Early podocyte injury and elevated levels of urinary podocyte-derived extracellular vesicles in swine with metabolic syndrome: role of podocyte mitochondria

Abstract

Metabolic syndrome (MetS) is associated with nutrient surplus and kidney hyperfiltration, accelerating chronic renal failure. The potential involvement of podocyte damage in early MetS remains unclear. Mitochondrial dysfunction is an important determinant of renal damage, but whether it contributes to MetS-related podocyte injury remains unknown. Domestic pigs were studied after 16 wk of diet-induced MetS, MetS treated with the mitochondria-targeted peptide elamipretide (ELAM; 0.1 mg·kg⁻¹·day⁻¹ sc) for the last month of diet, and lean controls (n = 6 pigs/group). Glomerular filtration rate (GFR) and renal blood flow (RBF) were measured using multidetector computed tomography, and podocyte and mitochondrial injury were measured by light and electron microscopy. Urinary levels of podocyte-derived extracellular vesicles (pEVs; nephrin positive/podocalyxin positive) were characterized by flow cytometry. Body weight, blood pressure, RBF, and GFR were elevated in MetS. Glomerular size and glomerular injury score were also elevated in MetS and decreased after ELAM treatment. Evidence of podocyte injury, impaired podocyte mitochondria, and foot process width were all increased in MetS but restored with ELAM. The urinary concentration of pEVs was elevated in MetS pigs and directly correlated with renal dysfunction, glomerular injury, and fibrosis and inversely correlated with glomerular nephrin expression. Additionally, pEV numbers were elevated in the urine of obese compared with lean human patients. Early MetS induces podocyte injury and mitochondrial damage, which can be blunted by mitoprotection. Urinary pEVs reflecting podocyte injury might represent early markers of MetS-related kidney disease and a novel therapeutic target.

INTRODUCTION

Metabolic syndrome (MetS) comprises a group of major cardiovascular risk factors associated with dyslipidemia, obesity, and insulin resistance (IR). A previous study (40) has linked MetS with the development of chronic kidney disease (CKD), and their coexistence is associated with increased cardiovascular mortality (14). However, the mechanisms of MetS-related kidney disease are not fully understood.

Podocytes are terminally differentiated, energy-requiring cells with limited replicative ability and typically do not proliferate after injury (47). Patients with obesity-related glomerulopathy often have reduced podocyte density, widened foot processes, and proteinuria (8). Similar glomerular findings have been reported in mice after a high-fat diet even before a significant increase in blood glucose (35, 46). Podocytes have a complex and unique cellular architecture, and their intersecting foot processes are mainly maintained by an energy-dependent arrangement of actin filaments. For example, ATP-dependent nephrin phosphorylation is required for the formation of slit membranes in foot processes (42).

Mitochondria play an important role in providing ATP for maintaining the actin cytoskeleton in podocytes (1, 17), and mitochondrial dysfunction is involved in several diseases in which podocyte injury is a key event (18, 20). The mitochondria-targeted peptide elamipretide (ELAM or SS-31) is a synthetic tetrapeptide that selectively targets cardiolipin on the inner mitochondrial membrane to protect cristae curvature, stabilizes mitochondrial structure, facilitates electron transport, and minimizes the output of reactive oxygen species (ROS) (5, 38, 48). ELAM reduces CKD by protecting mitochondria in renal ischemia and with high-fat diet (37, 39), and we have previously shown that it restores renal function in swine with renovascular disease (11, 12). Furthermore, early changes in medullary tubular segments induced by MetS in pigs are associated with mitochondrial abnormalities and protection of mitochondria decreased tubular injury (13). However, the involvement of glomerular mitochondrial injury in early MetS has not been explored.

Extracellular vesicles (EVs) are a heterogeneous group of spherical structures bounded by a lipid bilayer and carrying a cargo of proteins and nucleic acids and include exosomes, microvesicles, and apoptotic bodies. EVs may serve for intercellular communication and are often released into the extracellular space, including in plasma, urine, and saliva, by different kinds of cells (44) damaged or stressed by a hostile microenvironment, such as hypoxia (2). Recent studies have suggested urinary podocyte-derived EVs (pEVs) as biomarkers of podocyte-specific injury, as their levels increased in early diabetic nephropathy (7), renovascular hypertension (22), and preeclampsia (16). It is yet unclear whether pEV levels are altered at the early stage of MetS-related kidney disease. The present study tested the hypothesis that early MetS induces damage of podocytes and their mitochondria in pigs, which may be reflected in increased level of urinary pEVs, and that protection of mitochondria with ELAM would attenuate the damage.

MATERIALS AND METHODS

Experimental protocol.

The present study was approved by the Institutional Animal Care and Use Committee. Eighteen (3 mo old) female domestic pigs were randomly divided into two groups studied after 16 wk of observation (Fig. 1A). Lean pigs (n = 6) were fed standard diet (Purina Animal Nutrition, Arden Hills, MN), whereas 12 MetS pigs were fed a high-fat/high-fructose diet (5B4L, protein 16.1%, ether extract fat 43.0%, and carbohydrates 40.8%, Purina Test Diet, Richmond, IN) (10). Twelve weeks after the initiation of the diet, six MetS pigs started treatment with subcutaneous injections of ELAM (d-Arg-2969-dimethyl-Tyr-Lys-Phe-NH₂, Stealth BioTherapeutics, Newton Centre, MA), 0.1 mg/kg in 1 ml PBS once daily 5 days/wk for 4 consecutive weeks (10, 11). PBS vehicle was administered to the remaining six MetS pigs and six lean pigs.

Fig. 1.

A: schematic of the experimental protocol. B–I: elamipretide (ELAM) prevents glomerular mesangial expansion in glomeruli and reduces glomerular injury scores of metabolic syndrome (MetS) pigs. B−D: representative images of glomeruli stained with hematoxylin and eosin (B), periodic acid-Schiff (PAS; C), or trichrome (D) in lean, MetS, and MetS + ELAM-treated pigs. E: representative transmission electron microscopic images of glomeruli from lean, MetS, and MetS + ELAM-treated pigs showing podocytes (P), endothelial cells (E), capillary loops (CL), and the mesangium (M). F–I: quantification of images. Glomerular size was elevated in both MetS and MetS + ELAM-treated pigs (F), whereas the significant mesangial expansion in MetS was reduced by ELAM (G). H: MetS pigs had a higher mean glomerular injury score than lean pigs, which was lowered in MetS pigs treated with ELAM. I: percentage of uninjured glomeruli (score = 0) was lowest in MetS pigs, whereas the percentage of glomeruli with stage 1 injury was higher than lean pigs, and both were restored by ELAM. There were no differences in the groups for glomerular scores of 2 and 3 (n = 6 each). SC, subcutaneous; NS, not significant.

After 4 wk of treatment, animals were anesthetized and maintained with intravenous ketamine and xylazine (Telazol, Zoetis, and Ketaset, distributed by Zoetis, Kalamazoo, MI). Single kidney hemodynamics and function were measured using multidetector computed tomography (MDCT) and renal oxygenation by blood oxygen level-dependent magnetic resonance imaging (BOLD-MRI). Before MDCT experiments, urinary samples (50 ml) were collected from the bladder using a catheter and stored at −80°C for the measurement of urinary albumin and creatinine to calculate the albumin-to-creatinine ratio (ACR). To collect urinary EVs, fasting blood samples were collected to assess the levels of lipids and serum creatinine (SCr). For measurements of glucose and insulin, levels were collected for the calculation of the homeostasis model assessment of IR (HOMA-IR) (10). Analyze software (Biomedical Imaging Resource, Mayo Clinic, Rochester, MN) was used to analyze the cross-sectional MDCT images and to measure renal volume, renal blood flow (RBF), and glomerular filtration rate (GFR), as previously described (21). Mean arterial pressure (MAP) was monitored during MDCT experiments using an arterial catheter.

One week after the completion of all experiments, pigs were euthanized with a lethal intravenous dose of pentobarbital sodium (Fetalplus, Fort Dodge, Fort Dodge, IA) (11). The kidneys were immediately dissected, and sections were frozen in liquid nitrogen or preserved in formalin or Trump’s fixative for ex vivo experiments.

In addition, to assess the clinical relevance of pEVs, their numbers were measured in urine collected from obese and lean human patients.

Renal histopathology.

Paraffin sections were stained with hematoxylin and eosin, Masson’s trichrome, and periodic acid-Schiff (PAS) and examined by microscopy to determine glomerular size, glomerulosclerosis, and mesangial expansion, respectively, in ten ×400 glomeruli in each section. In each section stained with Masson’s trichrome, 10 glomeruli were ranked, describing the hallmarks of metabolic-specific glomerulosclerosis on a scale of 0−3, where 0 = no injury; 1 = mesangial thickening up to one-third of the tuft cross section and limited or partial Bowman capsule thickening; 2 = mesangial thickening >50% of the tuft, extensive loss of capillary loop structure and podocytes, and more extensive Bowman capsule thickening; and 3 = entirely sclerotic tufts with little or no podocytes or capillaries visible, the mesangium expanded beyond 75%, and Bowman capsule remnant. The total mean score was then calculated (36). Podocyte markers were determined by nephrin, podocalyxin, and Wilms’ tumor 1 (WT1) staining. Podocyte and mitochondrial morphology was examined by a digital electron microscopy, and all were quantified by Image-Pro Plus (Media Cybernetics, Rockville, MD). All image analyses were performed by an experienced nephrologist, in a blinded fashion, and all samples including controls were processed in the same way.

Immunohistochemistry and immunofluorescence.

Immunohistochemistry imaging used frozen sections (8 µm) and the following antibodies: nephrin, rabbit polyclonal 1:100 (orb157949, Biorbyt, San Francisco, CA); and podocalyxin, rabbit monoclonal 1:250 (anti-PODXL, ab150358, Abcam, Cambridge, MA). Immunofluorescence imaging used formalin-fixed, paraffin-embedded sections (5 µm) for WT1 [rabbit polyclonal 1:50 (MBS9203569, MyBioSource, San Diego, CA)]. Furthermore, double staining of WT1 and cardiolipin, an important component of the inner mitochondrial membrane ELAM restores, was performed to investigate podocyte mitochondrial metabolism. For nephrin and podocalyxin staining, ten ×400 glomerular images for each sample were quantified with positive staining threshold visually determined, and the average threshold was then applied to all other images automatically. In WT1-stained images, the number of nuclei of positive podocytes was counted from ten × 400 glomerular images for each sample, and the percentage was averaged.

Transmission electron microscopy.

Podocyte and mitochondrial morphology in glomeruli were assessed using transmission electron microscopy (TEM; Philips CM10, Philips, Amsterdam, The Netherlands) on samples fixed with Trump’s fixative solution. For analysis, three glomeruli in each pig were evaluated with four to six images of ×3K, ×10K, and ×50K each. ×3K images were used to analyze podocyte size (area/number), ×10K images were used for foot process width, slit pore number per 10 µm of glomerular basement membrane, and mitochondrial number in each podocyte, and ×50K images were used to measure the area and density of mitochondria in podocytes.

Podocyte area quantified with a visually determined podocyte threshold was determined in two to three glomeruli in each sample, and the number of podocyte nuclei was counted. Podocyte size was calculated by dividing the entire podocyte area by the number of podocyte nuclei in each glomerulus. Podocyte morphology analysis followed previously published methods (45). Straight basement membrane with uniform thickness and upright podocyte foot process was used to rule out bias sample cuts. One micrometer was calibrated using the image legend and randomly placed on a straight glomerular basement membrane to measure the number of slit pores. The arithmetic means of their widths and numbers were calculated. Mitochondrial structure was assessed by TEM in at least 100 mitochondria/sample. Mitochondrial area and density (number per unit area) were measured in 15–20 representative images in these podocytes, with only mitochondria fully contained within the borders of the images included. Mitochondria were counted per podocyte, and their density was calculated as number of mitochondria per 10 μm² podocyte area. Mitochondrial matrix density was calculated from mitochondrial intensity values (41). TEM images were analyzed using Image-Pro Plus.

EV isolation and analysis.

Podocyte injury was assessed by the concentrations of urinary EVs containing podocyte-specific proteins using digital flow cytometry (19). EVs were isolated from whole urine using Total Exosome Isolation reagent (no. 4484452, Invitrogen, Waltham, MA) according to the manufacturer’s guideline, as we have previously described (22). Isolated EVs (20 µl) were separated and stained with 0.5 µl Tag-it Violet (TIV) cell tracking dye (no. 425101, BioLegend, San Diego, CA) and subsequently with nephrin (4 µl, MyBioSource) and podocalyxin (6 µl, R&D Systems, Minneapolis, MN). EVs were quantified using a FlowSight (Amnis, Seattle, WA) imaging flow cytometer, as previously described (22), by acquiring at least 100,000 TIV-positive events identified as EVs. The flow-gating strategy included a positive gate for TIV events and either a double- or single-positive gate for both nephrin and podocalyxin. The concentration of single or double fluorescence-labeled urinary EVs was expressed as a count of pEVs per microliter of urine, using previously described methods (16).

Human urine EV isolation and analysis.

Urine was collected from 31 patients, of which 16 patients (body mass index >30) were considered obese and 15 patients (body mass index < 25) were lean controls. These patients were either healthy volunteers enrolled through the Mayo Biobank (4 lean and 4 obese patients) or patients with hypertension (11 lean and 12 obese patients). Patients with a previous history of stroke, coronary artery diseases, CKD, cancer, and diabetes were excluded. All urine samples were collected in the morning. Patients with hypertension were all treated with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers. The experiments were approved by the institutional review board, and informed consent was obtained.

Statistical analysis.

Results are expressed as means ± SE. Statistical analysis was performed using SPSS 16.0 statistical software (SPSS, Chicago, IL). Multiple-group comparisons were performed using one-way ANOVA followed by post hoc tests. Correlations were analyzed using Pearson or Spearman correlation coefficients according to data distribution. P values of <0.05 were considered statistically significant.

RESULTS

After 16 wk of diet, body weight, MAP, HOMA-IR, lipid fractions, renal volume, GFR, and RBF were significantly increased in pigs with MetS compared with lean pigs, consistent with MetS-induced hyperfiltration. ELAM decreased GFR per milliliter of tissue (but not absolute GFR) compared with untreated MetS and had no effect on other variables. Fasting glucose, SCr, and ACR did not differ among the groups (Table 1).

Table 1.

Systemic characteristics and single-kidney function in pigs after 16 wk of MetS or lean diet

Parameter	Lean	MetS	MetS + ELAM
Body weight, kg	73.8 ± 4.2	92.4 ± 0.8*	93.7 ± 4.4*
Mean arterial pressure, mmHg	104.8 ± 1.9	124.2 ± 3.9*	115.1 ± 4.3*
Fasting glucose, mg/dl	128.5 ± 14.9	121.0 ± 15.1	116.8 ± 13.5
Fasting insulin, µU/ml	0.4 ± 0.1	0.7 ± 0.1*	0.8 ± 0.1*
HOMA-IR score	0.7 ± 0.1	1.7 ± 0.4*	1.9 ± 0.1*
Total cholesterol, mg/dl	82.4 ± 6.8	423.3 ± 82.9*	447.1 ± 73.4*
HDL-cholesterol, mg/dl	47.6 ± 3.4	135.1 ± 32.3*	130.4 ± 32.5*
LDL-cholesterol, mg/dl	32.8 ± 6.7	416.8 ± 119.8*	404.5 ± 134.6*
Triglycerides, mg/dl	8.1 ± 1.3	18.3 ± 5.7*	18.5 ± 6.5*
Serum creatinine, mg/dl	1.57 ± 0.34	2.07 ± 0.57	1.66 ± 0.37
Albumin-to-creatinine ratio, mg/g	8.3 ± 7.2	6.4 ± 8.1	5.5 ± 1.4
Renal volume, ml	143.3 ± 6.2	226.1 ± 11.5*	241.3 ± 8.0*
Renal blood flow, ml/min	515.8 ± 38.3	881.9 ± 134.3*	834.5 ± 61.8*
GFR, ml/min	78.5 ± 7.5	145.1 ± 19.0*	132.9 ± 7.8*
GFR, ml·min−1·ml tissue−1	0.71 ± 0.04	0.85 ± 0.06*	0.63 ± 0.08†

Values are means ± SD; n = 6 kidneys/group. MetS, metabolic syndrome; ELAM, elamipretide; HOMA-IR, homeostasis model assessment of insulin resistance; GFR, glomerular filtration rate.

^*P ≤ 0.05 vs. the lean group;

^†P ≤ 0.05 vs. the MetS group.

ELAM prevented MetS-related early glomerulopathy.

Glomerular size significantly increased in pigs with MetS compared with lean pigs and was similar to that in MetS + ELAM-treated pigs (Fig. 1, B and F). PAS staining revealed in MetS capillary collapse, mesangial expansion, and increased matrix, which ELAM attenuated (Fig. 1, C and G). Trichrome staining showed glomerulosclerosis and increased glomerular injury score in MetS, which was again reduced by a 4-wk course of ELAM (Fig. 1, D, H, and I). Mesangial expansion in MetS was confirmed by TEM (Fig. 1E). Therefore, ELAM treatment abrogated glomerulosclerosis despite unchanged body weight, MAP, or glomerular size.

ELAM attenuated MetS-related podocyte injury.

Expression of nephrin, which is specific to mature podocytes, was decreased in MetS compared with lean pigs but increased in MetS + ELAM-treated pigs compared with MetS pigs, although it was not fully normalized (Fig. 2, A and C). Expression of podocalyxin followed a similar pattern (Fig. 2, B and D). Absolute podocyte number, as determined by nuclear staining of WT1, was not different among the groups, but their percentage out of all glomerular cells was lower than normal in MetS but not with ELAM treatment (Fig. 2, E and F). Representative high-magnification images of podocyte staining with WT1 and nephrin (yellow arrows) are shown in Fig. 7.

Fig. 2.

Elamipretide (ELAM) improves the reduction of expression of podocyte markers in metabolic syndrome (MetS) pigs. A and B: representative images showing the reduction of nephrin and podocalyxin expression in MetS pigs compared with lean and MetS + ELAM-treated pigs. Quantification confirmed that nephrin (C) and podocalyxin (D) expression was significantly decreased in MetS compared with lean glomeruli but significantly increased in MetS + ELAM-treated glomeruli compared with MetS glomeruli. Quantification of nuclei staining for Wilms’ tumor 1 (WT1) showed unchanged absolute podocyte number (E), whereas the percentage of podocytes out of glomerular cells was lower than normal in MetS but not with ELAM treatment (F). Data were averaged from 10 glomeruli/sample (n = 6 in each group). NS, not significant.

Fig. 7.

a–c: Urinary concentrations of podocyte-derived extracellular vesicles (pEVs) were elevated in metabolic syndrome (MetS) pigs and decreased after elamipretide (ELAM) treatment. A–F: correlations of urinary pEV concentration with clinical and renal parameters of pigs. a: Representative flow cytometry scatterplots of podocalyxin/nephrin double-positive EVs (yellow) in lean, MetS, and MetS + ELAM-treated pigs. Urinary concentrations of pEVs were higher in MetS compared with lean or MetS + ELAM-treated pigs (b). Urinary concentrations of total EVs in MetS were not significantly higher than in the lean groups but were higher than in the MetS + ELAM-treated group (c). pEV fractions correlated directly with serum creatinine (SCr; A), glomerular filtration rate (GFR; B), glomerular injury score (C), and periodic acid-Schiff (PAS)-positive matrix area (D) and inversely correlated with nephrin intensity (E). NS, not significant.

TEM showed in MetS podocytes with irregular shapes, flattened foot processes, and some areas of effacement, which were prevented by ELAM (Fig. 3A). Podocyte foot process base width in MetS pigs was greater than in lean pigs, and there were fewer slit pores, both of which were improved by ELAM (Fig. 3, B and C). In addition, podocytes were smaller in MetS, and their size was restored after ELAM treatment (Fig. 3D). Hence, MetS induced early podocyte injury, which mitochondrial protection attenuated.

Fig. 3.

Elamipretide (ELAM) protects podocyte size and foot processes in metabolic syndrome (MetS) pigs. A: representative transmission electron microscopic images of podocyte foot processes. Foot processes were flattened in MetS pigs (white arrows; A), and their base width was greater than in lean pigs but decreased after ELAM treatment (B). The number of slit pores per 10 µm was lower in MetS than lean animals and improved by ELAM (C). Podocyte size was reduced in MetS, a difference that was eliminated with ELAM (D). NS, not significant.

ELAM protected podocyte mitochondria from structural damage in MetS.

TEM showed that podocyte mitochondria were smaller in MetS than lean pigs (Fig. 4, A and B). Furthermore, the number, spatial density, and matrix density of podocyte mitochondria were significantly lower in MetS pigs but preserved in MetS + ELAM-treated pigs (Fig. 4, A and C–E). Therefore, MetS damaged podocyte mitochondria, which ELAM improved. Furthermore, WT1/cardiolipin costaining confirmed their colocalization and that cardiolipin-positive cells were also deceased in MetS and improved with MetS + ELAM (Fig. 5A). Representative high-magnification images of podocyte staining with WT1 and nephrin (yellow arrows) are shown in Fig. 6.

Fig. 4.

Elamipretide (ELAM) protects podocyte mitochondria in metabolic syndrome (MetS) pigs. A: representative transmission electron microscopic images of podocyte mitochondria (white arrows) in lean, MetS, and MetS + ELAM-treated pigs. Mitochondria were smaller in the MetS group and similar in size to those in the MetS + ELAM-treated group (B). The number (C), spatial density (D), and matrix density (E) of mitochondria in podocytes were significantly lower in MetS pigs than in lean pigs and preserved in MetS + ELAM-treated pigs. NS, not significant.

Fig. 5.

A: double staining showed that cardiolipin colocalized with Wilms’ tumor 1 (WT1) and that cardiolipin-positive cells were also deceased in metabolic syndrome (MetS) and improved with MetS + elamipretide (ELAM) treatment. B: nephrin-positive/podocalyxin-positive extracellular vesicles (EVs) were significantly higher in obese patients (n = 16) than in lean patients (n = 15), supporting the pathophysiology of obesity-induced renal abnormalities. *P < 0.05 vs. the lean group. BMI, body mass index.

Fig. 6.

Representative high-magnification images of podocyte staining with Wilms’ tumor 1 (WT1; green), cardiolipin (red), and nephrin (brown). The very bright spots are red blood cells; positively stained podocytes are marked with yellow arrows.

Urinary concentrations of pEVs were elevated in MetS but eliminated by ELAM.

Urinary concentrations (count/µl urine) of podocalyxin and nephrin double-positive pEVs were elevated in MetS pigs compared with lean pigs and declined in MetS + ELAM-treated pigs compared with MetS pigs (Fig. 7, a and b). Furthermore, the concentrations of all urinary EVs in MetS were not different from in Lean but were higher than MetS + ELAM (Fig. 7c).

Correlations of urinary pEV concentration with clinical and renal parameters.

Among all pigs, urinary pEV concentrations correlated directly with SCr, GFR, glomerular injury score, and PAS-positive matrix area and correlated inversely with expression of nephrin but not with ACR, MAP, or body weight. PEV concentration also strongly tended to inversely correlate with expression of podocalyxin (P = 0.057; Fig. 7, A–F).

pEVs were significantly higher in obese patients.

Nephrin-positive, podocalyxin-positive EVs were significantly higher in obese patients than in lean patients, supporting the clinical relevance of pEV in obese human patients (Fig. 5B).

DISCUSSION

Our study demonstrates the development of injury to podocytes and their mitochondria at the early stage of MetS-related kidney disease. Protection of podocyte mitochondria reduced podocyte injury, suggesting that mitochondria contribute to podocyte damage in MetS. Moreover, urinary pEVs, which reflect podocyte injury, may represent an early marker of MetS-related kidney disease and a novel therapeutic target.

MetS is a global epidemic and a major public health problem and represents a cluster of risk factors for CKD, including IR, obesity, and hypertension. The increased risk for CKD is reflected by the development of microalbuminuria and renal dysfunction in advanced stages of MetS. We have also previously demonstrated that early MetS (without vascular lesions or cardiac dysfunction) induced early structural and functional changes in tubular segments in pigs (13), without loss of renal function. However, whether podocyte injury develops in the early stage of MetS, before albuminuria or renal dysfunction, remains unclear.

In the present study, we used an early MetS swine model that recapitulates common features of human MetS (28) as well as sophisticated techniques that permit studying single kidney function and structure. After 16 wk of diet, our MetS pigs developed obesity, dyslipidemia, hypertension, and IR, indicating successful development of MetS, but not diabetes or albuminuria. We found that early renal dysfunction in pigs with MetS was characterized by hyperfiltration, evidenced by increased renal volume, RBF, and GFR.

The podocyte is the most differentiated cell type within the glomerular complex and an integral component of the glomerular basement membrane and slit pore diaphragm, essential for maintaining the glomerular filtration barrier (27). Podocyte injury destabilizes the foot process/slit pore diaphragm complex with resultant loss of glomerular permselectivity and, in turn, albuminuria and glomerulosclerosis (31). Glomerular enlargement with mechanical stretch and activation of the renin-angiotensin system is considered to be an important contributor in podocyte injury (26). Metabolic disorders such as lipid dysregulation are also involved in the aggravation of podocyte injury (6), and obesity-related glomerulopathy may decrease podocyte density and increase foot process width (8, 9). In our MetS swine model, we found glomerular hypertrophy, mesangial matrix expansion, and mild glomerulosclerosis. These were accompanied by decreased expression of nephrin, podocalyxin, and WT1, indicating injury and a decreased relative number of podocytes in MetS. The percent mesangial area in MetS glomeruli was elevated, suggesting that mesangial area expansion in MetS exceeded glomerular enlargement. The glomerular ultrastructure revealed decreased podocyte size, increased foot process width, and decreased slit pore numbers in MetS. Because podocytes are terminally differentiated cells, the increase in mesangial area and decrease in podocyte size can force the podocytes to extend their foot processes to maintain the covered area, predisposing them to progressive injury. Ultimately, these changes may result in deformation and flattening of the foot process attributable to the breakdown of their actin cytoskeleton, as ATP is required for actin assembly (1) and may eventuate in glomerulosclerosis.

High-fat diets increase mitochondrial ROS (33), and cardiolipin is particularly susceptible to lipid peroxidation. In the presence of ROS, cytochrome c peroxidase causes cardiolipin peroxidation and cristae degradation (4). Mitochondrial health is important for podocyte function, actin assembly (1), and protection from apoptosis (23) as well as for expression and trafficking of nephrin and WT1 (24, 32). In our study, morphological analysis indicated structural damage to podocyte mitochondria in MetS, and cardiolipin colocalized well with WT1 staining, implicating mitochondrial injury in the podocyte pathology associated with MetS.

ELAM is a tetrapeptide that protects cardiolipin from oxidation and inhibits cytochrome c peroxidase activity (3, 31, 48). Our previous study (13) demonstrated that protection of mitochondria with ELAM decreased medullary tubular injury in pigs with MetS, but its impact on glomerular podocyte lesions has not been fully elucidated. In mice fed a high-fat diet (37), representative TEM images suggested that ELAM protects mitochondrial morphology in renal cells. Our present study, focusing on podocytes in a large preclinical animal model, showed that ELAM induces a quantitative attenuation of MetS-related podocyte injury by protecting their mitochondria. This was reflected in reduced loss of podocyte-specific proteins, foot process abnormalities, podocyte shrinkage, and mesangial expansion. We also correlated these changes with renal function, although the phase of MetS was too early to induce albuminuria. Notably, ELAM had no effect on weight gain, IR, hypertension, or hyperlipidemia caused by MetS, suggesting that it directly protects the kidney from metabolic stress. ELAM is the first experimental drug shown to prevent early kidney injury in MetS by specifically targeting and protecting renal mitochondria in animal models. Our data indicate that it may represent a novel approach to the treatment of metabolic kidney disease beyond blood pressure and blood lipid control.

Our study also demonstrated that podocyte injury may be disclosed by changes in urinary levels of pEV. Urine can be easily obtained, constituting an ideal fluid for biomarker analysis in the renal system (29). Urinary EVs originate from different cell types along the nephron facing the urinary space and carry markers and cargo (protein, lipids, and microRNA) representative of their parent cells (34). A recent study (25) demonstrated that podocyte-derived microparticles (100- to 1,000-nm vesicles) increase profibrotic proteins in human proximal epithelial cells, suggesting that pEV may also have biological activity. Other investigators (30) performed proteomic analysis on podocyte EVs, although the role of these proteins in normal and disease conditions remains to be determined. Analysis of urinary EVs may therefore provide a noninvasive window into the site-specific physiological and pathological state of the kidney. In glomerular diseases, podocytes are frequent casualties, of which urinary pEVs may be a promising index. Podocalyxin is a good marker of urinary podocytes but can also be expressed on nonrenal or other glomerular cells. On the other hand, nephrin is podocyte specific but may be downregulated in proteinuria nephropathy (15). Hence, in the present study, we used strict criteria of double positivity of podocalyxin and nephrin in flow cytometry analysis of size-specific (<1,000 nm) pEVs.

Early detection of podocyte injury can lead to early intervention to prevent progression of kidney disease (43). Our study demonstrated that the urinary concentration of pEVs was elevated in MetS compared with lean groups in both pigs and human patients, suggesting podocyte injury. The increased pEVs in human patients are like attributable to obesity because the two groups of patients had similar blood pressure, supporting the clinical relevance of our pig observations and highlighting the pathophysiology of obesity-induced renal abnormalities. Importantly, treatment with ELAM relieved the injury and decreased pEV levels, in line with the reduced podocyte injury observed using TEM. The correlations of pEV levels with renal function, mesangial expansion, and especially expression of podocyte markers in the kidney again support the pEV as a marker of early podocyte damage in MetS.

Our study has certain limitations. The study groups were relatively small, and larger studies are needed to confirm our results. The present study could not link ELAM treatment and pEV levels to the progression of MetS-related kidney disease, which would require future longitudinal studies. Assessments of urinary pEVs with other or more severe pathologies are also needed. Studies in human patients compared obese with lean individuals, but all features of MetS could not be confirmed. Finally, other cell-derived EVs may also be shed in the urine in MetS. The nephrin/podocalyxin-positive staining in flow analysis represents mostly pEVs in urine, but we cannot rule out that it also captures some membrane debris within the same size range. Porcine-specific antibodies for podocyte markers are very limited in availability. Clearly, further studies are needed to identify the cellular source of all urinary EVs in MetS.

In conclusion, this study showed that early MetS induces mitochondrial abnormalities in podocytes, which is associated with glomerular hyperfiltration and podocyte injury. Chronic protection of mitochondria can restore the mitochondrial structure of podocytes and reduce podocyte injury, suggesting the mitochondria as therapeutic targets. Larger longitudinal studies are needed to determine whether an early intervention with ELAM can prevent more advanced MetS-related kidney damage. Furthermore, we found that urinary pEVs reflect podocyte injury in MetS, which may represent an early marker of renal injury and a novel therapeutic target. Increased pEVs were also observed in a group of obese human patients. These observations may have important implications for the management of patients with MetS.

GRANTS

This work was supported by a research grant from Stealth Biotherapeutics and by National Institutes of Health Grants DK-120242, DK-100081, DK-106427, DK-104273, HL-123160, and DK-102325.

DISCLOSURES

L. O. Lerman and S.C.T. hold a joint patent for the use of ELAM in renovascular disease. None of the other authors have any conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

L.-H.Z., A.A.N., A.S., I.O.S., and S.C.T. performed experiments; L.-H.Z., A.E., A.A.N., I.O.S., and J.R.W. analyzed data; L.-H.Z., X.Y.Z., S.C.T., and L.O.L. interpreted results of experiments; L.-H.Z., J.R.W., and X.Y.Z. prepared figures; L.-H.Z. and X.Y.Z. drafted manuscript; L.-H.Z. and L.O.L. edited and revised manuscript; L.-H.Z., X.Y.Z., and L.O.L. conceived and designed research, and approved final version of manuscript.