Introduction

Protein kinase C alpha (PKCα) is a member of the classic PKC family1, which is widely expressed in the mammalian kidney2, 3, 4 and contributes to various kidney functions, including substrate absorption and urine concentration5, 6. PKCα-mediated cytoskeleton remodeling results in endocytosis in epithelial cells, which increases the uptake of various substrates, such as NHE3 and albumin7, 8, 9. PKCα is localized in glomeruli, the intercalated cells of the cortical collecting duct, and the medullary-collecting duct in the mouse kidney4. Genetic knockout mice that lack PKCα exhibit decreased glomerular filtration rate, increased urinary output and lower urinary osmolarity, accompanied by normal water intake and normal levels of the plasma antidiuretic hormone arginine vasopressin (AVP)5. These results strongly suggest that PKCα-mediated urine concentration primarily occurs within the medullary collecting duct.

Aquaporin-2 (AQP2) is a vasopressin-regulated water channel in the principal cells of the connecting tubule and the collecting duct in the kidney10, 11. AQP2 is stored in an intracellular compartment and plays an important role in the regulation of urine concentration12, 13, 14, 15. AVP stimulates AQP2 translocation to the plasma membrane for the re-absorption of water. AQP2 is removed from the plasma membrane and returned to the intracellular compartment when the stimulation in terminated16, 17, 18. In addition, AVP also regulates AQP2 mRNA and protein levels, which mediates long-term regulation19, 20. Both short- and long-term regulation mechanisms are involved in the pathophysiology of AQP2-mediated urine concentration21. Although PKC is involved in angiotensin II-mediated AQP2 expression and trafficking22, our previous study demonstrated that AQP2 expression is only slightly different between PKCα knockout mice and wild-type mice5, which strongly suggests that the long-term AQP2 regulation mechanism does not contribute to PKCα-mediated urine concentration. AQP2 transportation is partially regulated by the cytoskeleton23, 24, 25, and PKCα-mediated remodeling of the actin cytoskeleton is involved in constitutive albumin uptake in the renal proximal tubule9. However, the contribution of PKCα to AQP2 trafficking, the role of the cytoskeleton in PKCα-mediated AQP2 transportation and the mechanism of PKCα regulation of AQP2 translocation are less well understood.

We hypothesized that PKCα mediates AQP2 trafficking via cytoskeleton distribution because of the important role of PKCα in urine concentration and cytoskeleton remodeling. Therefore, this study revealed the crosstalk among PKCα, AVP and AQP2 in trafficking and cytoskeletal remodeling in mIMCD3 cell.

Materials and methods

Constructs, antibodies, and reagents

The pEF-nero-PKCα A/25E vector was a gift from Dr Gottfried BAIER (Institute for Medical Biology and Human Genetics, University of Innsbruck, Innsbruck, Austria). The AQP2-GFP-pCMV6 construct was purchased from Origene (Rockville, MD, USA). Lipofectamine plus and geneticin (G418) were purchased from Invitrogen (Shanghai, China). Protein A/G-agarose beads and primary antibodies against phosphor-S256-AQP2, AQP2, PKCα, and GAPDH were purchased from Santa Cruz (Heidelberg, Germany). The primary antibody against α-tubulin was purchased from Calbiochem (Darmstadt, Germany). The secondary goat anti-rabbit IgG conjugated with Cy3 antibody and the analogue 1-desamino-8-D-arginine vasopressin (DdAVP) were purchased from Sigma (Shanghai, China). DAPI was purchased from Vector Burlingame (San Diego, CA, USA). Sulfo-Link NHS-LC-biotin and streptavidin-agarose beads were purchased from Pierce (Beijing, China). Major apparatuses included a Mini-PROTEAN II Electrophoresis Cell (Bio-Rad, Shanghai, China) and a confocal laser-scanning microscope (Olympus FV500, Japan).

Cells and cell culture

Immortalized mouse inner medullary collecting duct 3 (mIMCD3) cells were kindly provided by Dr John M LUK (Department of Urology, University of Hong Kong, Hong Kong, China). mIMCD3 cells were maintained in Dulbecco's modified Eagle's medium/Ham's F12 (1:1) (GIBCO, Invitrogen, Shanghai, China) supplemented with 10% fetal bovine serum (GIBCO, Invitrogen, Shanghai, China) and 2% penicillin-streptomycin (Amresco, Shanghai, China) in a humidified atmosphere with 5% CO2 at 37 °C.

Generation of pGCsi-U6/Neo PKCα shRNA plasmid

Three siRNAs were designed and synthesized by Invitrogen (Shanghai, China) according to the cDNA sequence of PKCα (NM_011101): #1, GTCCTTCACGTTCAAATTA; #2, GTGCAGTATGAAACTCAAA; and #3, CCATCCAACAACCTGGACA. The siRNAs were cloned into the eukaryotic expression plasmid pGCsi-U6/Neo (Genechem, Shanghai, China). The constructed pGCsi-U6/Neo-PKCα-siRNA vector was transfected into mIMCD3 cells. Western blotting was used to evaluate the suppression of PKCα expression in different cell groups.

Transfection

Transfection of mIMCD3 cells with the AQP2-GFP construct was performed using Lipofectamine 2000 (Shanghai, China) according to the manufacturer's instructions. The cells were maintained in medium containing 400 μg/mL geneticin for 24 h after transfection. Individual neomycin-resistant colonies were selected and expanded for 14 d after transfection.

AQP2-GFP was stably expressed in mIMCD3 cells (AQP2-mIMCD3 cells) that were transiently transfected with a eukaryotic expression vector encoding the constitutively activated form of PKCα A/25E (pEF-nero-PKCα A/25E)26 using Lipofectamine 2000 according to the manufacturer's instructions. AQP2-mIMCD3 cells were transfected with three pGCsi-U6/Neo-PKCα shRNAs and scrambled shRNA using Lipofectamine 2000 according to the manufacturer's instructions to evaluate the inhibition efficiency of the siRNAs. The cells were harvested 48 h after transfection, and Western blot and immunofluorescence were used to evaluate the transfection efficiency.

Immunocytochemical staining

AQP2-mIMCD3 cells were divided into 3 groups: (1) Scrambled shRNA transfected; (2) PKCα A/25E vector transfected; and (3) PKCα shRNA transfected. The cells in each group were treated with 100 μmol/L DdAVP for 30 min. DdAVP was removed, and the cells were incubated with culture medium for 2 h (washout). The cells were then fixed in 4% paraformaldehyde for 10 min, rinsed twice in PBS, and blocked for 15 min in a blocking/permeabilization solution (PBS containing 0.1% BSA and 0.3% Triton X-100). The cells were washed with PBS and pre-incubated in PBS containing 1% BSA for 20 min. The cells were incubated with an α-tubulin antibody (at a final concentration of 2 μg/mL) at 4 °C overnight. The cells were washed with PBS and incubated in the secondary goat anti-rabbit IgG conjugated with Cy3 antibody (at a final concentration of 2 μg/mL) for 1.5 h at RT. Finally, the cells were washed twice with PBS and mounted with DAPI for confocal laser-scanning microscopy equipped with a CoolSNAP HQ camera.

Biotinylation of surface membrane proteins

Identifiably transfected AQP2-mIMCD3 cells were cultured on polylysine-coated 60-mm dishes for 48 h and treated with DdAVP for 0 min, 30 min, or 30 min followed by a 2-h washout. The cells were washed three times with PBS/glycine (137 mmol/L NaCl, 2.7 mmol/L KCl, 10 mmol/L Na2HPO4, 1.8 mmol/L KH2PO4, pH 8.0, and 5 mmol/L glycine) and twice with PBS27. An ice-cold biotinylation reagent (0.5 mg/mL in PBS) was immediately added to cells. The biotinylation reagent was removed after a 10-min incubation on ice, and the cells were washed three times with PBS/glycine.

Immunoblotting and immunoprecipitation

Transfected and DdAVP-treated cells were lysed in RIPA buffer (10 mmol/L Tris-HCl, 0.15 mol/L NaCl, 1% NP-40, 1% Na-deoxycholate, 0.5% SDS, 0.02% sodium azide, and 1 mmol/L EDTA, pH 7.4). The cell lysates were clarified by centrifugation at 1000×g for 10 min at 4 °C. Protein concentrations were measured using the bicinchoninic acid protein assay reagent kit (Pierce, Thermo Fisher Scientific, Beijing, China) and adjusted to the same concentration using lysine buffer. A 0.5-mL aliquot of the cell lysates was incubated with 50 μL of streptavidin-agarose beads at 4 °C overnight to capture biotinylated proteins, 50 μL of protein A/G-agarose beads plus the AQP2 antibody to capture total AQP2, or 50 μL of protein A/G-agarose beads plus the PKCα antibody to capture total PKCα. The beads were pelleted, washed, and boiled for 5 min in 50 μL of cracking buffer (50 mmol/L Tris-HCl [pH 7.0], 10% glycerol, 2% SDS, and 2% β-mercaptoethanol). The eluted immunoprecipitates were separated on 10% polyacrylamide gels and transferred electrophoretically to PVDF membrane. The membranes were blocked for 90 min with blocking buffer during gentle shaking and incubated overnight with antibodies against AQP2 (0.5 μg/mL), phospho-S256-AQP2 (0.5 μg/mL), PKCα (0.5 μg/mL) or GAPDH (0.5 μg/mL). The membranes were incubated with an HRP-conjugated IgG (0.1 μg/mL) for 2 h at room temperature. Finally, the immunoreactive signals were visualized using an ECL reagent (Beyotime, Shanghai, China) and quantified using the Image J program (NIH, Bethesda, MD, USA).

Statistical analyses

The values are presented as the means±SEM. The data were analyzed using SPSS software. Statistical significance of differences was assessed using unpaired t-tests or two-way ANOVAs. A value of P<0.05 was viewed as statistically significant.

Results

Inhibition of PKCα and the overexpression of AQP2 in mIMCD3 cells

All three PKCα shRNA plasmids (#1, #2, and #3) and PKCα scrambled shRNA plasmids were transfected into mIMCD3 cells to select the shRNAs with the highest inhibition efficiency to PKCα. Crude proteins were extracted from the harvested cells 48 h after transfection and subjected to Western blotting using the PKCα antibody. PKCα shRNA #1 plasmids exhibited the greatest inhibitory effect on PKCα, but all three PKCα shRNAs demonstrated visible inhibition (P<0.01) (Figure 1A).

Figure 1
figure 1

Inhibition of PKCα and the overexpression of AQP2 in mIMCD3 cells. (A) PKCα shRNA plasmid #1, #2, #3, and PKCα scramble shRNA plasmid (control) were transfected into mIMCD3 cells respectively. Forty-eight hours after transfection, crude proteins extracted from harvested cells were subjected to Western blot using antibodies against PKCα. Representative blot was shown (left) and quantitative analysis of PKCα protein levels were normalized to control group (n=3 each group) (right). bP<0.05 vs control group. (B) AQP2-mIMCD3 stable cell line was generated by transfecting to express AQP2-GFP (See Materials and methods). The expression of AQP2-GFP was detected in selected WT7 stable cell lines. (C) AQP2-GFP stably expressed WT7 cells were transfected with scramble shRNA (Con), PKCα A/25E constructs or PKCα shRNA, respectively. Forty-eight hours after transfection, crude proteins extracted from harvested cells were subjected to Western blot using antibodies against PKCα (left). Quantitative analysis of PKCα protein levels were normalized to control group (n=3 each group) (right). bP<0.05 vs control group.

The transfection of mIMCD3 with the AQP2-GFP-pCMV6 plasmid produced 12 geneticin-resistant clones, 10 of which were subjected to Western blot analysis (data not shown). The clone that amply expressed AQP2 was selected and named WT-7. A strong signal at 30 kDa in the selected AQP2-mIMCD3 stable cell line (WT-7) was observed, but no signal was detected in nontransfected mIMCD3 cells (Figure 1B). These results suggest the successful preparation of AQP2-mIMCD3 stable cell lines.

The WT-7 stable cell line was used to investigate PKCα expression following different transfections of PKCα scrambled shRNA, PKCα A/25E, and PKCα shRNA (#1) using Western blot. The transfection of PKCα A/25E significantly enhanced PKCα expression, but PKCα shRNA (#1) dramatically suppressed PKCα expression (P<0.05) (Figure 1C). The WT-7 cell line and PKCα A/25E and PKCα shRNA (#1) plasmids were utilized in subsequent experiments.

DdAVP enhanced the phosphorylation of AQP2 and PKCα in vitro

The physical association of AQP2 with endogenous PKCα was examined in vitro in WT7 cell lines to reveal the direct role of PKCα on AQP2. Cells with or without DdAVP treatment for 30 min were lysed, and PKCα was immunoprecipitated using anti-PKCα antibodies. The immunoprecipitates were analyzed using Western blot and AQP2 antibodies. Strong signals for AQP2 and PKCα were detected in whole cell lysates and immunoprecipitates, but no AQP2 signal was detected in PKCα immunoprecipitates (Figure 2). These results suggest that PKCα did not interact with AQP2 in vitro. However, DdAVP treatment enhanced AQP2 phosphorylation at serine 256. Down-regulation of PKCα by shRNA did not alter DdAVP-mediated AQP2 phosphorylation at serine 256. These data indicated that PKCα did not influence DdAVP-induced AQP2 phosphorylation at serine 256 (Figure 2).

Figure 2
figure 2

DdAVP enhanced the phosphorylation of AQP2 and AQP2 did not physically interact with PKCα in vitro. AQP2-GFP stably expressed WT7 cells were transfected with indicated constructs. Forty-eight hours after transfection, cells were treated with or without DdAVP for 30 min. Cells were then lysed and immunoprecipitation was performed with anti-PKCα and anti-AQP2 antibodies. Crude proteins from whole cell lysates and the immunoprecipitates were subjected to Western blot with indicated antibodies.

PKCα expression influenced the distribution of α-tubulin and DdAVP-mediated AQP2 trafficking in mIMCD cells

An increase in PKCα activity induces the re-organization of microtubules in proximal tubular cells9. Therefore, PKCα-induced alterations in microtubule architecture were investigated. Most cells that were transfected with scrambled shRNA plasmid or PKCα shRNA plasmids exhibited a prominent dense network of microtubules around the nucleus with a considerably lower microtubule content in the cell periphery under basal conditions. The constitutive activation of PKCα increased microtubule formation in the cell periphery (Figure 3, upper panel and middle panel). AQP2-mIMCD3 cells (WT7 cells) were incubated with DdAVP for 30 min followed by a 2-h washout to evaluate the involvement of PKCα in AVP-induced microtubule formation in the cell periphery. Cells that were transfected with the control and PKCα A/25E plasmids displayed increased microtubule formation in the cell periphery following DdAVP stimulation. In contrast, the down-regulation of PKCα expression using shRNA inhibited DdAVP-induced microtubule depolymerization (Figure 3, lower panel). Taken together, these data strongly indicated that the expression of PKCα regulated microtubule distribution in mIMCD cells.

Figure 3
figure 3

PKCα expression influenced the distribution of α-tubulin and DdAVP-mediated AQP2 trafficking in mIMCD3 cells. AQP2-GFP stably expressed WT7 cells were transfected with PKCα cramble shRNA (Con), PKCα A/25E constructs or PKCα shRNA constructs. Forty-eight hours after transfection, cells were treated with or without DdAVP for 0 min, 30 min or 30 min followed by 2 h washout. Cells were then fixed, permeablized and stained with antibodies aganist PKCα (green) and α-tubulin (red) for immunofluorescence. Nuclear was stained by DAPI (blue).

The role of PKCα in DdAVP-induced AQP2 trafficking was investigated because PKCα influences microtubule reorganization, and the DdAVP-induced depolymerization of microtubules affects the redistribution of AQP2 to the plasma membrane25, 28, 29. Immunofluorescence analyses indicated that AQP2 was primarily localized around the nucleus under basal condition. Treatment with DdAVP for 30 min produced a pronounced increase in AQP2 localization in the plasma membrane. DdAVP washout restored the perinuclear localization of AQP2. The expression of constitutively activated PKCα produced a wide distribution of AQP2 throughout the cytoplasm in contrast to AQP2 localization around the nucleus under basal conditions. DdAVP also induced the translocation of AQP2 to the plasma membrane in the PKCα A/25E transfected group. The down-regulation of PKCα expression by shRNA interference inhibited DdAVP-induced AQP2 translocation to the plasma membrane (Figure 3).

Regulation of PKCα expression influenced DdAVP-stimulated plasma location of AQP2

mIMCD-3 surface membrane proteins were biotinylated following DdAVP treatment for 0 min, 30 min, and 30 min followed by a 2-h washout to confirm the immunofluorescence results. Biotinylated proteins were captured from whole-cell lysates using streptavidin beads, separated by SDS-PAGE and probed using an AQP2 antibody. Immunoprecipitation was performed in whole cell lysates in parallel using AQP2-specific antibodies. The DdAVP-induced increase in biotin-AQP2 was partially inhibited by PKCα down-regulation with shRNA transfection, but biotin-AQP2 was enhanced by the overexpression of PKCα A/25E (Figure 4). Moreover, different treatments only marginally affected the total amount of AQP2 as determined by AQP2 immunoprecipitation (Figure 4). These results were consistent with the AQP2 translocation observed by immunofluorescence detection and demonstrated that PKCα participated in DdAVP-mediated AQP2 translocation in mIMCD cells.

Figure 4
figure 4

Regulation of PKCα expression influenced DdAVP stimulated plasma location of AQP2. (A) AQP2-GFP stably expressed WT7 cells were transfected with indicated constructs. Forty-eight hours after transfection, surface membrane proteins were biotinylated at 0 min, DdAVP treated 30 min, and DdAVP treated 30 min followed by 2 h washout. Whole cell lysates were subjected to avidin pulldown assay using streptavidin-agarose beads or immunoprecipitation (IP) using anti-AQP2 antibodies. The recovered proteins were separated by SDS-PAGE and analyzed by Western blot for AQP2 as described in Materials and Methods. For each condition, the detected signals from avidin-pulldown assay and AQP2 immunoprecipitation were from same SDS-PAGE gel. (B) Signals from 3 independent experiments were quantified by densitometry. The intensity of the AQP2 signals from the avidin pulldowns were normalized to the corresponding AQP2 IP signals under each condition. Data were represented as mean±SEM. bP<0.05 vs DdAVP 0 min group. eP<0.05 vs DdAVP 30 min+2 h washout group. hP<0.05 vs DdAVP 30 min con group. kP<0.05 vs DdAVP 30 min PKCα A/25E group.

Discussion

PKCα exhibits different expression patterns in mouse and rat kidneys. For example, PKCα localizes in the proximal tubule of mice but not rats3, 4. Our previous study demonstrated that PKCα contributes to urine concentration in the mouse kidney5. The use of cell lines that originate from mouse IMCD cells is reasonable because AQP2 is primarily expressed in collecting duct principal cells, and the effect of PKCα on urine concentrating occurs in the inner medulla. In the present study, AQP2 trafficking was examined in an mIMCD3 cell line that was derived from mouse IMCD and exhibited the properties of IMCD cells. However, immortalized mIMCD3 cells lack AQP2 expression30, and AQP2 is only briefly expressed in primary cultured IMCD cells31. Therefore, we generated AQP2-mIMCD3 stable cell lines for the following experiments. Our results demonstrated that AQP2 was located in intracellular vesicles in non-stimulated mIMCD3 cells. DdAVP treatment induced the translocation of AQP2 from the cytoplasm to the plasma membrane, and DdAVP washout reversed AQP2 localization from the plasma membrane to the cytoplasm (Figure 4). Therefore, AQP2-mIMCD3 cells provide an optimal cell model to investigate AQP2 trafficking.

AQP2 phosphorylation at serine 256 by PKA plays an important role in AQP2 trafficking32, 33. However, the phosphorylation of AQP2 at serine 256 is not sufficient to maintain the presence of the water channel at the plasma membrane34, and the PKC-induced internalization of AQP2 in collecting duct cells does not depend on the phosphorylation state of AQP235, 36. Our study demonstrated that DdAVP increased AQP2 phosphorylation, and the inhibition of PKCα expression did not influence DdAVP-mediated AQP2 phosphorylation at serine 256, which is consistent with previous reports. These results strongly suggest that AQP2 phosphorylation at serine 256 is PKCα independent. However, the ability of PKCα to phosphorylate AQP2 at other phosphorylation sites, such as serine 231, remains unclear. Our results indicated that PKCα did not interact with AQP2 in vitro. Therefore, the ability of PKCα to mediate the trafficking of phosphorylated AQP2 requires further investigation.

The disruption of microtubules induces AQP2 translocation to the plasma membrane13, 23, 29, and PKCα is involved in cytoskeletal remodeling during cell motility, phagocytosis, neurite outgrowth and the regulation of cytoskeleton-associated proteins37, 38, 39, 40. Therefore, the impact of PKCα on the cytoskeleton in AQP2-mIMCD3 cells was investigated. Microtubules are part of the cytoskeleton and consist of α, β, and γ tubulins41. α-Tubulin is the major cytoskeletal protein related to cell trafficking. The role of PKCα in the assembly of α-tubulin was determined. PKCα overexpression or inhibition produced a constitutively active form of PKCα or down-regulated expression of PKCα, respectively, which resulted in the de- and re-polymerization of α-tubulin, respectively. Moreover, PKCα down-regulation prevented DdAVP-mediated α-tubulin depolymerization. These data indicated that PKCα is involved in the assembly of α-tubulin in AQP2-mIMCD3 cells, which is consistent with previous reports9.

The microtubule-dependent transport of AQP2 is predominantly responsible for AQP2 trafficking and localization inside of the cell after its internalization, but it is not responsible for the exocytic transport of the water channel25. In the present study, the overexpression of constitutively active PKCα produced a wide distribution of AQP2 throughout the cytoplasm, which is in contrast to its translocation to the plasma membrane. This result indicates that other factors/proteins rather than α-tubulin are required for AQP2 translocation to the plasma membrane40. Furthermore, the consistent inhibition of α-tubulin depolymerization by the down-regulation of PKCα expression produced a loss in DdAVP-mediated AQP2 translocation. This result confirms that α-tubulin is a key element for the proper localization of AQP2 in the cytoplasmic compartment.

In summary, the use of stably expressing AQP2-mIMCD3 cells demonstrated that PKCα did not functionally associate with AQP2. The down-regulation of PKCα expression altered the distribution of α-tubulin and inhibited DdAVP-mediated AQP2 trafficking. However, constitutively activated PKCα rescued or aggravated these changes. These data directly demonstrated that PKCα mediates AQP2 trafficking by influencing the assembly of α-tubulin and underscore the complexity of the molecular events of PKCα-mediated urine concentration.

Author contribution

Dr Li-jun YAO designed the research and wrote the paper; Dr Hong ZHAO performed biotinlyation of cell surface protein experiment; Xi YAO performed Western blot and immunocytochemistry experiment; Tao-xia WANG prepared plasmids and analysed data; Wen-min JIN performed cell culture experiment; Qian-qian JI contributed reagents preparation; Dr Xiao YANG wrote the paper; Dr Qiu-hong DUAN performed the research.