is current as of August 11, 2010 ):

result in gB fragments that are translocated into the cytoplasm. To determine whether cytosolic access is required, we examined the roles of TAP and proteasomes in gB cross-presentation. When DCs from Tap−/− mice were incubated with necrotic infected cells, gB cross-presentation was completely eliminated (Fig. 3F). In addition, cross-presentation of gB, as well as ICP6, was inhibited by lactacystin, indicating dependence on proteasomal processing (Fig. 3E). Cross-presentation thus depends on cytosolic processing of gB fragments generated in the phagosome by GILT-mediated reduction and cathepsin-mediated proteolysis. A requirement for GILT in the induction of the CD8+ T cell response to gB498-505 during an infection would argue that cross-priming is important for the in vivo anti–HSV-1 immune response. Wild-type and Ifi30−/− mice were infected with HSV-1, and the draining lymph nodes (LNs) were examined for the induction of Kb-gB498-505–specific and Kb-ICP6822-829–specific CD8+ T cells. Although mice lacking GILT generated the same average percentage of ICP6822-829-specific CD8+ T cells when infected with HSV-1 as wild-type mice, the number of gB498-505-specific CD8+ T cells was significantly reduced (Fig. 4, A to C). There was no difference in the survival of the infected mice. Responses to GILT-independent epitopes such as ICP6822-829 may make up for any deficiency. To determine whether GILT-dependent crosspresentation is a more general phenomenon, we examined the CD8+ T cell response of mice infected with the PR8 strain of influenza A virus. LN cells from naïve and infected mice were restimulated with wild-type DCs pulsed with peptides that correspond to a variety of H2-Kb– and H2-Db–restricted epitopes from hemagglutinin (HA), neuraminidase (NA), polymerase (PA), and nucleoprotein (NP) (www.immuneepitope.com/ home.do) (22). HA and NA contain six and eight disulfide bonds, respectively, whereas PA and NP have none (23–25). A similar percentage of wildtype and GILT-negative CD8+ T cells responded to Db-restricted PA and NP epitopes upon restimulation (Fig. 4, D and E). In contrast, the responses of CD8+ T cells from mice lacking GILT were significantly reduced for four out of five of the HA epitopes and for two out of three of the NA epitopes. The two HA epitopes to which almost no CD8+ T cells develop in the Ifi30−/− mice contain or are immediately adjacent to a cysteine (C480) involved in a disulfide bond (C21-C480). For both HA and NA, one epitope is GILT independent, strongly arguing against the possibility that any GILT requirement reflects GILT-dependent MHC class II–restricted responses that mediate CD4+ T cell help (26). Although the epitope specificity of the CD4+ T cells in the Ifi30−/− mice may be different from that of wild-type mice, the total numbers of CD4+ T cells that they generate during a viral immune response are similar (fig. S2), as are the numbers of CD4+ T cells in the spleens of uninfected wild-type and Ifi30−/− mice. The data show that GILT-dependent cross-presentation is not restricted to gB, and that cross-priming is important

1398

in the CD8+ T cell response to influenza virus. The residual CD8+ T cell responses observed to gB and the HA and NA epitopes by Ifi30−/− animals may reflect priming by directly infected APCs. The only known function of GILT is to reduce disulfide bonds, and we have shown that GILT is essential for cross-presentation of many peptides from disulfide-containing proteins. We suggest that reduction in the acidic environment of the phagosome facilitates partial proteolysis into fragments that are translocated into the cytosol where they are further degraded by the proteasome to generate peptides. These are transported by TAP and bind in a conventional manner, possibly after amino-terminal trimming (27), to MHC class I molecules. This latter step is likely to occur in the ER, but could occur in phagosomes that have recruited ER membrane components, although this issue remains contentious (28, 29). For gB, the inability to cross-present is reflected in a reduction in Kb-gB498-505–specific CD8+ T cells in vivo, indicating the importance of cross-priming in CD8+ T cell responses to HSV-1 infection. The similar reduction in HA- and NAspecific CD8+ T cells suggests that cross-priming is also important during influenza A infection. The role played by GILT in cross-priming, combined with its established involvement in MHC class II– restricted CD4+ T cell responses (30), indicates the importance of the enzyme in the immune system. This may have implications for vaccine design and approaches to tumor immunotherapy that involve peptide-based vaccines, in that linear peptides may not be the optimal vehicles for the expression of GILT-dependent epitopes, and for autoimmunity to self-antigens that contain multiple disulfide bonds. References and Notes 1. M. J. Bevan, J. Exp. Med. 143, 1283 (1976). 2. L. J. Sigal, S. Crotty, R. Andino, K. L. Rock, Nature 398, 77 (1999).

3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.

K. L. Rock, L. Shen, Immunol. Rev. 207, 166 (2005). R. S. Allan et al., Immunity 25, 153 (2006). S. Bedoui et al., Nat. Immunol. 10, 488 (2009). R. Belizaire, E. R. Unanue, Proc. Natl. Acad. Sci. U.S.A. 106, 17463 (2009). M. Kovacsovics-Bankowski, K. L. Rock, Science 267, 243 (1995). A. L. Ackerman, A. Giodini, P. Cresswell, Immunity 25, 607 (2006). A. Schäfer, D. H. Wolf, EMBO J. 28, 2874 (2009). M. L. Lin et al., Proc. Natl. Acad. Sci. U.S.A. 105, 3029 (2008). C. C. Norbury, B. J. Chambers, A. R. Prescott, H. G. Ljunggren, C. Watts, Eur. J. Immunol. 27, 280 (1997). A. Giodini, P. Cresswell, EMBO J. 27, 201 (2008). B. Arunachalam, U. T. Phan, H. J. Geuze, P. Cresswell, Proc. Natl. Acad. Sci. U.S.A. 97, 745 (2000). U. T. Phan, B. Arunachalam, P. Cresswell, J. Biol. Chem. 275, 25907 (2000). T. Hanke, F. L. Graham, K. L. Rosenthal, D. C. Johnson, J. Virol. 65, 1177 (1991). See supporting material on Science Online. E. E. Heldwein et al., Science 313, 217 (2006). R. Singh, A. Jamieson, P. Cresswell, Nature 455, 1244 (2008). D. S. Collins, E. R. Unanue, C. V. Harding, J. Immunol. 147, 4054 (1991). K. L. McCoy et al., J. Immunol. 143, 29 (1989). A. Savina et al., Cell 126, 205 (2006). W. Zhong, P. A. Reche, C. C. Lai, B. Reinhold, E. L. Reinherz, J. Biol. Chem. 278, 45135 (2003). J. N. Varghese, W. G. Laver, P. M. Colman, Nature 303, 35 (1983). I. A. Wilson, J. J. Skehel, D. C. Wiley, Nature 289, 366 (1981). Q. Ye, R. M. Krug, Y. J. Tao, Nature 444, 1078 (2006). N. K. Rajasagi et al., J. Virol. 83, 5256 (2009). T. Serwold, F. Gonzalez, J. Kim, R. Jacob, N. Shastri, Nature 419, 480 (2002). I. Jutras, M. Desjardins, Annu. Rev. Cell Dev. Biol. 21, 511 (2005). N. Touret et al., Cell 123, 157 (2005). M. Maric et al., Science 294, 1361 (2001). This work was supported by the Howard Hughes Medical Institute and NIH grant R37AI23081 (P.C.).

Supporting Online Material www.sciencemag.org/cgi/content/full/328/5984/1394/DC1 Materials and Methods Figs. S1 and S2 References 5 March 2010; accepted 4 May 2010 10.1126/science.1189176

Endosomal Chloride-Proton Exchange Rather Than Chloride Conductance Is Crucial for Renal Endocytosis Gaia Novarino, Stefanie Weinert, Gesa Rickheit,* Thomas J. Jentsch† Loss of the endosomal anion transport protein ClC-5 impairs renal endocytosis and underlies human Dent’s disease. ClC-5 is thought to promote endocytosis by facilitating endosomal acidification through the neutralization of proton pump currents. However, ClC-5 is a 2 chloride (Cl–)/proton (H+) exchanger rather than a Cl– channel. We generated mice that carry the uncoupling E211A (unc) mutation that converts ClC-5 into a pure Cl– conductor. Adenosine triphosphate (ATP)–dependent acidification of renal endosomes was reduced in mice in which ClC-5 was knocked out, but normal in Clcn5unc mice. However, their proximal tubular endocytosis was also impaired. Thus, endosomal chloride concentration, which is raised by ClC-5 in exchange for protons accumulated by the H+-ATPase, may play a role in endocytosis.

L

11 JUNE 2010

uminal acidification along the endocytic pathway serves many purposes (1), including the progression of endocytosis itself VOL 328

SCIENCE

(2). It is performed by endosomal H+-transporting adenosine triphosphatases (H+-ATPases) that need a countercurrent for electroneutrality. Because this

www.sciencemag.org


REPORTS

Leibniz-Institut für Molekulare Pharmakologie (FMP) and Max-Delbrück-Centrum für Molekulare Medizin (MDC), 13125 Berlin, Germany. *Present address: TaconicArtemis GmbH, 51063 Köln, Germany. †To whom correspondence should be addressed. E-mail: [email protected]

proximal tubular endocytosis (4, 7), which was studied in chimeric tubules resulting from random X-chromosomal inactivation in female Clcn5+/− mice (4). In those tubules, WT and KO cells were distinguished by means of antibodies to ClC-5 (3, 4), but this approach cannot differentiate between cells expressing the WT or uncoupled ClC-5 in Clcn5unc/+ tubules. Rather than epitopetagging the E211A mutant, which might interfere with its function, we generated mice in which the C terminus of ClC-5 was converted to that of ClC-3 (fig. S1, B and C, and S5). The generation of this Clcn5* allele required only two amino acid exchanges and changed neither ClC-5 currents (fig. S5D) nor its abundance (Fig. 1A) and localization (fig. S6A). Our antibodies against the C terminus of ClC-5 recognized ClC-5 and ClC-5unc but not ClC-5* (Figs. 1A and 2, C and D, and fig. S1, B and C). In vivo endocytosis experiments were performed by injecting into the bloodstream labeled endocytic cargo that can pass the glomerular filter. Endocytosis in cells expressing the Clcn5* allele or WT ClC-5 was indistinguishable (fig. S6B). However, cells expressing the Clcn5unc allele accumulated much less fluorescently labeled b-lactoglobulin, which is a marker for receptor-mediated endocytosis (Fig. 2C), or the fluid-phase marker dextran (Fig. 2D) than did neighboring cells that express the 2Cl–/H+–exchanger ClC-5*. Thus, uncoupling anion transport from protons resulted in a cellautonomous impairment of both receptor-mediated and fluid-phase endocytosis akin to Clcn5– mice (4). Similarly as in KO mice (4, 20), cells expres-

sing the uncoupled E211A mutant displayed reduced levels of the endocytic receptors megalin (Fig. 3, A and C) and cubilin (Fig. 3B)—a finding ascribed to impaired recycling to the plasma membrane (4, 20). In Clcn5unc kidney, the sodiumphosphate cotransporter NaPi-2a was shifted from the apical membrane to intracellular vesicles, and its overall abundance was reduced (fig. S7), which explains the observed hyperphosphaturia. Similarly increased endocytosis of NaPi-2a in Clcn5– mice was attributed to reduced endocytosis of filtered parathyroid hormone (PTH), entailing a luminal increase of PTH and excessive stimulation of apical PTH receptors (4, 5). Because the E211A mutation did not abolish ClC-5 currents and maintained endosomal acidification (Fig. 1C), it might have affected endocytosis less than a loss of ClC-5. We thus compared Clcn5unc and Clcn5– cells side by side in chimeric tubules of Clcn5unc/– females. No differences were detected in receptor-mediated (Fig. 2E) or fluid-phase endocytosis (fig. S8), nor in the localization and abundance of megalin (Fig. 3C). Thus unexpectedly, the anion conductance of ClC-5(E211A) could not even partially substitute for Cl–/H+ exchange in the support of proximal tubular endocytosis and normal localization of apical receptors. Because the intramembrane E211A point mutation was unlikely to have changed ClC-5 protein interactions, the pathology of Clcn5unc mice is caused neither by the disruption of a macromolecular ClC-5–containing endocytic complex (21, 22) nor by the loss of interaction with KIF3B

B kDa 82

ClC-5NT

Clcn5+

Clcn5unc

+/ y un c/ y */y -/y

-/y

+/ y un c/ y */y

A

ClC-5CT

64 49

-actin

-actin

37

ClC-5CT

ClC-5CT

D

FCCP

ATP

1.2

1.2

1.1

1.1

1.0

0

50

100 150 200 250

1.0

0.8

E 0 .4

**

*

0 .3 0 .2 0 .1

0.9

0.9 0.8

FCCP

F

ATP

F/F0

C

F/F0

current depends on chloride, conventional wisdom suggests (1) that endosomal Cl– channels are involved (fig. S1A). ClC-5 was thought to embody this channel in proximal tubular endosomes (3–5). Disruption of ClC-5 impairs renal endosomal acidification in vitro (5) and drastically reduces proximal tubular endocytosis in mice and humans (4, 6, 7). The hyperphosphaturia and hypercalciuria that lead to kidney stones in Dent’s disease have been attributed to impaired tubular endocytosis of calciotropic hormones (4). However, it has recently been shown (8–10) that ClC-5 is a 2Cl–/H+ exchanger rather than a Cl– channel. It seems counterintuitive that such an exchanger should neutralize pump currents because it mediates H+-efflux during ATP-driven acidification. The biological consequence of proton coupling has remained enigmatic (11). If the ClC-5 Cl–/H+ exchanger could be converted into an uncoupled Cl– conductor, it should efficiently facilitate endosomal acidification. Phenotypes of mice carrying such a mutation cannot be attributed to impaired endosomal acidification, but can be ascribed specifically to a loss of coupling chloride gradients to proton gradients. A mutation in the “gating” glutamate of CLC exchangers (12) suffices to convert them into pure anion conductors (8, 9, 13–15). We inserted the corresponding, well-characterized E211A mutation (8, 9, 14, 16, 17) into the Clcn5 gene on mouse chromosome X and created mice in which ClC-5 was converted into an uncoupled Cl– conductor (Clcn5unc mice) (figs. S1A and S2). The mutant protein was expressed at wild-type (WT) levels (Fig. 1A). No change was observed in its subcellular localization in kidney proximal tubular and intercalated cells (Fig. 1B and fig. S3). The renal expression of the related ClC-3 and ClC-4 proteins also was not affected (fig. S4). To test whether the uncoupled ClC-5unc mutant supported endosomal acidification, we added ATP to endosomal fractions from renal cortex (containing mainly proximal tubules) of WT or Clcn5unc/y mice and monitored vesicular pH using acridine orange fluorescence. H+-ATPase–driven acidification of WT and Clcn5unc vesicles occurred with similar efficiency but was severely reduced with endosomes from mice in which ClC-5 was knocked out (KO mice), as expected (5, 18) (Fig. 1, C to E). Despite maintaining active endosomal acidification, Clcn5unc mice displayed abnormalities found in ClC-5 KO (Clcn5 – ) mice and patients with Dent’s disease (4, 7, 19), such as lowmolecular-weight proteinuria (Fig. 2, A and B), hyperphosphaturia, and hypercalciuria (table S1). Proteinuria of Clcn5– mice results from impaired


REPORTS

0

50

100 150 200 250

0 .0

Time (s)

Time (s)

Fig. 1. Renal endosomal acidification of mice converting ClC-5 into a chloride conductor. (A) Immunoblot for ClC-5 using kidney membranes of different genotypes with antibodies against the (left) N terminus or the (right) C terminus (C5/05A). Asterisk indicates mutated C terminus of ClC-5 (fig. S5). (B) Identical staining pattern using the C-terminal PEP5E antibody to ClC-5 (3) in proximal tubules of WT and Clcn5unc/y mice. Scale bar, 5 mm. (C and D) Averaged traces of acridine orange fluorescence comparing ATP-driven acidification of renal cortical vesicles from (C) WT (green) and Clcn5unc (red) and (D) WT (green) and Clcn5– (black) mice (with identical WT traces). Reduced fluorescence reflects vesicular acidification. The protonophore FCCP was added as a control. F0, fluorescence at time (t) = 0. (E) Averaged total change in fluorescence. Data averaged from 25 (+/y), 13 (–/y), and 10 (unc/y) experiments with material obtained from more than four independent vesicle preparations per genotype. Error bars, SEM. *P ≤ 0.05; **P ≤ 0.01.

www.sciencemag.org

SCIENCE

VOL 328

11 JUNE 2010

1399

pling of Cl– gradients from H+ gradients. ClC-5 might drive H+-ATPase–independent acidification by exchanging cytosolic H+ for intravesicular Cl– at an early step of endocytosis (9). Shortly after pinching off from the plasma membrane,

1400

11 JUNE 2010

-/y

kDa 55

27

Merge

Merge

Merge

Merge

Dextran

Merge

Megalin

VOL 328

SCIENCE

+/ y un c

-/y

y

/y

Megalin

+/

Fig. 3. Effect of uncou- A Clcn5unc pling ClC-5 on endocytic receptors. (A) Reduced megalin levels in Clcn5uncexpressing cells of a chimeric tubule of Clcn5unc/* Clcn5* mice. (B) Reduced cubilin ClC-5CT expression in Clcn5uncexpressing cells that were B Clcn5* identified through reduced endocytosis of Alexa 488–dextran. (C) Similar reduction of megalin in cells expressing Clcn5 – Clcn5uncCubilin or Clcn5unc. To compare weak megalin immunostaining in Clcn5unc/– tu- C Clcn5 bules, image intensity was boosted in the center panel as compared with that of (A). (Right) Immunoblot for megalin Clcn5unc using kidney membranes ClC-5CT from WT, Clcn5unc/y, and –/y Clcn5 mice. Scale bars, 4 mm [(A) and (B)] and 5 mm (C).

+/ y un c/ y +/ y -/y

y un c/ y -/y

+/

+/ y

Fig. 2. Proteinuria and im- A B kDa paired endocytosis in Clcn5 – 70 unc DBP Clcn5 mice. (A) Urinary proteins analyzed by means of SDS–polyacrylamide RBP 27 gel electrophoresis and silver staining. (B) Immunoblot for vitamin D–binding protein (DBP) and retinolbinding protein (RBP) in urine. (C) In vivo 10-min upClcn5 unc C take of Alexa Fluor 546– labeled (Invitrogen, Carlsbad, CA) b-lactoglobulin (red) showed decreased receptor-mediated endocytosis in Clcn5unc-expressing Clcn5* Lactoglobulin ClC-5CT cells in a chimeric proximal tubule from a Clcn5unc/* fe- D unc Clcn5 male. ClC-5unc, but not the – + ClC-5* Cl /H exchanger, is recognized by the antibody C5/05A (green). (Right) Overlay with additional Clcn5* brush-border staining for ClC-5CT Dextran villin (blue). (D) Reduced fluid-phase endocytosis of E Clcn5 Alexa Fluor 488–labeled (Invitrogen) dextran in ClC5unc–expressing cells in an experiment similar to (C). (E) Similar b-lactoglobulin– uptake of cells lacking ClC-5 Clcn5unc Lactoglobulin ClC-5CT or expressing Clcn5unc in unc/– a Clcn5 female. Scale bars, 5 mm (C), 3.2 mm (D), and 6 mm (E).

+/ y un c/ y +/ y

(23), which are mechanisms that have been suggested to underlie Dent’s disease. The impairment of endocytosis in Dent’s disease is not likely to result only from reduced ATP-dependent endosomal acidification (3) but is related to an uncou-

Megalin

endocytic vesicles may contain the high extracellular chloride concentration. A selective lack of such an initial Cl– gradient–driven endosomal acidification might explain the impaired endocytosis of both Clcn5– and Clcn5unc mice. However, endosomal [Cl–] was found to be initially low—a finding ascribed to surface charge effects (18). Furthermore, a lack of a Cl– gradient–driven acidification cannot be invoked for the severe lysosomal pathology of Clcn7−/− (24, 25) and Clcn7unc/unc mice (26) because their lysosomal pH is normal (24–26) and because this mechanism would require a substantial Cl– supply to lysosomes. The surprisingly similar pathologies of Clcn5 – and Clcn5unc mice might be due to reduced endosomal Cl– accumulation in each mouse model. Whereas Cl– channels operating in parallel to an H+-ATPase would raise intravesicular Cl– during active acidification, stoichiometric 2Cl–/H+ exchange would maintain high vesicular Cl– concentration also under steady state. Indeed, lysosomes containing the WT ClC-7 Cl–/H+ exchanger display higher [Cl–] than their Clcn7−/− or Clcn7unc/unc counterparts (26). Furthermore, endosomal [Cl–] might regulate endosomal Ca++ channels (27). The analysis of Clcn5unc and Clcn7unc (26) mice suggests that luminal anion concentration is important all along the endosomal-lysosomal pathway. References and Notes 1. I. Mellman, R. Fuchs, A. Helenius, Annu. Rev. Biochem. 55, 663 (1986). 2. M. J. Clague, S. Urbé, F. Aniento, J. Gruenberg, J. Biol. Chem. 269, 21 (1994). 3. W. Günther, A. Lüchow, F. Cluzeaud, A. Vandewalle, T. J. Jentsch, Proc. Natl. Acad. Sci. U.S.A. 95, 8075 (1998). 4. N. Piwon, W. Günther, M. Schwake, M. R. Bösl, T. J. Jentsch, Nature 408, 369 (2000). 5. W. Günther, N. Piwon, T. J. Jentsch, Pflu¨gers Arch. 445, 456 (2003). 6. S. E. Lloyd et al., Nature 379, 445 (1996). 7. S. S. Wang et al., Hum. Mol. Genet. 9, 2937 (2000). 8. A. Picollo, M. Pusch, Nature 436, 420 (2005). 9. O. Scheel, A. A. Zdebik, S. Lourdel, T. J. Jentsch, Nature 436, 424 (2005). 10. G. Zifarelli, M. Pusch, EMBO J. 28, 175 (2009). 11. T. J. Jentsch, J. Physiol. 578, 633 (2007). 12. R. Dutzler, E. B. Campbell, R. MacKinnon, Science 300, 108 (2003). 13. A. Accardi, C. Miller, Nature 427, 803 (2004). 14. A. A. Zdebik et al., J. Biol. Chem. 283, 4219 (2007). 15. E. Y. Bergsdorf, A. A. Zdebik, T. J. Jentsch, J. Biol. Chem. 284, 11184 (2009). 16. In the mutants, other amino acids were substituted at certain locations; for example, E211A indicates that glutamic acid at position 211 was replaced by alanine. Single-letter abbreviations for the amino acid residues are as follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr. 17. Materials and methods are available as supporting material on Science Online. 18. M. Hara-Chikuma, Y. Wang, S. E. Guggino, W. B. Guggino, A. S. Verkman, Biochem. Biophys. Res. Commun. 329, 941 (2005). 19. S. J. Scheinman, Kidney Int. 53, 3 (1998). 20. E. I. Christensen et al., Proc. Natl. Acad. Sci. U.S.A. 100, 8472 (2003). 21. D. H. Hryciw et al., J. Biol. Chem. 278, 40169 (2003). 22. D. H. Hryciw et al., J. Biol. Chem. 281, 16068 (2006). 23. A. A. Reed et al., Am. J. Physiol. Renal Physiol. 298, F365 (2010).

www.sciencemag.org


REPORTS

REPORTS W. Blaner, and C. Wagner for antibodies against cubilin, megalin, retinol binding protein, and AE1, respectively. This work was supported by the Deutsche Forschungsgemeinschaft (grants Zd58/1, Je164/6, and Je164/9).

+ +/ +

+/

gl

-/-

-/u

c/

un

un c/

+

un

Ostm1

27

-/-

Clcn7

Tibia

c

c

nc /u

kD 35

Kidney

ClC-7

64

Clcn7

Brain

C

nc

Tibia +/

+/ + un c

/u

un c/

kD 82

-/-

unc/unc

un

Clcn7

nc

c

Kidney

-/-

Brain

B

+/ + un c

A

-/-

C

currents mediated by ClC-7 have been deemed necessary for shunting lysosomal proton-pump currents (1). However, lysosomal pH was normal in cells lacking either ClC-7 or Ostm1 (2, 3). ClC-7 now seems likely to be a Cl–/H+ exchanger rather than a Cl– channel (6, 7). Because H+-pump currents may be neutralized by both Cl– channels and electrogenic Cl–/H+ exchangers (6), it is unclear whether lysosomal Cl–/H+ exchange confers functional

-/-

During lysosomal acidification, proton-pump currents are thought to be shunted by a chloride ion (Cl–) channel, tentatively identified as ClC-7. Surprisingly, recent data suggest that ClC-7 instead mediates Cl–/proton (H+) exchange. We generated mice carrying a point mutation converting ClC-7 into an uncoupled (unc) Cl– conductor. Despite maintaining lysosomal conductance and normal lysosomal pH, these Clcn7unc/unc mice showed lysosomal storage disease like mice lacking ClC-7. However, their osteopetrosis was milder, and they lacked a coat color phenotype. Thus, only some roles of ClC-7 Cl–/H+ exchange can be taken over by a Cl– conductance. This conductance was even deleterious in Clcn7+/unc mice. Clcn7–/– and Clcn7unc/unc mice accumulated less Cl– in lysosomes than did wild-type mice. Thus, lowered lysosomal chloride may underlie their common phenotypes.

nc

Stefanie Weinert,1,2 Sabrina Jabs,1,2,6 Chayarop Supanchart,3 Michaela Schweizer,4 Niclas Gimber,1,2 Martin Richter,1,6 Jörg Rademann,1,6* Tobias Stauber,1,2 Uwe Kornak,3,5 Thomas J. Jentsch1,2†

advantages over the simple Cl– conductance in the textbook model for vesicular acidification. We created knock-in mice in which the ClC-7 “gating” glutamate (E) was mutated to alanine (A) (fig. S1) (8). On the basis of results from other CLC Cl–/H+ exchangers (9–12), this Glu245 → Ala245 (E245A) mutation should lead to Cl– transport that is uncoupled (unc) from protons, hence our designation of this allele as Clcn7unc. Homozygous Clcn7 unc/unc mice showed severe growth retardation (Fig. 1A and fig. S2) and died within 5 weeks. ClC7unc and wild-type (WT) ClC-7 were expressed at similar levels (Fig. 1B) and similarly localized to lysosomes (Fig. 1D). Neither the abundance, nor the lysosomal localization of Ostm1 was changed in Clcn7 unc/unc mice, contrasting with its strongly reduced protein level (3) and mislocalization in Clcn7 –/– cells (Fig. 1, C and D). In neurons, however, ClC-7unc staining was more diffuse (fig. S3B), reflecting changed lysosomal compartments like in Clcn7 –/– neurons (2). The abundance of other CLC exchangers was unchanged in Clcn7unc/unc mice (fig. S4). In an agouti genetic background, the coat color of Clcn7 –/– and Ostm1–/– (grey-lethal) mice is grey (3, 4), whereas it was brownish in WT and Clcn7 unc/unc mice (Fig. 1A). Clcn7 unc/unc mice were osteopetrotic (Fig. 2A and fig. S5), although less severely than Clcn7 –/– (1) or Ostm1 –/– (4) mice. ClC-7 and Ostm1 were detected at the ruffled border of Clcn7unc/unc osteoclasts (fig. S3A). This

-/u

www.sciencemag.org/cgi/content/full/science.1188070/DC1 Materials and Methods

49 37


8 February 2010; accepted 21 April 2010 Published online 29 April 2010; 10.1126/science.1188070 Include this information when citing this paper.

Supporting Online Material

Lysosomal Pathology and Osteopetrosis upon Loss of H+-Driven Lysosomal Cl– Accumulation

lC-7 is the only member of the CLC gene family of anion transporters substantially expressed on lysosomes (1–3), where it resides together with its b-subunit Ostm1 (3). Inactivation of either subunit leads to lysosomal storage disease and osteopetrosis in mice and humans (1–4). Cellular defects include slowed degradation of endocytosed proteins (5) and impaired acidification of the osteoclast resorption lacuna (1). Cl–

Figs. S1 to S8 Table S1 References

un

24. D. Kasper et al., EMBO J. 24, 1079 (2005). 25. P. F. Lange, L. Wartosch, T. J. Jentsch, J. C. Fuhrmann, Nature 440, 220 (2006). 26. S. Weinert et al., Science 328, 1401 (2010); published online 29 April 2010 (10.1126/science.1188072). 27. M. Saito, P. I. Hanson, P. Schlesinger, J. Biol. Chem. 282, 27327 (2007). 28. We thank R. Pareja-Alcaraz and S. Rode for technical assistance and P. Verroust, S. Bachmann,

β-actin 27

VAMP2

+/+

D

Fig. 1. ClC-7 and Ostm1 in mice carrying different Clcn7 alleles. (A) Clcn7unc/unc, Clcn7 –/–, and WT mice at postnatal day 22 (P22) in an agouti background. (B) ClC-7 immunoblot of tissues from Clcn7 unc/unc, Clcn7 –/–, and WT mice. (C) The mature form of Ostm1 (doublet) was absent from Ostm1 –/– (grey-lethal, gl) and Clcn7 –/– brains but showed similar abundance in Clcn7 +/+, Clcn7unc/unc (0.97 T 0.20), and Clcn7 –/unc (0.85 T 0.32) brains (normalized to WT, T SEM, six mouse pairs, three immunoblots). kD, kilodaltons. (D) (Top) Immunofluorescence of ClC-7 or ClC-7unc (green) and the lysosomal marker Lamp-1 (red) in WT, Clcn7 – / –, and Clcn7 unc/unc fibroblasts. (Bottom) Costaining for Ostm1 (red) and ClC-7 or ClC-7unc (green). Normal localization of ClC-7unc and Ostm1 in Clcn7 unc/unc cells, but reticular Ostm1 staining in Clcn7 – / – cells.

www.sciencemag.org

SCIENCE

Clcn7

+/+

Clcn7

-/-

Clcn7

unc/unc

ClC-7 Lamp-1

ClC-7 Lamp-1

ClC-7 Lamp-1

ClC-7 Ostm1

ClC-7 Ostm1

ClC-7 Ostm1

VOL 328

11 JUNE 2010

1401

www.sciencemag.org/cgi/content/full/science.1188070/DC1

Supporting Online Material for

Endosomal Chloride-Proton Exchange Rather Than Chloride Conductance is Crucial for Renal Endocytosis Gaia Novarino, Stefanie Weinert, Gesa Rickheit, Thomas J. Jentsch*

*To whom correspondence should be addressed. E-mail: [email protected]

Published 29 April 2010 on Science Express DOI: 10.1126/science.1188070 This PDF file includes: Materials and Methods Figs. S1 to S8 Table S1 References

Supporting Online Material for

Endosomal Chloride-Proton Exchange Rather Than Chloride Conductance is Crucial for Renal Endocytosis

Gaia Novarino, Stefanie Weinert, Gesa Rickheit & Thomas J. Jentsch

* To whom correspondence should be addressed E-mail: [email protected]

This PDF file includes: Materials and Methods Figs. S1 to S8 Table S1 References

1

Materials and Methods Mice The generation of Clcn5 - mice has been described previously (1). For the generation of knock-in mice genomic Clcn5 DNA isolated from a mouse λFixII 129/Sv library (Stratagene) was cloned into a pKO 901 Scrambler vector (Lexicon Genetics Incorporated) and used to create the targeting constructs shown in Fig. S1A and S4A. Pgk promoter-driven diphtheria toxin A fragment (dtA) was used to select against random integration. Additionally, a neomycin selection cassette flanked by FRT sites (Clcn5 *) or loxP sites (Clcn5.unc) was inserted. The linearized vectors were electroporated into E14 (Clcn5.unc) or R1 (Clcn5 *) embryonic stem cells. DNA from neomycin-resistant clones was digested using Acc65I and analyzed by Southern blot using an external 400 bp (Clcn5.unc) or 600 bp (Clcn5 *) probe. Cells from a correctly targeted clone were injected into C57Bl/6J blastocysts that were implanted into foster mothers. To remove the neomycin cassette chimeric males were bred with cre-recombinase expressing ‘deleter’ mice (2), resulting in the uncoupled Clcn5.unc allele, or with FLPeR mice (3) resulting in the Clcn5 * allele. Clcn5.unc and Clcn5 * mice were born at approximately Mendelian ratio and survived normally. Experiments were performed with mice in a mixed C57Bl/6-129/Svj genetic background, always using littermates as controls. Sequencing Targeting constructs were completely sequenced on ABI 3730 DNA analyzer (Applied Biosystems). The entire open reading frame of Clcn5.unc was amplified by RT-PCR in 5 pieces from mRNA extracted from kidneys of Clcn5.unc/y mice and sequenced to verify the presence of the E211A mutation and the absence of fortuitous other mutations. RNA expression analysis Total kidney RNA was extracted from adult Clcn5.unc and wild-type littermates using Trizol (Invitrogen) reagent and the High-Pure RNA purification kit (QIAGEN). RNA was transcribed into cDNA using the SuperScript II cDNA kit (Invitrogen) with random hexamers. Real-time PCR was performed using the 7900 HT cycler from Applied Biosystems and the SYBR Green Power PCR master mix (Applied Biosystems) and normalized to HPRT (hypoxanthine phosphoribosyltransferase). An initial denaturation step (10 min at 95°C) was followed by 40 cycles with two steps: 95°C for 15 s, followed by 60°C for 60 s. Each sample was amplified in duplicate and gave consistent results in two independent experiments. The value of the wild-type animals was set to 100. RNA values from Clcn5.unc animals are shown as % of the wild-type. Contamination with genomic DNA was negligible. Western blot 30 µg of kidney membrane preparation (140,000 x g pellet) were separated by SDS-PAGE using 8.5% acrylamide gels. For megalin analysis we used 3-8% NuPAGE Tris-acetate gels with NuPAGE Tris-Acetate SDS Running Buffer (Invitrogen).

2

In vivo endocytosis experiments Bovine β-lactoglobulin (Sigma Aldrich) labeled with Alexa Fluor 546 (Invitrogen) or Alexa Fluor 488-dextran (10 kDa; Invitrogen) was injected into the vena cava of anesthetized mice. After 7 minutes kidneys were perfused with PBS for 3 minutes and subsequently with 4% PFA. Immunohistochemistry and antibodies Mice were fixed by perfusion with 4% (w/v) PFA, tissues dissected and sucrose embedded for cryosectioning. The following primary antibodies were used: rabbit PEP5A (N-terminus, ClC-5NT) and PEP5E (C-terminus, ClC5CT) against ClC-5 (4), rabbit anti-ClC-3 (ab#1035) and rabbit anti-ClC-4 (ab#1261) as described (5), guinea pig anti-AE1 (6), mouse anti-villin (Acris), guinea pig anti-megalin (7), rabbit anti-cubilin (8), rabbit anti-β-actin (Sigma Aldrich). The antibody against NaPi-2a was generated in rabbits against the two peptides CYARPEPRSPQLPPRV and CPSPRLALPAHHNATRL, and the rabbit C5/05A antibody against the peptide CQDPDSILFN representing the Cterminus of ClC-5 (ClC-5CT). Peptides were coupled to keyhole limpet hemocyanin (KLH) via the C-terminalcysteine residues. Guinea pig ClC-5 antibody was raised against the same amino-terminal peptide as PEP-5A (1). Specificity of affinity-purified antibodies was checked using Clcn5-- mice. Secondary antibodies conjugated to Alexa Fluor 488, 546 or 633 (Invitrogen) or HRP (Jackson ImmunoResearch) were used. Image acquisition All images were acquired using a confocal microscope (LSM510, Zeiss). Serum and urine analysis Blood was drawn from the retro-orbital sinus under light ether anesthesia. Mice were kept in metabolic cages for urine collection. Blood and urine proteins, glucose and salts were determined by Dr. D. Becker (Institut für Labormedizin, Helios Klinikum Berlin-Buch). Urine pH was measured with a standard glass electrode. Serum 25(OH)-D3 and 1,25(OH)2-D3 were determined using ELISA assays (Immunodiagnostic Systems GmbH). Urine samples were normalized to creatinine values and analyzed by SDS–PAGE followed by silver staining or Western blotting using polyclonal antibodies against retinol binding protein (from W.S. Blaner, Cleveland) or vitamin D binding protein (Dako). Acidification assay For ATP-driven acidification assay endosome-enriched vesicles were prepared as described previously (9). Briefly, vesicles were prepared from kidneys from 2-4 mice of the same genotype and pooled. Animals were killed, kidneys were taken out and the renal cortex (consisting of ~80% proximal tubules) was dissected in ice cold PBS. All the subsequent steps were performed at 4 °C. When pooling kidneys from 4 anim als, the tissue was homogenized in 30 ml of homogenization buffer (300 mM mannitol, 12 mM HEPES/Tris pH 7.4) with 20 strokes in a loose-fitting glass/Teflon Potter homogenizer (1,200 rpm) (B. Braun). After homogenization, the suspension was centrifuged at 2,500 x g for 15 min. The pellet was discarded and the supernatant was centrifuged at 20,000 x g for 20 min. Most of the resulting

3

supernatant (S) was decanted and saved. The rest of the supernatant (about 1 ml) was used to disperse the fluffy upper part of the pellet by careful swirling of the tube. Supernatant S and the dispersed fluffy pellet were combined and centrifuged at 48,000 x g for 30 min. After centrifugation supernatant was discarded. Pellets containing crude plasma membranes and endocytic vesicles were resuspended and centrifuged at 48,000 x g for 30 min in a 16% Percoll gradient to separate endocytic vesicles from other membranes. After centrifugation the lowest 3 ml were taken and 27 ml of a potassium buffer (300 mM mannitol, 100 mM K+ gluconate, 5 mM MgSO4, 5 mM HEPES/Tris pH 7.0) were added. This vesicle suspension was incubated for 30 minutes on ice and subsequently centrifuged at 48,000 x g for 30 min. Supernatant was discarded. The pellet was resuspended in 1 ml potassium buffer and centrifuged at 2,500 x g for 15 min. The pellet, which contains endocytic vesicles, was resuspended in 50 µl of potassium buffer. The vesicle suspension was kept at 0-4° C at a concentration of 4 mg/ml for a maximum of 2 hours before performing the acidification assay. A vesicle suspension aliquot containing 200 µg protein (as determined by BCA assay) was suspended in a chloride buffer (maintained at 37 oC) containing: 300 mM mannitol, 100 mM KCl, 5 mM MgSO4, 6 µM acridine orange, 5 mM HEPES/Tris, pH 7.0. The pH-sensitive fluorescence of acridine orange was monitored during continuous stirring with a Xenius spectrofluorometer (Safas) (excitation 492 nm, emission 530 nm). Once fluorescence baseline was stable, the assay was started. Na-ATP (final concentration 1.5 mM) and FCCP (final concentration 10 µM) were injected in the cuvette using an automatic injector system. The immediate increase in acridine orange fluorescence observed after addition of ATP is an artifact due to the interaction of the dye and ATP (10). For evaluation and comparing acidification (Fig. 1C, D), we included only measurements in which the peak upon ATP addition was between 1.0 and 1.3 fluorescence ratios, indicating that speed of mixing and other parameters were in a similar range. Most of the acidification assay experiments were performed by experimenters blinded to the genotype. Electrophysiological analysis of ClC-5 mutants expressed in Xenopus oocytes Pieces of ovary were obtained by surgery from deeply anesthetized (0.1% tricaine; Sigma-Aldrich) Xenopus laevis frogs. Oocytes were prepared by manual dissection and collagenase A (Roche Applied Science) digestion. 5 ng of wild-type or negative-tagged ClC-5 cRNA were injected into oocytes which were kept in ND96 solution (containing 96 mM NaCl, 2 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2, 5 mM HEPES, pH 7.5) at 17 °C for 2 days. Two-electrode voltage-clamping was performed at room temperature (20–24 °C) using a TEC10 amplifier (npi Electronics) and pClamp9 software (Molecular Devices). The standard bath solution contained 96 mM NaCl, 2 mM K+ gluconate, 5 mM Ca2+ D-gluconate, 1.2 mM MgSO4, 5 mM HEPES, pH 7.5. Data analysis and statistical methods For figure 1 data were analyzed using software SAFAS SP2000 (SAFAS) and Origin 7.5 (OriginLab Corporation). Significance was determined using a twotailed Student’s t-test

4

Supporting figures and legends

Fig. S1. Knock-in mice converting ClC-5 into a chloride conductor. (A) Models for endosomal acidification. In the classical model (a) a Cl- channel supplies countercurrent for the proton pump, but in reality a Cl-/H+-exchanger performs this role (b). An uncoupling point mutation converts ClC-5 into a pure conductor (c), going back to ‘classics’ (a). In the knock-out (d) luminal charge accumulation blocks acidification. (B) CLC topology model indicating positions of the uncoupling E211A (unc) mutation and the negative tag (*) in which the extreme C-terminus of ClC-5 is converted into that of ClC-3 (fig S5). (C) ClC-5 variants expressed in WT and knock-in mice and their recognition by our Nterminal and C-terminal antibodies.

5

Fig. S2. Generation of knock-in mice carrying the uncoupling Clcn5.unc allele. (A) Targeting strategy. The targeting construct (top) contained 13 kb of mouse genomic sequence that was modified by inserting the uncoupling E211A mutation into exon 6 and a neomycin selection cassette (flanked by loxP sites) between exons 6 and 7. The Acc65I restriction site that was used for neomycin cassette insertion was destroyed in the cloning process. A diphtheria toxin A (dtA) cassette was added at the 5’ end to select against random integration. Removal of the neomycin cassette in vivo resulted in the uncoupled Clcn5.unc allele (bottom). (B) Southern blot analysis of Acc65Idigested genomic DNA (left) and genotyping PCR (right) using the hybridization probe and PCR primers, respectively, indicated in (A). The sequences of PCR primers were: 5´-ATGTGTGCAGCAGATGTGTGCC-3´, 5´CAACCTTCAGCACCACAAAAGC-3´. (C) DNA sequence obtained from RTPCR-amplified kidney mRNA confirms the presence of the E211A mutation. (D) Quantitative real-time PCR on mRNA extracted from WT and Clcn5.unc/unc kidneys shows similar mRNA levels for the WT and Clcn5.unc allele. RNA levels are shown as % ± SD of WT values that were set to 100 (2 animals per genotype). Primer pairs used: 5´-CTTCATGTACGTCCTCTGGGCTCTTCTGTT-3´ and 5´-CCAGGGGGCCCGCCTTGCCCAGGCTCAAGCC-3´; 5´-GCCTGGGCAAGGCGGGCCCCCTGGTGCACGTGGC and 5´-GGAAGAGGTCAGCTACTAC-3´. 6

Fig. S3. Localization of ClC-5 in intercalated cells. Identical staining pattern using the C-terminal PEP5E ClC-5 antibody (4) in distal tubular intercalated cells of WT and Clcn5.unc/y mice. Intercalated cells show basolateral staining (green) for the anion exchanger AE1. Scale bar 2.7 µm.

Fig. S4. Expression of ClC-3 and ClC-4 in kidneys of Clcn5 +, Clcn5 - and Clcn5.unc mice. Membrane proteins from kidneys of the indicated genotypes were separated by SDS-PAGE and analyzed in Western blot experiments with antibodies against ClC-3 and ClC-4 (ab#1035 and #1261 described by Maritzen et al. (5)). There is no upregulation of these close homologues of ClC-5 neither in Clcn5 - nor in Clcn5.unc kidneys.

7

Fig. S5. Generation of knock-in mice carrying the ‘negatively tagged’ Clcn5 * allele. (A) Targeting strategy. Top, targeting construct containing 12.5 kb of mouse genomic DNA, which has been modified by changing four base pairs in exon 12 that mutate two amino-acids of the extreme carboxyterminus (indicated by an asterisk "*"). An Acc65I site was inserted between exons 9 and 10 for screening purposes. Additionally, a neomycin selection cassette flanked by FRT sites has been inserted into the intron preceding exon 12, and a diphtheria toxin A (dtA) cassette has been added to the 5’ end to select against non-homologous recombination. In vivo excision of the selection cassette resulted in the Clcn5** allele (bottom). PCR primers used for genotyping were 5´-GCTCTCGTAAAGCATTTACTATGC-3´ and 5´-GTATAGGCTGAGGAGAGTGTT-3´. (B) Sequence comparison of the extreme C-

8

terminus of mouse ClC-3, -4, and -5 proteins that form a distinct branch of the CLC gene family. Non-conserved amino-acids are shown in blue. Since two antibodies generated against C-terminal peptides of ClC-5 (the PEP5E antibody (4) and the newly generated C5/05A antibody) recognize ClC-5, while showing almost no cross-reactivity against ClC-3, we created a ‘negatively tagged’ ClC-5 (ClC-5*) by minimally changing its C-terminal sequence by converting it to that of ClC-3. This minimal alteration of ClC-5 did not affect its function. (C) Confirmation of changed C-terminus by sequencing genomic DNA from WT and Clcn5** mice. (D) Unchanged electrophysiological properties of the ClC-5* mutant transporter. Xenopus oocytes were injected with cRNA (5 ng) and examined by two-electrode voltage-clamping as described previously (11), using a protocol that clamped the voltage to values between -125 and + 100 mV. Typical original traces are shown at left, and mean currents at + 100 mV at right (n=10, ± SEM). These currents reflect electrogenic 2Cl-/H+-exchange. Mean currents and original traces are modified from (12).

9

Fig. S6. Analysis of ‘negatively tagged’ Clcn5** cells. (A) Subcellular localization of WT (left) and the negatively tagged ClC-5* (right) protein in proximal tubular cells. No difference in localization was observed. (B) In vivo endocytosis of Alexa Fluor 546-labeled β-lactoglobulin in proximal tubules of female Clcn5:+/* mice expressing side by side either WT ClC-5 (labeled Clcn5:+) or ‘negatively tagged’ ClC-5* (labeled Clcn5**). Only ClC-5+ is recognized by the C-terminal ClC-5 antibody C5/05A, resulting in green staining (left panels). Clcn5**- and Clcn5:+-expressing cells show indistinguishable accumulation of β-lactoglobulin (stained in red, center panels) which was stopped after a 10 min uptake period. Right panels show overlay, with additional staining for the brush border protein villin in blue. Scale bar, 6 µm for (A) and 5 µm for (B).

10

Fig. S7. NaPi-2a expression in proximal tubule cells. Reduced expression and intracellular localization of NaPi-2a in proximal tubular cells expressing ClC-5unc instead of WT ClC-5. Right, similar reduction of NaPi-2a in Western blot of Clcn5 -/y and ClC-5unc/y kidney membranes. Scale bar, 5 µm.

Fig. S8. Fluid-phase endocytosis in Clcn5.unc and Clcn5 - proximal tubule cells. No difference in uptake of labeled dextran between cells lacking ClC-5 (Clcn5 -) or expressing Clcn5.unc in a chimeric tubule from a Clcn5.unc/- female. Scale bar, 5 µm.

11

Table S1 Serum and urinary parameters

A Clcn5 +/y

Clcn5.unc/y

serum concentrations (mmol/l) Ca Pi Na Cl K glucose creatinine (µmol/l) insulin (pg/ml)

2.56 ± 0.13 (10) 2.33 ± 0.38 (10) 153.07 ± 5.31 (10) 115.84 ± 4.73 (10) 5.28 ± 0.50 (10) 7.88 ±1.04 (10) 14.75 ± 4.06 (10) 2490 ± 935 (10)

2.64 ± 0.16 (10) 2.77 ± 0.64 (10) 157.48 ± 6.13 (10) 111.12 ± 5.38 (10) 5.42 ± 0.26 (10) 5.96 ±1.49 (10) 15.57 ± 6.97 (10) 2261 ± 1103 (10)

B Clcn5.+/y

Clcn5.unc/y

Clcn5 +/y

Clcn5 -/y

urine concentrations (mmol/mmol creatinine) Ca/creatinine Pi/creatinine Na/creatinine Cl/creatinine K/creatinine Mg/creatinine glucose/creatinine

1.77 ± 0.15 (40) 6.04± 0.87 (40) 56.90 ± 2.63 (20) 112.84 ± 7.53 (20) 108.33 ± 4.30 (20) 6.46 ± 0.69 (20) 1.30± 0.21 (40)

3.19 ± 0.20 ***(40) 11.49± 0.88 ***(40) 59.39 ± 2.44 (20) 93.52 ± 8.30 (20) 129.07 ± 5.63 *(20) 9.28 ± 0.87 (20) 4.53 ± 0.82 ***(40)

1.88 ± 0.12 (20) 5.01 ± 0.79 (20) 55.86 ± 1.57 (20) 128.08 ± 4.06 (20) 125.95 ± 2.76 (20) 8.84 ± 0.41 (20) 1.36 ± 0.16 (20)

2.39 ± 0.21 (20) 7.98 ± 1.13* (20) 53.18 ± 1.78 (20) 106.88 ± 2.67 (20) 110.42 ± 3.44 * (20) 8.44 ± 0.74 (20) 2.74 ±0.44 * (20)

creatinine (mmol/l) insulin (pg/ml) pH urinary volume (ml/day)

2.65 ± 0.27 (40) 161.6 ± 147 (20) 7.27 ± 0.28 (20)

1.05 ± 0.27 **(40) 5160 ± 1761* (20) 6.25 ± 0.33 * (20)

2.23 ± 0.29 (20) 223 ± 79 (20) 7.20 ± 0.09 (20)

2.11 ± 0.56 (20) 5166 ± 288* (20) 6.81 ± 0.07 * (20)

1.92 ± 0.20 (40)

4.90 ± 0.46 ***(40)

1.92 ± 0.12 (20)

2.36 ± 0.41 (20)

*** p< 0.0005 , ** p