Polyclonal rabbit anti-SGK3, anti-phospho-Nedd4-2 (Ser342), anti-Nedd4-2 and anti-phospho-GSK3(α/β) antibodies for immunoblotting and immunofluorescence were purchased from Cell Signaling Technology, Inc., (Beverly, MA, United States). Polyclonal rabbit anti-SGK3 antibody for immunohistochemistry was purchased from Abnova (Taipei, Taiwan, China). Monoclonal mouse anti–phospho-GSK3(α/β) antibody for immunoblotting, Polyclonal rabbit anti-SGK3 antibody for immunofluorescence, and monoclonal mouse anti-nephrin antibody for ubiquitination and immunofluorescence were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, United States). Polyclonal rabbit nephrin for antibody immunoblotting and immunohistochemistry was purchased from boster (Wuhan, China). Monoclonal mouse anti-ubiquitin (UB) antibody was purchased from Covance Inc., (UT, United States). Monoclonal mouse anti-GAPDH, rabbit polyclonal anti-Na,K-ATPase, rabbit polyclonal anti-flag tag, horseradish peroxides-conjugated anti-rabbit and anti-mouse secondary antibodies for immunoblotting, Cy3–conjugated Affinipure Goat Anti-Mouse IgG (H + L) and Fluorescein (FITC)–conjugated Affinipure Goat Anti-Rabbit IgG (H + L) were purchased from Proteintech (Wuhan, China). 4,6-diamidino-2-phenylindole (DAPI) was from Beyotime. The pcDNA3.1/mSGK3-S486D plasmid, pcDNA3/mNedd4-2 plasmid and pMO/mSGK3-K191M plasmid were gifts from David Pearce (Departments of Medicine, and Molecular and Cellular Pharmacology, University of California, San Francisco, CA, United States). The pCMV/HA/mGSK3β plasmid was purchased from Origene Technology, Inc., (Rockville, MD, United States). The ubiquitin-Flag (UB-Flag) plasmid was a gift from Prof. Duan Qiuhong, Biochemistry Laboratory, Tongji Medical College, Huazhong University of Science and Technology. Recombinant lentivirus SGK3-shRNA and Nedd4-2-shRNA were purchased from Genechem (Shanghai, China). MG132 were purchased from APExBIO Technology (Houston, TX, United States). Phosphatase inhibitor mixture and proteinase inhibitor cocktail were purchased from Roche Applied Science (Indianapolis, IN, United States).

Male BALB/c mice were purchased from the Animal Experiment Center of Wuhan University (Wuhan, China). Baseline urine was collected in the male BALB/C mice (20–25 g, 8 weeks old). 8-week-old mice were treated with a single-dose tail vein injection of 10.5 mg/kg ADR (Sigma, USA, D1515) after 1 week of adaptive breeding (ADR group, n = 6). Control mice were treated with an equivalent volume of phosphate-buffered saline (PBS) (control group, n = 5). All mice were euthanized at 4 weeks after ADR treatment, and blood samples and kidneys were collected. The harvested kidneys were collected for western blotting, immunohistochemical and immunofluorescence analysis. All the animal experiments were approved by the Animal Care and Use Committee of Tongji Medical College. All methods were performed in accordance with the relevant guidelines and regulations.

Blood urea nitrogen (BUN) and urinary and serum creatinine levels were determined by the respective kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China), and total protein levels in urine were assessed by Urinary Total Protein Determination Kit (Ningbo Elejech Biologic Technology, Zhejiang, China), according to the manufacturer’s instruction. The urinary albumin excretion rate was expressed as the ratio of albumin to creatinine.

Renal tissues were fixed in 10% formalin neutral fixative overnight after removal, embedded in paraffin and sliced to make 5-μm-thick sections that were stained with periodic acid-Schiff (PAS) reagent for histopathology. Immunohistochemistry and immunofluorescence staining proceeded and analyzed essentially as previously described (Peng et al., 2018). The sections was subject to rehydration by soaking in xylene for 30 min for two intervals and then placed into progressively lower concentrations of ETOH (100, 95, 70, and 50%) for 5 min each and finally placed in ddH20 for 5 min. Antigen retrieval was performed by incubating slides in a citrate buffer solution for 2 min in a 95°C water bath. Slides were cooled at room temperature for 20 min and washed 3 times with PBS. Endogenous peroxidase activity was blocked with 3% H2O2 in methanol for 30 min. Non-specific binding was blocked using 10% normal goat serum for 30 min. The slides were washed with PBS and incubated with polyclonal anti-SGK3, polyclonal anti-nephrin and anti-Nedd4-2 overnight at 4°C.

For immunohistochemistry, the sections were incubated with polyperoxidase anti-rabbit and anti-goat IgG secondary antibodies for 30 min at room temperature The bound antibody was visualized using a DAB Kit. Positive expression for SGK3 and nephrin in 50 random glomeruli per mouse were counted and expressed as average per glomerulus. All images were acquired using light microscopy. Positive expression for SGK3 and nephrin were semi-quantitative using the Image Pro Plus software.

For immunofluorescence, the sections were stained with Cy3–conjugated Affinipure Goat Anti-Mouse IgG (H + L) (SA00009-1; Proteintech, Wuhan, China) and Fluorescein (FITC)–conjugated Affinipure Goat Anti-Rabbit IgG (H + L) (SA00003-2, Proteintech, Wuhan, China). Nuclei were co-stained with 4,6-diamidino-2-phenylindole (DAPI) for 5 min. All sections were observed under a fluorescent microscope (Nikon, Tokyo, Japan).

The conditional immortalized mouse podocytes were cultivated as described earlier (Peng et al., 2018). Briefly, cells were provided with RPMI 1640 medium (HyClone, United States) supplemented with 10% (v/v) fetal bovine serum (GIBCO, Thermo Fisher Scientific Inc., United States), 100 μg/ml streptomycin and 100 U/ml penicillin. Cells were cultured at 33°C under 5% CO2 and 95% air in culture medium supplemented with 10 U/ml recombinant interferon-γ (IFN-γ) in order to proliferation. Podocytes were transferred at 37°C in IFN-γ-free medium cultivated 10–14 days for inducing differentiation. Mouse podocytes cells were treated with ADR (0.2 μg/ml) for 24 h or MG132 (10 μM) for 8 h at 37°C before collected.

Mouse podocytes were transfected with various plasmids for 48 h using Lipofectamine 2000 reagent in line with the manufacturer’s instructions (Invitrogen, Carlsbad, CA, United States), and the cell lysates were prepared for immunoblot and ubiquitination experiment measurements. Mouse podocytes were infected with SGK3 and Nedd4-2 short hairpin RNAs (shRNAs) or scramble lentiviruses for seventy 2 h and then cell lysates were collected for immunoblot.

Protein expression was measured by western blot analysis. Mouse podocytes were lysed in Lysis buffer added with protease inhibitors cocktail and dithiotreitol (DTT). Membranous protein was extracted according to the manufacturer’s instructions (KeyGEN BioTECH, Jiangsu, China). In brief, the suspension kept shaking for 30 s 6 times after every 1 min, and then discarded after centrifugation. The precipitate was added with extraction buffer and kept shaking for 30 s six times after every 5 min. The supernatant protein was membrane protein. For total protein, the mouse podocytes and kidney tissues were lysed in RIPA buffer supplemented with a protease inhibitor cocktail, PMSF and phosphatase inhibitor for 1 h, and the supernatant was extracted as total protein. Equal amounts of proteins were separated by electrophoresis on SDS-polyacrylamide gel electrophoresis (SDS-PAGE) gels then transferred to a polyvinylidene fluoride membrane (Millipore, Bedford, MA, United States). Membranes were blocked with 5% milk for 2 h at room temperature and incubated with appropriate primary antibody at 4°C overnight. The following day membranes were incubated with corresponding horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature and detected by ECL (Millipore, Germany).

Ubiquitination assays were performed as described (Yuan et al., 2018). Mouse podocytes were treated with ADR (0.2 μg/ml) or transfected with plasmids and MG132 (10 μM) was added for 8 h before collected to prevent proteasomal degradation. Cells were washed with PBS three times and dissociated in 1 ml lysis buffer [50 mM HEPES (pH 7.5), 150 mM NaCl, 1 mM EGTA, 1% TritonX-100, 25 mM sodium fluoride, 40 mM β-glycerolphosphate, 10 mM sodium pyrophosphate, 1.5 mM MgCl2, 10% glycerol] supplemented with PMSF, cocktail, and phosphatase inhibitors. The lysates were centrifuged at 12,000 rpm for 15 min at 4°C. The supernatant proteins (about 1 mg) were formulated into an equal volume system and added to the protein A/G agarose beads for co-immunoprecipitation using the appropriate primary antibody (anti-Flag antibody, anti-nephrin antibody) overnight at 4°C. Agarose beads were gathered and loading buffers (5×) were added to the co-immunoprecipitates. The samples were separated by SDS-polyacrylamide gels for western blot and immunoblotted with anti-nephrin antibody or anti-ubiquitin (UB) antibody.

Mouse podocytes were grown on 10-cm dishes, and transfected with 10 μg of the plasmids DNA encoding mouse Nedd4-2 or empty vector using Lipofectamine 2000 reagent for 48 h. After transfection, cells were washed twice at 4 °C in cold PBS and dissociated in 500 μl IP lysis buffer (Beyotime Biotechnology, Shanghai, China). After centrifugation (12,000 rpm, 4°C, 15 min), the supernatant was quantified by Bradford assay and 1 mg of lysate was incubated with the anti-nephrin antibody and protein A/G agarose beads. Then the samples were subjected to western blot analysis.

UbiBrowser¹ is an integrated bioinformatics platform, which is used to predict proteome-wide human E3-substrate networks. Totally 20,220 human proteins’ corresponding E3 ligases can be explored. Furthermore, 4,560 human proteins’ 19,451 ubiquitination sites can also be explored (Li et al., 2017). We used it to predict the potential E3 ligases of NPHS1 (nephrin).

All data were expressed as the mean ± SE and SPSS 22.0 software was used for analyzed. An unpaired Student’s t test was used for comparison of two groups and a one-way ANOVA with Dunnett’s post hoc test was applied for comparison of more than 2 groups. The correlation of SGK3 expression with proteinuria or with nephrin expression was analyzed by Spearman’s rank correlation coefficient. A value of P < 0.05 was defined statistically significant. Analyses were performed using Graphpad Instat software (GraphPad Software, San Diego, CA, United States).

Our previous study demonstrated that the absence of SGK3 expression leads to podocyte dysfunction and proteinuria (Peng et al., 2018). To determine the relationship between SGK3 and proteinuria in ADR nephritis, an ADR nephritis mouse model was established as previously described (Peng et al., 2018). As shown in Figure 1A, on the 28th day after modeling, urinary protein excretion was significantly increased in the ADR mice compared with that in the control group. However, no significant difference in blood urea nitrogen and serum creatinine levels was noted. Furthermore, the renal tissue presented obvious pathological damage by periodic acid-Schiff staining after ADR injection. Protein casts were found in the renal tubules and glomerular focal segmental sclerosis was visible (Figure 1B). Western blotting analysis revealed a significant decrease in the protein level of nephrin and SGK3 in the kidneys of ADR nephritis mice (Figure 1C). Immunohistochemical analysis of kidney tissue sections revealed that SGK3 and nephrin levels were significantly lower in the glomeruli of ADR mice than those of the control mice (Figure 1D). Immunofluorescence examination presented that SGK3 was expressed in the podocytes of mouse kidney, and ADR treatment decreased the co-localization of SGK3 and nephrin (Figure 1E). Correlation analysis showed that the levels of SGK3 were negatively correlated with the urinary output of the protein (Figure 1F). Furthermore, there was a significant positive correlation between the levels of SGK3 and nephrin (Figure 1G). Thus, our results showed that SGK3 participated in ADR-induced podocyte injury, possibly by affecting nephrin levels.

We have previously reported that SGK3 is involved in the downregulation of podocin and CD2AP expression and podocyte viability is induced by puromycin aminonucleoside (PAN) (Peng et al., 2018). However, it remains unclear whether SGK3 participates in an ADR-mediated decrease in nephrin levels. To address this question, podocytes were transfected with an SGK3 inactivated plasmid, SGK3-K191M, or SGK3-specific shRNAs in vitro. We found that the total and membrane protein levels of nephrin were significantly downregulated after SGK3 inactivation (Figures 2A,B) and knockdown (Figure 2C), indicating that downregulation of SGK3 led to decreased nephrin protein levels. We also transfected mouse podocyte cells (MPCs) with an SGK3 constitutively activated plasmid, SGK3-S486D, followed by ADR treatment. We noted that the increase in SGK3 activity reversed the ADR-induced decrease in nephrin levels (Figure 2D), implying that SGK3 was involved in the ADR-mediated decrease in nephrin levels.

Multiple studies have shown that ADR stimulation leads to podocyte injury (Liu et al., 2018; Angeletti et al., 2020). As illustrated by western blotting, ADR treatment indeed resulted in increased levels of podocyte damage marker, desmin, and decreased levels of podocin (Figure 3A), thereby validating the podocyte damage model. As shown in Figure 3B, the levels of nephrin and SGK3 decreased after ADR treatment. We further explored the effect of ADR on nephrin degradation. The cultured MPCs were treated with ADR and the proteasome inhibitor, MG132. Compared with the ADR + /MG132- group, the levels of nephrin in the ADR + /MG132 + group were significantly increased to the control level. Thus, ADR-induced downregulation of nephrin protein levels could be reversed by MG132 (Figure 3C). Next, we determined the ubiquitination level of nephrin after ADR treatment. MPCs were transfected with a ubiquitin-Flag (UB-Flag) plasmid for 48 h and treated with ADR for 24 h. As expected, the ubiquitin-mediated protein degradation of nephrin was enhanced in the ADR-treated MPCs (Figure 3D). These data suggested that ADR treatment increased the ubiquitination mediated degradation of nephrin.

To further investigate the mechanism by which SGK3 regulates nephrin protein levels, we transfected podocytes with SGK3-K191M and then treated the cells with or without MG132. The decreased nephrin levels in the K191M + /MG132- group were reversed in the K191M + /MG132 + group (Figure 4A), suggesting that SGK3 inactivation decreased nephrin protein levels by promoting the proteasome degradation pathway of nephrin. Next, we examined the effect of SGK3 on nephrin ubiquitination. Cultured MPCs were co-transfected with SGK3-K191M and ubiquitin-Flag (UB-Flag) plasmids, and the ubiquitination level of nephrin was assessed by a ubiquitination assay. As shown in Figure 4B, the ubiquitination level of nephrin was significantly increased in the SGK3-K191M transfection group compared with that in the control group. These studies suggested that the downregulation of SGK3 activity led to an increased ubiquitin-proteasome-mediated protein degradation of nephrin.

Previous report showed that Nedd4-2, a common E3 ubiquitin ligase, was abundantly expressed in the cytoplasm of podocytes of mouse kidney and contributed to dendrin degradation, which is also a component of the slit diaphragms (SDs) complex (Shirata et al., 2017). To explore the specific molecular mechanism underlying the regulation of nephrin by SGK3, we first examined whether the SGK3 target protein Nedd4-2 was involved in SGK3 triggered ubiquitin-mediated degradation of nephrin. We confirmed that the phosho-Nedd4-2/Nedd4-2 ratio was significantly downregulated after ADR treatment or SGK3-K191M plasmid transfection, which indicated that the activity of Nedd4-2 increased because phospho-Nedd4-2 was in an inactivated form (Figures 5A,B). This demonstrated the involvement of the SGK3/Nedd4-2 signaling pathway in the ADR-induced podocytes injury in vitro. We found that the protein levels of nephrin increased after Nedd4-2 knockdown (Figure 5C) and the total and membrane protein levels of nephrin decreased after Nedd4-2 overexpression (Figures 5D,E). Furthermore, MG132 administration could reverse the Nedd4-2 overexpression-induced decline in nephrin protein levels (Figure 5F), suggesting that Nedd4-2 was involved in the ubiquitin-mediated nephrin degradation. As there were no canonical WW-binding PPxY motifs, a classical target sequence of Nedd4-2, present in nephrin, we queried NPHS1 (nephrin) as a substrate in the web tool of UbiBrowser² to explore the detailed mechanism of Nedd4-2 on nephrin degradation. We found that nephrin could interact with some of the E3 ligases, wherein Nedd4-2 showed the strongest interaction with nephrin (Figure 5G). Furthermore, Ig-like domains (Ig2 and Ig8) were predicted as the binding sites of NPHS1 (nephrin) to the HECT domain of Nedd4-2 (Figure 5H). Using a co-immunoprecipitation technique, we demonstrated that nephrin could indeed interact with Nedd4-2 (Figure 5I), and the ubiquitination levels of nephrin was significantly enhanced in the Nedd4-2-transfected mouse podocytes compared with that in the control mouse podocytes (Figure 5I). Immunofluorescence examination presented that Nedd4-2 was slightly expressed in the podocytes of mouse kidney, and ADR treatment resulted in increase co-localization of Nedd4-2 and nephrin (Figure 5J). These data suggested that the SGK3/Nedd4-2 signaling pathway was involved in ubiquitin-mediated protein degradation of nephrin.

We had previously reported that PAN stimulation and SGK3 shRNA infection of podocytes significantly decreased the phosphorylation of GSK3β, and the SGK3/GSK3β signaling pathway inhibited podocin expression during PAN-induced podocyte injury (Peng et al., 2018). In this study, we further explored whether the SGK3/GSK3β signaling pathway regulates nephrin levels. Consistent with our previous report, phosho-GSK3β/GSK3β ratio was significantly downregulated after SGK3-K191M plasmid transfection, which indicated that the activity of GSK3β increased because phospho-GSK3β was in an inactivated form (Figure 6A). Moreover, the levels of nephrin decreased in the GSK3β transfection group compared with that in the control group (Figure 6B), and the protein levels of nephrin were also significantly increased after treatment with LiCl, a GSK3β inhibitor (Peng et al., 2018), for 24 h (Figure 6C). In addition, when MG132 was added after GSK3β transfection, the protein levels of nephrin did not decrease (Figure 6D). However, nephrin ubiquitination levels were similar between the GSK3β-transfected mouse podocytes and the control mouse podocytes (Figure 6E). These results indicated that the SGK3/GSK3β signaling pathway regulated the protein levels of nephrin.