Astragaloside IV protects against podocyte injury via SERCA2-dependent ER stress reduction and AMPKα-regulated autophagy induction in streptozotocin-induced diabetic nephropathy

Aberrant endoplasmic reticulum (ER) stress and autophagy are associated with diabetic nephropathy. Here we investigated the effect of astragaloside IV (AS-IV) on the progression of diabetic nephropathy (DN) and the underlying mechanism involving ER stress and autophagy in streptozotocin (STZ)-induced diabetic mice and high glucose (HG)-incubated podocytes. The diabetic mice developed progressive albuminuria and glomerulosclerosis within 8 weeks, which were significantly ameliorated by AS-IV treatment in a dose-dependent manner. Moreover, diabetes or HG-induced podocyte apoptosis was markedly attenuated by AS-IV, paralleled by a marked remission in ER stress and a remarkable restoration in impaired autophagy, which were associated with a significant improvement in the expression of sarcoendoplasmic reticulum Ca2+ ATPase 2b (SERCA2b) and AMP-activated protein kinase α (AMPKα) phosphorylation, respectively. Knockdown of SERCA2 in podocytes induced ER stress and largely abolished the protective effect of AS-IV, but had no obvious effect on the expression of autophagy-associated proteins. On the other hand, blockade of either autophagy induction or AMPKα activation could also significantly mitigate AS-IV-induced beneficial effect. Collectively, these results suggest that AS-IV prevented the progression of DN, which is mediated at least in part by SERCA2-dependent ER stress attenuation and AMPKα-promoted autophagy induction.


Results
AS-IV prevented the progression of DN in STZ-induced diabetic mice. C57BL/6J mice were rendered diabetic by using a two-dose regimen of STZ as reported 31 . The STZ-injected mice developed hyperglycemia (blood glucose > 350 mg·dL −1 ) 1 week after STZ treatment. At this point the mice were randomly allocated into 4 groups and treated respectively with vehicle, or 3, 6, and 12 mg −1 ·kg −1 day −1 AS-IV for 8 weeks, with nondiabetic mice receiving vehicle or 12 mg −1 ·kg −1 day −1 AS-IV as controls. The hyperglycemia persisted during the 8-week period and all STZ-induced diabetic mice failed to gain weight during the 8 weeks (Table 1). As expected, DN-vehicle mice developed robust progressive albuminuria (expressed as urinary albumin to creatinine ratio (ACR)) and AS-IV prevented the development of diabetes-induced albuminuria in a dose-dependent manner (Fig. 1A) without altering the blood glucose levels and body weight (Table 1). Consistently, both blood urea nitrogen (BUN) and serum creatinine levels were markedly elevated in DN-vehicle group relative to NC-vehicle group, which were significantly reduced by AS-IV treatment in a dose-dependent manner (Fig. 1B,C).
Histological examination showed that DN-vehicle mice displayed larger glomerular size, an obvious mesangial matrix expansion, and severe glomerulosclerosis at week 9 compared with NC-vehicle mice. 12 mg −1 ·kg −1 day −1 AS-IV treatment markedly attenuated glomerular hypertrophy, mesangial matrix expansion and glomerulosclerosis in DN mice (Fig. 1D,E; Supplementary Fig. S1). The renin-angiotensin system (RAS) has been suggested as a major mediator of glomerular injury in DN 32 . We found that intrarenal RAS activity was upregulated in DN mice, as evidenced by elevated intrarenal Renin and angiotensinogen (AGT) mRNA expression and AGT protein expression, which was reversed by 12 mg −1 ·kg −1 day −1 AS-IV treatment ( Supplementary Fig. S1). Inflammation is implicated in DN 33 . DN-vehicle mice exhibited intense renal inflammation, as indicated by immunohistochemistry staining and quantitative RT-PCR (qRT-PCR) detection of monocyte chemotactic protein-1 (MCP-1) and Groups Blood glucose (mg·dL −1 ) Body weight (g) Week  Table 1. Fasting blood glucose and body weight during the treatment period. 8-week-old C57BL/6 J mice were injected intraperitoneally with 100 mg·kg −1 STZ for 2 consecutive days in the first week (week 0). 1 week after STZ injection (week 1), the mice were treated with AS-IV at the indicated doses of AS-IV for 8 consecutive weeks. Data are expressed as mean ± SEM. n = 8-12 per group. ** P < 0.01 compared to NC-vehicle group.   Figure S2. Data are expressed as mean ± SEM. n = 10-12 for A-C and n = 4-6 for D-J. ** P < 0.01 compared to NC-vehicle group; # P < 0.05, ## P < 0.01 compared to DN-vehicle group. One-way ANOVA and Newman-Keuls multiple comparisons test (A-C,E,G,H,J). NC, non-diabetic control. DN, diabetic nephropathy.
tumor necrosis factor-α (TNF-α), two inflammatory cytokines. However, AS-IV dose-dependently suppressed diabetes-induced MCP-1 and TNF-α upregulation ( Supplementary Fig. S1). Podocyte injury plays a key role in the development of albuminuria in diabetic nephropathy 3 . The number of apoptotic cells within the glomeruli significantly increased and podocyte number dramatically decreased in DN-vehicle mice compared with NC-vehicle mice, which was partially rescued by AS-IV administration, as evidenced by TUNEL staining and the podocyte nucleus marker Wilms' tumor 1 (WT1) labeling ( Fig. 1F-H). Podocyte foot process markers Podocin and Nephrin were also downregulated in DN-vehicle mice, as confirmed by immunofluorescence staining and western blot, indicating a damaged glomerular filtration barrier. However, AS-IV restored the expression of Podocin and Nephrin in a dose-dependent manner (Fig. 1F-J).
AS-IV alleviated ER stress in STZ-induced diabetic mice. Next, we investigated if the ER stress pathway was activated. As shown in Fig. 2A-C, compared with NC-vehicle mice, there were significant increases in both mRNA and protein levels of GRP78, an ER stress marker primarily localized to the glomeruli, in DN-vehicle mice, which were abrogated by AS-IV in a dose-dependent manner ( Fig. 2A-C). Diabetes-induced activation of ER stress signaling transducers ATF6 and PERK, and their downstream targets, including phospho-eukaryotic translation initiation factor 2α (p-eIF2α) and pro-apoptotic transcription factor CHOP were significantly attenuated by AS-IV treatment (Fig. 2D,E). Likewise, AS-IV markedly reduced diabetes-induced phosphorylation of IRE1α and its downstream target JNK, an apoptosis mediator, and decreased spliced X-box binding protein 1 (XBP1) expression (Fig. 2F,G). In addition, the apoptosis marker protein cleaved caspase 12 was upregulated in DN mice but was attenuated by AS-IV (Fig. 2F,G). These findings supported that AS-IV mitigated ER stress and ER stress-induced apoptosis.

AS-IV restored SERCA2b expression and activity in STZ-induced diabetic mice. Dysfunction of
SERCA has emerged as a major trigger for ER stress 34 . Our previous work has identified that SERCA2b was the predominant SERCA2b isoform expressed in the kidney and conditionally immortalized mouse podocyte cell line, while both SERCA2a and SERCA3 were expressed at extremely low levels. SERCA1 expression could not be detected 35 . Here, qRT-PCR analyses showed that SERCA2a and SERCA3 mRNA expression did not change significantly between DN mice and NC mice (Fig. 3A,B). SERCA2b mRNA and protein expression were dramatically downregulated in DN group compared with NC group, both of which were rescued by AS-IV in a dose-dependent manner ( Fig. 3C-E). In parallel to a marked decrease in SERCA2 expression, SERCA activity almost decreased by 50% in DN mouse kidney compared to NC mice. However, treatment with AS-IV at 6 mg·kg −1 ·day −1 and 12 mg·kg −1 ·day −1 remarkably restored SERCA activity in the kidneys of DN mice (Fig. 3F).

AS-IV restored the defective autophagy in STZ-induced diabetic mice.
Autophagy is an adaptive response in cells under environmental stress and participates in the renal diseases 18 . To investigate the role of autophagy in DN, we examined the alterations of autophagy in DN by immunofluorescence staining of LC3. As shown in Fig. 4A, glomeruli exhibited high levels of autophagy activity under basal conditions while tubuli had extremely low levels of autophagy activity. At 9 weeks after STZ injection, autophagy activity was suppressed. However, AS-IV treatment could dose-dependently enhance the glomerular autophagy activity as evidenced by improved autophgosome formation (Fig. 4A,B). Concomitant with the immunofluorescence staining alterations, western blot analysis showed that the expression of autophagy-related proteins such as LC3A/B, Beclin-1 and Atg12 was markedly decreased and p62, a specific target of the autophagy degradation, was increased in kidney tissues at 9 weeks post-STZ injection, indicating a suppressed autophagy level. However, AS-IV treatment for 8 weeks largely reversed the expression of these components of the autophagy pathway (Fig. 4C,D). Moreover, we found that DN-vehicle mice displayed enhanced activation of mTOR signaling and impaired AMPK activity, as indicated by increased phosphorylation of mTOR and p70S6 kinase (p70S6K) and decreased phosphorylation of AMPKα. However, this alternation was prevented by AS-IV treatment (Fig. 4E,F). The above results indicated that AS-IV was able to induce autophagy possibly via activating AMPK in the kidneys of DN mice.
AS-IV protected against HG-induced podocyte apoptosis. Immortalized mouse podocytes were cultured in low glucose (LG), mannitol (M) or HG medium and the effect of high glucose on podocyte apoptosis was monitored. Flow cytometry analysis using Annexin V-FITC/PI staining revealed that HG induced apoptosis in a time-dependent manner (Fig. 5A,B), accompanied by a decline in the expression of podocyte markers Podocin and Nephrin as determined by western blot (Fig. 5C,D). Pretreatment with AS-IV attenuated HG-induced podocyte apoptosis in a concentration-dependent manner (Fig. 5E,F), paralleled by the restoration of podocyte Podocin and Nephrin as confirmed by western blot and immunofluorescence staining (Fig. 5G,H).

AS-IV attenuated HG-triggered ER stress in a SERCA2-dependent mananer in podocytes.
To determine whether ER stress was involved in HG-induced podocyte apoptosis, ER stress-related proteins were detected by western blot at various time points. Incubation with HG for 12-24 h significantly decreased the expression of SERCA2 and upregulated ER stress-related proteins such as GRP78, cleaved ATF6, p-PERK, p-IRE1 and CHOP, which indicated a strong induction of ER stress (Fig. 6A,B). However, pre-incubation of AS-IV improved HG-decreased SERCA2 expression and reversed the expression of HG-induced ER stress-associated proteins at 24 h post-HG stimulation in a concentration-dependent manner (Fig. 6C,D). Immunofluorescence staining of GRP78 also confirmed the repression of HG-induced ER stress by AS-IV (Fig. 6E). Additionally, HG-elevated expression of apoptosis mediators, including CHOP, p-JNK, cleaved caspase 12 and caspase 3, was significantly downregulated by AS-IV ( Fig. 6A-G). These data demonstrated that AS-IV ameliorated HG-induced podocytes apoptosis partially through mitigating ER stress-mediated apoptotic pathway.
Since SERCA2 has been recognized as a key regulator of ER stress, we next investigated whether SERCA2 mediated the beneficial effect of AS-IV on ER stress in podocytes. As shown in Fig. 7A, transfection of  Figure S3. Data are expressed as mean ± SEM. n = 4-6 per group. * P < 0.05, ** P < 0.01 compared to NC-vehicle group; # P < 0.05, ## P < 0.01 compared to DN-vehicle group. One-way ANOVA and Newman-Keuls multiple comparisons test (A,B,E,G). NC, non-diabetic control. DN, diabetic nephropathy.
SERCA2b-specific siRNA (MSS202247), but not control siRNA, significantly reduced SERCA2 expression in podocytes under basal condition. Gene silencing of SERCA2 markedly induced ER stress in podocytes under LG condition and enhanced HG-induced ER stress, suggesting that SERCA2b inhibition triggered ER stress in podocytes (Fig. 7A,B). In addition, SERCA2b silencing abrogated AS-IV-alleviated ER stress (Fig. 7A,B). Flow cytometry analysis showed that down regulating SERCA2 expression could significantly attenuated the anti-apoptotic effect of AS-IV (Fig. 7C,D). These data confirmed the critical role of SERCA2 in mediating the protective effect of AS-IV on HG-triggered ER stress and apoptosis.

AS-IV restored HG-inhibited autophagy through AMPKα activation in podocytes.
To delineate the role of autophagy in HG-stimulated podocytes, we examined the expression pattern of autophagy-related proteins after HG stimulation. HG incubation for 12-24 h remarkably induced the expression of autophagy-related proteins, which then began to decline after treatment with HG for 48 h (Fig. 8A,B). This observation confirmed a short-term inductive effect of HG on autophagy and a long-term suppressive effect of HG on autophagy in podocytes. The transfection of fluorescent reporter GFP-LC3 to monitor autophagic flux in podocytes also validated defective autophagy at 72 h post-HG incubation, while AS-IV could restore the insufficient autophagy in a concentration-dependent manner as indicated by the increased number of dot-like GFP-LC3 puncta (Fig. 8C). Western blot analysis showed that AS-IV increased HG-diminished abundance of LC3A/B, Atg12 and Beclin, and reduced the accumulation of p62 (Fig. 8D,E). In addition, HG stimulation for 72 h markedly enhanced the phosphorylation of mTOR and p70S6K and reduced the phosphorylation of AMPKα, which were also reversed by AS-IV in a concentration-dependent manner (Fig. 8F,G). While SERCA2b silencing did not alter the expression of autophagy markers in HG-treated podocytes (Fig. 9A,B), autophagy inhibitor 3-methyladenine (3-MA) or AMPKα inhibitor compound C (CC) markedly antagonized the beneficial effect of AS-IV on autophagy induction and HG-induced apoptosis (Fig. 9C-F), suggesting the crucial role of AMPKα-promoted autophagy induction in mediating the protective effect of AS-IV in HG-induced podocytes and that SERCA2 is not required for autophagy induction.

Discussion
In the present study, we used a STZ-induced diabetic model to investigate the effect of AS-IV on DN progression and gained further insight into the mechanisms underlying the beneficial action of AS-IV in vivo and in vitro. We demonstrated that AS-IV exhibited a prominent renoprotection in DN development, as evidenced by enhanced  renal function, ameliorated glomerulosclerosis and renal inflammation. The protective effect of AS-IV was associated with ER stress attenuation via SERCA2b restoration and autophagy enhancement through AMPKα activation, which subsequently protected podocytes from hyperglycemia or HG-induced apoptosis. This investigation highlights AS-IV as an antiproteinuric drug targeting podocytes for DN therapy. DN is a long-term complication of diabetes that develops in about 30% of patients with type 1 and 10% of patients with type 2 diabetes, with proteinuria and glomerulosclerosis as the pathological hallmarks 3,36 . Microalbuminuria is the earliest sign and a major risk factor for progressive renal function decline in DN 37 . In our study, STZ-treated mice developed progressive albuminuria and elevated serum BUN and creatinine, as well as significant glomerular hypertrophy, mesangial matrix expansion and glomerulosclerosis similar to human DN, indicating the impaired renal function and the development of DN in our animal model. Interestingly, AS-IV treatment markedly attenuated albuminuria, reduced serum BUN and creatinine, and ameliorated glomerular histopathology and inflammation, suggesting AS-IV effectively prevented the progression of DN. However, AS-IV treatment had no obvious effects on body weight and fasting blood glucose, which is consistent with the investigation by Chen et al. 30 . Our previous work conducted in db/db mice, a mouse model for type 2 diabetes, showed that AS-IV could decrease hyperglycemia 35 . In the present study, AS-IV could not lower hyperglycemia in the type 1 diabetes model, which indicated that the improved renal function might be a direct effect of AS-IV on podocytes. As expected, hyperglycemia or HG-induced podocyte apoptosis was significantly attenuated by AS-IV, confirming the protective effect of AS-IV on podocytes. Glomerular hypertrophy is one of the earliest pathological alterations found in DN 38 . To further elucidate the mechanism underlying AS-IV-mediated amelioration of diabetes-induced glomerular hypertrophy despite the unaffected glycemic control, we examined renal components of RAS given the key role of the intrarenal RAS in DN development, including increasing glomerular capillary pressure and permeability, stimulation of renal cell proliferation and hypertrophy, and synthesis of cytokines and extracellular matrix 39 . We found that intrarenal Renin and AGT mRNA levels as well as AGT protein levels were markedly induced in DN group and downregulated after AS-IV treatment. Thus, our study suggested that the inhibtion of renal RAS activation was most likely the basis for AS-IV-suppressed glomerular hypertrophy in DN.
SERCA, an imperative maintainer for ER Ca 2+ 9 , is a major regulator of ER stress. Emerging evidence reveal that the activity or expression of SERCA was compromised in metabolic syndromes, resulting in ER stress and apoptosis, while the restoration of SERCA2 activity or expression ameliorated ER stress and in turn improved metabolic abnormalities 14,40 . In our study, both SERCA2 expression and activity were significantly decreased in renal cortex of DN mice and HG-incubated mouse podocytes, paralleled by a marked induction of ER stress and the activation of both UPR signaling pathway and ER stress-mediated apoptotic pathway. AS-IV significantly restored hyperglycemia or HG-impaired expression of SERCA2 with a concomitant remission of ER stress. The data are in agreement with the report that AS-IV has pharmacological activities against podocyte injury through ER stress inhibition 30 . Furthermore, SERCA2 knockdown dramatically triggered ER stress and abolished AS-IV-reduced ER stress in podocytes, indicating SERCA2 mediated the inhibitory effect of AS-IV on  HG-induced ER stress. Our results suggest that AS-IV can protect diabetes/HG-inflicted podocyte apoptosis at least in part via an ER stress-inhibiting mechanism involving SERCA2.
Though several investigations have been performed to define function of autophagy in the kidneys 20,41,42 , the role of autophagy in DN is still incompletely understood. There are some reports suggesting that autophagy is protective [22][23][24]43 , while some indicate that autophagy might be pathogenic 44 . Our data showed that autophagy was expressed at high levels under basal conditions and compromised in 8-week-old diabetic mice, and short-term HG incubation induced autophagy while long-term HG incubation suppressed autophagy in podocytes in vitro.
These observations were similar to other studies 22,23,45 . Interestingly, the deficient autophagy was restored by AS-IV in a dose-dependent manner. Autophagy is regulated by the nutrient-sensing pathways, including AMPK and mTOR 16 . AMPK activity decreases and mTORC1 activity increases in human and experimental type 1 and type 2 DN 20 . As expected, our study revealed a suppressed activity of AMPKα and an enhanced activity of mTOR in DN mice and HG-treated podocytes, which were significantly reversed by AS-IV. Either autophagy inhibitor 3-MA or AMPKα inhibitor CC could partly antagonize the protective effect of AS-IV against HG-induced autophagy defect and podocyte apoptosis, suggesting the crucial role of autophagy induction and AMPKα activation in mediating the anti-apoptotic effect of AS-IV in HG-treated podocytes. However, SERCA2 silencing had no obvious effect on AS-IV-improved autophagy activity, indicating autophagy repairment was not dependent on SERCA2-mediated ER stress reduction.
The UPR-associated proteins ATF6 and PERK have been reported to induce autophagy, while IRE1α acts a negative regulator of autophagy 46,47 . Autophagy is activated during ER stress to supplement endoplasmic reticulum-associated degradation (ERAD), which mitigates ER stress and reduces cell death under conditions of ER stress 48,49 . Autophagy inhibition can induce ER stress and podocyte apoptosis by activating the pro-apoptotic pathway 50 . Podocyte-specific deficiency of autophagy exacerbates proteinuria in diabetic nephropathy 24 and leads to a glomerulopathy in aging mice that was related to ER stress 23 . Collectively, these findings suggest that autophagy represents an adaptive response of podocytes to ER stress that is renoprotective in DN. On the other hand, another study revealed that ER stress mediates HG-induced defective autophagy in podocytes 22 . The establishment of a causal relationship between autophagy and ER stress reduction needs further studies.
In conclusion, this study provides evidence that AS-IV protects against diabetes/HG-induced podocyte apoptosis by a mechanism involving ER stress attenuation mediated by SERCA2 restoration and autophagy enhancement promoted by AMPKα activation, which subsequently attenuates the albuminuria and prevents the progression of DN. This study highlights the role of SERCA and autophagy in the beneficial action of AS-IV on DN development, which may elicit a new therapeutic strategy for DN.

Methods
Animals and treatment. 6-week old male C57BL/6J mice were purchased from Shanghai laboratory animal center (Shanghai, China). After 2 weeks of acclimation, the animals were induced to diabetes by intraperitoneal injection of freshly prepared STZ (Sigma-Aldrich, St Louis, MO, USA, dissolved in 0.01 M citrate buffer, pH 4.5) at 100 mg·kg −1 ·day −1 for 2 consecutive days. 1 week after STZ injection, fasting blood glucose was measured to verify the development of diabetes. Mice with blood glucose >350 mg·dL −1 were randomly separated into 4 groups and treated respectively with vehicle, 3 mg·kg −1 AS-IV, 6 mg·kg −1 AS-IV and 12 mg·kg −1 AS-IV by daily gavage for 8 weeks. NC mice without STZ treatment were randomly divided into 2 groups and administered respectively with vehicle and 12 mg·kg −1 AS-IV as controls. AS-IV was purchased from Shanghai Bogoo Biotechnology company, Limited (purity at 98%, Shanghai, China) and suspended in 0.5% carboxymethyl cellulose as a vehicle for the administration. All the work was carried out in accordance with the approved guidelines for the use of experimental animals in Putuo Hospital, Shanghai University of Traditional Chinese Medicine.

Measurement of metabolic and biochemical parameters.
Body weight, fasting glucose, and urine volume were measured at 4-week intervals. Fasting blood glucose was monitored with Omron HEA-230 Glucometer (Omron Corporation, Kyoto, Japan) by using one drop of tail blood. Animals were placed in individual metabolic cages every 4 weeks for 24 h urine collection and drinking volume calculation. Blood samples were drawn from orbit for serum BUN and creatinine detection at the end of the study. All mice were killed at 9 weeks after the start of STZ treatment, and kidneys were immediately harvested for protein or RNA extraction or for histological analysis. Urinary albumin and creatinine were measured with commercial ELISA kits (Biovision, Milpitas, CA, USA). Urinary albumin excretion was expressed as ACR. BUN was measured using Urea Nitrogen Colorimetric Detection Kit (Arbor Assays, Ann Arbor, MI, USA). All the biochemical parameters were determined according to the manufacturer's instructions.
Renal histology and immunohistochemistry. The kidneys were fixed in 4% paraformaldehyde (PFA), dehydrated, embedded in paraffin and cut into 5 μm-thick sections. Renal sections were stained with periodic acid-Schiff (PAS). Semiquantitative scoring of glomerular sclerosis was performed using a five-grade method described previously 51 . At least 50 glomeruli per section were evaluated by an examiner masked to the experimental conditions. For immunohistochemistry, paraffin-embedded sections were stained with primary antibodies against WT1 (ab89901, Abcam, Cambridge, MA, USA), MCP-1 (ab25124, Abcam, Cambridge, MA, USA), SERCA2 (ab2861, Abcam, Cambridge, MA, USA), and TNF-α (sc-52746, Santa Cruz Biotechnology, Santa Cruz, CA, USA) overnight at 4 °C. After incubation with biotinylated secondary antibody (Vector Laboratories, Burlingame, CA, USA), the sections were incubated with VECTASTAIN ABC reagent (Vector Laboratories, Burlingame, CA, USA) and color development was achieved using 3, 3′ diaminobenzidine (Vector Laboratories, Burlingame, CA, USA). After several PBS rinses, Cy TM 2-conjugated secondary antibodies (Jackson ImmunoResearch, West Grove, PA, USA) were applied for 1 h at room temperature. Podocytes cultured on coverslips were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 for 5 min. Then cells were blocked with 5% BSA in PBS for 1 h at room temperature and incubated with the specific primary antibodies against Podocin (ab50339, Abcam, Cambridge, MA, USA), Nephrin (ab58968, Abcam, Cambridge, MA, USA), and GRP78 (ab21685, Abcam, Cambridge, MA, USA) followed by staining with Cy TM 2 or Cy TM 3-conjugated secondary antibodies (Jackson ImmunoResearch, West Grove, PA, USA). The nuclei were visualized by 4′,6-diamidino-2-phenylindole (DAPI) staining. Images were taken using a LEICA laser scan microscope (LEICA DM IRB; Leica, Wetzlar, Germany). The number of autophagosomes per cell in the glomeruli was quantified based on LC3-positive staining foci randomly selected from 20 glomeruli per mouse. GFP-LC3 transfection. Podocytes were seeded into 35 mm glass bottom cell culture dish and grown to 50% confluence. Cells were transfected with Premo ™ Autophagy Sensor LC3B-GFP (Invitrogen, Carlsbad, CA, USA) for 16 hours as per the manufacturer's protocol. The transfected cells were treated under the indicated conditions and images of GFP-LC3 fluorescence were captured using a LEICA laser scan microscope (LEICA DM IRB; Leica, Wetzlar, Germany).
Apoptosis assay by flow cytometry. Podocytes were plated into 12-well dishes and cultured under the indicated conditions. Cell apoptosis was assessed using FITC Annexin V Apoptosis Detection Kit I (Cat: 556547, BD Biosciences, Franklin Lakes, NJ, USA) following the manufacturer's protocol. Apoptotic podocytes were defined as annexin V-positive/PI-negative (early apoptotic) and annexin V-positive/PI-positive (late apoptotic) cells.

Forward primer (5′-3′)
Reverse primer (5′-3′) SERCA activity measurement. ER fraction from kidney tissue was isolated using the ER isolation kit (Cat: ER0100, Sigma Aldrich, St. Louis, MO, USA). Briefly, fresh kidney tissue was homogenized in isotonic extraction buffer in a glass tube homogenizer. After a series of centrifugation (1,000 × g for 10 min, 12,000 × g for 15 min, 100,000 × g for 1 h), the microsomes were obtained and then homogenized with isotonic extraction buffer. SERCA activity was measured based on the inorganic phosphate production using a commercially available kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) according to the manufacturer's instruction. SERCA activity was normalized to protein concentration.
Statistical analysis. Data are expressed as the means ± SEM, with n representing the number of animals for in vivo experiments and independent experiments for in vitro experiments. Statistical analysis was conducted with GraphPad Prism software 5 software (GraphPad Software Inc., San Diego, CA, USA). Unpaired two-tailed t test or one-way ANOVA followed by the Newman-Keuls multiple comparisons test was used for statistical comparisons among experimental groups, with a value of P < 0.05 being considered statistically significant.