Polydatin impairs mitochondria fitness and ameliorates podocyte injury by suppressing Drp1 expression

Polydatin (PD), a resveratrol glycoside, has been shown to protect renal function in diabetic nephropathy (DN), but the underlying molecular mechanism remains unclear. This study demonstrates that PD stabilize the mitochondrial morphology and attenuate mitochondrial malfunction in both KKAy mice and in hyperglycemia (HG)‐induced MPC5 cells. We use Western blot analysis to demonstrate that PD reversed podocyte apoptosis induced by HG via suppressing dynamin‐related protein 1 (Drp1). This effect may depend on the ability of PD to inhibit the generation of cellular reactive oxygen species (ROS). In conclusion, we demonstrate that PD may be therapeutically useful in DN, and that, podocyte apoptosis induced by HG can be reversed by PD through suppressing Drp1 expression.

important factor contributing to podocyte injury in KKAy mice (unpublished data). Drp1 is known to play an essential role in mitochondrial fusion and in promoting high glucose (HG)-induced podocyte apoptosis (Katusic & Austin, 2014). Many reports indicate that under hyperglycemic conditions, Drp1 activity, mitochondrial fission, cytochrome C release, and ROS production are enhanced, thereby promoting cell apoptosis (Minai & Yeheskely-Hayon, 2013).
Polydatin (PD), a resveratrol glycoside extracted from the roots of Polygonum cuspidatum, is widely applied in traditional Chinese remedies (Jiang et al., 2017). PD exhibits more potent antioxidant effects than resveratrol due to its specialized biological properties resulting from conformational differences relative to resveratrol (Lanzilli et al., 2012). Previous studies have demonstrated that PD can improve heart function, protect against Alzheimer's disease, and ameliorate renal injury via mitochondrial protection (Jin, Xu, Wang, Xu, & Zhang, 2009). Furthermore, many studies have suggested that PD suppresses fibronectin accumulation in hyperglycemic glomerular mesangial cells and ameliorates renal function in diabetic rats, thereby inhibiting renal fibrosis in DN (Gao et al., 2015). Although, PD exhibits demonstrable mitochondrial protective effects, its regulation of HG-induced activation of Drp1 in DN remains to be further elucidated.
KKAy mice with spontaneous type 2 diabetes are a widely used animal model in DN research. These mice feeding on a high-fat diet would show clinical manifestations of hyperglycemia, impaired glucose tolerance, hyperinsulinemia, moderate obesity, hyperlipidemia, and proteinuria. Kidney damage in these mice is very similar to that which occurs during human DN.
In the current study, we used KKAy mice and a conditionally immortalized mouse podocyte cell line (MPC5) to study the protective effect of PD. We demonstrate for the first time that PD strongly protects podocytes in DN by inhibiting Drp1 activation and Drp1-mediated mitochondria fission.

| Cell culture and lentivirus gene transfer
Conditionally immortalized mouse podocytes (MPC5) were purchased from Rantai Company (Shanghai, China). MPC5 cells were cultured according to the previously described method (Mundel, Reiser, & Kriz, 1997). In brief, podocytes were cultured in collagen I-coated dishes (BD Biosciences, Bedford, MA)  The lentiviruses were added to the cells together with Polybrene (Sigma-Aldrich) at a final concentration of 10 mM for 6 hr (at 37°C) and cultured in F-12 Ham's medium.

| Animals and experimental protocols
All animal study protocols for the present study were reviewed and approved by the Institutional Animal Care and Use Committee at Nanfang Medical University, China. In brief, KKAy mice and their nondiabetic control C57BL/6Jmice (9-11 weeks of age) were purchased from the Institute of Laboratory Animal Sciences, Chinese Academy of Medical Science (Beijing, China). Mice were individually housed in stainless steel cages in a room at a constant temperature and humidity with 12 hr light/dark cycles. During the experiment, the KKAy mice were provided high-fat diet and water ad libitum.
Male C57BL/6J mice were pair-fed with normal diet and water ad libitum. At 14 weeks of age, urine samples were collected, and KKAy mice with a urine albumin-creatinine ratio (ACR) ≥300 µg/mg were considered to have DN. The KKAy mice were randomly divided into a PD treatment group (DIAB + PD, n = 10), which received 100 mg/kg PD for 8 weeks, or a DN control group (DIAB, n = 10), which was administered the same volume of vehicle (containing 0.5% sodium carboxymethylcellulose). Age-matched healthy C57BL/6J mice receiving vehicle were used as a normal control group (CTRL, n = 10), and mice receiving 100 mg/kg of PD for 8 weeks served as the PD control group (CTRL + PD, n = 10). At the end of the study, urine was collected from individual mice housed in metabolic cages for 24 hr. All mice were sacrificed by carbon dioxide asphyxiation. Blood samples were collected, and plasma and serum were separated and stored at −20°C until final biochemical analysis.
Kidneys samples were fixed in 10% buffered formalin and frozen at −80°C until use.
Briefly, 10 mice were orally administered a fixed dose of PD (5000 mg/kg body weight) after fasting overnight, and any signs and symptoms of toxicity and mortality were recorded, if any, up to a period of 72 hr. NI ET AL. | 2777 2.5 | Biochemical analysis A total of 24 hr urine protein (UP), urinary albumin excretion (UAE), blood urea nitrogen (BUN), and serum creatinine (SCr) levels were analyzed by the Department of Clinical Laboratory at the Shenzhen Hospital of Nanfang Medical University. Renal hypertrophy was assessed using the kidney weight to body weight ratio (KW/BW).

| Isolation of kidney glomeruli and mitochondria
Glomeruli were isolated using a previously described method (Takemoto et al., 2002) using a kit purchased from Sigma (Sigma-Aldrich) according to the manufacturer's protocol. Briefly, mice were anesthetized and the kidney was removed and perfused with 5 ml of phosphate-buffered saline, then kidneys were minced into small pieces, digested by collagenase and DNase, and filtered. After washing for three times, the glomeruli were collected using a magnet and the purity of glomeruli was confirmed to be about 95% by phase-contrast microscopy. Mitochondria from glomeruli were isolated using a kit purchased from Sigma, according to the manufacturer's protocol.
Isolated mitochondria were resuspended in radioimmunoprecipitation assay buffer at 4°C and then centrifuged at 12,000g for 10 min at 4°C.
The protein concentration was determined using the Bradford method.

| Histopathological analysis
Mice kidneys were excised, fixed in 4% paraformaldehyde, dehydrated, and embedded in paraffin, sectioned at 3 µm, and stained with hematoxylin and eosin.

| Evaluation of podocyte apoptosis in renal tissue
Podocyte apoptosis was evaluated using a commercially available terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) assay kit (CHEMICON International, Temecula, CA) according to the manufacturer's protocol. The sections of frozen kidney tissue were counterstained with DAPI, anti-synaptopodin antibody, and TUNEL. TUNEL-positive cells (Green) were analyzed via fluorescence microscopy (Nikon Corporation, Tokyo, Japan).

| Hoechst 33258 staining
Podocytes were fixed with 4% formaldehyde in phosphate-buffered saline (PBS) for 10 min, and the cells were stained with Hoechst 33258 for 1 hr. Apoptotic cells were identified by the condensation and fragmentation of their nuclei, and images were captured using a fluorescence microscope (Nikon Corporation, Tokyo, Japan).

| Analysis of mitochondrial cytochrome C release
MPC5 cells were cultured for 72 hr in 5.3 mM or 30 mM D-glucose, harvested by gentle scraping into the medium, and washed with icecold phosphate-buffered saline (PBS). Mitochondrial fractions containing cytochrome C were obtained by using a Mitochondria Isolation kit (Sigma-Aldrich). The pellet contained the mitochondrial fraction, and the supernatant was collected as the cytosolic fraction. The release of cytochrome C from the mitochondria into the cytoplasm was detected by Western blot quantification.
The activities of caspase-3 and -9 are expressed as the absorbance of the cleaved substrate (pNA) at 405 nm.

| Transmission electron microscopy (TEM)
Podocytes or renal cortex were fixed in 2.5% glutaraldehyde and then fixed in 1% O s O 4 /0.1 mol/L phosphate buffer (Sigma-Aldrich) for 1 hr.
Ultrathin sections, 50 nm thick, were cut with an Ultracut S Ultramicrotome, placed on copper grids, double-stained with uranyl acetate and lead citrate, and examined by TEM (LSM 510; Carl Zeiss, Germany). The ultrastructural changes to the podocyte slit diaphragm of the mice (Ruotsalainen et al., 1999) were measured as previously described.

| Mitochondrial membrane potential (MMP) by florescent JC1
The MMP in the podocytes and isolated mitochondria were determined using JC-1 fluorescent staining according to manufacturer's protocol as described previously (Daehn et al., 2014). The isolated mitochondria and cells were washed twice with ice-cold Hank's balanced salt solution (HBSS) (Gibco BRL) centrifuged at 400 × g for 1 min at 4°C. Cells or mitochondria were washed with PBS twice; the results were read using a microplate reader (FLUOStar Omega, BMG Labtech, Ortenberg, Germany). The ratio was calculated by relative aggregate fluorescence (red) to JC-1 monomer (green). . ATP assays were performed as previously described, and ATP levels were measured and normalized to total protein concentration.

| Mitochondrial ROS determination
Intracellular ROS generation was detected as described previously (Zhang, Wang, & Chen, 2012). Isolated glomeruli and MPC5 cells were stimulated with 30 mM D-glucose or 25 mM PD for 3 days. The collected cells were washed twice with wash buffer and were directly analyzed using a fluorescence microplate reader. More than 10,000 cells were acquired and analyzed for each sample, and the ROS generation was normalized to the protein concentration of each treated sample relative to control.

| PD improves renal function in KKAy mice
To investigate the effect of PD in KKAy mice kidneys during DN, we administered vehicle or 100 mg/kg/day PD orally to KKAy and C57BL/6J mice for 8 weeks. The KW/BW ratio, 24 hr UP, UAE, SCr, and BUN levels significantly increased in the DIAB mice when compared with CTRL mice. The oral administration of PD effectively decreased the fasting blood glucose (FBG), KW/BW ratio, 24 hr UP, UAE, SCr, and BUN levels relative to DIAB mice (Table 2). However, PD treatment did not affect the FBG levels or renal function of CTRL mice.

| PD attenuates podocyte apoptosis in KKAy mice
In order to study the effect of PD on podocyte in diabetic milieu. We studied KKAy mice that had been treated with PD for 8 weeks. The kidneys of the mice were examined histopathologically. KKAy mice exhibited significant mesangial expansion compared with C57BL/6J mice, and PD treatment prevented these changes and restored normal structure to the glomeruli (Figure 1a). In the kidneys of KKAy mice, the foot processes exhibited extensive fusion and filtration slits, and pore density was reduced. By contrast, KKAy mice treated with PD

| PD attenuates apoptosis in hyperglycemic MPC5 cells
As shown in Figure 2, podocyte apoptosis was measured using the TUNEL assay, Hoechst 33258 staining, and flow cytometry. Treatment with PD significantly prevented the decrease of cell viability. We further investigated the effects of PD on cytochrome C release and the activation of the caspase cascade. We observed a significant release of cytochrome C from the mitochondria after exposure to HG, whereas PD treatment blocked cytochrome C release (Figure 2c). HG activated caspase-3 (cleaved form) and -9, which suggested that HG-induced podocyte apoptosis was mediated through the intrinsic mitochondrial proapoptotic pathway, and that the activation of both caspase-3 and -9 were substantially reduced by PD (Figure 2d). Furthermore, real-time PCR and Western blot analysis revealed that HG-induced reduction of podocin and nephrin was restored by PD treatment (Figure 2e,f).

| PD inhibits HG-induced mitochondria fission
Compared with the mitochondria in NG cells, the mitochondria in hyperglycemic MPC5 cells exhibited a more punctate pattern. However, this fragmentation could be prevented by PD treatment. We also analyzed MMP ( Figure 3a) and ATP production (Figure 3b) in hyperglycemic MPC5 cells after PD treatment. Subsequently, we investigated the mechanism underlying the reversion effect of PD on podocyte mitochondria. As shown in Figure 3c,d, the ultrastructural electron microscopy analysis revealed marked differences in the shape of the mitochondria in HG cells. After PD treatment, the mitochondria maintained a long, filamentous structure. The podocyte mitochondria from KKAy mice were notably swollen, deformed, and vesicular compared with those from C57BL/6J mice. By contrast, KKAy mice treated with PD exhibited relatively few swollen, deformed, and vesicular mitochondria (Figure 3e).

| PD blocks Drp1 expression in hyperglycemic podocytes
Drp1-mediated fission is involved in mitochondrial fragmentation (Filichia, Hoffer, Qi, & Luo, 2016) and acts with other cofactors to induce cytochrome C release, caspase activation, and apoptosis (Oettinghaus et al., 2016). Hence, we investigated whether HG or PD could affect the expression of Drp1 in podocyte mitochondria. We isolated and cultured primary podocytes from KKAy mice. As shown in There are several key molecules that modulate mitochondria morphology. In addition to Drp1, which mediates mitochondria  (Kushnareva et al., 2013). We thus analyzed the Opa1 levels in primary podocytes and determined that the level of Opa1 was not affected by HG or PD (Figure 4e).

| PD inhibits mitochondrial fission and cell apoptosis in podocytes by suppressing Drp1 expression
We further examined the mechanism that links PD and Drp1. The level of Drp1 expression and mitochondrial fragmentation in podocytes were significantly elevated or reduced when cells were infected with a Drp1-GFP

| PD prevents cell apoptosis and ROS production in Drp1 siRNA-infected primary podocytes
As shown in Figure 7, primary podocytes treated with HG + Drp1-siRNA + PD exhibited decreased apoptosis compared with cells treated with HG + Drp1-siRNA, and ROS levels in primary podocytes were affected by HG but not Drp1-siRNA.

| DISCUSSION
Murine models of diabetes mellitus suggest that podocyte apoptosis is a key mediator in the pathogenesis of DN (Zhou et al., 2012). In the PD is an active stilbene compound isolated from the roots of Polygonumcuspidatum. PD has been manifested to possess antioxidative and anti-inflammatory activities (Mohan et al., 2013). PD possesses antioxidative and anti-inflammatory activities (Mohan et al., 2013). Previous studies have demonstrated the therapeutic effects of PD in HG-induced injury in multiple organs, including heart, liver, and brain (Xu et al., 2016;Zhang, Tan, Zhang, & Yao, 2015). To test the safety of PD, we administered a single large PD dose of 5000 mg/kg body weight to mice and observed no adverse effects. The LD50 value of PD is presumed to be greater than 5000 mg/kg body weight in mice, and PD can be effectively considered a non-toxic substance.
PD has been shown to possess antioxidant and anti-inflammatory activities (Mohan et al., 2013). Previous studies have demonstrated the therapeutic effects of PD on HG-induced injury in multiple organs, including the heart, liver, and brain (Xu et al., 2016;Zhang et al., 2015).
In the present study, we observed that PD decreased all of the analyzed markers of renal dysfunction (Table 2) in KKAy mice, suggesting that PD possesses renoprotective effects.
Recent studies have demonstrated that podocyte loss occurs in patients with DN (Maezawa, Takemoto, & Yokote, 2015). We observed that the shape of the foot processes, shape of the filtration slits, and the pore density of podocytes in KKAy mice can be restored after PD treatment (Figure 1a-c). Nephrin and podocin are podocytespecific proteins that play crucial roles in the function of the glomerular filtration barrier. Nephrin is a key factor in the glomerular slit diaphragm. The absence of nephrin leads to proteinuria and foot process effacement (Teng et al., 2016). The expression of renal nephrin and podocin was significantly reduced in diabetic KKAy mice. Our data demonstrate that PD treatment restores the levels of nephrin and podocin proteins (Figure 1d). These results strongly indicate that PD can reverse podocytes apoptosis in diabetic KKAy mice glomeruli. To explore the protective effect of PD on podocytes further, we took advantage of MPC5 cells exposed to HG conditions to assess the effect of PD on cell apoptosis. HG has also been shown to induce podocyte apoptosis, and this effect has been attributed to the dephosphorylation of cytosolic phospho-Bad and the associated accumulation of cytosolic cytochrome C (Yuan et al., 2017). In the present study, PD protected against HG-induced MPC5 cell injury by inhibiting both cell apoptosis and the loss of slit diaphragm proteins.
Furthermore, podocyte apoptosis induced by Aldo was accompanied by the loss of two silt diaphragm proteins, nephrin, and podocin. Our findings, demonstrate that PD protects against HG-induced podocyte injury in vitro by inhibiting both cell apoptosis and the loss of slit diaphragm proteins (Figure 2).
During the progression of type 2 diabetes, mitochondria play a major role in podocyte apoptosis and the development of diabetes (Wang et al., 2012;Xu et al., 2016). Mitochondrial fission is a highly regulated process, and mitochondrial fragmentation is involved in the leakage of mitochondrial membrane proteins and the early stages of cell apoptosis (Zou, Roth, Younis, Burgoon, & Ganey, 2010). It is now widely recognized that mitochondrial dynamics are crucial for mitochondrial function, maintenance, and quality control (Westermann, 2010). Altered | 2785 mitochondrial dynamics are often accompanied by increased ROS production, decreased cellular ATP and defects in cellular respiration evoked by oxidative damage (Ma, 2013). Mitochondrial dysfunction in podocytesis is increasingly recognized as a factor contributing to the pathogenesis of DN (Haas et al., 2016).
To evaluate mitochondrial function, we selected a group of indicators, such as mitochondrial morphology, MMP, and ATP production. PD strongly prevented mitochondria fragmentation in HG-induced podocytes, which suggests the therapeutic effect of PD on DN (Figure 3).
Drp1, a large GTPase, plays a critical role in mitochondrial fission.
In agreement with previous findings (Liu et al., 2015), the level of Drp1 was markedly increased in hyperglycemic MPC5 cells, but this increase was significantly attenuated by PD treatment (Figure 4a). We  (Figure 5a,b). Goyal, Fell, Sarin, Youle, and Sriram, (2007) suggested that knockdown of Drp1 by siRNA delays but does not prevent HG-induced cell apoptosis. In the present study, we We also demonstrated that PD prevents primary podocytes apoptosis secondary to antioxidant activity ( Figure 6). A growing body of evidence suggests that excess cellular ROS plays a key role in mitochondrial perturbations, the pathogenesis of diabetes complications, and podocyte apoptosis (Khamaneh, Alipour, Sheikhzadeh Hesari, & Ghadiri Soufi, 2015). In the current study, we demonstrated that HG-induced ROS can be significantly blocked by PD treatment. Furthermore, the expression of Drp1 was positively associated with the cellular ROS level (Figure 7), suggesting that PD may protect mitochondria through inhibition of ROS generation.
Collectively, our findings provide new insights into the pathogenic process of HG-induced podocyte injury and also identify a new