High glucose reduces expression of podocin in cultured human podocytes by stimulating TRPC6

Abstract

The transient receptor potential canonical 6 (TRPC6) channel and podocin are colocalized in the glomerular slit diaphragm as an important complex to maintain podocyte function. Gain of TRPC6 function and loss of podocin function induce podocyte injury. We have previously shown that high glucose induces apoptosis of podocytes by activating TRPC6; however, whether the activated TRPC6 can alter podocin expression remains unknown. Western blot analysis and confocal microscopy were used to examine both expression levels of TRPC6, podocin, and nephrin and morphological changes of podocytes in response to high glucose. High glucose increased the expression of TRPC6 but reduced the expression of podocin and nephrin, in both cultured human podocytes and type 1 diabetic rat kidneys. The decreased podocin was diminished in TRPC6 knockdown podocytes. High glucose elevated intracellular Ca²⁺ in control podocytes but not in TRPC6 knockdown podocytes. High glucose also elevated the expression of a tight junction protein, zonula occludens-1, and induced the redistribution of zonula occludens-1 and loss of podocyte processes. These data together suggest that high glucose reduces protein levels of podocin by activating TRPC6 and induces morphological changes of cultured podocytes.

INTRODUCTION

The role of transient receptor potential canonical 6 (TRPC6) in podocyte pathology has been highlighted by recent discoveries that TRPC6 is a channel in the glomerular slit diaphragm (24) and that a mutation in the TRPC6 gene causes familial focal segmental glomerulosclerosis (35). Since overexpression of TRPC6 by in vivo gene delivery causes proteinuria (16), which is one of the characteristic symptoms of glomerulosclerosis, the abnormalities in TRPC6 for causing familiar focal segmental glomerulosclerosis are likely to be gain-of-function mutations (20). Using a cultured human podocyte cell line, we have previously shown that activation of TRPC6 by a pharmacologically high concentration of glucose can induce podocyte apoptosis (15). This is consistent with other reports that loss of podocytes represents an early change in patients with diabetes (19, 23) and animal models (29) and that hyperglycemia induces podocyte loss via apoptosis by elevating intracellular reactive oxygen species (32). Since depletion of podocytes results in focal segmental glomerulosclerosis (2, 33), loss of podocytes due to apoptosis may also account for the pathogenesis of glomerulosclerosis at the late stages of diabetes. Indeed, podocyte injury is known to play an important role in the development and progression of diabetic glomerulosclerosis (6, 8). However, even a high concentration of glucose (33 mM) only induces <30% of podocytes to undergo apoptosis (15). In addition to apoptosis, podocyte pathology in diabetes mellitus also shows the unique effacement of podocyte foot processes (22, 28). Recent studies have shown that deletion of TRPC6 can reduce podocyte foot process effacement induced by puromycin aminonucleoside (13). Therefore, enhanced TRPC6 activity or expression should not only induce apoptosis to reduce the number of podocytes in the glomerulus but also promote podocyte foot process effacement.

Podocin, another slit diaphragm protein in podocytes that is encoded by the gene NPSH2, also plays an important role in podocyte injury. However, opposite to gain-of-function mutation of TRPC6, nonsense, frameshift, or missense of podocin accounts for the focal segmental glomerulosclerosis found in familial idiopathic nephrotic syndromes (3). A missense mutation in podocin also leads to early and severe renal disease in mice (21). Molecular deletion of podocin in the mouse mature kidney causes focal segmental glomerulosclerosis and nephritic syndrome (17). These studies have suggested that reduced podocin activity or expression also mediates podocyte injury. In contrast to podocin, loss of TRPC6 in TRPC6-deficient mice does not induce any obvious renal phenotype (4). Therefore, increased TRPC6 expression has more pathological significance than decreased TRPC6 expression. Conversely, reduced podocin expression has more pathological significance than increased podocin expression for podocyte injury. In the present study, we show that high glucose and type 1 diabetes mellitus increase the expression of TRPC6 but decrease the expression of podocin and that high glucose also induces morphological changes of podocytes. These changes may mediate the podocyte pathology caused by hyperglycemia in type 1 diabetes mellitus.

MATERIALS AND METHODS

Cell culture, TRPC6 knockdown, and the type 1 diabetic rat model.

An immortalized human podocyte cell line was provided by Dr. Moin Saleem (Children's Renal Unit and Academic Renal Unit, University of Bristol, Southmead Hospital, Bristol, UK). Cells were cultured on petri dishes as previously described (26). Briefly, cells were cultured in regular RPMI-1640 supplemented with 10% FBS and 100 U/ml penicillin-streptomycin in humidified 5% CO₂ incubators and first incubated at 33°C. In some experiments, regular RPMI-1640 was replaced with RPMI-1640 containing low NaCl (120 mM) to examine how activation of TRPC6 by hypoosmotic stretch affects the expression of podocin. After the cells became 70% confluent, they were incubated at 37°C for 2 wk before any experiment manipulations. To knock down TRPC6 expression, podocytes were stably transfected with TRPC6 silencing shRNA carried by a lentiviral vector (Santa Cruz Biotechnology). Podocytes with reduced TRPC6 expression (TRPC6 knockdown podocytes) were further selected in the presence of G418. Before TRPC6 knockdown podocytes were used, the reduction of TRPC6 expression was confirmed by Western blot experiments. All experiments in this study were performed at room temperature. In some experiments, Sprague-Dawley rats were given a single intraperitoneal injection of streptozotocin (STZ; 55 mg/kg). Previous studies have shown that within 5 days, the rats are only hyperglycemic without any hyperlipidemia or kidney injury (5, 11), which is a good model for type 1 diabetes mellitus. With the use of this STZ rat model, experiments were performed on day 5 after the STZ injection. All animal protocols and procedures were approved by the Animal Care and Use Committee of Emory University and were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Western blot analysis.

Cell lysates (100 μg) were loaded and electrophoresed on 10% SDS-PAGE gels for 60–90 min. Gels were blotted onto PVDF membranes for 1 h at 90 V. After 1 h of being blocked with 5% BSA, PBS, and Tween 20 buffer, PVDF membranes were incubated with primary antibodies (1:1,000 dilution) to TRPC6 (SAB2102583, Sigma; St. Louis, MO), zonula occludens-1 (ZO-1; SC-33725, Santa Cruz Biotechnology), podocin (P0372, Sigma), or nephrin (PRS2265, Sigma) overnight at 4°C and then incubated with horseradish peroxidase-conjugated sheep anti-rabbit IgG secondary antibody (1:5,000 dilution, GE Healthcare) for 1 h after four vigorous washes. Finally, blots were visualized with chemiluminescence using the ECL Plus Western Blotting Detection System (GE Healthcare).

Confocal microscopy imaging.

To evaluate protein levels of podocin, TRPC6, and ZO-1, podocytes were incubated with rabbit polyclonal podocin antibody (1:500, P0372, Sigma) and rabbit polyclonal TRPC6 antibody (1:500, SAB2102583, Sigma) overnight, washed three times, and then incubated with both goat anti-rabbit Alexa Fluor 488 antibody (2 μg/mL, A-11034, Invitrogen) and rat Alexa Fluor 594-conjugated ZO-1 antibody (1:500, SC-33725 AF594, Santa Cruz Biotechnology) for 2 h. Confocal microscopy experiments were performed, as we have previously described (34), to determine how high glucose affect both protein expression levels of podocin and TRPC6 and the distribution of ZO-1 in podocytes. Alexa Fluor 488 (TRPC6) was excited with a 488-nm laser and visualized through a 515-nm emission filter, shown in green (see Fig. 4). Alexa Fluor 594 (ZO-1) was excited with a 543-nm laser and visualized through a 618-nm emission filter, shown in red (see Fig. 4). Confocal microscopy XY scanning of podocytes was accomplished using an upright confocal microscope (Fluoview 1000, Olympus, Tokyo, Japan). To determine intracellular Ca²⁺, podocytes were incubated with 5 μM Fluo-4 AM, a fluorescent Ca²⁺ indicator, for 30 min. Fluo-4 was excited with a 488-nm laser and visualized through a 515-nm emission filter. In each set of experiments, images were taken using the same parameter settings. Data were analyzed using Olympus Fluoview FV1000 (version 3.1) software.

Fig. 4.

High glucose (HG) reduces the protein levels of podocin in a transient receptor potential canonical 6 (TRPC6)-dependent manner. A and B: confocal microscopy images showing that HG (22 mM for 7 days) reduced podocin protein levels in control podocytes (A) but not in TRPC6 knockdown podocytes (B). White boxes are the zoom-in areas after treatment with HG as shown on the right. Podocin is shown in green; zonula occludens-1 (ZO-1) is shown in red. Images represent 5 individual experiments showing consistent results.

Chemicals.

Most chemicals were purchased from Sigma-Aldrich. All solutions were either premade and stored in a −20°C freezer or freshly made before the experiments. All concentrations listed in this study represent final concentrations.

Statistical analysis.

Data are reported as means ± SD. Statistical analysis was performed with SigmaPlot and SigmaStat software (Jandel Scientific, San Rafael, CA). A paired t-test was used for comparisons between pre- and posttreatment activities. ANOVA was used for comparisons among various treatment groups. A Z test and χ²-test were used for comparisons between the changes in percentages. Results were considered significant if P < 0.05.

RESULTS

High glucose and type 1 diabetes mellitus increase the expression of TRPC6 and decrease the expression of podocin and nephrin.

It has been previously shown by the co-author’s laboratory that STZ can elevate blood glucose up to 300–450 mg/dL (18–27 mM) (11). Therefore, in the present study, we treated cultured human podocytes with 22 mM glucose for 7 days to determine whether high glucose regulates the expression of TRPC6, podocin, and nephrin. The data showed that high glucose significantly increased the expression of both membrane-bound and whole cell TRPC6 levels but decreased the expression of podocin and nephrin (Fig. 1). To confirm the in vitro data from cultured human podocytes in a diabetic rat model, we used kidneys from type 1 diabetes mellitus STZ rats in which high blood glucose has been previously confirmed (11). Consistently, the in vivo data also showed that the expression of TRPC6 was increased but that the expression of podocin and nephrin was decreased in hyperglycemic STZ rats (Fig. 2). These data suggest that high glucose increases the expression of TRPC6 but decreases the expression of podocin and nephrin.

Fig. 1.

High glucose increases the expression of transient receptor potential canonical 6 (TRPC6) and decreases the expression of podocin and nephrin in cultured human podocytes. A: representative Western blots from control podocytes and podocytes treated with 22 mM glucose for 7 days, designated as high glucose through the whole study. B: summary plots of protein levels of biotinylated TRPC6, whole cell TRPC6, podocin, and nephrin in control podocytes and podocytes treated with high glucose. In A and B, the density of each band representing a protein levels was normalized with the density of β-actin in the matching lane. Data represent 5 individual experiments showing consistent results.

Fig. 2.

Type 1 diabetes mellitus increases the expression of transient receptor potential canonical 6 (TRPC6) and decreases the expression of podocin and nephrin in streptozotocin (STZ) rats. A: representative Western blots from the kidney of control and STZ rats. B: summary plots of normalized protein levels of TRPC6, podocin, and nephrin in control and STZ rats. Data represent 5 individual experiments showing consistent results.

High glucose and activation of TRPC6 decrease the expression of podocin in control podocytes but not in TRPC6 knockdown podocytes.

To further determine whether glucose decrease the expression of podocin by activating TRPC6, experiments were performed using both control podocytes and TRPC6 knockdown podocytes. Successful TRPC6 knockdown has been previously confirmed and published (15). The Western blot data showed that high glucose decreased the expression of podocin in control podocytes but not in TRPC6 knockdown podocytes (Fig. 3). In this set of experiments, we also tested that if activation of TRPC6 with hypoosmotic stretch either alone or plus hyperforin produced similar effects on the expression of podocin. The data showed that hyperforin did not produce any additive effects, indicating that hypoosmotic stretch reduces the expression of podocin by activating TRPC6. In addition, we also performed confocal microscopy experiments. Consistently, the data showed that high glucose decreased the expression of podocin in control podocytes but not in TRPC6 knockdown podocytes (Fig. 4). These data suggest that high glucose decreases the protein expression of podocin by activating TRPC6.

Fig. 3.

High glucose reduces the expression of podocin via a transient receptor potential canonical 6 (TRPC6)-dependent pathway. A: representative Western blots of podocin from control podocytes or TRPC6 knockdown podocytes. Cells were either under control conditions or treated with 22 mM glucose for 7 days, hypoosmotic solution (to activate TRPC6), and hyperforin (10 μM, a TRPC6 activator). B: summary plots of normalized protein of podocin protein levels under each condition listed in A. Data represent 5 individual experiments showing consistent results. NS, not significant.

High glucose increases intracellular Ca²⁺ and the expression of ZO-1 by activating TRPC6.

Since TRPC6 is a Ca²⁺-permeable channel (7), theoretically, stimulation of TRPC6 by high glucose should increase intracellular Ca²⁺. Indeed, our confocal microscopy data showed that high glucose strongly elevated intracellular Ca²⁺ in control podocytes but not in TRPC6 knockdown podocytes (Fig. 5). Previous studies have shown that the expression of ZO-1 is increased in response to elevation of intracellular Ca²⁺ (31). Therefore, high glucose should promote the expression of ZO-1. Indeed, our data showed that high glucose increased the expression of ZO-1 in cultured human podocytes (Fig. 6). The effects were mimicked by activation of TRPC6 with hyperforin and abolished by blockade of TRPC6 with SKF-96365. These data suggest that high glucose increases intracellular Ca²⁺ and the expression of ZO-1 by activation of TRPC6.

Fig. 5.

High glucose increases intracellular Ca²⁺ in a transient receptor potential canonical 6 (TRPC6)-dependent manner. A and B: confocal microscopy images showing that high glucose (22 mM for 7 days) increased intracellular Ca²⁺ in control podocytes (A) but not in TRPC6 knockdown podocytes (B). Relative intracellular Ca²⁺ levels were labeled with Fluor-4 AM (brighter indicates higher intracellular Ca²⁺ levels). C: summary plots of changes in green fluorescence intensity in podocytes after treatment with high glucose normalized with that in podocytes under control conditions. Images represent 5 individual experiments showing consistent results.

Fig. 6.

High glucose increases the expression of zonula occludens-1 (ZO-1) via a transient receptor potential canonical 6 (TRPC6)-dependent pathway. A: representative Western blots from control podocytes and podocytes treated with high glucose (22 mM for 7 days), hyperforin (10 μM), and high glucose (22 mM for 7 days) plus SKF-96365 (10 μM). B: summary plots of normalized protein of ZO-1 under each condition listed in A. Data represent 5 individual experiments showing consistent results.

High glucose causes loss of podocyte processes via a TRPC6-dependent pathway.

Podocyte food process effacement is the early change in the glomerulus found in diabetes. However, there are no in vitro data to show that the change can be initiated by high glucose challenge. Our data from confocal microscopy experiments showed that ZO-1 was predominantly located in the processes of cultured human podocytes under control conditions, but after high glucose, it was accumulated in the connection between two podocytes to form a single straight line (Fig. 7). These data suggest that stimulation of TRPC6 by high glucose causes cultured human podocytes to lose both their processes and normal distribution of ZO-1.

Fig. 7.

High glucose (HG) causes loss of podocyte processes and redistribution of zonula occludens-1 (ZO-1) probably by stimulating transient receptor potential canonical 6 (TRPC6). Confocal microscopy images show podocytes either under control conditions or after treatment with HG (22 mM for 7 days) in the absence or presence of SKF-96365 (SKF; 10 μM). TRPC6 is shown in green; ZO-1 is shown in red. White boxes are zoom-in areas between two podocytes under each condition as shown on the right. The white arrow indicates the tight connection between two podocytes after treatment with HG. Images represent 6 separate experiments showing consistent results.

DISCUSSION

The present study shows that high glucose and type 1 diabetes mellitus increase the expression of TRPC6 but decrease the expression of podocin and nephrin. We have previously shown that high glucose stimulates TRPC6 by elevating intracellular reactive oxygen species (15). Previous studies have shown that high glucose significantly decreases mRNA and protein expressions of podocin and nephrin (14). Here, we show that high glucose reduces podocin protein levels via a TRPC6-dependent mechanism. However, the underlying mechanism remains unclear. Since we also show that high glucose elevates intracellular Ca²⁺ by stimulating TRPC6, elevation of intracellular Ca²⁺ may play a role in mediating the reduction of podocin induced by high glucose. To our knowledge, there is no evidence that an elevation of intracellular Ca²⁺ can downregulate mRNA and protein expression of podocin, but intracellular Ca²⁺ is known to regulate protein degradation by modulating ubiquitin ligases (18). Recent studies have suggested that protein levels of podocin can be regulated via a protein ubiquination pathway (25). Therefore, elevation of intracellular Ca²⁺ due to activation of TRPC6 may reduce podocin protein levels by promoting its degradation. However, how activation of TRPC6 leads to the decrease in podocin will serve as interesting directions for our future studies.

Previous studies have suggested a reverse interaction between podocin and TRPC6 in that podocin antagonizes the gating of TRPC6 evoked by membrane stretch and diacylglycerol (1, 9, 12). However, our new discovery is not contradictory to previous reports but complements previous studies by suggesting a positive feedback mechanism for podocyte injury. That is, activation of TRPC6 reduces protein levels of podocin; conversely, reduced podocin expression due to activation of TRPC6 further activates TRPC6 by increasing the sensitivity of TRPC6 to membrane stretch and diacylglycerol. Therefore, high glucose may initiate a positive feedback loop to strongly activate TRPC6 and reduce podocin to cause podocyte injury. This positive feedback mechanism should play an important role in diabetic glomerular nephropathy. Recent studies have suggested that vitamin D can reduce TRPC expression in podocytes (30). Further studies may suggest the clinical use of TRPC6 channel blockers or vitamin D for diminishing podocin degradation, which breaks the positive feedback loop, and subsequently preventing the glomerular damage in patients with type 1 diabetes.

Since it is known that TRPC6 is a Ca²⁺-permeable channel (7) and that intracellular Ca²⁺ promotes ZO-1 biogenesis (31), we argue that increases in intracellular Ca²⁺ may account for the increased ZO-1 expression. However, it remains to be determined whether the redistribution of ZO-1 results in the formation of tight connections between podocytes or if the formation of tight connections caused the redistribution of ZO-1 we observed. These topics will also serve as interesting directions for our future studies. We also show that high glucose induces podocytes to lose their processes and form tight cell-to-cell connections and that high glucose also promotes ZO-1 accumulation along the tight connections. It is known that intracellular Ca²⁺ is required for ZO-1 trafficking (31), but podocin also mediates the assembly of tight junctions between foot processes in nephritic podocytes (27). However, whether high glucose induces tight connections between podocytes via TRPC6-mediated elevation of intracellular Ca²⁺ or the subsequent decrease of podocin expression remains to be further determined. Also, STZ-treated Sprague-Dawley rats may have limited kidney injury. In contrast, recent studies have suggested that significant podocyte injury can occur in STZ-treated Dahl rats (10), suggesting that STZ-treated Dahl rats may be a better model for studying diabetic nephropathy.

GRANTS

This work was supported by the National Natural Science Foundation of China (NSFC) Grants 81701155 (to X.-Y. Lu) and 81670381 (to B.-C. Liu) and NSFC Grants 81571166 and 81771361 and NSFC International Collaboration and Exchange Program Grant 81820108014 (to L.-H. Wang) and National Institute of Diabetes and Digestive and Kidney Diseases Grants 5R01-DK-067110 and R01-DK-100582 (to H.-P. Ma).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

X.-Y.L., B.-C.L., Y.-Z.C., L.W., and H.-P.M. conceived and designed research; X.-Y.L. and B.-C.L. performed experiments; X.-Y.L., B.-C.L., Y.-Z.C., J.D.K., H.-X.Z., L.W., and H.-P.M. analyzed data; X.-Y.L., B.-C.L., Y.-Z.C., C.S., H.S., G.C., J.D.K., H.-X.Z., L.W., and H.-P.M. interpreted results of experiments; X.-Y.L., B.-C.L., and Y.-Z.C. prepared figures; X.-Y.L. and H.-P.M. drafted manuscript; X.-Y.L., B.-C.L., and H.-P.M. edited and revised manuscript; X.-Y.L., L.W., and H.-P.M. approved final version of manuscript.