WNK4 Enhances the Degradation of NCC through a Sortilin-Mediated Lysosomal Pathway

WNK kinase is a serine/threonine kinase that plays an important role in electrolyte homeostasis. WNK4 significantly inhibits the surface expression of the sodium chloride co-transporter (NCC) by enhancing the degradation of NCC through a lysosomal pathway, but the mechanisms underlying this trafficking are unknown. Here, we investigated the effect of the lysosomal targeting receptor sortilin on NCC expression and degradation. In Cos-7 cells, we observed that the presence of WNK4 reduced the steady-state amount of NCC by approximately half. Co-transfection with truncated sortilin (a dominant negative mutant) prevented this WNK4-induced reduction in NCC. NCC immunoprecipitated with both wild-type sortilin and, to a lesser extent, truncated sortilin. Immunostaining revealed that WNK4 increased the co-localization of NCC with the lysosomal marker cathepsin D, and NCC co-localized with wild-type sortilin, truncated sortilin, and WNK4 in the perinuclear region. These findings suggest that WNK4 promotes NCC targeting to the lysosome for degradation via a mechanism involving sortilin. WT, NCC protein expression is significantly reduced in NCC (cid:1) WNK4 (cid:1) GFP Vect group as compared with either control group or NCC (cid:1) WNK4 (cid:1) sortilin TRU group in D (** P (cid:5) 0.001). NCC protein expression in NCC (cid:1) WNK4 (cid:1) sortilin WT is also significantly reduced as compared with the control group (* P (cid:5) 0.01). The level of NCC protein expression in NCC (cid:1) WNK4 (cid:1) sortilin TRU group was significantly higher as compared with the NCC (cid:1) WNK4 (cid:1) sortilin WT group (# P (cid:5) 0.05).

WNK (with no lysine [K]) kinase is a subfamily of serine/threonine kinases. 1 Mutations in two members of this family, WNK1 and WNK4, result in pseudohypoaldosteronism type II, 2 featuring hypertension, hyperkalemia, and metabolic acidosis. Previous studies showed that wild-type (WT) WNK4 inhibits the activity and surface expression of sodium chloride co-transporter (NCC) in Xenopus oocytes. 3,4 Interestingly, one study showed that a NCC harboring five different Gitelman-type mutations exhibits low activity that is mainly due to a reduction of functional NCC inserting into the plasma membrane. 5 Our previous study 6 also indicated that NCC surface expression is regulated by altering its degradation through the lysosomal pathway. These combined studies suggest that alteration of NCC function can result from perturbing its protein synthesis, 7,8 glycosylation and processing, 9,10 and delivery to the plasma membrane. 8, 10 Golbang et al. 11 showed that WNK4 also blocks the forward trafficking of NCC; however, the exact molecular basis of interference of WNK4 kinase with NCC forward trafficking and enhancing its degradation through a lysosomal pathway remains to be clarified.
There are two major sorting mechanisms involving targeting lysosomal proteins to lysosomes. 12,13 One is the mannose-6-phosphate receptor (M6PR)-mediated mechanism, 14 -16 and the other is mediated by sortilin. 17 The trans-Golgi network (TGN) is the major site of sorting for newly synthesized proteins that are targeted to lysosomes for degradation. These newly synthesized proteins first bind to either M6PR 18 -20 or sortilin 21 in the TGN and subsequently recruit the Golgi-localized, ␥-earcontaining, ADP-ribosylation factor-binding proteins (GGAs) [22][23][24] clathrin and adaptor protein 1 (AP1) or AP3 25,26 to form clathrin-coated vesicles to initiate their trafficking to endosomes, following a secretory pathway, or to lysosomes for degradation.
Sortilin is a newly identified lysosomal targeting receptor that is involved in the alternative sorting of the lysosomal sphingolipid activator protein prosaposin 27,28 and the GM2 activator protein. 28,29 It is a homolog of the yeast vacuolar sorting receptor Vps10p and belongs to the type I Vps10p superfamily. 17 The human sortilin gene encodes 833 amino acids. It consists of an N-terminal propeptide, a furin cleavage site, a large luminal domain, a single transmembrane region, and a short cytoplasmic tail. 17 The major pool of sortilin accumulates in the TGN and vesicles, whereas 10% of sortilin is present in the plasma membrane. The truncation of the cytoplasmic tail of sortilin (sortilin TRU) leads to a disruption of the lysosomal sorting function, causing a majority of sortilin TRU to be retained in the TGN; however, a small portion of sortilin TRU may leak to the plasma membrane or vesicles. Sortilin TRU can serve as a dominant negative mutant. 28 Sortilin seems to have multiple functions. Sortilin not only binds different ligands such as neurotensin, receptor associated protein, and prosaposin, 17,30 but also is involved in intracellular sorting, endocytosis, and signal transduction. 31 Studies have shown that sortilin binds the glucose transporter Glut4 and is translocated with Glut4 to the plasma membrane in response to insulin stimulation, [32][33][34] suggesting that sortilin may be involved in the regulation of membrane transporters; therefore, we speculated that WNK4 might promote the degradation of NCC through the sortilin-mediated lysosomal pathway. Here, we report that WNK4 downregulates the steady-state protein levels of NCC in Cos-7 cells. Sortilin TRU reverses the degradation of NCC promoted by WNK4. The N-terminus of NCC binds sortilin WT as well as sortilin TRU, to a lesser extent. WNK4 co-localizes with NCC and sortilin and increases NCC co-localization with cathepsin D, a lysosomal marker. These data suggest that WNK4, NCC, and sortilin interact with one another to facilitate the WNK4-promoted degradation of NCC through a lysosomal pathway by a sortilin-mediated targeting mechanism.

Effect of Sortilin on WNK4-Mediated Inhibition of NCC Protein Expression
We have shown that WNK4 WT enhances NCC degradation through the lysosomal pathway. 6 To determine whether a lysosomal sorting mechanism such as sortilin is involved in WNK4-promoted degradation, we co-transfected Cos-7 cells with hemagglutinin (HA)-NCC in combination with either green fluorescence protein (GFP)-sortilin WT or sortilin TRU ( Figure 1) in the absence (Figure 2, A and B) or the presence of myc-WNK4 WT (Figure 2, C and D). There was no significant change in the steady-state amount of NCC in the absence of WNK4 in both NCC ϩ sortilin WT and NCC ϩ sortilin TRU groups as compared with NCC alone (control) group (P Ͼ 0.05; n ϭ 4). In the presence of WNK4 WT, NCC protein level was significantly reduced by 52% (47.9 Ϯ 8.7% in NCC ϩ WNK4 [WNK4 WT group] versus 100% in the NCC alone [control group]; P Ͻ 0.001; n ϭ 6). There was also a significant reduction in NCC expression in NCC ϩ WNK4 ϩ sortilin WT (sortilin WT group, 60.2 Ϯ 5.0%) as compared with the control group (P Ͻ 0.01; n ϭ 6); however, there was no significant difference in NCC expression between sortilin WT group and WNK4 WT group. In contrast, in NCC ϩ WNK4 ϩ sortilin TRU (sortilin TRU group), NCC protein expression (92.9 Ϯ 6.8%) was significantly increased as compared with either WNK4 WT group (P Ͻ 0.001; n ϭ 6) or sortilin WT group (P Ͻ 0.05; n ϭ 6), indicating that the sortilin TRU likely reverses the inhibitory effect of WNK4 on NCC protein expression through its dominant negative effect on endogenous sortilin, because Cos-7 cells express endogenous sortilin protein. These data suggest that WNK4 enhances the degradation of NCC through a lysosomal pathway via sortilin.

NCC Interacts with Sortilin WT and Its Truncated Sortilin
Because sortilin TRU reverses the inhibitory effect of WNK4 on NCC protein expression, we next tested whether NCC in- teracts with sortilin using co-immunoprecipitation (co-IP) experiments in Cos-7 cells. As shown in Figure 3C, A and B, in the cells co-transfected with HA-NCC WT in combination with either GFP-tagged sortilin WT or TRU, the anti-GFP antibody immunoprecipitated both sortilin WT and TRU. In addition, both sortilin WT and TRU co-immunoprecipitated with NCC (samples 1 and 2 in Figure 3, A and B), although sortilin TRU bound to NCC with much less avidity. In contrast, in the cells co-transfected with a GFP vector and HA-NCC WT without sortilin, the anti-GFP antibody did not immunoprecipitate NCC (sample 5 in Figure 3A and sample 3 in Figure 3B). The reciprocal co-IP also showed that NCC co-precipitated sortilin WT as well as sortilin TRU to a lesser extent (samples 1 and 2 in Figure 3C). These data suggest that NCC interacts with both sortilin WT and TRU. Because sortilin TRU without its cytoplasmic tail still binds NCC, albeit with a lesser affinity, the luminal domain of sortilin could be the binding site. We also investigated which region of NCC contains the binding site for sortilin. As shown in Figure 4, in Cos-7 cells transiently transfected with either GST-NCC (1 through 135) or GST-NCC (605 through 1031) truncation mutants in combination with either GFP-sortilin WT or TRU, NCC (1 through 135) was able to pull down sortilin WT (sample 1) as well as sortilin TRU (sample 2) but to a lesser extent, whereas NCC (605 through 1031) did not pull down either sortilin WT (sample 3) or sortilin TRU (sample 4). These results suggest that the amino terminus of NCC is the site of interaction with sortilin.

Distribution of Sortilin and Truncated Sortilin
We then examined the distribution of sortilin WT and TRU in Cos-7 cells. Sortilin WT is seen in the perinuclear region, peripheral granular structures, and plasma membrane ( Figure  5AA), whereas sortilin TRU is mainly seen in the perinuclear region with some in plasma membrane ( Figure 5AD). Sortilin WT co-localizes with TGN 38, a TGN marker, in the perinuclear region ( Figure 5AC). Sortilin TRU also co-localizes with Figure 2. The effect of sortilin on NCC protein expression is shown. Two days after transfection of Cos-7 cells with HA-NCC WT and either GFP-sortilin WT or truncated sortilin TRU in the absence or presence of myc-WNK4 WT as indicated, cells were lysed. Lysates were subjected to 5% SDS-PAGE followed by immunoblotting with anti-HA, anti-GFP, and anti-Myc antibodies. (A and C) Representative Western blot results. The top blot shows the steady-state protein level of NCC in both A and C. The bottom blot in A shows actin for protein loading control. The middle and bottom blots in C indicate WNK4 WT and sortilin protein expression, respectively. (B and D) Summary of the results of steady-state NCC protein expression for A (n ϭ 4) and C (n ϭ 6), respectively. Data are presented as the ratio of change from control groups (normalized to 100%) in which NCCϩGFP Vect is for A and NCCϩCD4ϩ GFP Vect is for C. There is no statistical difference in NCC protein expression in the absence of WNK4 in either NCCϩsortilin WT or NCCϩsortilin TRU groups as compared with control group in B. In the presence of WNK4 WT, NCC protein expression is significantly reduced in NCCϩWNK4ϩGFP Vect group as compared with either control group or NCCϩWNK4ϩsortilin TRU group in D (**P Ͻ 0.001). NCC protein expression in NCCϩWNK4ϩsortilin WT is also significantly reduced as compared with the control group (*P Ͻ 0.01). The level of NCC protein expression in NCCϩWNK4ϩsortilin TRU group was significantly higher as compared with the NCCϩWNK4ϩsortilin WT group (#P Ͻ 0.05).
TGN 38 in the perinuclear region ( Figure 5AF), but, unlike sortilin WT, the expression in the peripheral granular structure is lost. These data indicate that sortilin WT is distributed in the TGN in addition to vesicles and the plasma membrane, and sortilin TRU is mainly retained in the TGN.

NCC Co-localizes with Sortilin WT and Its Mutant Sortilin TRU
We further investigated the cellular location where NCC interacts with sortilin in Cos-7 cells transiently transfected with HA-NCC WT in combination with either GFP-sortilin or TRU. NCC in red ( Figure 5BB) is co-localized with sortilin WT in green ( Figure 5BA) in the perinuclear region and peripheral granular structure ( Figure 5BC), whereas a majority of NCC in red ( Figure 5BE) is co-localized with sortilin TRU ( Figure  5BD) in the perinuclear region and only a small amount of NCC is co-localized in the peripheral region ( Figure 5BF), suggesting that sortilin TRU complexed with NCC is mainly re-tained in the TGN. These findings suggest that the truncated sortilin TRU may bind the endogenous sortilin via a dominant negative effect and interfere with its interaction with NCC and therefore prevent NCC from WNK4-mediated lysosomal degradation.

Sortilin Expresses and Co-localizes with NCC in the Renal Distal Convoluted Tubule
To confirm sortilin expression in native kidney tissue, particularly in the distal convoluted tubule (DCT), where NCC is exclusively expressed, we performed immunohistochemistry, Western blot analysis, and reverse transcriptase-PCR. As shown in Figure 6A, endogenous sortilin was expressed in the cytoplasm and subapical region of the mouse DCT (mDCT) and cortical collecting duct and co-localized with NCC mainly in the subapical region of mDCT. Sortilin mRNA and protein were also expressed in Cos-7, mouse kidney and mDCT cell lines ( Figure 6, C and D). In addition, NCC and WNK4 were . Sortilin interacts with NCC. Cos-7 cells were co-transfected with either GFP-sortilin WT or truncated GFP-sortilin TRU alone or HA-NCC in combination with GFP-sortilin WT or GFP-sortilin TRU as indicated. Forty-eight hours after transfection, cells were lysed, IP and co-IP were performed, and cells were subjected to Western blot analysis. (A and B) Representative IP and co-IP results. The top blots in the IP section indicate the co-IP results using a polyclonal anti-GFP antibody for IP and probing with a monoclonal anti-HA antibody. The bottom blots in the IP section indicate IP results using an anti-GFP antibody for IP and probing with anti-GFP antibody. The blots in the lysate input section indicate the lysates expressing NCC and sortilin, respectively. In both samples 1 and 2 in A and B, anti-GFP antibody precipitates both sortilin WT and TRU, and both sortilin WT and TRU are able to pull down HA-NCC. In sample 5 in A and sample 3 in B, anti-GFP antibody cannot pull down NCC as the negative control. Samples 3 and 4 in A serve as negative controls as well, because cells were not transfected with HA-NCC. (C) Reciprocal co-IP results. Again, the top blot in the IP section indicates co-IP results using monoclonal anti-HA antibody for IP and probing with a polyclonal anti-GFP antibody. The bottom blot in the IP section indicates IP results using an anti-HA antibody and probing with anti-HA antibody. The blots in the lysate input section indicate the lysates expressing sortilin and NCC, respectively. Again, in samples 1 and 2, anti-HA antibody precipitates HA-NCC, and NCC is able to pull down sortilin WT as well as sortilin TRU, although to a lesser extent. Samples 3 and 4 serve as negative control. Ϫ, Lysates incubated with preimmune rabbit serum; ϩ, lysates incubated with rabbit anti-GFP antibody.
BASIC RESEARCH www.jasn.org expressed in the kidney and mDCT cells ( Figure 6, B and D). These data suggest that sortilin, NCC, and WNK4 are coexpressed in native DCT.

WNK4 Promotes NCC Targeting to Lysosome for Degradation
We further investigated whether WNK4 increases NCC targeting to the lysosomal compartment by immunofluorescence confocal microscopy. Cos-7 cells were transiently transfected with GFP-NCC with or without myc-WNK4. The cells were pretreated with lysosomal inhibitors leupeptin (30 M) and E64 (50 M) for 14 h before immunostaining; therefore, the pool of NCC that would normally be degraded in lysosomes could be visualized. As shown in Figure 7A, in the NCC alone group, NCC is distributed in plasma membrane, cytoplasm, and perinuclear region and has some degree of co-localization with lysosomal marker cathepsin D. In contrast, in the NCC plus WNK4 group, NCC seems to be mainly distributed in cytoplasm and the perinuclear region, and the degree of co-localization of NCC with cathepsin D seems to be increased. To verify these findings, we quantified the overlapping signals between NCC and cathepsin D in both groups. As shown in Figure 7B, the percentage of NCC co-localization with cathepsin D in NCC plus WNK4 group significantly increased as compared with NCC alone group (16.4 Ϯ 1.3 versus 32.3 Ϯ 3.2%; P Ͻ  In A, sortilin WT in green seems to be distributed in the perinuclear region and peripheral granular structure as well as plasma membrane. In B, TGN 38 is shown in red. C shows the merged picture, indicating the co-localization of sortilin WT with TGN 38 in the perinuclear region in yellow. In D, the truncated sortilin TRU in green is mainly retained in the perinuclear region and some in the plasma membrane, but its expression pattern seems to be different from sortilin WT's, because the peripheral granular structure is lost. In E, TGN 38 is shown in red. F shows the merged picture, indicating the co-localization of sortilin TRU with TGN 38 mainly in the perinuclear region in yellow. (B) In A and D, sortilin WT and sortilin TRU seem to be distributed in the similar patterns as described in A, respectively. In B, NCC in red is distributed in the perinuclear, cytoplasmic, and plasma membrane regions. C shows the merged picture, indicating the co-localization of NCC with sortilin WT in the perinuclear and cytoplasmic regions in yellow. In E, NCC in red is mainly distributed in the perinuclear region with some in the plasma membrane. F shows the merged picture, indicating that the co-localization of NCC with sortilin TRU in yellow is mainly in the perinuclear region, and its peripheral co-localization with NCC seems be less. These findings suggest that the truncated sortilin TRU may prevent NCC from WNK4-promoted degradation through the lysosomal pathway by retaining NCC in the perinuclear region (TGN) or diverting NCC to other vesicular compartments other than lysosomes. Bar ϭ 10 M. 0.001; n ϭ 19). These data suggest that WNK4 promotes NCC accumulation in lysosomal compartment for degradation.

NCC Co-localizes with Sortilin and WNK4
We previously showed that WNK4 interacts with NCC. 6 We now showed that NCC interacts with sortilin. To examine whether NCC, sortilin, and WNK4 coexpress in the same subcellular compartment, we evaluated the co-distribution of these proteins in Cos-7 cells co-transfected with HA-NCC, GFP-sortilin WT, and myc-WNK4 WT. As shown in Figure 8, NCC, sortilin, and WNK4 are co-localized in the perinuclear region and peripheral granular structures. These combined data suggest that NCC, sortilin, and WNK4 might be associated together and NCC might sequentially bind sortilin and WNK4 that ultimately facilitates the WNK4-promoted NCC targeting to the lysosome for degradation via a sortilin-mediated mechanism.

DISCUSSION
Emerging evidence has shown WNK kinase plays an important role in maintaining homeostasis of ion channels and trans-porters. [35][36][37] Our previous study showed that WNK4 enhances NCC degradation through a lysosomal pathway. 6 In this study, we showed that sortilin TRU reverses the inhibitory effect of WNK4 on NCC. We also showed that NCC interacts with sortilin WT as well as TRU, albeit to a lesser extent, via its Nterminus. Sortilin WT is co-localized with NCC throughout the cytoplasm, whereas the sortilin TRU is co-localized with NCC mainly in the perinuclear region. Sortilin is endogenously expressed in the kidney, especially in DCT cells, and co-localized with NCC in vivo. WNK4 increases NCC targeting to lysosomal compartment. In addition, WNK4, sortilin, and NCC seem to be co-localized in the perinuclear region. These data suggest that WNK4, NCC, and sortilin interact with one another and facilitate NCC targeting to the lysosome for degradation.
Previous studies showed that sortilin TRU is mainly retained in the perinuclear region (Golgi and TGN). 28 We showed that sortilin TRU bind NCC even though its affinity to NCC is less than sortilin WT; therefore, the reversal of the inhibitory effect of WNK4 on NCC could be due to sortilin TRU's dominant negative effect on endogenous sortilin, which interferes with its interaction with NCC and therefore prevents NCC from moving to the lysosome. We also showed that BASIC RESEARCH www.jasn.org WNK4 significantly increases NCC accumulation in the lysosomal compartment. These data suggest that WNK4 downregulates NCC by promoting NCC to the lysosome via a sortilin-mediated mechanism, which represents a novel mechanism for the regulation of NCC trafficking by WNK4 kinase.
Protein trafficking in the eukaryotic systems involves a series of protein-protein interactions between the cargo and protein transport carriers. Cargo proteins are selected for sorting to specific destinations by coat proteins or by receptors that interact with coat proteins. 38 For a cargo protein to exit a sorting compartment, a receptor needs to interact with cytoplasmic coat proteins, such as adaptor proteins and clathrin; then small-coated vesicles are formed from donor membranes in the TGN and traffic the cargo to acceptor membranes such as endosomes and lysosomes. Clathrin is a structural protein that is the main constituent in clathrin-coated vesicles (CCV). These vesicles also include the GGAs and either of two heterotetrameric AP complexes, AP1 or AP3. 25,26 The GGAs are newly identified monomeric clathrin APs, [22][23][24] which have similar function as the AP complexes. The VHS domain of the GGA protein functions as a recognition module for sorting signals. 39 The cytoplasmic tail of M6PR or sortilin binds GGA via its DXXLL motif and subsequently recruits clathrin to form CCVs designated to endosomes and lysosomes. In this study, we showed that NCC interacts with sortilin, most likely in the TGN. It is probable that NCC binds sortilin, subsequently recruiting GGA and clathrin to form CCV. 24,38,39 The CCV con- Figure 7. WNK4 promotes NCC targeting to the lysosomal compartment. Two days after Cos-7 cells transiently co-transfected with GFP-NCC alone or with myc-WNK4 WT, immunostaining and confocal microscopy were performed. Sixteen hours before fixation, Cos-7 cells were pretreated with lysosomal inhibitors leupeptin (30 M) and E64 (50 M); therefore, the pool of NCC that would normally be degraded in lysosomes could be visualized and the percentage of NCC co-localized with lysosomal marker cathepsin D was quantified for individual cells. (A) GFP-NCC is seen in green. Cathepsin D is detected by a polyclonal rabbit anti-cathepsin D followed by Cy3-conjugated secondary antibody in red. WNK4 is shown in cyan. DAPI in blue shows the nucleus staining. The merged pictures show the co-localization of NCC with cathepsin D in yellow. In the absence of WNK4, NCC is distributed in plasma membrane, cytoplasmic, and perinuclear regions, and it shows some extent of co-localization with cathepsin D, whereas in the presence of WNK4, NCC is seen mainly in cytoplasmic and perinuclear regions as well as less in plasma membrane, and the degree of its co-localization with cathepsin D seems to be increased. (B) To quantify the percentage of NCC co-localization with cathepsin D in these two groups, the images of 19 cells from three separate experiments in each group were obtained by z axis scanning under confocal microscopy. The percentage of the co-localization of NCC with cathepsin D in yellow was quantified using the manufacturer's software (Zen; Zeiss). In the presence of WNK4, the percentage of co-localization of NCC with lysosomal marker cathepsin D is significantly higher than that in the absence of WNK4, indicating that WNK4 promotes NCC targeting to lysosomal compartments for degradation. Figure 8. NCC co-localizes with sortilin and WNK4. Cos-7 cells were co-transfected with myc-WNK4 WT, GFP-sortilin WT, and HA-NCC WT. Forty-eight hours after transfection, immunostaining and confocal microscopy were performed. The distribution pattern for sortilin WT in green seems to be similar to sortilin WT as described in Figure 5A, and NCC expression pattern in red seems to be similar to the NCC pattern as described in Figure 5B. WNK4 in blue is expressed in the perinuclear and cytoplasmic regions. The merged picture shows the co-localization of WNK4, sortilin, and NCC in the perinuclear region in white, indicating that WNK4, sortilin, and NCC might be associated together.
taining NCC may also associate with AP3 to target to the lysosome for degradation. A recent study also showed that WNK4 increases the association of NCC with AP3 to promote NCC delivery to the lysosomal compartment for degradation, 40 which is consistent with our conclusion. Our findings further support the notion that NCC is regulated through affecting its forward trafficking in addition to other mechanisms such as altering its abundance, 41-43 glycosylation, 10 and phosphorylation. 44,45 In the kidney, NCC is exclusively expressed in DCT and responsible for 5 to 7% of sodium reabsorption. 46 It is likely that, under normal physiologic conditions, NCC constantly traffics to both the plasma membrane via a secretory pathway and the lysosome for degradation to maintain its basal steady protein level, which is subject to regulation by many mediators. 3,6,10,[41][42][43][44][45] Renal DCT cells contain endogenous WNK4 kinase. 2 In the presence of WNK4, the rate of NCC targeting to the lysosomal compartment is much higher than that of NCC without WNK4. We also showed that WNK4, NCC, and sortilin co-localize together in the perinuclear region, indicating that they might be associated together and/or interact in the same vesicles, traffic together, and facilitate the WNK4-promoted NCC targeting to the lysosome for degradation through a sortilin receptor-mediated mechanism; however, because we used overexpression system, overall protein degradation will inevitably be increased to cope with the excess protein, which is a potential limitation in our study. We believe that, overall, increased protein degradation would not necessarily alter the effect of WNK4 on the regulation of NCC. Taken together, our data suggest that WNK4 kinase could play a role in redirecting NCC's destination. The exact mechanism of how WNK4 alters NCC trafficking needs to be further explored.
Sortilin not only is involved in sorting of lysosomal hydrolases 13 but also is engaged in intracellular sorting of Glut4, 32,34,47 endocytosis, 48 and signal transduction. 49,50 In this study, we showed that NCC interacts with sortilin WT as well as TRU, to a lesser extent. Sortilin TRU, which lacks the cytoplasmic tail, is still able to bind NCC, indicating that NCC interacts with the luminal domain of sortilin. Because sortilin TRU has low binding affinity to NCC, other potential binding sites in the cytoplasmic tail of sortilin cannot be excluded. The exact reason that sortilin TRU has a low affinity for NCC needs to be clarified. We also found that the N-terminus of NCC is responsible for its interaction with sortilin, but we do not know which sequence or motif in the N-terminus is involved in this interaction. Previous studies demonstrated that WNK4 inhibits NCC surface expression and its function in Xenopus oocytes 3,4 and mammalian cells 6,11 by a kinase-dependent mechanism. Because the N-terminus of NCC is responsible for interaction with sortilin, it is speculated that WNK4 might phosphorylate the N-terminus of NCC directly or indirectly and alter its interaction with sortilin. Although the mechanism remains to be clarified, it is clear that WNK4 affects NCC sorting direction and targeting. Further experiments will be needed to examine whether the phosphorylation status at the N-terminus of NCC affects its trafficking or final sorting destination. Unlike WNK4, WNK3 stimulates NCC activity in Xenopus oocytes. 51,52 WNK3 and WNK4 share a high degree of amino acid identity within the kinase domain but a much lower degree of amino acid identity in the amino and carboxyl terminal domains. San-Cristobal et al. 53 reported that amino terminus of WNK3 or WNK4 plays an important role in regulation of NCC activity. How the amino terminal domain of WNK4 affects NCC function or trafficking, affects the interaction with sortilin and NCC, and ultimately changes the direction of NCC sorting remains unknown. Elucidating these molecular mechanisms will definitely provide a novel view on NCC regulation by WNK kinase.
Sortilin has been shown to play an important role in participating in sorting membrane proteins or transporters such as NCC. This study showed that WNK4 promotes NCC degradation through a lysosomal pathway via a sortilin-mediated mechanism, which provides a novel mechanism underlying NCC regulation by WNK kinase.

Plasmids and Constructs
Human WNK4 WT and human NCC WT were amplified by PCR from a human kidney cDNA library and then subcloned into pCMVtaq 3B vector (Stratagene, La Jolla, CA) and pCMV-HA vector (Clontech, Palo Alto, CA), respectively, as described previously. 6 The GSTtagged full-length NCC, N-terminal intracellular NCC (1 through 135), and C-terminal intracellular NCC (605 through 1031) constructs were generated by PCR amplification using pCMV-HA-NCC as template and by insertion of respective cDNA into modified PRK5-GST vectors. The pEGFP-sortilin WT and its truncated sortilin lacking cytoplasmic tail (pEGFP-sortilin TRU), a dominant negative mutant, were provided by Dr. Carlos R. Morales (McGill University, Hamilton, Ontario, Canada). 27 All constructs, shown in Figure 1, were confirmed by sequence analysis. These constructs have been successfully expressed in African green monkey kidney cells (Cos-7) and confirmed by Western blot analysis.

Cell Culture, Transfection, and Animals
Cos-7 cells obtained from American Type Culture Collection (Manassas, VA) were maintained in DMEM supplemented with 10% FCS, L-glutamine (2 mM), penicillin (100 U/ml), and streptomycin (100 g/ml). mDCT cells were a gift of Peter Friedman (University of Pittsburgh, Pittsburgh, PA). 43,54 mDCT cells were maintained in DMEM/ F-12 medium containing 5% FCS, 1% penicillin, 1% streptomycin, and 1% neomycin. All other media and components were purchased from Invitrogen (Carlsbad, CA). Cells were transiently transfected using LipofectAMINE 2000 (Invitrogen) according to the manufacturer's instructions. Forty-eight hours after transfection, cells were used for Western blot analysis, IP, or immunostaining. All animal protocols were approved by the Emory University Institutional Animal Care and Use Committee. The kidneys from adult male Sprague-Dawley rats (Charles River Laboratory, Wilmington, MA) and C57BL/6J mice (Jackson Laboratories, Bar Harbor, ME), were harvested for Western blot analysis and immunohistochemistry.

Western Blotting Analysis
Cos-7 cells were harvested and processed as described previously. 6 Briefly, cells transiently transfected with various DNA constructs as indicated were lysed in lysis buffer containing 20 mM HEPES (pH 7.5), 120 mM NaCl, 5.0 mM EDTA, 1.0% Triton X-100, 0.5 mM dithiothreitol, 1.0 mM PMSF, and complete protease inhibitor (Roche Diagnostics, Mannheim, Germany; 1 tablet per 50 ml of solution). The lysates were spun at 6000 ϫ g for 5 min to pellet the insoluble material, and the proteins from supernatant were quantified using a Pierce BCA Protein Assay kit (Pierce, Rockford, IL). After mixing in Laemmli buffer (Bio-Rad, Hercules, CA) and incubating at 37°C for 30 min, the protein sample was separated by SDS-PAGE and electrophoretically transferred to polyvinylidene difluoride (PVDF) membranes (Amersham Biosciences, Piscataway, NJ) for Western blot. The probing with specific antibodies and subsequent detection with ECL plus system (Amersham Biosciences) or Super signal (Pierce) were performed according to standard procedures as described previously. 6,55 IP and Co-IP IP and co-IP were preformed as described previously. 55 In brief, Cos-7 cells were transfected with plasmids as indicated. Forty-eight hours after transfection, the cells were lysed. The cell lysates were incubated with primary antibodies for 2 h, then Protein G/A Sepharose beads were added, and the lysates were further mixed and incubated overnight at 4°C. After washing with lysis buffer twice and PBS once, beads were then eluted with Laemmli buffer (Bio-Rad). The eluted proteins were separated by SDS-PAGE, transferred onto PVDF membrane, and probed with appropriate antibodies. For co-IP experiments, blots were first probed with an antibody to the first binding partner. The PVDF membranes were then stripped and reprobed with an antibody to the second binding partner. Reciprocal IP was also performed with respective antibodies.

GST Pull-Down Assay
GST pull-down assay was modified and carried out as described previously. 56 In brief, Cos-7 cells were co-transfected with GST-tagged N-terminus or C-terminus of NCC in combination with either sortilin WT or sortilin TRU. Forty-eight hours after transfection, cells were lysed in lysis buffer, and cell lysates were prepared as described previously. 6 The resultant supernatants were incubated with Glutathione-Sepharose 4B (Amersham Biosciences) overnight at 4°C. The Glutathione-Sepharose 4B beads were then washed five times with lysis buffer. The bound protein complexes were resolved on a 7% SDS-PAGE and transferred to a PVDF membrane. Subsequently, the blots were probed with the indicated antibodies.

Immunostaining, Immunohistochemistry, and Confocal Microscopy
Immunostaining and confocal microscopy were performed as described previously. 6 In brief, for immunostaining experiments, the transfected Cos-7 cells were fixed and blocked with 5% nonfat dry milk in PBS for 1 h, then incubated with primary antibodies for 30 min at 37°C. Cells were then washed with PBS three times followed by 30 min of incubation at 37°C with appropriate secondary antibodies conjugated to either fluorescein (FITC) or isothiocyanate Cy3 fluorescent dye (Jackson ImmunoResearch Laboratory, West Grove, PA). After staining, the coverslips were washed thoroughly with PBS three times, mounted with antiquenching medium (Vector Laboratories, Burlingame, CA), and sealed. For immunohistochemistry, in situ fixation of mouse kidneys was performed as described previously. 57 For paraffin embedding, tissues were dehydrated in a series of graded ethyl alcohol followed by xylene and embedded in paraffin and then cut into 2-m sections. Sections were deparaffinized and rehydrated. Endogenous peroxidase was blocked by 0.5% H 2 O 2 in absolute methanol for 30 min at room temperature. For revealing antigens, sections were incubated in 1 mM Tris solution (pH 9.0) supplemented with 0.5 mM EGTA and heated in a microwave oven for 10 min. Nonspecific binding of IgG was prevented by incubating the sections in 50 mM NH 4 Cl-PBS for 30 min, followed by blocking in PBS supplemented with 1% BSA, 0.05% saponin, and 0.2% gelatin. Sections were incubated for 30 min at 37°C with primary antibodies (monoclonal antisortilin antibody [1:200] or rabbit anti-NCC antibody [1:200] from Dr. Mark Knepper, National Institutes of Health) diluted in PBS supplemented with 0.1% BSA and 0.3% Triton X-100. After rinsing with PBS, sections were incubated with either Cy-3 or FITC-conjugated secondary antibody (1:200) for 30 min at 37°C. After washing with PBS, sections were mounted with antiquenching medium (Vector Laboratories) and sealed. The slides were examined using a confocal laser scanning microscope (model LSM510; Zeiss). The images were acquired using the manufacturer's software and prepared for publication with Adobe Photoshop. The co-localization analyses were performed using the manufacturer's software (Zen; Zeiss).

Statistical Analysis
The data are presented as means Ϯ SE. Statistical significance was determined by t test when two groups were compared or by one-way ANOVA, followed by Bonferroni post hoc tests when multiple groups were compared. We assigned significance at P Ͻ 0.05.