Dimeric Architecture of the Human Bumetanide ... - Semantic Scholar

1 downloads 0 Views 252KB Size Report
Haven, CT) or an influenza hemagglutinin (HA) epitope (“YPYDVP-. DYA”) (29), was cloned at .... g for 1 h at 4°C to pellet undissolved membranes. The solubilized ..... Mount DB, Mercado A, Song L, Xu J, George AL Jr, Delpire E,. Gamba G: ...
J Am Soc Nephrol 14: 3039–3046, 2003

Dimeric Architecture of the Human Bumetanide-Sensitive Na-K-Cl Co-transporter PATRICK G.J.F. STARREMANS,* FERRY F.J. KERSTEN,* LAMBERTUS P.W.J. VAN DEN HEUVEL,† NINE V.A.M. KNOERS,‡ and RENE´ J.M. BINDELS* Departments of *Physiology, †Pediatrics, and ‡Human Genetics, University Medical Centre Nijmegen, Nijmegen, The Netherlands

Abstract. The primary mediator of NaCl reabsorption in the renal distal tubule is the human bumetanide-sensitive Na⫹-K⫹2Cl⫺ co-transporter (hNKCC2), located at the apical membrane of the thick ascending limb of Henle’s loop. The physiologic importance of this transporter is emphasized by the tubular disorder Bartter syndrome type I, which arises from the functional impairment of hNKCC2 as a result of mutations in the SLC12A1 gene. The aim of the present study was to investigate the oligomeric state of hNKCC2 to understand further its operational mechanism. To this end, hNKCC2 was heterologously expressed in Xenopus laevis oocytes. Chemical cross-linking with dimethyl-3,3-dithio-bis-propionamidate indicated that hNKCC2 subunits can reversibly form high molecular weight complexes. Co-immunoprecipitation of tagged hNKCC2 subunits further substantiated a physical interaction

between individual hNKCC2 subunits. The size of the hNKCC2 multimers was determined by sucrose gradient centrifugation, and a preference for dimeric complexes (approximately 320 kD) was demonstrated. Finally, concatemeric constructs consisting of two wild-type subunits or a wild-type and a functionally impaired hNKCC2 subunit (G319R) were expressed in oocytes. Subsequently, the concatemers were functionally characterized, resulting in a significant bumetanidesensitive 22Na⫹ uptake of 2.5 ⫾ 0.2 nmol/oocyte per 30 min for the wild-type–wild-type concatemer, which was reduced to 1.3 ⫾ 0.1 nmol/oocyte per 30 min for the wild-type–G319R concatemer. In conclusion, this study suggests that hNKCC2 forms at least functional dimers when expressed in Xenopus laevis oocytes of which the individual subunits transport Na⫹ independently.

The primary mediator of NaCl reabsorption in the renal distal tubule is the bumetanide-sensitive Na⫹-K⫹-2Cl⫺ co-transporter (hNKCC2), located at the apical membrane of the thick ascending limb of Henle’s loop (TAL) (1). This co-transporter is responsible for ⬎25% of the active sodium reabsorption in the kidney. It is, therefore, an important factor in the regulation of the circulating fluid volume and in long-term BP control. Through the innate sensitivity for loop diuretics, NKCC2 has been pinpointed as a prime target in the treatment of hypertension. The physiologic importance of TAL is further illustrated by the fact that this nephron segment plays a significant role in the urinary concentrating mechanism as a result of the uncoupling of water and NaCl reabsorption (2). In addition, malfunctioning of this nephron segment can have severe clinical consequences. Impairment of the key transport proteins in TAL, including NKCC2, the ATP-sensitive K⫹ channel (ROMK), the Cl⫺ channel (ClC-Kb), and its associated subunit

Barttin, results in the severe tubular transport disorder Bartter syndrome (3–7). NKCC2 (SLC12A1) is a member of the SLC12A superfamily of electroneutral cation-coupled co-transporters encompassing two Na⫹-K⫹-2Cl⫺ co-transporters (NKCC1 and NKCC2) (8,9), the Na⫹-Cl⫺ co-transporter (NCC) (10,11), and at least four K⫹-Cl⫺ co-transporters (KCC1 to 4) (12–16). Genes encoding these transmembrane proteins are highly homologous and share a common predicted membrane topology of 12 transmembrane domains with both N- and C-terminus located intracellularly (17,18). In addition, the SLC12A1 gene encodes six different isoforms of NKCC2. These splice variants are created by a combination of three separate exon 4 cassettes (a, b, and f) and two alternative C-termini. This results in three long (1099 aa) and three short or truncated isoforms (770 aa) (19,20). These different isoforms display axial expression along the TAL and show significant differences in kinetic behavior, consistent with their spatial distribution as they reabsorb Na⫹, K⫹, and Cl⫺ from progressively diluted luminal fluid (21,22). The short isoforms, when expressed separately, are able to exhibit only a K⫹-independent mode of NaCl transport under hypotonic conditions (23). However, when coexpressed with their longer relatives in Xenopus laevis oocytes, they exert a negative effect on the latter, which in turn can be abolished by cAMP (24). These studies suggest that an interaction between individual NKCC2 subunits can occur. It has been shown that other family members, including the

Received April 19, 2003. Accepted September 5, 2003. Correspondence to Dr. René J.M. Bindels, 160 Cell Physiology, University Medical Centre Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. Phone: ⫹31-24-3614211; Fax: ⫹31-24-3616413; E-mail: [email protected] 1046-6673/1412-3039 Journal of the American Society of Nephrology Copyright © 2003 by the American Society of Nephrology DOI: 10.1097/01.ASN.0000097370.29737.5B

3040

Journal of the American Society of Nephrology

basolateral Na⫹-K⫹-2Cl⫺ co-transporter NKCC1 (SLC12A2) and the K⫹-Cl⫹ co-transporter KCC1 (SLC12A4), can form multimeric proteins (25,26). Taken together, these findings indicate that NKCC2 could function as a multimeric protein. This could have important implications for the interpretation of identified heterozygous mutations in Bartter syndrome, which shows a recessive mode of inheritance (3,27). The aim of the present study was to assess the oligomeric state of hNKCC2 to understand further its operational mechanism. To this end, HA- and FLAG-tagged NKCC2 constructs were generated and heterologously expressed in Xenopus laevis oocytes. Subsequently, four independent techniques, including chemical cross-linking, co-immunoprecipitation, density gradient centrifugation, and the analysis of concatemeric proteins, were used to determine the quaternary structure of this Na⫹-K⫹-2Cl⫺ co-transporter.

J Am Soc Nephrol 14: 3039–3046, 2003

Immunoblot Analysis Aliquots of dissolved total membranes were subjected to SDSPAGE electrophoresis (7% wt/vol) and electroblotted onto polyvinylidene difluoride membranes (Millipore, Bedford, MA). Blots were incubated with 5% (wt/vol) nonfat dried milk (NFDM) in Trisbuffered saline (TBS-T; pH 7.4 containing 0.2% [wt/vol] tween-20). Immunoblots were incubated overnight at 4°C with primary antibodies including mouse anti–FLAG-peroxidase coupled antibodies (Sigma Chemical Co., St. Louis, MO), 1:2000, 5% (wt/vol) NFDM in TBS-T, mouse anti-HA (Roche, Indianapolis, IN), 1:4000, 1% (wt/ vol) NFDM in TBS-T or mouse anti–HA-peroxidase (Roche), 1:1000, 1% (wt/vol) NFDM in TBS-T. After washing, HA blots were incubated at room temperature with the corresponding secondary antibody sheep anti-mouse IgG peroxidase (Sigma), 1:2000, in TBS-T, and immunopositive bands were visualized using an enhanced chemiluminescence system (Pierce, Rockford, IL).

Cross-Linking of Proteins

Materials and Methods Synthesis of Tagged hNKCC2a Constructs NKCC2a cDNA was obtained from a human kidney cDNA library (Clontech Laboratories, Palo Alto, CA) by means of PCR and cloned into a pGEM-Teasy vector (Promega, Madison, WI). The coding sequence was subcloned into a custom oocyte expression vector, pTLN (28). A FLAG-epitope (“DYKDDDDK”; IBI, Kodak, New Haven, CT) or an influenza hemagglutinin (HA) epitope (“YPYDVPDYA”) (29), was cloned at the 5' end of the wild-type construct replacing the original ATG to allow distinction between subunits during the immunoprecipitation experiments. All constructs were checked by double-stranded sequence analysis. No cross-reactivity of HA-tagged hNKCC2a with FLAG-antibody or FLAG-tagged hNKCC2a with HA-antibody was observed during the immunoblotting experiments.

Preparation and Injection of Oocytes Oocytes were obtained from Xenopus laevis and defolliculated by incubation for 2 h in modified Barth’s solution (MBS; 88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO3, 10 mM HEPES-Tris [N-2-hydroxyethylpiperazine-N'-2-ethane-sulfonic acid] pH 7.4, 0.8 mM MgSO4, 0.3 mM Ca[NO3]2, 0.4 mM CaCl2, and 25 ␮g/ml gentamycin) containing 2 mg/ml collagenase A (Roche Molecular Biochemicals, Mannheim, Germany). Stage V and VI oocytes were selected and stored at 18°C in MBS. G-capped RNA transcripts were synthesized in vitro from Mlu I-linearized human NKCC2 templates using SP6 RNA polymerase. cRNA integrity was checked by agarose gel electrophoresis, and their concentrations were determined using spectrophotometric analysis. Defolliculated oocytes were injected with 50 nl of water containing 0 to 25 ng of cRNA and incubated 72 h at 18°C in MBS.

Isolation of Total Membranes For isolation of total membranes (plasma and subcellular membranes), 10 to 50 oocytes were homogenized in 1 ml of homogenization buffer (20 mM Tris-HCl [pH 7.4], 5 mM MgCl2, 5 mM NaH2PO4, 1 mM EDTA, 80 mM sucrose, 1 mM PMSF, 5 ␮g/ml leupeptin, and 5 ␮g/ml pepstatin) and centrifuged two times for 10 min at 100 ⫻ g and 4°C to remove yolk proteins. Next, the total membranes were pelleted by centrifugation at 16,200 ⫻ g at 4°C for 30 min and subsequently dissolved in Laemmli-buffer (2 ␮l/oocyte) and incubated 30 to 60 min at 37°C as described previously (30).

Total membrane preparations of oocytes expressing FLAGhNKCC2 were resuspended and incubated for 30 min at 37°C in cross-linking buffer (0.5% [wt/vol] sodium desoxycholate, 20 mM HEPES-NaOH [pH 7.2], 5 mM KCl, 130 mM NaCl, 10% [vol/vol] glycerol, 5 mM EDTA, and protease inhibitors). Samples were divided into three equal amounts. Two parts were treated with 2 mM dimethyl-3,3'-dithio-bispropionamidate (DTBP), a cleavable crosslinker with an 11.9 Å arm, in cross-linking buffer and incubated for 60 min on ice. Subsequently, cross-linking was terminated by the addition of 100 mM Tris-HCl (pH 6.8), and samples were incubated for 30 min on ice. Samples were incubated in Laemmli buffer for 30 min at 37°C with and without 100 mM dithiothreitol (DTT). As a control, the third part was not treated with DTBP.

Co-immunoprecipitation of FLAG-NKCC2 with HA-NKCC2 Twenty-microliter equivalents of protein A– coupled agarose beads (Pharmacia, Uppsala, Sweden) were preincubated for 16 h at 4°C with 2 ␮l of monoclonal HA antibody (Roche) in 0.7 ml of IPP500 (500 mM NaCl, 10 mM Tris-HCl [pH 8.0], 0.1% [vol/vol] NP-40, 0.1% [vol/vol] Tween-20, 1 mM PMSF, and 5 mg/ml leupeptin and pepstatin) and 0.1% (wt/vol) BSA. The beads were washed three times with IPP100 (100 mM NaCl, 10 mM Tris-HCl [pH 8.0], 0.1% [vol/ vol] NP-40, 0.1% [vol/vol] Tween-20, 1 mM PMSF, and 5 mg/ml leupeptin and pepstatin). Isolated total membranes of 15 oocytes expressing HA-NKCC2 or FLAG-NKCC2 or coexpressing both were incubated for 1 h at 37°C in 50 ␮l of solubilization buffer (20 mM Tris-HCl [pH 6.8], 10% [vol/vol] glycerol, 5 mM EDTA, 1% (wt/vol) sodium-desoxycholate, 1 mM PMSF, and 5 mg/ml leupeptin and pepstatin) and centrifuged at 16,000 ⫻ g for 1 h at 4°C to pellet undissolved membranes. The solubilized membranes were diluted with 700 ␮l of sucrose buffer (100 mM NaCl, 20 mM Tris-HCl [pH 6.8], 0.5 mM EDTA, 0.1% [vol/vol] Triton X100, 10% [wt/vol] sucrose, 1 mM PMSF, and 5 ␮g/ml leupeptin and pepstatin), added to the washed antibody-bound protein A beads and incubated for 16 h at 4°C. After incubation, the beads were washed three times with IPP100, incubated in 25 ␮l of Laemmli buffer for 1 h at 37°C, and subjected to immunoblotting (30).

Co-immunoprecipitation of HA-NKCC2 with FLAG-NKCC2 A total of 25 ␮l of anti-FLAG M2 affinity Gel Freezer-Safe beads (Sigma Chemical Co.) was washed three times with IPP100 (100 mM

J Am Soc Nephrol 14: 3039–3046, 2003

Dimeric Architecture of hNKCC2

NaCl, 10 mM Tris-HCl [pH 8.0], 0.1% [vol/vol] NP-40, 0.1% [vol/ vol] Tween-20, 1 mM PMSF, and 5 mg/ml leupeptin and pepstatin). Isolated total membranes of 23 oocytes expressing HA-NKCC2 or FLAG-NKCC2 or coexpressing both were incubated for 1 h at 37°C in 50 ␮l of solubilization buffer (20 mM Tris-HCl [pH 6.8], 10% [vol/vol] glycerol, 5 mM EDTA, 0,2% [vol/vol] Triton X-100, 1 mM PMSF, and 5 mg/ml leupeptin and pepstatin) and centrifuged at 16,000 ⫻ g for 1 h at 4°C to pellet undissolved membranes. The solubilized membranes were diluted with 700 ␮l of sucrose buffer (100 mM NaCl, 20 mM Tris-HCl [pH 6.8], 0.5 mM EDTA, 0.1% [vol/vol] Triton X100, 10% (wt/vol) sucrose, 1 mM PMSF, and 5 ␮g/ml leupeptin and pepstatin), added to the antibody-bound beads, and incubated for 16 h at 4°C. After incubation, the beads were washed three times with IPP100, incubated in 25 ␮l of Laemmli buffer for 1 h at 37°C, and subjected to immunoblotting.

Sedimentation by Sucrose Gradient Centrifugation Total membranes of 100 oocytes injected with 25 ng of FLAGhNKCC2 cRNA were incubated in solubilization buffer (1% [wt/vol] sodium-desoxycholate, 20 mM Tris-HCl [pH 6.8], 5 mM EDTA, 10% [wt/vol) glycerol, 1 mM PMSF, and 5 mg/ml leupeptin and pepstatin) for 1 h at 37°C and subsequently centrifuged for 1 h at 100,000 ⫻ g and 4°C to pellet undissolved membranes (31). Samples were supplemented with gradient buffer (20 mM Tris-HCl [pH 6.8], 5 mM EDTA, 0.1% [vol/vol] Triton X-100, 1 mM PMSF, and 5 mg/ml leupeptin and pepstatin) to 300 ␮l. Sedimentation by gradient centrifugation was done essentially as described by Jung et al. (32). Solutions of 10, 15, 20, 25, 30, and 35% (wt/vol) sucrose in gradient buffer were prepared. The dissolved membrane samples were loaded onto the gradient and subjected to 150,000 ⫻ g centrifugation for 16 h at 8°C. Then 200-␮l fractions, designated 1 to 16, were collected carefully and analyzed by immunoblotting. As sedimentation markers, a mixture of phosphorylase B (97 kD), yeast alcohol dehydrogenase (150 kD), ␤-amylase (200 kD), catalase (232 kD), and apoferritin (443 kD) was used. All markers were obtained from Sigma.

Expression of Concatemeric hNKCC2 cDNA Constructs

3041

30 min, with or without 100 ␮M bumetanide (21). Oocytes were then washed five times with ice-cold uptake medium without inhibitors, subsequently solubilized individually in 200 ␮l of 10% (wt/vol) SDS and counted in a Beckmann LS-600 liquid scintillation counter (Beckmann-Coulter, Fullerton, CA). The data were expressed as the mean ⫾ SEM. The statistical significance was determined by t test, and P ⬍ 0.05 was considered significant.

Immunocytochemistry After removal of the follicle membranes, oocytes were fixed in 1% (wt/vol) paraformaldehyde solution for 2 h (35,36), washed twice in 80% (vol/vol) ethanol, and embedded in paraffin using a Citadel Tissue Processor and Histocenter 2 (Shandon Southern Products Ltd, Cheshire, UK). Seven-micron sections were cut, deparaffinized, and incubated for 30 min in TN (100 mM Tris-HCl [pH 7.6], 150 mM NaCl) and subsequently blocked in TNB (TN containing 0.5% [wt/ vol] blocking reagent from Renaissance TSA-direct kit; NEN Lifescience Products, Boston, MA) for 1 h at room temperature. Sections were subsequently incubated overnight at 4°C with 1:200 diluted mouse M2 anti-FLAG monoclonal (Sigma-Aldrich, St. Louis, MO) in TNB. After three washes in TNT (TN containing 0.05% [wt/vol] Tween-20), sections were stained for 1 h at room temperature with a 1:300 diluted Alexa 488 conjugated anti-mouse IgG (Molecular Probes, Eugene, OR) in TNB. Finally, sections were washed three times in TNT, dehydrated in subsequently 50% (vol/vol) and 100% (vol/vol) methanol, and mounted in Vectashield (Vector Laboratories, Burlingame, CA). Digital images were made using a MRC-1000 confocal laser scanning microscope (Biorad, Richmond, VA).

Results Cross-Linking of hNKCC2 Subunits Chemical cross-linking of hNKCC2, heterologously expressed in oocytes, showed that hNKCC2 monomers disappeared upon treatment with DTBP, whereas oligomeric complexes with a molecular mass of 300 to 400 kD appeared (Figure 1). DTBP contains a cleavable spacer, allowing the

Concatemeric constructs were produced by linking the coding sequences of two FLAG-hNKCC2a subunits in a head-to-tail manner. By replacing the stop codon of one FLAG-hNKCC2a with a linker of six glycines followed by a unique EcoRV restriction site, the in-frame insertion of a second FLAG-hNKCC2a subunit was enabled. Two FLAG-hNKCC2a dimers were constructed, a wild-type–wild-type and a wild-type–mutant configuration. For the wild-type–mutant configuration, the G319R mutant was selected. This mutant is functionally impaired but properly routed to the plasma membrane as described previously (33). The dimeric constructs were verified by restriction digestion and were heterologously expressed in oocytes as mentioned above.

Na⫹ Uptake Assay

22



The oocytes where transferred to a Cl -free medium (96 mM Na-gluconate, 2 mM K-gluconate, 1.8 mM Ca-gluconate, 2.5 mM Na-pyruvate, 5 mM HEPES-Tris [pH 7.4], 1 mM Mg[NO3]2, and 50 ␮g/ml gentamycin) 16 to 20 h before the uptake experiment (34). Ten to 15 Cl⫺-depleted oocytes were then transferred to 500-␮l uptake medium (41 mM N-methyl-D-glucamine–HCl [pH 7.4], 38 mM NaCl, 10 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2, and 5 mM HEPES-Tris [pH 7.4]; 100 ␮M amiloride; 100 ␮M hydrochlorothiazide; and 100 ␮M ouabain) containing 3 ␮Ci of 22Na⫹/ml and incubated at 30°C for

Figure 1. Cross-linking of human bumetanide-sensitive Na⫹-K⫹2Cl⫺ co-transporter (hNKCC2) heterologously expressed in oocytes. Total membranes of 45 noninjected (A to C) and 25 ng of FLAGhNKCC2 injected oocytes (D to F) were isolated as described in Materials and Methods. One aliquot of each group was treated with the chemical cross-linker dimethyl-3,3'-dithio-bispropionamidate (DTBP; B and E), treated with DTBP and DTT (C and F), or used as untreated control (A and D). Samples were subjected to immunoblotting and stained with the mouse M2 anti-FLAG antibody (n ⫽ 3 independent experiments).

3042

Journal of the American Society of Nephrology

conjugate to be broken by DTT. Indeed, incubation of the cross-linked hNKCC2 complexes with DTT revealed recurrence of the monomers. Under similar conditions, no signal was detected for total membranes of noninjected oocytes.

Co-immunoprecipitation of HA-NKCC2 with FLAG-NKCC2 The results from the cross-link experiments indicated that hNKCC2 might be able to form multimeric complexes. To substantiate further these results, we investigated whether hNKCC2 subunits carrying HA or FLAG epitope tags could be co-immunoprecipitated when co-injected into oocytes. First, total membranes were isolated from oocytes expressing HAhNKCC2, FLAG-hNKCC2, or both to demonstrate protein expression and specificity of the applied antibodies. Immunoblotting confirmed expression of both proteins, which were specifically detected by the HA and FLAG antibodies, respectively (Figure 2A). Subsequently, HA-hNKCC2 and FLAGhNKCC2 proteins were coexpressed and immunoprecipitated with HA antibodies. Immunoblots containing the complexes

Figure 2. Co-immunoprecipitation of FLAG-hNKCC2 and HANKCC2. cRNA of HA-hNKCC2 and/or FLAG-hNKCC2 was (co-)injected in oocytes. Subsequently, total membranes were isolated from injected oocytes and noninjected controls (Ni) and processed (n ⫽ 3 independent experiments). (A) Immunoblot analysis demonstrating that both tagged co-transporters are expressed and that the applied antibodies do not cross-react. Five oocyte equivalents were loaded on the blot. (B) Co-immunoprecipitations were performed with the HA antibody, and subsequently immunoblots were stained with a peroxidase-coupled anti-FLAG antibody. Fifteen oocyte equivalents were used in the co-immunoprecipitation experiment and loaded on the immunoblot. (C) Co-immunoprecipitations were performed with the FLAG antibody, and subsequently immunoblots were stained with a peroxidase-coupled anti-HA antibody. Twentythree oocyte equivalents were used in the co-immunoprecipitation experiment and loaded on the immunoblot.

J Am Soc Nephrol 14: 3039–3046, 2003

were probed with a peroxidase coupled FLAG antibody. The results shown in Figure 2B demonstrate that FLAG-hNKCC2 can be co-immunoprecipitated with the HA antibody. Similar results were obtained when the immunoprecipitation was performed with FLAG antibodies and the corresponding immunoblot was probed with HA antibodies (Figure 2C), thus confirming the formation of a multimeric hNKCC2 complex from both sites.

Determination of the hNKCC2 Complex Size Because the aforementioned experiments suggest that hNKCC2 can form a multimeric complex, the size of these complexes was subsequently estimated. To this end, total membranes were isolated from oocytes expressing hNKCC2, solubilized in 1% (wt/vol) sodium-desoxycholate, and subjected to sucrose gradient centrifugation. Immunoblotting of 16 fractions collected from the gradient revealed that the peak intensity of hNKCC2 complexes was located in fraction 10 (Figure 3A). The sedimentation marker proteins, phosphorylase B (97), alcohol dehydrogenase (150), ␤-amylase (200 kD), catalase (232 KD), and apoferritin (443 kD) were loaded on a parallel sucrose gradient (peak marker fractions are indicated by arrows in Figure 3A). A semilogarithmic plot of the peak marker fractions versus their molecular mass yielded the following formula to calculate the mass of the hNKCC2 complex:

Figure 3. Density centrifugation of hNKCC2 complexes. (A) Immunoblot loaded with sequentially collected fractions (5 to 16) of a sucrose gradient containing total membranes of oocytes expressing hNKCC2 and subjected to ultracentrifugation. Increasing band intensities with an optimum in fraction 10 was observed, which corresponds to a complex of approximately 320 kD. Also shown is a control of total membranes expressing hNKCC2 (WT). Arrows indicate peak fractions of marker proteins (n ⫽ 3). (B) Immunoblot loaded with sequentially collected fractions (5 to 16) of a sucrose gradient containing total membranes of oocytes expressing hNKCC2 in the presence of SDS and subjected to ultracentrifugation. A shift in peak intensity is seen from fraction 10 (in A) to fraction 5, corresponding to a molecular mass of approximately 150 kD (in B) as SDS dissolves the formed complexes. Also shown is a control of total membranes expressing hNKCC2 (WT; n ⫽ 3).

J Am Soc Nephrol 14: 3039–3046, 2003

Dimeric Architecture of hNKCC2

3043

M ⫽ 47.38e0.1974*fraction nr with R2 ⫽ 0.97, which indicated that hNKCC2 migrates as a complex with a molecular mass of approximately 340 kD, suggesting the formation of a dimeric complex. Sucrose gradient centrifugation in the presence of 0.1% (wt/vol) SDS reduced the molecular mass to approximately 160 kD, corresponding to the estimated size of the complex-glycosylated hNKCC2 monomer (Figure 3B).

Construction and Functional Analysis of Concatemeric hNKCC2 Dimers To investigate the functionality of dimeric hNKCC2, we constructed a concatemeric protein. Expression of wild-type hNKCC2 and the dimeric concatemer was confirmed by immunoblot, demonstrating specific bands at approximately 160 kD for the monomer and approximately 320 kD for the dimer (Figure 4A). Subsequently, functional analysis was performed, and the bumetanide-sensitive 22Na⫹ uptake was 2.6 ⫾ 0.2 and 2.5 ⫾ 0.2 nmol/oocyte per 30 min for the injected monomeric hNKCC2 and the wild-type–wild-type concatemer, respectively (Figure 4B). To gain additional information on the function of dimeric hNKCC2, we constructed a dimeric concatemer in which one of the subunits carried a mutation previously identified in Bartter syndrome type I. The G319R mutation was selected because it is normally processed in oocytes but functionally impaired as shown previously (33). Protein expression of both concatemers was confirmed by immunoblot (Figure 5A), and the bumetanide-sensitive Na⫹ transport activity of oocytes expressing the wild-type–wild-type or the wild-type–G319R dimer was measured, resulting in an uptake of 2.5 ⫾ 0.2 and 1.3 ⫾ 0.1 nmol/oocyte per 30 min, respectively (Figure 5B). Thus, the Na⫹ uptake was significantly reduced (P ⬍ 0.05) in the wild-type–G319R concatemer. To verify subcellular localization of the WT-G319R concatemer, we performed immunocytochemistry on oocytes injected with 25 ng WT-G319R

Figure 5. Functional analysis of a wild-type–mutant concatemer. (A) Total membranes of noninjected (Ni) controls, oocytes expressing wild-type–wild-type concatemer (WT-WT), or wild-type–G319R concatemer (WT-G319R) were separated on a 6% (vol/vol) SDSPAGE gel and immunoblotted. One oocyte equivalent was loaded in every lane. (B) Bumetanide-sensitive 22Na⫹ uptake of oocytes, injected with wild-type–wild-type (WT-WT) or wild-type–mutant (WTG319R) cRNA, measured after 30 min of incubation at 30°C and normalized for protein expression levels (n ⫽ 15 for three independent experiments; *P ⬍ 0.05). (C) Immunocytochemical analysis of sections of oocytes expressing the WT-G319R concatemer or noninjected controls (D).

cRNA (Figure 5C) or noninjected controls (Figure 5D). The results clearly show a plasma membrane localization for the WT-G319R concatemer similar to the previously published results for WT or G319R alone (33).

Discussion

Figure 4. Functional analysis of a dimeric hNKCC2 concatemer. (A) Total membranes of noninjected (Ni) controls, oocytes expressing FLAG-tagged hNKCC2a (WT), or hNKCC2 concatemer (WT-WT) were separated on a 6% (vol/vol) SDS-PAGE gel and immunoblotted. One oocyte equivalent was loaded in every lane. (B) Bumetanidesensitive 22Na⫹ uptake of oocytes, injected with 3 ng of wild-type hNKCC2 (WT) or 25 ng of concatemeric (WT-WT) cRNA, measured after 30 min of incubation at 30°C (n ⫽ 15 for three independent experiments).

The present study suggests that the human bumetanidesensitive Na⫹-K⫹-2Cl⫺ co-transporter is functionally present at the plasma membrane as a homodimeric complex. Four independent methods were combined to assess the multimeric state of the co-transporter complex. First, chemical crosslinking experiments revealed protein band shifts from monomeric hNKCC2 to multimeric compositions. Second, co-immunoprecipitation experiments demonstrated that hNKCC2 subunits are physically linked. Third, sucrose gradient centrifugation substantiated that hNKCC2 complexes have a molecular weight corresponding to a dimeric configuration. Fourth, tracer analysis of concatemeric proteins revealed that dimeric hNKCC2 complexes, when expressed in oocytes, exhibit a significant bumetanide-sensitive Na⫹ transport activity. The first indication that multimerization of hNKCC2 subunits occurs was obtained by cross-linking experiments. Addition of the chemical cross-linker DTBP resulted in a band shift of the hNKCC2 monomeric band to a larger configuration

3044

Journal of the American Society of Nephrology

(⬎300 kD). This indicated that hNKCC2 subunits are present within 11.9 Å (the maximal bridging length of the cross-linker) of each other. Earlier, a study by Moore-Hoon et al. (25) suggested that the secretory Na⫹-K⫹-2Cl⫺ co-transporter (NKCC1) forms a homodimeric complex. Recently, we applied a similar approach to show that the related thiazide-sensitive NaCl co-transporter (NCC) forms homodimers (37). Additional evidence of oligomerization was obtained by performing co-immunoprecipitation experiments using hNKCC2 subunits carrying different epitope tags. This proved that hNKCC2 subunits could physically interact when heterologously expressed in oocytes. In addition, sucrose gradient centrifugation experiments revealed that hNKCC2 sedimented predominantly as an approximately 320-kD complex, suggesting at least a dimeric configuration. The finding that dimerization occurs offers an explanation for the observations made by Meade et al. (20,38) stating that the short or truncated isoforms of NKCC2 are able to affect negatively the longer isoforms by an unknown mechanism. Similar results have been reported for KCC1, for which several truncated mutants where shown to co-immunoprecipitate with the wild-type proteins (26). For other family members, such as NCC and KCC3, it has been stated that dimerization occurs for wild-type and mutant subunits, whereas several of the latter can also be detected at the plasma membrane (37,39). The different isoforms of NKCC2 are expressed differentially along the distal tubule of the kidney, with the a isoform present in both the medullary and cortical segments of TAL, whereas the f and b isoforms are expressed predominantly in the medullary region and the macula densa, respectively (19). Together with our results, this strengthens the possibility that NKCC2 a isoforms could dimerize with either b or f in those cells where they are coexpressed. It is interesting that recent studies have shown differences in ion affinity for the a, b, and f NKCC2 isoforms, creating multiple possibilities for the physiologic regulation and modulation of NaCl transport in TAL (21,22,38). To investigate the dimeric function more closely, we analyzed two hNKCC2 concatemers. The bumetanide-sensitive 22 Na⫹ uptake of the wild-type–wild-type concatemer was comparable to that of complexes formed upon injecting of monomeric hNKCC2 constructs. Next, we introduced the G319R mutation in one of the subunits of the concatemer. This mutation, identified in a study by Vargas et al. (40), is normally processed and trafficked to the plasma membrane but functionally impaired when expressed in Xenopus oocytes (33). The wild-type–G319R concatemer heterologously expressed in oocytes was abundantly present at the plasma membrane. The bumetanide-sensitive 22Na⫹ uptake of the wild-type–G319R was half that of the wild-type–wild-type concatemer. This could suggest that both hNKCC2 subunits in a dimer can function as separate transporters, although it is also possible that wild-type subunits in the wild-type–G391R mutant dimerize with adjacent wild-type subunits. Similar results have been reported for a lactose permease dimer, a transmembrane protein that shares several structural traits with the SLC12A family, in having a permease domain and 12 membrane-spanning

J Am Soc Nephrol 14: 3039–3046, 2003

␣-helices. Sahin-Toth et al. (41) showed that introduction of an inactivating mutation in a concatemeric permease dimer reduced function by 50%, indicating that the subunits function as separate entities. Our results are in line with the fact that Bartter syndrome is a recessive disorder and thus no pathogenic phenotype would be present in a heterozygous situation. Thus, apparently enough transport capacity remains or wildtype–wild-type dimers are preferentially formed. It would, however, be interesting to study heterozygous carriers of Bartter mutations in more detail to investigate whether they present a phenotype under normal or challenged conditions, including a reduced BP or an increased susceptibility to diuretic-induced hypokalemia and hypercalciuria (3,27). In conclusion, our data suggest that human NKCC2, when expressed in oocytes, forms at least homodimers of which the individual subunits can function independently. This might be applicable to all members of the SLC12A family, because also the close relatives NKCC1 and KCC1 form functional dimers. This knowledge substantiates our insight into the functional mechanism of the SLC12A co-transporter family and contributes to our understanding of NaCl reabsorption in TAL.

Acknowledgments This study was supported by a grant of the Dutch Kidney Foundation (C97.1662). We thank Susan Hoefs and Fieke Mooren for superb technical assistance.

References 1. Simon DB, Lifton RP: Ion transporter mutations in Gitelman’s and Bartter’s syndromes. Curr Opin Nephrol Hypertens 7: 43– 47, 1998 2. de Rouffignac C: The urinary concentrating mechanism: A model of integrative physiology. Nephrol Dial Transplant 16: 2127–2130, 2001 3. Simon DB, Karet FE, Hamdan JM, DiPietro A, Sanjad SA, Lifton RP: Bartter’s syndrome, hypokalaemic alkalosis with hypercalciuria, is caused by mutations in the Na⫹-K⫹-2Cl⫺ cotransporter NKCC2. Nat Genet 13: 183–188, 1996 4. Bartter FC, Pronove P, Gill JR, McCardle R: Hyperplasia of the juxtaglomerular complex with hyperaldosteronism and hypokalaemic alkalosis. Am J Med 33: 811– 828, 1962 5. Simon DB, Karet FE, Rodriguez-Soriano J, Hamdan JH, DiPietro A, Trachtman H, Sanjad SA, Lifton RP: Genetic heterogeneity of Bartter’s syndrome revealed by mutations in the K⫹ channel. ROMK. Nat Genet 14: 152–156, 1996 6. Simon DB, Bindra RS, Mansfield TA, Nelson-Williams C, Mendonca E, Stone R, Schurman S, Nayir A, Alpay H, Bakkaloglu A, Rodriguez-Soriano J, Morales JM, Sanjad SA, Taylor CM, Pilz D, Brem A, Trachtman H, Griswold W, Richard GA, John E, Lifton RP: Mutations in the Cl⫺ channel gene, CLCNKB, cause Bartter’s syndrome type III. Nat Genet 17: 171–178, 1997 7. Birkenhager R, Otto E, Schurmann MJ, Vollmer M, Ruf EM, Maier-Lutz I, Beekmann F, Fekete A, Omran H, Feldmann D, Milford DV, Jeck N, Konrad M, Landau D, Knoers NV, Antignac C, Sudbrak R, Kispert A, Hildebrandt F: Mutation of BSND causes Bartter syndrome with sensorineural deafness and kidney failure. Nat Genet 29: 310 –314, 2001

J Am Soc Nephrol 14: 3039–3046, 2003

8. Gamba G, Miyanoshita A, Lombardi M, Lytton J, Lee WS, Hediger MA, Hebert SC: Molecular cloning, primary structure, and characterization of two members of the mammalian electroneutral Na⫹-(K⫹)-(2)Cl⫺ cotransporter family expressed in kidney. J Biol Chem 269: 17713–17722, 1994 9. Xu JC, Lytle C, Zhu TT, Payne JA, Benz E Jr, Forbush B 3rd: Molecular cloning and functional expression of the bumetanidesensitive Na⫹-K⫹-2Cl⫺ cotransporter. Proc Natl Acad Sci U S A 91: 2201–2205, 1994 10. Gamba G, Saltzberg SN, Lombardi M, Miyanoshita A, Lytton J, Hediger MA, Brenner BM, Hebert SC: Primary structure and functional expression of a cDNA encoding the thiazide-sensitive, electroneutral Na⫹-Cl⫺ cotransporter. Proc Natl Acad Sci U S A 90: 2749 –2753, 1993 11. Simon DB, Nelson-Williams C, Bia MJ, Ellison D, Karet FE, Molina AM, Vaara I, Iwata F, Cushner HM, Koolen M, Gainza FJ, Gitleman HJ, Lifton RP: Gitelman’s variant of Bartter’s syndrome, inherited hypokalaemic alkalosis, is caused by mutations in the thiazide-sensitive Na⫹-Cl⫺ cotransporter. Nat Genet 12: 24 –30, 1996 12. Payne JA, Stevenson TJ, Donaldson LF: Molecular characterization of a putative K⫹-Cl⫺ cotransporter in rat brain. A neuronalspecific isoform. J Biol Chem 271: 16245–16252, 1996 13. Gillen CM, Brill S, Payne JA, Forbush B 3rd: Molecular cloning and functional expression of the K-Cl cotransporter from rabbit, rat, and human. A new member of the cation-chloride cotransporter family. J Biol Chem 271: 16237–16244, 1996 14. Hiki K, D’Andrea RJ, Furze J, Crawford J, Woollatt E, Sutherland GR, Vadas MA, Gamble JR: Cloning, characterization, and chromosomal location of a novel human K⫹-Cl⫺ cotransporter. J Biol Chem 274: 10661–10667, 1999 15. Mount DB, Gamba G: Renal potassium-chloride cotransporters. Curr Opin Nephrol Hypertens 10: 685– 691, 2001 16. Mount DB, Mercado A, Song L, Xu J, George AL Jr, Delpire E, Gamba G: Cloning and characterization of KCC3 and KCC4, new members of the cation-chloride cotransporter gene family. J Biol Chem 274: 16355–16362, 1999 17. Mount DB, Delpire E, Gamba G, Hall AE, Poch E, Hoover RS, Hebert SC: The electroneutral cation-chloride cotransporters. J Exp Biol 201: 2091–2102, 1998 18. Delpire E, Mount DB: Human and murine phenotypes associated with defects in cation-chloride cotransport. Annu Rev Physiol 64: 803– 843, 2002 19. Payne JA, Forbush B 3rd: Alternatively spliced isoforms of the putative renal Na⫹-K⫹-2Cl⫺ cotransporter are differentially distributed within the rabbit kidney. Proc Natl Acad Sci U S A 91: 4544 – 4548, 1994 20. Mount DB, Baekgaard A, Hall AE, Plata C, Xu J, Beier DR, Gamba G, Hebert SC: Isoforms of the Na⫹-K⫹-2Cl⫺ cotransporter in murine TAL I. Molecular characterization and intrarenal localization. Am J Physiol 276: F347–F358, 1999 21. Plata C, Meade P, Vazquez N, Hebert SC, Gamba G: Functional properties of the apical Na⫹-K⫹-2Cl⫺ cotransporter isoforms. J Biol Chem 277: 11004 –11012, 2002 22. Gimenez I, Isenring P, Forbush B: Spatially distributed alternative splice variants of the renal Na⫹-K⫹-2Cl⫺ cotransporter exhibit dramatically different affinities for the transported ions. J Biol Chem 277: 8767– 8770, 2002 23. Plata C, Meade P, Hall A, Welch RC, Vazquez N, Hebert SC, Gamba G: Alternatively spliced isoform of apical Na⫹-K⫹-2Cl⫺ cotransporter gene encodes a furosemide-sensitive Na⫹-Cl⫺ cotransporter. Am J Physiol Renal Physiol 280: F574 –F582, 2001

Dimeric Architecture of hNKCC2

3045

24. Plata C, Mount DB, Rubio V, Hebert SC, Gamba G: Isoforms of the Na⫹-K⫹-2Cl⫺ cotransporter in murine TAL II. Functional characterization and activation by cAMP. Am J Physiol 276: F359 –F366, 1999 25. Moore-Hoon ML, Turner RJ: The structural unit of the secretory Na⫹-K⫹-2Cl⫺ cotransporter (NKCC1) is a homodimer. Biochemistry 39: 3718 –3724, 2000 26. Casula S, Shmukler BE, Wilhelm S, Stuart-Tilley AK, Su W, Chernova MN, Brugnara C, Alper SL: A dominant negative mutant of the KCC1 K⫹-Cl⫺ cotransporter: Both N- and Cterminal cytoplasmic domains are required for K⫹-Cl⫺ cotransport activity. J Biol Chem 276: 41870 – 41878, 2001 27. Simon DB, Lifton RP: The molecular basis of inherited hypokalemic alkalosis: Bartter’s and Gitelman’s syndromes. Am J Physiol 271: F961–F966, 1996 28. Steinmeyer K, Schwappach B, Bens M, Vandewalle A, Jentsch TJ: Cloning and functional expression of rat CLC-5, a Cl⫺ channel related to kidney disease. J Biol Chem 270: 31172– 31177, 1995 29. Chen YT, Holcomb C, Moore HP: Expression and localization of two low molecular weight GTP-binding proteins, Rab8 and Rab10, by epitope tag. Proc Natl Acad Sci U S A 90: 6508 – 6512, 1993 30. Kamsteeg EJ, Wormhoudt TA, Rijss JP, van Os CH, Deen PM: An impaired routing of wild-type aquaporin-2 after tetramerization with an aquaporin-2 mutant explains dominant nephrogenic diabetes insipidus. EMBO J 18: 2394 –2400, 1999 31. Neely JD, Christensen BM, Nielsen S, Agre P: Heterotetrameric composition of aquaporin-4 water channels. Biochemistry 38: 11156 –11163, 1999 32. Jung JS, Preston GM, Smith BL, Guggino WB, Agre P: Molecular structure of the water channel through aquaporin CHIP. The hourglass model. J Biol Chem 269: 14648 –14654, 1994 33. Starremans PGJF, Kersten FFJ, van den Heuvel LPWJ, Knoers NVAM, Bindels RJM: Mutations in the human Na⫹-K⫹-2Cl⫺ cotransporter (NKCC2) identified in Bartter syndrome type I consistently result in nonfunctional transporters. J Am Soc Nephrol 14: 1419 –1426, 2003 34. Gamba G, Saltzberg SN, Lombardi M, Miyanoshita A, Lytton J, Hediger MA, Brenner BM, Hebert SC: Primary structure and functional expression of a cDNA encoding the thiazide-sensitive, electroneutral Na⫹-Cl⫺ cotransporter. Proc Natl Acad Sci U S A 90: 2749 –2753, 1993 35. Mulders SM, Rijss JP, Hartog A, Bindels RJ, van Os CH, Deen PM: Importance of the mercury-sensitive cysteine on function and routing of AQP1 and AQP2 in oocytes. Am J Physiol 273: F451–F456, 1997 36. Bindels RJ, Timmermans JA, Hartog A, Coers W, van Os CH: Calbindin-D9k and parvalbumin are exclusively located along basolateral membranes in rat distal nephron. J Am Soc Nephrol 2: 1122–1129, 1991 37. De Jong JC, Willems PH, Mooren FJ, Van Den Heuvel LP, Knoers NV, Bindels RJ: The structural unit of the thiazidesensitive NaCl cotransporter is a homodimer. J Biol Chem 278: 24302–24307, 2003 38. Meade P, Hoover RS, Plata C, Vazquez N, Bobadilla NA, Gamba G, Hebert SC: cAMP-dependent activation of the renalspecific Na⫹-K⫹-2Cl⫺ cotransporter is mediated by regulation of cotransporter trafficking. Am J Physiol Renal Physiol 284: F1145–F1154, 2003 39. Howard HC, Mount DB, Rochefort D, Byun N, Dupre N, Lu J, Fan X, Song L, Riviere JB, Prevost C, Horst J, Simonati A, Lemcke B, Welch R, England R, Zhan FQ, Mercado A, Siesser WB, George AL, McDonald MP, Bouchard JP, Mathieu J,

3046

Journal of the American Society of Nephrology

Delpire E, Rouleau GA: The K⫹-Cl⫺ cotransporter KCC3 is mutant in a severe peripheral neuropathy associated with agenesis of the corpus callosum. Nat Genet 32: 384 –392, 2002 40. Vargas-Poussou R, Feldmann D, Vollmer M, Konrad M, Kelly L, van den Heuvel LP, Tebourbi L, Brandis M, Karolyi L, Hebert SC, Lemmink HH, Deschenes G, Hildebrandt F, Seyberth HW, Guay-Woodford LM, Knoers NV, Antignac C: Novel molecular

J Am Soc Nephrol 14: 3039–3046, 2003

variants of the Na⫹-K⫹-2Cl⫺ cotransporter gene are responsible for antenatal Bartter syndrome. Am J Hum Genet 62: 1332–1340, 1998 41. Sahin-Toth M, Lawrence MC, Kaback HR: Properties of permease dimer, a fusion protein containing two lactose permease molecules from Escherichia coli. Proc Natl Acad Sci U S A 91: 5421–5425, 1994