- Email: [email protected]
Regulation of apical localization of the thiazide-sensitive NaCl cotransporter by WNK4 in polarized epithelial cells
BBRC Biochemical and Biophysical Research Communications 330 (2005) 410–414 www.elsevier.com/locate/ybbrc
Regulation of apical localization of the thiazide-sensitive NaCl cotransporter by WNK4 in polarized epithelial cells Sung-Sen Yang, Kozue Yamauchi, Tatemitsu Rai, Atsushi Hiyama, Eisei Sohara, Tatsunori Suzuki, Tomohiro Itoh, Shin Suda, Sei Sasaki, Shinichi Uchida * Department of Nephrology, Graduate School of Medicine, Tokyo Medical and Dental University, 1-5-45 Yushima Bunkyo, Tokyo 113-8519, Japan Received 16 February 2005 Available online 14 March 2005
Abstract Missense mutations in the WNK4 gene have been postulated to cause pseudohypoaldosteronism type II (PHAII), an autosomaldominant disorder characterized by hyperkalemia and hypertension. Previous reports using Xenopus oocytes showed that wild-type WNK4 expression inhibited surface expression of the thiazide-sensitive NaCl cotransporter (NCC), while a disease-causing mutant lost the inhibitory effect on NCC surface expression. To determine if these changes observed in oocytes really occur in polarized epithelial cells, we generated stable MDCK II cell lines expressing NCC alone or NCC plus wild-type WNK4 or a disease-causing (D564A) WNK4. In contrast to the apical localization of NCC without co-expression of WNK4, immunofluorescence microscopy and biotin surface labeling revealed that this apical localization was equally decreased by both the wild-type and the mutant WNK4 expression. Apical localizations of two PHAII-unrelated apical transporters, sodium-independent amino acid transporter, BAT1 and bile salt export pump, Bsep, were also found to be decreased by both wild-type and mutant WNK4 expression. These results indicate that the regulation of NCC was not related to the disease-causing mutation and not restricted to the PHAII-related specific transporters. The regulation of intracellular localization of NCC by WNK4 might not be involved in the pathogenesis of PHAII. 2005 Elsevier Inc. All rights reserved. Keywords: Madin–Darby canine kidney II cells; Pseudohypoaldosteronism type II; Sodium chloride cotransporter; WNK4 kinase
Pseudohypoaldosteronism type II (PHAII) [OMIM 145260] is an autosomal-dominant disorder characterized by hyperkalemia and hypertension with a normal glomerular filtration rate and aldosterone response [1]. Based on clinical observations, an increased activity of thiazide-sensitive sodium chloride cotransporter (NCC) [2] or an increase in distal nephron chloride permeability, known as the ‘‘chloride shunt’’ [3], has been proposed to explain the pathogenesis of PHAII. Recently, Wilson et al. [4] reported that four missense mutations in the WNK4 gene and one long deletion in the first intron of the WNK1 kinase gene caused PHAII. Three of the PHAII-causing WNK4 mutations (E562K, D564A, *
Corresponding author. Fax: +81 3 5803 5215. E-mail address: [email protected] (S. Uchida).
0006-291X/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.02.172
and Q565E) were clustered in a segment distal to the first coil domain. This segment is highly conserved among all members of the WNK family in humans, as well as in mice and rats, suggesting that these highly clustered mutations would have a common pathogenic effect. As expected from the symptoms experienced by PHAII patients, WNK1 and WNK4 were found to be present in distal nephron segments in mice [4] where NCC is present. However, WNK4 was either cytosolic or co-localized with ZO-1 but not with NCC in the distal convoluted tubules (DCT). Since thiazide was previously identified as an effective treatment for this disease, Wilson et al. [5] and Yang et al. [6] examined whether WNK4 regulated the function of thiazide-sensitive NCC. They measured the transport activity of NCC that was co-expressed in Xenopus oocytes with either
S.-S. Yang et al. / Biochemical and Biophysical Research Communications 330 (2005) 410–414
wild-type WNK4 or disease-causing mutant WNK4. Wilson et al. [5] found that wild-type WNK4 decreased the surface expression of NCC; but, the mutant WNK4 lacked this effect. They speculated that the apical protein level of NCC would be increased in patients carrying this mutation in one allele because of the lack of the inhibitory effect. In addition to the regulation of NCC, Kahle et al. [7] reported that wild-type WNK4 inhibits the surface expression of renal K channel (ROMK). Moreover, the surface expressions of Na–K–Cl cotransporter (NKCC1) and the Cl-base exchanger SCL26A6 (CFEX) were also shown to be negatively regulated by wild-type WNK4 in Xenopus oocytes [8]. To examine whether these regulations are really physiologically important and involved in the pathogenesis of PHAII, or just the consequence of overexpression in the Xenopus oocyte expression system, we examined the regulation of NCC by WNK4 in polarized epithelial cells (MDCK II cells), which contain tight junctions. Since WNK4 is localized to the tight junctions in vivo, an expression system lacking the tight junction, i.e., the oocyte system, might not be suitable for the assessment of WNK4 function. Here, we performed the first examination of the localization of NCC in mammalian polarized epithelia that express wild-type or mutant human WNK4 (D564A). The regulation of apical localization of NCC was indeed observed in cultured mammalian polarized epithelial cells, but the lack of difference in the inhibitory effect between wild-type and mutant WNK4, as well as the fact that PHAII-unrelated transporters were also regulated, supports the notion that the regulation of apical membrane localization of NCC by WNK4 might not be involved in the pathogenesis of PHAII. Materials and methods Generation of stable MDCK II cell lines expressing HA-tagged NCC alone or NCC plus flag-tagged WNK4 (wild-type or D564A mutant). NCC cDNA was isolated by RT-PCR using human kidney mRNA as a template. The NCC cDNA was cloned into the pHM6 vector (Roche) and an N-terminal hemagglutinin (HA) tag was added. We subcloned the HA-NCC cDNA into the multiple cloning sites I of pBI vector (Clontech). N-terminal Flag-tagged human wild-type or D564A WNK4 cDNA [9] was cloned into the NotI site of pBI-HA-NCC. The pBI vector containing HA-NCC alone or HA-NCC plus Flag-WNK4 (wild-type or D564A) was co-transfected with linear hygromycin marker (Clontech) into Tet-off MDCK II cells (Clontech) using Lipofectamine 2000 (Invitrogen). Stable cell lines selected by hygromycin B (200 lg/ml) were screened by immunoblotting. Rat anti-HA mAb (9Y10) (Roche Diagnostics) and mouse M2 anti-Flag mAb (Sigma) were used to detect HA-tagged and Flag-tagged proteins, respectively. Immunofluorescence staining. Stable cells (1 · 105) were seeded onto a 12-mm diameter polyester Transwell filter (Corning, Corning, NY) with doxycycline (2 lg/ml), and the transgene expression was induced for four days by the removal of doxycycline after cells became confluent. Cells were fixed in 2% paraformaldehyde for 15 min and then permeabilized for 5 min in phosphate-buffered saline (PBS) containing
411
0.1% Triton X-100. Fixed cells were incubated with a primary antibody followed by an Alexa-labeled secondary antibody (Molecular Probes). Filters were mounted on glass slides with aqueous mounting medium (Daido Sangyo, Japan) and examined with a Carl-Zeiss LSM 510 confocal laser scanning microscope using a 63· water-immersion objective. Isolation of apical plasma membranes by biotinylation. Stable cells (3 · 105) were seeded on a 24-mm diameter polyester Transwell filter (Corning, Corning, NY), and the transgene expression was induced as described above. After the cells were washed twice with ice-cold PBSCM (PBS with 0.1 mM CaCl2 and 1 mM MgCl2), they were incubated twice for 20 min at 4 C with 6.0 mg/ml Sulfo-NHS-SS-Biotin (Pierce) in a biotinylation buffer (10 mM triethanolamine, 2 mM CaCl2, and 125 mM NaCl, pH 8.9) on the apical side. Subsequently, the cells were incubated for 5 min with a quenching solution (50 mM NH4Cl in PBSCM) at 4 C and rinsed twice with ice-cold PBS-CM. Cells on the filter were then lysed in 200 ll RIPA lysis buffer (1· PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% SDS) with protease inhibitors (Complete, Roche) for 30 min at 4 C. About 30 ll of the lysates was reserved for Western blot to check the total expression of NCC and WNK4 within the cells. Streptavidin beads (60 ll/sample) (Pierce) were added to the remainder of the lysates and incubated for 16 h at 4 C. After the incubation, the beads were centrifuged for 5 min, washed four times with lysis buffer, and washed once with 10 mM Tris–HCl (pH 7.5). Finally, the beads were resuspended in 30 ll Laemmli buffer and denatured for 30 min at 37 C. Transient expression of BAT1 or Bsep in the stable MDCK II cells. After stable cells seeded onto a 12-mm diameter polyester Transwell filter (Corning, Corning, NY) became confluent, the cells were cultured in medium without doxycycline for two days. Then, pAdeno-X-Bsep vector (a gift from Dr. Sugiyama, Tokyo University) or pGFP-BAT1 C2 with pCDNA3.1-rBAT (gifts from Dr. Kanai, Kyorin University) was transfected into the cells. Forty-eight hours after transfection, the cells were fixed for immunofluorescence staining.
Results Both the wild-type and a mutant WNK4 (D564A) inhibit the apical localization of NCC in MDCK II cells We generated stable MDCK II cell lines expressing NCC alone or NCC with either the wild-type WNK4 or a disease-causing mutant (D564A) using a tetracycline-inducible system. Inducible expression of NCC and WNK4 was confirmed by Western blot analysis (Fig. 1) and immunofluorescence. In some cell lines, leaky expression of WNK4 was observed. Accordingly, we performed the following experiments in cells where the expression was induced (doxycycline-negative). The wild-type and the mutant WNK4 mainly co-localized with a tight-junction protein, occludin, as we previously reported (data not shown) [9]. To determine the cellular localization of NCC, established stable cell lines were grown on permeable supports, and the expression of NCC and WNK4 was induced by removal of doxycycline from the medium for 4 days after the cells became confluent. After mild permeabilization and fixation, cells were stained with anti-HA mAb (9Y10) to detect NCC. The NCC localization in three stable cell lines is shown in Fig. 2. In contrast to the NCC localization in the cells that expressed
412
S.-S. Yang et al. / Biochemical and Biophysical Research Communications 330 (2005) 410–414
Fig. 1. Generation of MDCK II cell lines that express NCC alone or co-express NCC and wild-type or mutant (D564A) WNK4. (A) HAtagged NCC proteins and (B) Flag-tagged WNK4 proteins were fractioned by SDS–PAGE and detected by using an anti-HA mAb (9Y10) and mouse M2 anti-Flag mAb, respectively. Expression of NCC and WNK4 protein was induced by removal of doxycycline (Dox) from the medium for 4 days. 1, host cell; 2 and 3, cell line expressing NCC only; 4 and 5, cell line co-expressing NCC and wild-type WNK4; and 6 and 7, cell line co-expressing NCC and mutant WNK4.
Fig. 3. Cell surface biotinylation assay. Apical cell surface expression of NCC in MDCKII cells expressing NCC only (1) or co-expressing NCC and wild-type WNK4 (2) or mutant WNK4 (3). Stable cells expressing NCC and WNK4 were subjected to apical cell surface biotinylation assay. Biotinylated proteins were isolated with streptavidin-agarose beads and immunoblotted for NCC by anti-HA mAb (9Y10). Apical biotinylated proteins (upper panel) and the total cell lysates (middle panel) were fractioned by SDS–PAGE and immunoblotted.
only NCC, the Z-scan clearly showed that NCC localization was broadly cytoplasmic in both the wild-type and the mutant WNK4-expressing cells. Thus, the dominant apical localization observed in cells expressing only NCC was lost by WNK4 expression. To confirm the results of the NCC immunofluorescence assay, we used biotin surface labeling. After transgene expression was induced in stable cells grown on a permeable support, the apical membranes were biotinylated and recovered with streptavidin-conjugated agarose. As shown in Fig. 3, when the NCC expression within the three different cell lines was almost equal, the amount of apical biotinylated NCC proteins in the wild-type and the mutant WNK4-expressing cell lines
was decreased compared with that in the cells that expressed only NCC. Apical localization of two other PHAII-unrelated apical transporters was also inhibited by WNK4 in MDCK II cells To determine if WNK4Õs negative regulatory effect on the apical localization of NCC is specific to these proteins or not, we transiently expressed two other PHAII-unrelated transporters, BAT1 and Bsep, which localize to apical membranes in vivo and in MDCK cells [10–12]. As shown in Fig. 4, both transporters were localized to apical membranes in the MDCK II cells
Fig. 2. Cellular localization of NCC in the established cell lines. Localization of stably expressed NCC in MDCK II cells expressing NCC only: (1) or co-expressing NCC and wild-type WNK4 (2) or mutant WNK4 (3). XY (A) and XZ (B) images were obtained by confocal laser scanning microscopy. Immunocytochemistry was performed with rat anti-HA mAb (9Y10) and mouse anti-b catenin mAb as primary antibodies, and anti-rat IgG Alexa 546 (red) and anti-mouse IgG Alexa 488 (green) as the secondary antibodies. The arrowhead indicates the apical cell surface.
S.-S. Yang et al. / Biochemical and Biophysical Research Communications 330 (2005) 410–414
413
Fig. 4. Localization of GFP-BAT1 and Bsep transiently expressed in MDCK II cells. Immunofluorescence of green fluorescent protein (GFP)-tagged BAT1 (upper panels) and Bsep (lower panels) in host cells (1) and MDCK II cells expressing wild-type WNK4 (2) or mutant WNK4 (3). BAT1 is shown in green and b-catenin is shown in red (upper panels). Bsep is shown in red and b-catenin is shown in green (lower panels). Arrowheads indicate the apical cell surface.
without WNK4 expression. However, the apical localization of both transporters was almost completely inhibited by either wild-type or mutant WNK4 expression. These results suggest that both wild-type and mutant WNK4 have an inhibitory effect on the apical localization of several proteins. This inhibitory effect did not differ between wild-type and mutant WNK4.
Discussion Previous studies using an oocyte expression system showed that a disease-causing mutant WNK4 [5,6] lacked the inhibitory effect on the plasma membrane localization of NCC. This has been postulated to be a pathogenesis of PHAII, i.e., an increased NCC-dependent NaCl reabsorption in kidney since thiazide, a specific inhibitor for NCC, is effective in the treatment for PHAII [1–3]. However, these two studies and other studies [7,8] used a Xenopus oocyte expression system to examine WNK4 function. Recently, we demonstrated that mutant WNK4 increased paracellular chloride permeability in MDCK II cells, and we proposed that the ‘‘chloride shunt’’ theory could explain the pathogenesis of PHAII [9]. Very recently, Kahle et al. [13] published almost identical data using the same cell line. Since
wild-type WNK4 was shown to be present mainly in tight junctions in vivo, we thought that experiments examining the role of WNK4 in NCC localization should also be performed in a polarized epithelium. Accordingly, we established stable cell lines expressing NCC and wild-type WNK4 or a disease-causing WNK4. We chose D564A mutant since we previously showed that this mutant had a pathogenic effect in increasing chloride permeability and claudin phosphorylation in the same cell line [9], and we assumed that the three highly clustered mutations would cause a common pathogenic effect. Although the cell lines were stable, NCC expression was not homogeneous, i.e., not all cells showed bright NCC fluorescence (Fig. 2). There has been only one previous report that describes the immunolocalization of NCC expressed in polarized epithelial cells [14]. de Jong et al. reported that NCC showed apical localization; but, NCC was not exclusively located in the apical membrane, as it was also found in the cytoplasm. Our observations regarding NCC localization in the cells that expressed only NCC were similar to this previous study. As clearly shown in immunofluorescence images and a biotin surface labeling experiment (Figs. 2 and 3), wild-type WNK4 decreased the apical surface expression of NCC. In addition, D564AWNK4 also decreased the surface expression as well. Although these
414
S.-S. Yang et al. / Biochemical and Biophysical Research Communications 330 (2005) 410–414
results confirmed the ability of WNK4 to decrease surface expression of NCC in polarized epithelial cells as previously observed in Xenopus oocytes, the difference between wild-type and the mutant WNK4 was not observed in MDCK II cells. In addition to NCC, such regulation also occurred on two PHAII-unrelated transporters (Fig. 4). Based on these results, we speculate that this inhibitory effect of WNK4 might be caused by an unknown function inherent to acutely overexpressed WNK4. Physiological relevance of this inhibitory effect in the pathogenesis of PHAII must be re-evaluated by a gene targeting or a gene knock-down of WNK4 in cells or tissues where both WNK4 and each transporter or channel are endogenously expressed. In conclusion, our findings clearly indicate that the inhibitory effect of WNK4 on NCC apical localization may be unrelated to the pathogenesis of PHAII, because mutant WNK4 did not show a different function compared to wild-type WNK4.
Acknowledgment
[5]
[6]
[7]
[8]
[9]
[10]
This study was supported in part by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
References [1] R.D. Gordon, Syndrome of hypertension and hyperkalemia with normal glomerular filtration rate, Hypertension 8 (1986) 93–102. [2] Z. Farfel, A. Iaina, T. Rosenthal, U. Waks, S. Shibolet, J. Gafni, Familial hyperpotassemia and hypertension accompanied by normal plasma aldosterone levels: possible hereditary cell membrane defect, Arch. Intern. Med. (Moscow) 138 (1978) 1828–1832. [3] M. Schambelan, A. Sebastian, F.C. Rector Jr., Mineralocorticoid-resistant renal hyperkalemia without salt wasting (type II pseudohypoaldosteronism): role of increased renal chloride reabsorption, Kidney Int. 19 (1981) 716–727. [4] F.H. Wilson,, S. Disse-Nicodeme, K.A. Choate, K. Ishikawa, C. Nelson-Williams, I. Desitter, M. Gunel, D.V. Milford, G.W. Lipkin, J.M. Achard, M.P. Feely, B. Dussol, Y. Berland, R.J.
[11]
[12]
[13]
[14]
Unwin, H. Mayan, D.B. Simon, Z. Farfel, X. Jeunemaitre, R.P. Lifton, Human hypertension caused by mutations in WNK kinases, Science 293 (2001) 1107–1112. F.H. Wilson, K.T. Kahle, E. Sabath, M.D. Lalioti, A.K. Rapson, R.S Hoover, S.C. Hebert, G. Gamba, R.P. Lifton, Molecular pathogenesis of inherited hypertension with hyperkalemia: the Na–Cl cotransporter is inhibited by wild-type but not mutant WNK4, Proc. Natl. Acad. Sci. USA 100 (2003) 680–684. C.L. Yang, J. Angell, R. Mitchell, D.H. Ellison, WNK kinases regulate thiazide-sensitive Na–Cl cotransport, J. Clin. Invest. 111 (2003) 1039–1045. K.T. Kahle, F.H. Wilson, Q. Leng, M.D. Lalioti, A.D. OÕConnell, K. Dong, A.K. Rapson, G.G. MacGregor, G. Giebisch, S.C. Hebert, R.P. Lifton, WNK4 regulates the balance between renal NaCl reabsorption and K+ secretion, Nat. Genet. 35 (2003) 372– 376. K.T. Kahle, I. Gimenez, H. Hassan, F.H. Wilson, R.D. Wong, B. Forbush, P.S. Aronson, R.P. Lifton, WNK4 regulates apical and basolateral Cl-flux in extrarenal epithelia, Proc. Natl. Acad. Sci. USA 101 (2004) 2064–2069. K. Yamauchi, T. Rai, K. Kobayashi, E. Sohara, T. Suzuki, T. Itoh, S. Suda, A. Hayama, S. Sasaki, S. Uchida, Disease-causing mutant WNK4 increases paracellular chloride permeability and phosphorylates claudins, Proc. Natl. Acad. Sci. USA 101 (2004) 4690–4694. R. Kleta, E. Romeo, Z. Ristic, T. Ohura, C. Stuart, M. ArcosBurgos, M.H. Dave, C.A. Wagner, S.R. Camargo, S. Inoue, N. Matsuura, A. Helip-Wooley, D. Bockenhauer, R. Warth, I. Bernardini, G. Visser, T. Eggermann, P. Lee, A. Chairoungdua, P. Jutabha, E. Babu, S. Nilwarangkoon, N. Anzai, Y. Kanai, F. Verrey, W.A. Gahl, A. Koizumi, Mutations in SLC6A19, encoding B0AT1, cause Hartnup disorder, Nat. Genet. 36 (2004) 999– 1002. B. Gao, M.V. St Pierre, B. Stieger, P.J. Meier, Differential expression of bile salt and organic anion transporters in developing rat liver, J. Hepatol. 41 (2004) 201–208. S. Mita, H. Suzuki, H. Akita, B. Stieger, P.J. Meier, A.F. Hofmann, Y. Sugiyama, Vectorial transport of bile salts across MDCK cells expressing both rat Na+/Taurocholate cotransporting polypeptide and rat bile salt export pump, Am. J. Physiol. Gastrointest. Liver Physiol. 288 (2005) 159–167. K.T. Kahle, G.G. Macgregor, F.H. Wilson, A.N. Van Hoek, D. Brown, T. Ardito, M. Kashgarian, G. Giebisch, S.C. Hebert, E.L. Boulpaep, R.P. Lifton, Paracellular Cl-permeability is regulated by WNK4 kinase: insight into normal physiology and hypertension, Proc. Natl. Acad. Sci. USA 101 (2004) 14877–14882. J.C. de Jong, P.H. Willems, L.P. van den Heuvel, N.V. Knoers, R.J. Bindels, Functional expression of the human thiazidesensitive NaCl cotransporter in Madin–Darby canine kidney cells, J. Am. Soc. Nephrol. 14 (2003) 2428–2435.
Regulation of apical localization of the thiazide-sensitive NaCl cotransporter by WNK4 in polarized epithelial cells Sung-Sen Yang, Kozue Yamauchi, Tatemitsu Rai, Atsushi Hiyama, Eisei Sohara, Tatsunori Suzuki, Tomohiro Itoh, Shin Suda, Sei Sasaki, Shinichi Uchida * Department of Nephrology, Graduate School of Medicine, Tokyo Medical and Dental University, 1-5-45 Yushima Bunkyo, Tokyo 113-8519, Japan Received 16 February 2005 Available online 14 March 2005
Abstract Missense mutations in the WNK4 gene have been postulated to cause pseudohypoaldosteronism type II (PHAII), an autosomaldominant disorder characterized by hyperkalemia and hypertension. Previous reports using Xenopus oocytes showed that wild-type WNK4 expression inhibited surface expression of the thiazide-sensitive NaCl cotransporter (NCC), while a disease-causing mutant lost the inhibitory effect on NCC surface expression. To determine if these changes observed in oocytes really occur in polarized epithelial cells, we generated stable MDCK II cell lines expressing NCC alone or NCC plus wild-type WNK4 or a disease-causing (D564A) WNK4. In contrast to the apical localization of NCC without co-expression of WNK4, immunofluorescence microscopy and biotin surface labeling revealed that this apical localization was equally decreased by both the wild-type and the mutant WNK4 expression. Apical localizations of two PHAII-unrelated apical transporters, sodium-independent amino acid transporter, BAT1 and bile salt export pump, Bsep, were also found to be decreased by both wild-type and mutant WNK4 expression. These results indicate that the regulation of NCC was not related to the disease-causing mutation and not restricted to the PHAII-related specific transporters. The regulation of intracellular localization of NCC by WNK4 might not be involved in the pathogenesis of PHAII. 2005 Elsevier Inc. All rights reserved. Keywords: Madin–Darby canine kidney II cells; Pseudohypoaldosteronism type II; Sodium chloride cotransporter; WNK4 kinase
Pseudohypoaldosteronism type II (PHAII) [OMIM 145260] is an autosomal-dominant disorder characterized by hyperkalemia and hypertension with a normal glomerular filtration rate and aldosterone response [1]. Based on clinical observations, an increased activity of thiazide-sensitive sodium chloride cotransporter (NCC) [2] or an increase in distal nephron chloride permeability, known as the ‘‘chloride shunt’’ [3], has been proposed to explain the pathogenesis of PHAII. Recently, Wilson et al. [4] reported that four missense mutations in the WNK4 gene and one long deletion in the first intron of the WNK1 kinase gene caused PHAII. Three of the PHAII-causing WNK4 mutations (E562K, D564A, *
Corresponding author. Fax: +81 3 5803 5215. E-mail address: [email protected] (S. Uchida).
0006-291X/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.02.172
and Q565E) were clustered in a segment distal to the first coil domain. This segment is highly conserved among all members of the WNK family in humans, as well as in mice and rats, suggesting that these highly clustered mutations would have a common pathogenic effect. As expected from the symptoms experienced by PHAII patients, WNK1 and WNK4 were found to be present in distal nephron segments in mice [4] where NCC is present. However, WNK4 was either cytosolic or co-localized with ZO-1 but not with NCC in the distal convoluted tubules (DCT). Since thiazide was previously identified as an effective treatment for this disease, Wilson et al. [5] and Yang et al. [6] examined whether WNK4 regulated the function of thiazide-sensitive NCC. They measured the transport activity of NCC that was co-expressed in Xenopus oocytes with either
S.-S. Yang et al. / Biochemical and Biophysical Research Communications 330 (2005) 410–414
wild-type WNK4 or disease-causing mutant WNK4. Wilson et al. [5] found that wild-type WNK4 decreased the surface expression of NCC; but, the mutant WNK4 lacked this effect. They speculated that the apical protein level of NCC would be increased in patients carrying this mutation in one allele because of the lack of the inhibitory effect. In addition to the regulation of NCC, Kahle et al. [7] reported that wild-type WNK4 inhibits the surface expression of renal K channel (ROMK). Moreover, the surface expressions of Na–K–Cl cotransporter (NKCC1) and the Cl-base exchanger SCL26A6 (CFEX) were also shown to be negatively regulated by wild-type WNK4 in Xenopus oocytes [8]. To examine whether these regulations are really physiologically important and involved in the pathogenesis of PHAII, or just the consequence of overexpression in the Xenopus oocyte expression system, we examined the regulation of NCC by WNK4 in polarized epithelial cells (MDCK II cells), which contain tight junctions. Since WNK4 is localized to the tight junctions in vivo, an expression system lacking the tight junction, i.e., the oocyte system, might not be suitable for the assessment of WNK4 function. Here, we performed the first examination of the localization of NCC in mammalian polarized epithelia that express wild-type or mutant human WNK4 (D564A). The regulation of apical localization of NCC was indeed observed in cultured mammalian polarized epithelial cells, but the lack of difference in the inhibitory effect between wild-type and mutant WNK4, as well as the fact that PHAII-unrelated transporters were also regulated, supports the notion that the regulation of apical membrane localization of NCC by WNK4 might not be involved in the pathogenesis of PHAII. Materials and methods Generation of stable MDCK II cell lines expressing HA-tagged NCC alone or NCC plus flag-tagged WNK4 (wild-type or D564A mutant). NCC cDNA was isolated by RT-PCR using human kidney mRNA as a template. The NCC cDNA was cloned into the pHM6 vector (Roche) and an N-terminal hemagglutinin (HA) tag was added. We subcloned the HA-NCC cDNA into the multiple cloning sites I of pBI vector (Clontech). N-terminal Flag-tagged human wild-type or D564A WNK4 cDNA [9] was cloned into the NotI site of pBI-HA-NCC. The pBI vector containing HA-NCC alone or HA-NCC plus Flag-WNK4 (wild-type or D564A) was co-transfected with linear hygromycin marker (Clontech) into Tet-off MDCK II cells (Clontech) using Lipofectamine 2000 (Invitrogen). Stable cell lines selected by hygromycin B (200 lg/ml) were screened by immunoblotting. Rat anti-HA mAb (9Y10) (Roche Diagnostics) and mouse M2 anti-Flag mAb (Sigma) were used to detect HA-tagged and Flag-tagged proteins, respectively. Immunofluorescence staining. Stable cells (1 · 105) were seeded onto a 12-mm diameter polyester Transwell filter (Corning, Corning, NY) with doxycycline (2 lg/ml), and the transgene expression was induced for four days by the removal of doxycycline after cells became confluent. Cells were fixed in 2% paraformaldehyde for 15 min and then permeabilized for 5 min in phosphate-buffered saline (PBS) containing
411
0.1% Triton X-100. Fixed cells were incubated with a primary antibody followed by an Alexa-labeled secondary antibody (Molecular Probes). Filters were mounted on glass slides with aqueous mounting medium (Daido Sangyo, Japan) and examined with a Carl-Zeiss LSM 510 confocal laser scanning microscope using a 63· water-immersion objective. Isolation of apical plasma membranes by biotinylation. Stable cells (3 · 105) were seeded on a 24-mm diameter polyester Transwell filter (Corning, Corning, NY), and the transgene expression was induced as described above. After the cells were washed twice with ice-cold PBSCM (PBS with 0.1 mM CaCl2 and 1 mM MgCl2), they were incubated twice for 20 min at 4 C with 6.0 mg/ml Sulfo-NHS-SS-Biotin (Pierce) in a biotinylation buffer (10 mM triethanolamine, 2 mM CaCl2, and 125 mM NaCl, pH 8.9) on the apical side. Subsequently, the cells were incubated for 5 min with a quenching solution (50 mM NH4Cl in PBSCM) at 4 C and rinsed twice with ice-cold PBS-CM. Cells on the filter were then lysed in 200 ll RIPA lysis buffer (1· PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% SDS) with protease inhibitors (Complete, Roche) for 30 min at 4 C. About 30 ll of the lysates was reserved for Western blot to check the total expression of NCC and WNK4 within the cells. Streptavidin beads (60 ll/sample) (Pierce) were added to the remainder of the lysates and incubated for 16 h at 4 C. After the incubation, the beads were centrifuged for 5 min, washed four times with lysis buffer, and washed once with 10 mM Tris–HCl (pH 7.5). Finally, the beads were resuspended in 30 ll Laemmli buffer and denatured for 30 min at 37 C. Transient expression of BAT1 or Bsep in the stable MDCK II cells. After stable cells seeded onto a 12-mm diameter polyester Transwell filter (Corning, Corning, NY) became confluent, the cells were cultured in medium without doxycycline for two days. Then, pAdeno-X-Bsep vector (a gift from Dr. Sugiyama, Tokyo University) or pGFP-BAT1 C2 with pCDNA3.1-rBAT (gifts from Dr. Kanai, Kyorin University) was transfected into the cells. Forty-eight hours after transfection, the cells were fixed for immunofluorescence staining.
Results Both the wild-type and a mutant WNK4 (D564A) inhibit the apical localization of NCC in MDCK II cells We generated stable MDCK II cell lines expressing NCC alone or NCC with either the wild-type WNK4 or a disease-causing mutant (D564A) using a tetracycline-inducible system. Inducible expression of NCC and WNK4 was confirmed by Western blot analysis (Fig. 1) and immunofluorescence. In some cell lines, leaky expression of WNK4 was observed. Accordingly, we performed the following experiments in cells where the expression was induced (doxycycline-negative). The wild-type and the mutant WNK4 mainly co-localized with a tight-junction protein, occludin, as we previously reported (data not shown) [9]. To determine the cellular localization of NCC, established stable cell lines were grown on permeable supports, and the expression of NCC and WNK4 was induced by removal of doxycycline from the medium for 4 days after the cells became confluent. After mild permeabilization and fixation, cells were stained with anti-HA mAb (9Y10) to detect NCC. The NCC localization in three stable cell lines is shown in Fig. 2. In contrast to the NCC localization in the cells that expressed
412
S.-S. Yang et al. / Biochemical and Biophysical Research Communications 330 (2005) 410–414
Fig. 1. Generation of MDCK II cell lines that express NCC alone or co-express NCC and wild-type or mutant (D564A) WNK4. (A) HAtagged NCC proteins and (B) Flag-tagged WNK4 proteins were fractioned by SDS–PAGE and detected by using an anti-HA mAb (9Y10) and mouse M2 anti-Flag mAb, respectively. Expression of NCC and WNK4 protein was induced by removal of doxycycline (Dox) from the medium for 4 days. 1, host cell; 2 and 3, cell line expressing NCC only; 4 and 5, cell line co-expressing NCC and wild-type WNK4; and 6 and 7, cell line co-expressing NCC and mutant WNK4.
Fig. 3. Cell surface biotinylation assay. Apical cell surface expression of NCC in MDCKII cells expressing NCC only (1) or co-expressing NCC and wild-type WNK4 (2) or mutant WNK4 (3). Stable cells expressing NCC and WNK4 were subjected to apical cell surface biotinylation assay. Biotinylated proteins were isolated with streptavidin-agarose beads and immunoblotted for NCC by anti-HA mAb (9Y10). Apical biotinylated proteins (upper panel) and the total cell lysates (middle panel) were fractioned by SDS–PAGE and immunoblotted.
only NCC, the Z-scan clearly showed that NCC localization was broadly cytoplasmic in both the wild-type and the mutant WNK4-expressing cells. Thus, the dominant apical localization observed in cells expressing only NCC was lost by WNK4 expression. To confirm the results of the NCC immunofluorescence assay, we used biotin surface labeling. After transgene expression was induced in stable cells grown on a permeable support, the apical membranes were biotinylated and recovered with streptavidin-conjugated agarose. As shown in Fig. 3, when the NCC expression within the three different cell lines was almost equal, the amount of apical biotinylated NCC proteins in the wild-type and the mutant WNK4-expressing cell lines
was decreased compared with that in the cells that expressed only NCC. Apical localization of two other PHAII-unrelated apical transporters was also inhibited by WNK4 in MDCK II cells To determine if WNK4Õs negative regulatory effect on the apical localization of NCC is specific to these proteins or not, we transiently expressed two other PHAII-unrelated transporters, BAT1 and Bsep, which localize to apical membranes in vivo and in MDCK cells [10–12]. As shown in Fig. 4, both transporters were localized to apical membranes in the MDCK II cells
Fig. 2. Cellular localization of NCC in the established cell lines. Localization of stably expressed NCC in MDCK II cells expressing NCC only: (1) or co-expressing NCC and wild-type WNK4 (2) or mutant WNK4 (3). XY (A) and XZ (B) images were obtained by confocal laser scanning microscopy. Immunocytochemistry was performed with rat anti-HA mAb (9Y10) and mouse anti-b catenin mAb as primary antibodies, and anti-rat IgG Alexa 546 (red) and anti-mouse IgG Alexa 488 (green) as the secondary antibodies. The arrowhead indicates the apical cell surface.
S.-S. Yang et al. / Biochemical and Biophysical Research Communications 330 (2005) 410–414
413
Fig. 4. Localization of GFP-BAT1 and Bsep transiently expressed in MDCK II cells. Immunofluorescence of green fluorescent protein (GFP)-tagged BAT1 (upper panels) and Bsep (lower panels) in host cells (1) and MDCK II cells expressing wild-type WNK4 (2) or mutant WNK4 (3). BAT1 is shown in green and b-catenin is shown in red (upper panels). Bsep is shown in red and b-catenin is shown in green (lower panels). Arrowheads indicate the apical cell surface.
without WNK4 expression. However, the apical localization of both transporters was almost completely inhibited by either wild-type or mutant WNK4 expression. These results suggest that both wild-type and mutant WNK4 have an inhibitory effect on the apical localization of several proteins. This inhibitory effect did not differ between wild-type and mutant WNK4.
Discussion Previous studies using an oocyte expression system showed that a disease-causing mutant WNK4 [5,6] lacked the inhibitory effect on the plasma membrane localization of NCC. This has been postulated to be a pathogenesis of PHAII, i.e., an increased NCC-dependent NaCl reabsorption in kidney since thiazide, a specific inhibitor for NCC, is effective in the treatment for PHAII [1–3]. However, these two studies and other studies [7,8] used a Xenopus oocyte expression system to examine WNK4 function. Recently, we demonstrated that mutant WNK4 increased paracellular chloride permeability in MDCK II cells, and we proposed that the ‘‘chloride shunt’’ theory could explain the pathogenesis of PHAII [9]. Very recently, Kahle et al. [13] published almost identical data using the same cell line. Since
wild-type WNK4 was shown to be present mainly in tight junctions in vivo, we thought that experiments examining the role of WNK4 in NCC localization should also be performed in a polarized epithelium. Accordingly, we established stable cell lines expressing NCC and wild-type WNK4 or a disease-causing WNK4. We chose D564A mutant since we previously showed that this mutant had a pathogenic effect in increasing chloride permeability and claudin phosphorylation in the same cell line [9], and we assumed that the three highly clustered mutations would cause a common pathogenic effect. Although the cell lines were stable, NCC expression was not homogeneous, i.e., not all cells showed bright NCC fluorescence (Fig. 2). There has been only one previous report that describes the immunolocalization of NCC expressed in polarized epithelial cells [14]. de Jong et al. reported that NCC showed apical localization; but, NCC was not exclusively located in the apical membrane, as it was also found in the cytoplasm. Our observations regarding NCC localization in the cells that expressed only NCC were similar to this previous study. As clearly shown in immunofluorescence images and a biotin surface labeling experiment (Figs. 2 and 3), wild-type WNK4 decreased the apical surface expression of NCC. In addition, D564AWNK4 also decreased the surface expression as well. Although these
414
S.-S. Yang et al. / Biochemical and Biophysical Research Communications 330 (2005) 410–414
results confirmed the ability of WNK4 to decrease surface expression of NCC in polarized epithelial cells as previously observed in Xenopus oocytes, the difference between wild-type and the mutant WNK4 was not observed in MDCK II cells. In addition to NCC, such regulation also occurred on two PHAII-unrelated transporters (Fig. 4). Based on these results, we speculate that this inhibitory effect of WNK4 might be caused by an unknown function inherent to acutely overexpressed WNK4. Physiological relevance of this inhibitory effect in the pathogenesis of PHAII must be re-evaluated by a gene targeting or a gene knock-down of WNK4 in cells or tissues where both WNK4 and each transporter or channel are endogenously expressed. In conclusion, our findings clearly indicate that the inhibitory effect of WNK4 on NCC apical localization may be unrelated to the pathogenesis of PHAII, because mutant WNK4 did not show a different function compared to wild-type WNK4.
Acknowledgment
[5]
[6]
[7]
[8]
[9]
[10]
This study was supported in part by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
References [1] R.D. Gordon, Syndrome of hypertension and hyperkalemia with normal glomerular filtration rate, Hypertension 8 (1986) 93–102. [2] Z. Farfel, A. Iaina, T. Rosenthal, U. Waks, S. Shibolet, J. Gafni, Familial hyperpotassemia and hypertension accompanied by normal plasma aldosterone levels: possible hereditary cell membrane defect, Arch. Intern. Med. (Moscow) 138 (1978) 1828–1832. [3] M. Schambelan, A. Sebastian, F.C. Rector Jr., Mineralocorticoid-resistant renal hyperkalemia without salt wasting (type II pseudohypoaldosteronism): role of increased renal chloride reabsorption, Kidney Int. 19 (1981) 716–727. [4] F.H. Wilson,, S. Disse-Nicodeme, K.A. Choate, K. Ishikawa, C. Nelson-Williams, I. Desitter, M. Gunel, D.V. Milford, G.W. Lipkin, J.M. Achard, M.P. Feely, B. Dussol, Y. Berland, R.J.
[11]
[12]
[13]
[14]
Unwin, H. Mayan, D.B. Simon, Z. Farfel, X. Jeunemaitre, R.P. Lifton, Human hypertension caused by mutations in WNK kinases, Science 293 (2001) 1107–1112. F.H. Wilson, K.T. Kahle, E. Sabath, M.D. Lalioti, A.K. Rapson, R.S Hoover, S.C. Hebert, G. Gamba, R.P. Lifton, Molecular pathogenesis of inherited hypertension with hyperkalemia: the Na–Cl cotransporter is inhibited by wild-type but not mutant WNK4, Proc. Natl. Acad. Sci. USA 100 (2003) 680–684. C.L. Yang, J. Angell, R. Mitchell, D.H. Ellison, WNK kinases regulate thiazide-sensitive Na–Cl cotransport, J. Clin. Invest. 111 (2003) 1039–1045. K.T. Kahle, F.H. Wilson, Q. Leng, M.D. Lalioti, A.D. OÕConnell, K. Dong, A.K. Rapson, G.G. MacGregor, G. Giebisch, S.C. Hebert, R.P. Lifton, WNK4 regulates the balance between renal NaCl reabsorption and K+ secretion, Nat. Genet. 35 (2003) 372– 376. K.T. Kahle, I. Gimenez, H. Hassan, F.H. Wilson, R.D. Wong, B. Forbush, P.S. Aronson, R.P. Lifton, WNK4 regulates apical and basolateral Cl-flux in extrarenal epithelia, Proc. Natl. Acad. Sci. USA 101 (2004) 2064–2069. K. Yamauchi, T. Rai, K. Kobayashi, E. Sohara, T. Suzuki, T. Itoh, S. Suda, A. Hayama, S. Sasaki, S. Uchida, Disease-causing mutant WNK4 increases paracellular chloride permeability and phosphorylates claudins, Proc. Natl. Acad. Sci. USA 101 (2004) 4690–4694. R. Kleta, E. Romeo, Z. Ristic, T. Ohura, C. Stuart, M. ArcosBurgos, M.H. Dave, C.A. Wagner, S.R. Camargo, S. Inoue, N. Matsuura, A. Helip-Wooley, D. Bockenhauer, R. Warth, I. Bernardini, G. Visser, T. Eggermann, P. Lee, A. Chairoungdua, P. Jutabha, E. Babu, S. Nilwarangkoon, N. Anzai, Y. Kanai, F. Verrey, W.A. Gahl, A. Koizumi, Mutations in SLC6A19, encoding B0AT1, cause Hartnup disorder, Nat. Genet. 36 (2004) 999– 1002. B. Gao, M.V. St Pierre, B. Stieger, P.J. Meier, Differential expression of bile salt and organic anion transporters in developing rat liver, J. Hepatol. 41 (2004) 201–208. S. Mita, H. Suzuki, H. Akita, B. Stieger, P.J. Meier, A.F. Hofmann, Y. Sugiyama, Vectorial transport of bile salts across MDCK cells expressing both rat Na+/Taurocholate cotransporting polypeptide and rat bile salt export pump, Am. J. Physiol. Gastrointest. Liver Physiol. 288 (2005) 159–167. K.T. Kahle, G.G. Macgregor, F.H. Wilson, A.N. Van Hoek, D. Brown, T. Ardito, M. Kashgarian, G. Giebisch, S.C. Hebert, E.L. Boulpaep, R.P. Lifton, Paracellular Cl-permeability is regulated by WNK4 kinase: insight into normal physiology and hypertension, Proc. Natl. Acad. Sci. USA 101 (2004) 14877–14882. J.C. de Jong, P.H. Willems, L.P. van den Heuvel, N.V. Knoers, R.J. Bindels, Functional expression of the human thiazidesensitive NaCl cotransporter in Madin–Darby canine kidney cells, J. Am. Soc. Nephrol. 14 (2003) 2428–2435.