Circulating extracellular vesicles of patients with steroid-sensitive nephrotic syndrome have higher RAC1 and induce recapitulation of nephrotic syndrome phenotype in podocytes

Abstract

Since previous research suggests a role of a circulating factor in the pathogenesis of steroid-sensitive nephrotic syndrome (NS), we speculated that circulating plasma extracellular vesicles (EVs) are a candidate source of such a soluble mediator. Here, we aimed to characterize and try to delineate the effects of these EVs in vitro. Plasma EVs from 20 children with steroid-sensitive NS in relapse and remission, 10 healthy controls, and 6 disease controls were obtained by serial ultracentrifugation. Characterization of these EVs was performed by electron microscopy, flow cytometry, and Western blot analysis. Major proteins from plasma EVs were identified via mass spectrometry. Gene Ontology classification analysis and Ingenuity Pathway Analysis were performed on selectively expressed EV proteins during relapse. Immortalized human podocyte culture was used to detect the effects of EVs on podocytes. The protein content and particle number of plasma EVs were significantly increased during NS relapse. Relapse NS EVs selectively expressed proteins that involved actin cytoskeleton rearrangement. Among these, the level of RAC-GTP was significantly increased in relapse EVs compared with remission and disease control EVs. Relapse EVs were efficiently internalized by podocytes and induced significantly enhanced motility and albumin permeability. Moreover, relapse EVs induced significantly higher levels of RAC-GTP and phospho-p38 and decreased the levels of synaptopodin in podocytes. Circulating relapse EVs are biologically active molecules that carry active RAC1 as cargo and induce recapitulation of the NS phenotype in podocytes in vitro.

NEW & NOTEWORTHY Up to now, the role of extracellular vesicles (EVs) in the pathogenesis of steroid-sensitive nephrotic syndrome (NS) has not been studied. Here, we found that relapse NS EVs contain significantly increased active RAC1, induce enhanced podocyte motility, and increase expression of RAC-GTP and phospho-p38 expression in vitro. These results suggest that plasma EVs are biologically active molecules in the pathogenesis of NS.

INTRODUCTION

Nephrotic syndrome (NS) is one of the most common causes of glomerular disease in children and is characterized by the triad of proteinuria, hypoalbuminemia, and edema. The major molecular event in the pathogenesis of NS is the disruption of the glomerular filtration barrier, which is primarily driven by podocyte injury (1). The most common clinical presentation of NS in children is steroid-sensitive NS (SSNS), characterized by complete remission within 4 wk of steroid therapy and no apparent glomerular change in the light microscopic evaluation of kidney biopsies, thereby named as minimal change disease (MCD) (2). The observation of diffuse foot process effacement without immune complex deposition by electron microscopy is the only morphological pattern of podocyte injury in MCD. SSNS has been suggested to be caused by a circulating plasma factor that causes reversible foot process effacement (3). The amelioration of proteinuria by glucocorticoids and other immunosuppressive drugs and the induction of proteinuria by plasma from patients with NS and supernatants from T cell hybridoma lines derived from patients with MCD suggest that this plasma factor is secreted by immune cells (4, 5). Certain cytokines and serum permeability factors have been implicated in MCD; however, to date, none of the proposed factors have caused reproducible in vitro and in vivo pathological effects (3). Recently, the antiproteinuric effects of immunosuppressive drugs, including glucocorticoids, cyclosporine, and rituximab, have been attributed to their direct modulation of the podocyte actin cytoskeleton and preservation of synaptopodin degradation, which has placed podocyte dysfunction, possibly triggered by an immunological insult, at the center of NS pathogenesis (3, 6).

In addition to classical signaling pathways, extracellular vesicles (EVs) have the potential to deliver a particularly diverse array of signals to cells at a level beyond that of soluble factor signaling (7). EVs are characterized as particles of submicron sizes that are delimited by a lipid bilayer that envelopes a diverse array of proteins, nucleic acids, bioactive molecules, and cargo contents that represent the status of the cell of origin. EVs are enriched in not only membrane proteins, such as adhesion molecules and membrane trafficking molecules, but also cytoskeleton molecules, heat shock proteins, cytoplasmic enzymes, signal transduction molecules, cytokines, chemokines, proteinases, and cell-specific antigens (8). These functional contents vary according to the cell type and reflect the particular physiological and pathological conditions of the producer cells at the time of EV packaging and secretion. The nomenclature of EVs has not yet been fully described (9). A number of different EV subpopulations have been described, and their classification is usually dependent on their size and specific mechanism of biogenesis. The most well-studied EVs are exosomes, which have sizes of <100 nm, are endosomal in origin, and are generated by the reverse budding of multivesicular bodies within cells before their secretion. Other notable EVs include microvesicles (100–1,000 nm), which are directly shed from the plasma membranes of cells, and apoptotic bodies (1–2 μm) (9). EVs interact with recipient cells, which may be in close proximity to or a significant distance from the cell of origin. Recent advances have shown that EV signaling can profoundly modulate infections, host immune responses, and various diseases, such as cancer (10).

As previous research suggests a role of a circulating factor in the development of SSNS, we speculated that EVs are a candidate source of such a soluble mediator. In this report, we aimed to characterize the circulating EVs of patients with SSNS and try to delineate the effects of these EVs on podocytes in vitro.

MATERIALS AND METHODS

Participants

This study was approved by the local ethics committee of the Dr. Sami Ulus Children’s Hospital and Human Ethics Committee of Bilkent University. The patient group consisted of 20 children with SSNS. Blood for the isolation of relapsed NS EVs was taken at the time of the patient’s first attack, and EVs were used in experiments after the patients experienced a complete response to steroids within 4 wk of the treatment. As shown in Table 1, on follow-up of mean 3.8 ± 0.8 yr, five of our patients with SSNS were diagnosed as frequent relapsers and 15 of them were diagnosed as infrequent relapsers. None of our patients developed secondary steroid resistance or had a high suspicion of different underlying pathology during the follow-up period. Remission NS EVs were isolated from the same patients when they were nonproteinuric with normal albumin levels, after the cessation of steroid treatment, or once they were prescribed low-dose steroids (<0.2 mg/kg/day). As disease controls, samples from three patients with membranous nephropathy and three patients with Henoch-Schönlein purpura nephritis confirmed by biopsy were used. Blood from these patients was taken at their initial admission to Dr. Sami Ulus Children’s Hospital, and these patients were not on immunosuppressive treatment at the time of blood withdrawal. All healthy controls (n = 10) had negative protein with urinary dipstick and normal urine microscopy. Fasting blood from patients and healthy donors was collected in EDTA-coated tubes after informed consent had been obtained. Clinical and demographic characteristics of patients and controls are shown in Table 1.

Table 1.

Clinical and demographic characteristics of patients and controls

	Healthy Controls	Patients With SSNS	Disease Control Patients
n	10	20	6
Diagnosis		SSNS [frequent relapser (n = 5) and infrequent relapser (n = 15)]	Membranous [nephropathy (n = 3) and Henoch-Schönlein purpura nephritis (n = 3)]
Mean follow-up time, yr		3.8 ± 0.8	1.7 ± 0.4
Sex (male/female)	5/5*	11/9*	3/3
Age (means ± SD)	6.8 ± 3.2†	5.9 ± 1.8†	8.8 ± 4.7
Laboratory findings at relapse
Albumin (means ± SD)	NA	1.62 ± 0.27	1.71 ± 0.25
Spot protein/creatinine (means ± SD), mg/mg creatinine	NA	6.97 ± 2.89	8.31 ± 2.16

NA, not applicable; SSNS, steroid-sensitive nephrotic syndrome.

^*The statistical significance of the sex ratio was insignificant between healthy controls and patients with steroid-sensitive nephrotic syndrome (SSNS) [P = 0.860 Fischer’s exact test].

^†The statistical significance of the mean values of ages was insignificant between healthy controls and patients with SSNS [P = 0.335 (t test)].

Isolation of Plasma and Plasma EVs From Human Blood

Whole blood was mixed with an equal volume of PBS (Biological Industries, Beit-HaEmek, Israel) and slowly layered onto the lymphocyte separation medium with a 3:2 (blood-to-medium) ratio and centrifuged at 540 g for 30 min. The upper plasma fraction and cloudy interface containing peripheral blood mononuclear cells (PBMCs) were separated. The plasma fraction was centrifuged at 2,000 g to get rid of platelet and dead cells. Supernatants were collected to new tubes, snap frozen in liquid nitrogen, and kept at −80°C for further use. At the time of ultracentrifugation, frozen tubes were thawed very slowly overnight at +4°C to prevent EV and/or exosome rupture before isolation and vigorously vortexed before proceeding to ultracentrifugation. If needed, to adjust the volume to a fixed amount before centrifugation, some plasma samples were supplemented with cold PBS to 14 mL and centrifuged at 10,000 g for 30 min at 4°C. Supernatants were transferred to another ultracentrifuge tube and centrifuged at 100,000 g for 90 min at 4°C. At the end of this step, EVs were sedimented. The supernatants were accepted as plasma-depleted EVs and used in further experiments. To get rid of contaminating proteins, EVs were further washed with 14 mL of PBS and recentrifuged at 100,000 g for 90 min at 4°C. The pellets containing EVs were dissolved with PBS.

Determination of EV Protein Content and Number

Protein content of EVs was determined with a Pierce BCA Protein Assay Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. Briefly, 25 μL/well BSA standard dilutions and 5 μL/well samples were placed in 96-well plates and mixed with 200 μL of the working reagent. The plate was incubated in the dark at 37°C for 30 min and then cooled down at room temperature for 15 min. Absorbance at 562 nm was measured with a Synergy HT microplate reader (BioTek, Winooski, VT). The particle number of 1 μg/µL EVs was measured by a tunable pulse resistive index system (QNano, Izon Biosciences). Protein amount and particle concentration of EVs are presented as normalized to plasma volume (mL) and plasma albumin concentration (g/dL).

Transmission Electron Microscopy

The morphology and size of EVs were evaluated by transmission electron microscopy (TEM). EV suspensions (5 μL) were dropped onto formvar/carbon-coated nickel mesh grids and incubated for 20 min. Excess suspension on the nickel mesh grids was then blotted with filter paper, and nickel mesh grids were negatively stained with 2.0% phosphotungstic acid and 2.0% uranyl acetate, respectively. After being washed, samples were air-dried for 15 min and visualized using a digital camera (Orius) connected to transmission electron microscope (JEM1400, Jeol, Tokyo, Japan).

Analysis of EV Surface Markers by Flow Cytometry

A bead-based detection method was used for surface protein characterization of EVs with flow cytometry. Latex beads were coated with purified anti-human CD63 or CD81 antibody to capture EVs. Carboxyl latex beads (Thermo Fisher Scientific) were mixed with anti-CD63 (clone H5C6, BioLegend, San Diego, CA) or anti-CD81 (clone 5A6, BioLegend) at a 1 μL:1 μg (bead-to-antibody) ratio to capture CD63- or CD81-positive EVs. The volume was completed to 50 μL with PBS, and the mixture was incubated for 30 min at room temperature. The volume was then increased to 500 μL with PBS, and the mixture was incubated on a rotator at a low speed overnight. The bead-antibody mixture was precipitated at 10,000 g for 10 min, blocked with 5% BSA (Capricorn Scientific, Ebsdorfergrund, Germany) for 4 h at room temperature, precipitated again at 10,000 g for 10 min, resuspended in PBS containing 1% BSA, and stored at 4°C. For each staining of EVs, 1 μg of EVs was mixed with 1 μL of the final coated bead solution. The volume was increased with PBS up to 50 μL, and the mixture was incubated at room temperature for 30 min. The volume was then increased to 500 μL, and the mixture was slowly rotated overnight to let the EVs and beads conjugate. After overnight conjugation, EV-bead conjugates were incubated with fluorochrome antibodies for human CD9 (clone HI9a, Biolegend), CD63, CD81 (Biolegend), and PE-podocalyxin (clone 3D3, Santa Cruz Biotechnology, Dallas, TX) at 1:50 dilution and their appropriate isotype controls at a concentration of 1 μg/mL in 100 μL volume for 1 h at room temperature in the dark. After 1 h of incubation, samples were washed with PBS, centrifuged at 10,000 g for 5 min, resuspended in 100 μL PBS, and then analyzed with a NovoCyte flow cytometer (ACEA Biosciences). For the quantification of cell surface-specific markers of plasma EV samples, EV-anti-CD81-coated bead conjugates were stained with an antibody (1:100 dilution) against one of the following cell surface markers: PE-CD42a (platelets, clone ALMA 16, BD PharMingen, Franklin Lakes, NJ), FITC-CD105 (endothelial cells, clone 43A3, Biolegend), BV421-CD3 (T cells, clone UCHT1, BD PharMingen), APC/Cy7-CD14 (monocytes, clone HCD14, Biolegend), and PE/Cy7-CD19 (clone SJ25C1, BD PharMingen) with the same method. The particle count after a very stringent FSC-SSC gating strategy was divided by the total bead-EV conjugate counts to calculate the percentage of antibody-positive beads. Every EV sample was stained in triplicate, and the mean percentage of each sample was used in the analysis.

Sample Preparation for Proteomics Experiments

Fifty micrograms of protein from each sample were transferred to microtubes. The proteins were reduced with dithiothreitol (DTT) to adjust to the 5 mM final concentration at room temperature and alkylated with iodoacetamide (IAA) to set the 50 mM final concentration at room temperature for 1 h in dark. Additional DTT was added to set the 10 mM final concentration. Proteins were precipitated using the methanol/chloroform precipitation protocol and dissolved in solution with 8 M urea and 50 mM Tris buffer at pH 8.5. The urea concentration was then diluted to 1 M with 50 mM Tris buffer at pH 8.5. Trypsin was dissolved in 50 mM Tris buffer at pH 8.5 and then added to the protein solution at a ratio of 1:100 (wt/wt, trypsin/protein). Finally, incubation was performed overnight. Trifluoroacetic acid was added to the incubated solution to stop the enzymatic digestion, setting the final concentration of trifluoroacetic concentration to 0.5% (vol/vol). Samples were cleaned using Sep-Pak (Waters) according to the manufacturer’s protocol. Samples were dried using a Speed-Vac centrifugal evaporator and then dissolved in MilliQ water (18. 2 MΩ⋅cm) to make the concentration of protein 0.5 µg/µL.

Mass Spectrometry

Peptides were run on a 15-cm column (EASY-Spray column, 15-cm length, 75-µm internal diameter, PepMap C18, 3-µm particles, 100-Å pore size) connected to an Ultimate 3000 RSLnano system (Dionex, Thermo Scientific) in a Q Exactive Plus mass spectrometer. Samples were loaded onto the column with buffer A (0.1% formic acid in water) and eluted with a 260-min gradient time with 5% to 95% buffer B (95% acetontrile, 0.1% formic acid, and 5% H₂O) at a flow rate of 250 nL/min. Mass spectra were acquired with an Orbitrap Q Exactive Plus mass spectrometer in the data-dependent mode, and the scan range was from 200 to 3,000 m/z for 4 h. Peptide fragmentation was performed via higher energy collision dissociation with the energy set at 29 normalized collision energy for peptide fragmentation. MS/MS spectra were acquired at a resolution of 17,500 for every sample. The analysis was run only once.

Proteomics Analysis

The raw data from mass spectrometry were run in Proteome Discoverer version 2.2 for peptide and protein identification. Data analysis was performed using Microsoft Excel. Gene symbols of the significant proteins identified were uploaded to the Ingenuity Pathway Analysis (IPA v10.2020) server for in-depth knowledge analysis using the “Core Analysis” function (Fisher’s exact test P value: 1 e⁻³). The Venn diagram was generated using the “venn()” function in R package gplots v3.0.1.1. The gene list enrichment analysis platform, EnrichR v01.07.2020, (https://amp.pharm.mssm.edu/Enrichr/), was used with the following library in this study: Gene Ontology (GO) Cellular Compartments. Information for the top 100 EV proteins from the database Vesiclepedia (http://www.microvesicles.org/extracellular_vesicle_markers) was accessed in July 2020 (11). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD024217 (12).

Human Podocyte Cell Culture

Human conditionally immortalized human podocytes (LY8H3) were kindly provided by Prof. Moin Saleem (Bristol, UK). For proliferation, they were cultivated at the permissive temperature of 33°C in podocyte medium containing RPMI-1640 supplemented with insulin transferrin selenium, penicillin-streptomycin, normocin, and 10% FBS. For differentiation, they were grown on rat tail collagen type I (100 µq/mL, Gibco, Thermo Fischer Scientific)-coated wells at 37°C for 10–14 days before being used experimentally. On some experiments with EVs, EV-free podocyte medium was used, which was prepared by FBS centrifuged at 100,000 g for 4 h. EVs from the podocyte culture supernatant were isolated with the protocol discussed above in Isolation of Plasma and Plasma EVs From Human Blood. Podocytes were grown on collagen-coated T75 plates and stimulated with IL-13 (40 µg/mL) or LPS (25 µg/mL) or with podocyte medium as a control for 24 h.

Determination of EV Binding and Uptake of Human Immortalized Podocytes

The EV surface was stained with the lipophilic dye SP-DiOC18 (Invitrogen, Carlsbad, CA), which can be analyzed on the FITC channel on flow cytometry. SP-DiOC was mixed with EVs at a ratio of 1 μg:10 μg (SP-DiOC to EV). The mixture was incubated for 60 min at 37°C in the dark. The solution was diluted to 14 mL with PBS and ultracentrifuged for 90 min at 100,000 RCF to remove unbound dye. The EV pellet was dissolved in PBS, and the volume was adjusted according to further applications. In some experiments, we attempted to further purify EVs following ultracentrifugation and applied to size exclusion columns (IZON qEV column, 70 nm). Human immortalized podocytes were seeded to 6-well plates. After 14 days of differentiation, cells were stimulated with 20 μg/mL SP-DiOC-stained NS and healthy EVs for 2, 4, 8, 16, or 24 h. After incubations were done, cells were scraped, washed with PBS, centrifuged at 300 RCF for 5 min, resuspended in 500 μL PBS, and then analyzed with a NovoCyte flow cytometer. In this reading, the first FITC-positive cells were accepted as bound and uptake (internalized) EVs. To determine the internalized EV fraction by the cells, the remaining cells were mixed with 30 µL trypan blue (0.1%) to quench membrane-bound not internalized EVs from the podocyte surface, and the FITC gate was measured again from the same tube. This value was recorded as internalized EV levels by podocytes. Also, fluorescence signals of SP-DiOC-stained EVs in podocytes were visualized with the EVOS FL Auto cell imaging system (Thermo Fisher Scientific).

PKH26 Labeling of EVs and Phalloidin Staining of Podocytes

Plasma EVs were labeled with the red lipophilic fluorescent dye PKH26 (Sigma-Aldrich, St. Louis, MO) according to the instructions. The suspension, containing 125 µg (125 µL) EVs, was transferred to a conical-bottom polypropylene tube. A 2× EV suspension was prepared by adding 125 µL of diluent C to the EV suspension, and a 2× dye solution was prepared by adding 0.5 μL of the PKH26 ethanolic dye solution to 250 µL of diluent C (labeling vehicle provided in the kit). Then, 250 mL of the 2× EV suspension was quickly added to 250 µL of the 2× dye solution, and the EVs/dye suspension was incubated for ∼1–5 min. Then, 500 µL of 1% BSA were added to bind excess dye. EVs were centrifuged for 90 min at 100,000 g, and pellets were suspended in podocyte medium. In some experiments, we attempted to purify EVs following ultracentrifugation and applied to size exclusion columns (qEVoriginal/70-nm SEC Columns, Izon Biosciences). Podocytes were incubated with 20 µg of PKH26-labeled EVs for 24 h. Podocytes were fixed with 4% paraformaldehyde for 15 min, washed with PBS, and then permeabilized with 0.3% Triton X-100 for 10 min. Wells were blocked with 5% BSA for 30 min at room temperature. Subsequently, podocytes were stained with 2 drops/mL fluorescent-conjugated phalloidin (ActinGreen 488 ReadyProbe Reagent, Thermo Fisher) and DAPI (NucBlue Fixed Cell ReadyProbes Reagent, Thermo Fisher) at 2 drops/mL in PBS for 15 min at room temperature. Wells were washed several times in PBS to remove unbound dye and observed using the EVOS FL Auto cell imaging system (Thermo Fisher Scientific).

Albumin Influx Assay

The albumin influx assay was performed to examine the filtration barrier function of podocyte monolayers as previously described (13). Transwell chambers (6.5 mm, 24 chambers) with a 0.4-μm pore size (Costar) were used in the albumin influx assay. Podocytes (1 × 10⁵ podocytes) were seeded on the collagen-coated upper permeable supports and cultured under differentiating conditions for 10 days. Cells were washed twice with PBS supplemented with 1 mM MgCl₂ and 1 mM CaCl₂ to preserve cadherin-based junctions. Podocytes were incubated with patient and healthy EVs (20 µg/mL) for 4 h. Cells were rewashed, the upper chamber was then filled with 0.1 mL serum-free RPMI-1640, and the lower chamber was then filled with 0.5 mL serum-free RPMI-1640 supplemented with 40 mg/mL BSA (Capricorn Scientific). The total protein concentration in the upper chamber was examined using the BCA method at 0, 2, and 4 h.

Scratch Assay

Podocytes were cultured on collagen-coated (100 µg/mL) 12-well plates and differentiated for 14 days at 37°C. The culture media were aspirated, and two scratches diagonally were inflicted with a 200-µL pipette tip. Podocytes were then washed with PBS three times to remove any debris and promigratory factors. EVs (20 µg/mL) in EV-depleted podocyte medium or plasma (10%) or EV-depleted plasma (10%) in RPMI-1640 were then added to the wells, and the scratch area was imaged immediately (time 0) and 12 h later using the EVOS FL Auto cell imaging system (Thermo Fisher Scientific). The areas of the clear zone at 0 and 12 h were measured by ImageJ, and podocyte migration was assessed by a reduction in the area, indicating more motile cells. The experiment was repeated three times with EVs of the same patients and healthy controls. Each repeat had two replicates.

G-LISA Rac Activation Assay

Immortalized cultured human podocytes were differentiated on collagen-coated 6-well plates by switching the temperature to 37°C for 10 days and serum starved for 24 days. The next day, they were stimulated with human plasma (10%) or EVs (20 µg/mL) for 4 h. Cells were then lysed, and their Rac1 activity was quantified using a colorimetric G-LISA Rac Activation Assay Biochem kit (Cytoskeleton, Denver, CO) according to the manufacturer’s instructions. EVs were lysed by cold lysis buffer of the kit, sonicated, and vortexed, and the supernatant after 10,000-g centrifugation was used for measuring RAC-GTP content.

Flow Cytometry of Human Podocytes for Phosho-p38 Staining

Immortalized human podocytes were differentiated on collagen-coated 6-well plates for 14 days. On the day of experiments, cells were serum starved for 4 h, and the media were then changed with RPMI-1640 media supplemented with 1) 10% FBS, 2) 10% human plasma, 3) 10% EV depleted-plasma, 4) patient or healthy EVs (20 µg/mL) and EV-depleted 10% FBS, and 5) 10% FBS with 40 ng/mL IL-13. After 30 min, the media were discarded, and cells were fixed with 1.5 mL fixation buffer (2% formaldehyde) for 15 min at 37°C. Cells were scraped, centrifuged at 800 g for 5 min, washed two times with PBS, and permeabilized with 80% cold methanol on ice for 1 h. After being washed, cells were incubated with anti-p38 MAPK (pT180/pY182) (Alexa Fluor 647 or PE/Cy7) (clone 36/p38, BD Biosciences,) at 1:50 dilution for 1 h at room temperature, washed once with FACS buffer (1% BSA and 0.1% sodium azide in PBS), and analyzed with a Novocyte flow cytometer.

Western Blot Analysis

Cell lysates or EV lysates were incubated with RIPA buffer containing 2 M NaCl, 1 M Tris (pH 8), Nonidet P-40, 10% SDS, protease inhibitor cocktail (Thermo Fisher), and fresh phosphatase inhibitors (Thermo Fisher). After 30 min of incubation with intermittent vortexing every 5 min, lysates were centrifuged at 10,000 g for 20 min, and supernatants were collected. Protein concentration was determined with a Pierce BCA Protein Assay Kit. Twenty-five micrograms of protein were loaded per well and separated by 10% or 12% SDS-PAGE, transferred to a PVDF membrane, and probed with 1:1,000 diluted primary antibodies against anti-phospho-p38 (p-p38; clone E-1, Santa Cruz Biotechnology), anti-tumor susceptibility gene 101 (TSG-101) antibody [clone EPR7130(B), Abcam, Cambridge, UK)], anti-α-tubulin (clone B-7, Santa Cruz Biotechnology), anti-synaptopodin (clone D-9, Santa Cruz Biotechnology), anti-flotillin (Cat. No. 3252, Cell Signaling Technology, Frankfurt am Main, Germany), anti-glucose-regulated protein 94 (GRP94; clone D6X2Q, Cell Signaling Technology), and anti-ALG-2-interacting protein X (ALIX; clone 3A9, Cell Signaling Technology). Blots were visualized by an Amersham Imager 600 (Little Chalfont, UK) after incubation of the membranes with 1:5,000 diluted horseradish peroxidase-linked secondary antibodies.

Statistical Analysis

Data were analyzed with GraphPad Prism 6 Software (San Diego, CA). Normality tests of the numerical variables were evaluated with a Kolmogorov-Smirnov test. Descriptive analyses are presented as means ± SD for numerical variables or means ± SE for repeated experiments. An unpaired Student’s t test or a Mann-Whitney U test were used to compare the two groups. An ANOVA test with Dunnet’s or Tukey’s post hoc test was used for analyses of repeated experiments. A chi-squared test or Fisher’s exact test (when chi-squared test assumptions did not hold due to low expected counts) was used to compare categorical variables. n represents the number of subjects. A P value of < 0.05 was considered significant.

RESULTS

EV Particle and Protein Concentrations Are Higher During Relapse and Remission Compared With Healthy Controls

EVs were isolated by serial ultracentrifugation from healthy controls and patients with NS in relapse and remission. The International Society of Extracellular Vesicles defines EVs as particles released from cells that are delimited by a lipid bilayer, cannot replicate, and do not have a functional nucleus (9). Our TEM images showed the bilayer lipid structure of our EVs (Fig. 1A). Relapse NS EVs were shown to carry ALIX, TSG-101, and flotillin by immunoblot analysis (Fig. 1B and Supplemental Fig. S1; all Supplemental Material is available at https://doi.org/10.6084/m9.figshare.15153141.v2) and were enriched in tetraspanins such as CD9, CD63, and CD81 (Supplemental Fig. S2), which play roles in membrane organization and are commonly used as cardinal EV surface markers. As a negative EV marker, glucose-regulated protein 94, which is an endoplasmic reticulum resident protein, was as expected positive in cell lysates and absent in EV samples (Fig. 1B). Plasma EVs constitute particles from a very different cell type. To elucidate the source of NS relapse EVs and to compare with healthy controls, bead-captured EVs were stained with cell surface molecules of platelets, endothelium, monocytes, T cells, and B cells. As previously reported, platelet EVs constituted the major EV pool of our samples (14), and their mean percentage was significantly higher than in healthy controls (P < 0.001). Similarly, mean percentages of endothelial cells, monocytes, and B cell EV percentages were significantly higher in relapse patients with NS compared with healthy controls (Fig. 2C). Moreover, the expression of podocalyxin, which is expressed on podocytes and endothelial cells, was significantly higher than that on the same amount of healthy control EVs (P < 0.001, Fig. 1D).

Figure 1.

Characterization of extracellular vesicles (EVs) obtained from plasma by serial ultracentrifugation. A: transmission electron microscopy images showing the typical bilayer membrane of EVs (arrowheads). Scale bars = 50 nm. B: Western blot images of expression of EV markers [ALG-2-interacting protein X (ALIX), tumor susceptibility gene 101 (TSG-101), and flotillin] on EVs from six relapse patients with nephrotic syndrome (NS) and podocyte cell lysates. Glucose-regulated protein 94 (GRP94) was used as a negative EV marker. C: circulating EVs of healthy controls (n = 10) and relapse patients with NS (n = 20) were stained with the corresponding cell type-specific markers. Data are presented as means ± SD (in %). An unpaired t test was used. ****P < 0.0001; ***P < 0.001; **P < 0.01. ns, nonsignificant. D: podocalyxin expression represented as mean fold induction (FI) on the matched isotype control on NS relapse EVs (n = 20) was significantly higher than in healthy controls (n = 10, P < 0.001). Data are presented as means ± SD. An unpaired Student’s t test was performed.

Figure 2.

Extracellular vesicle (EV) protein and particle concentration are higher in relapse and remission patients with nephrotic syndrome (NS) than in healthy controls. A: protein concentrations of relapse and remission EVs (n = 10) normalized to plasma volume were significantly higher compared with healthy EVs [n = 10, P < 0.05 (unpaired Student’s t test)]. B: particle concentrations of relapse and remission EVs normalized to plasma volume were significantly higher compared with healthy EVs (P < 0.05). Data are presented as means ± SD; columns represent means, and error bars are defined as the SD of individual data points. An unpaired Student’s t test was performed. C: representative images of particle size distribution and concentration EVs from a healthy control, a relapse patient with NS, and a remission patient with NS measured by tunable resistive pulse sensing (QNano, Izon Biosciences).

The protein contents of the EVs were measured by BCA, and their individual particle sizes and numbers were measured by tunable resistive pulse sensing. The protein amount and particle concentration per milliliter of plasma in the samples isolated during nephrotic relapse and remission were significantly higher than those in the samples from healthy controls (Fig. 2, A and B). Since samples from patients and healthy controls had different albumin levels and different plasma amounts during ultracentrifugation, the yields were also normalized based on plasma volume and albumin concentrations (9). The protein amount and particle concentration per milliliter of plasma per gram albumin in samples isolated during nephrotic relapse were significantly higher than those in samples isolated during remission and those in samples from healthy controls (Supplemental Fig. S3, A and B). Representative images of particle size distribution and concentration of a healthy control, a relapse patient with NS, and remission patient with NS are shown in Fig. 2C. The median size of the particles was 205 ± 31 nm in relapse patients, 209 ± 25 nm in remission patients, and 213 ± 25 nm in healthy controls, suggesting that the sizes of the EVs in the different disease states were not significantly different from each other (Supplemental Fig. S3C).

Kinetics of EV Binding and Uptake by Podocytes

The kinetics of EV binding and internalization by podocytes were assessed with the green lipophilic dye SP-DIOC and the red lipophilic dye PKH26, and the samples were analyzed by flow cytometry and immunofluorescence. Live cell imaging of the podocytes cocultured with SP-DIOC-stained EVs showed increased uptake of the EVs over time (Fig. 3A). The kinetics of the binding and internalization of the patient and healthy EVs by podocytes were similar in the flow cytometry analysis (Fig. 3B and Supplemental Fig. S4). At the end of 24 h, EVs stained with the red lipophilic dye PKH26 were located in the perinuclear region of the podocytes (Fig. 3D).

Figure 3.

Internalization kinetics and representative photomicrograph images of extracellular vesicle (EV) uptake by podocytes. A: time course (0-, 4-, 8-, and 16-h EV incubation) representative images of EV uptake stained with the green lipophilic dye SPDiOC visualized at the green fluorescent protein (GFP) channel. Scale bars = 1:200. B: SPDiOC-labeled EVs from two healthy controls and two relapse patients with nephrotic syndrome were incubated with podocytes, and time course uptake and binding and uptake (internalization) kinetics were determined by FITC-positive podocytes by flow cytometry. Nearly all SPDiOC-labeled EVs were taken by podocytes in 24 h. For uptake (internalization) kinetics, trypan blue (0.1%) was used to quench cell membrane-bound EVs. C: images of podocytes not exposed to EVs (phalloidin: green and DAPI: blue). D: EVs stained with the lipophilic red dye PKH26 were coincubated with podocytes for 24 h, and podocytes were then stained with the actin dye phalloidin (green). EVs were mostly located around the nuclear region (DAPI: blue) at the end of 24 h. Scale bar = 1:100.

EVs Isolated During Relapse Increase Podocyte Motility and Albumin Permeability In Vitro

Foot process effacement and increased motility of podocytes are the key events in the development of NS. In vitro podocyte motility was assessed by a scratch assay. After the well had been scratched, the podocytes were stimulated with plasma, plasma fractions from which EVs were depleted (supernatant collected after ultracentrifugation at 100,000 g) and EVs from patients and healthy controls (Fig. 4A). Relapse NS EVs significantly increased the motility of the podocytes compared with NS plasma, healthy plasma, and healthy EVs (Fig. 4, B and C). In addition, podocytes exposed to relapse NS EVs showed increased motility compared with podocytes exposed to disease control EVs (Fig. 4C). An albumin influx assay provides information about the integrity of the slit diaphragm in vitro (5). Podocytes seeded on collagen-coated Transwells were exposed to EVs from patients with NS and healthy controls for 4 h, and the albumin influx was then measured by a BCA assay. At the second hour, albumin concentrations were not significantly different among wells, but at the fourth hour, the wells treated with relapse NS EVs had higher albumin concentrations in the upper chamber than the wells treated with healthy EVs and remission NS EVs (P < 0.001; Fig. 4D).

Figure 4.

Plasma extracellular vesicles (EVs) from relapse patients with nephrotic syndrome (NS) significantly increase podocyte motility and permeability. A: representative set of well images showing the area of the clear zone at 0, 6, 12, and 24 h after the scratch that were treated with 10% FBS (unstimulated), relapse NS plasma, relapse NS EVs, and NS plasma depleted with EVs. Scale bars = 1:1,000 µm. B: time course of the scratch assay treated with 10% FBS (unstimulated), healthy plasma, healthy EVs, relapse NS plasma, relapse NS EVs, NS plasma depleted with EVs, disease control plasma, and disease control EVs. C: relapse NS EVs significantly increased podocyte motility compared with podocytes treated with FBS. Other treatment groups had nonsignificant differences compared with the FBS-treated group. The area of the clear zone was measured in pixels. The area of the clear zone at the postscratch 12th hour is expressed as the percent area of the 0th hour. Each experiment was performed at least in duplicate and repeated three times independently. Columns represent means; error bars are defined as the SEs of three independent experiments. One-way ANOVA with Dunnet’s post hoc test was performed versus FBS-treated podocytes. Data are presented as means ± SE. ****P < 0.0001. D: albumin influx assay of podocytes treated with healthy (n = 5) and remission NS (n = 5) and relapse NS plasma EVs (n = 5). At the end of the fourth hour, podocytes treated with NS relapse EVs had significantly higher albumin permeability compared with remission and healthy EVs and untreated controls [P < 0.0001 (two-way ANOVA)]. Means and SEs of three independent experiments are shown.

Proteomic Analysis of EVs Isolated From Patients With NS and Healthy Controls

To determine the protein cargo of the EVs, plasma EVs isolated from four patients with NS in relapse, four patients with NS in remission, and four healthy controls were subjected to mass spectrometry analysis. A total of 925 proteins from remission EVs, 588 proteins from relapse EVs, and 361 proteins from healthy control EVs were identified. Twenty-four of 588 proteins identified during relapse were among the top 100 proteins identified in EVs in the Vesiclepedia database (Fig. 5A) (11). The proteins that were most significantly expressed in relapse EVs based on GO analysis according to cellular compartment are shown in Fig. 5B. One of the most common cellular compartments is the EV and exosome compartment, which confirms the source of these proteins. Among the identified proteins, 272 proteins were only identified in the relapse NS EVs (Fig. 5C). Ingenuity pathway analysis of these 272 selectively expressed proteins revealed that the most significantly affected canonical pathways were RhoGD signaling, regulation of actin-based mobility by Rho, ERK/MAPK signaling, integrin-linked kinase (ILK) signaling, integrin signaling, signaling by Rho family GTPases, hepatocyte growth factor signaling, and clathrin-mediated endocytosis signaling (Fig. 5D). In addition, annotations of disease analysis revealed that six proteins [Rho GDP dissociation inhibitor-α (ARHGDIA), carbonic anhydrase 1 (CA1), carbonic anhydrase 2 (CA2), integrin subunit-β₁ (ITGB1), phosphodiesterase 4B (PDE4B), and phosphodiesterase 8 A (PDE8A)] were significantly related to nephrosis (P = 0.00496).

Figure 5.

Proteomics analysis of plasma extracellular vesicles (EVs) from relapse patients with nephrotic syndrome (NS), remission patients with NS, and healthy controls. A: 24 of 588 proteins identified during relapse were among the top 100 proteins identified in the Vesiclepedia database (11). B: Gene Ontology analysis of proteins expressed on relapse EVs showing the 10 most significantly expressed proteins according to cellular compartment. C: Venn diagram showing proteins expressed on EVs among patients and healthy controls. A total of 272 proteins were uniquely expressed on relapse EVs. D: the gene-gene interaction network was identified by ingenuity pathway analysis (IPA) software. The input was the list of selectively expressed relapse NS EV protein genes shown in Fig. 2C. Imported gene identifiers were mapped to the corresponding gene objects and overlaid onto the global molecular network identified by the IPA algorithm. Edge relationship and node type symbols illustrate the nature of the relationship among genes, considering the functionality of each gene supplied. Enrichment (FET) P value threshold: 1 e⁻³.

EVs From Relapse Patients With NS Induce Increased RAC1 and p-p38 Activation In Vitro

The protein cargo of NS relapse EVs was markedly different from that of NS remission EVs and healthy control EVs with respect to the ILK, MAPK, Rho, and RhoGD pathways. All these pathways converge at the Rac family small GTPase protein C3 botulinum toxin substrate 1 (RAC1). We measured the levels of active RAC1 and RAC-GTP in EVs by G-LISA after protein lysis and sonication and found that the RAC-GTP level normalized to EV particle number of relapse EVs was significantly higher than that of healthy EVs, remission EVs, and disease control EVs (Fig. 6A). Moreover, podocytes exposed to IL-13 and LPS secreted EVs with significantly higher levels of RAC-GTP than those secreted by unstimulated podocytes (Fig. 6C). Based on previous findings showing increased motility and albumin permeability of podocytes treated with NS relapse EVs, we measured RAC-GTP levels in podocytes stimulated with EVs. The results showed that relapse NS EVs significantly increased RAC-GTP levels in podocytes compared with healthy EVs, remission EVs, and disease control EVs (Fig. 6D). MAPK p-p38 has been shown to be downstream of RAC1 activation in podocytes (15). Relapse NS EVs significantly induced higher p-p38 expression in podocytes compared with healthy, remission NS, and disease control EVs, similar to RAC-GTP levels (Fig. 6D). In podocytes exposed to IL-13 (40 ng/mL), the peak RAC-GTP level was reached at time = 15 min and the peak p-p38 expression level was reached at time = 20 min (Supplemental Fig. S5, A and B).

Figure 6.

Relapse nephrotic syndrome (NS) extracellular vesicles (EVs) have higher RAC-GTP content and induce increased RAC-GTP expression in cultured podocytes. A: plasma relapse NS EVs (n = 10) have higher RAC-GTP content normalized to EV particle number compared with healthy EVs (n = 10), remission EVs (n = 10), and disease control EVs (n = 6) as measured by G-LISA. B: EVs derived from supernatants of podocytes after 24 h of IL-13 (40 µg/mL) and LPS (25 µg/mL) stimulation caused significantly increased RAC-GTP luminescence normalized to EV particle number compared with unstimulated podocytes. Columns represent means; error bars are defined as the SD of individual data points. An unpaired Student’s t test was performed. C and D: active RAC-GTP was measured by G-LISA in podocytes treated with IL-13 (40 ng/mL, positive control), healthy EVs, NS relapse EVs, and NS remission EVs (20 μg/mL). NS relapse EVs significantly increased RAC-GTP and phospho-p38 (p-p38) expression compared with healthy EVs (P < 0.001), NS remission EVs (P < 0.001), and other disease control EVs (P < 0.001). NS remission EVs and other glomerular disease control EVs did not induce significant RAC-GTP and p-p38 expression compared with healthy EVs. Anti-RAC1 antibody luminescence was normalized to untreated podocytes and presented as fold induction (FI). Each experiment was performed at least in triplicate and repeated three times independently. Columns represent means; error bars are defined as the SD of three independent experiments. Two-way ANOVA with Tukey’s post hoc test was performed. ****P < 0.0001. ns, nonsignificant.

Compared with relapse NS plasma, relapse NS EVs from the same patients induced significantly higher p-p38 stimulation (Fig. 7, A–C). Moreover, plasma after 100,000-g ultracentrifugation, which was considered EV-free plasma, induced significantly lower p-p38 expression than relapse NS plasma from the same patients (Fig. 7C). We also assessed the time course of p-p38 and synaptopodin expression after stimulation with relapse NS EVs and plasma from the same patient (Fig. 7C). As shown in Fig. 7D, according to immunoblot analysis, relapse NS EVs and plasma led to increased p-p38 expression and decreased synaptopodin expression over time (Supplemental Fig. S6).

Figure 7.

Nephrotic syndrome (NS) extracellular vesicles (EVs) stimulate significantly higher phosho-p38 (p-p38) expression compared with NS relapse plasma and relapse NS EV-depleted plasma on cultured podocytes. A: representative images of flow cytometry plots displaying p-p38 expression. Anti-p-p38 labeled with PE-Cy7 and the appropriate isotype control was used. The gating strategy after excluding cell aggregates with forward and side scatter parameters is shown. PE-Cy7-positive cells were prominently shifted with IL-13 stimulation for 30 min. Both relapse NS plasma and relapse EV stimulation caused increases in the p-p38-positive cell percentage (A) and mean fluorescent intensity (MFI; B). C: MFI ratio of p-p38 on podocytes was significantly higher with relapse EV stimulation compared with NS plasma and NS EV-depleted plasma stimulation from the same patients (n = 10). Data are presented as means ± SD. A paired Student’s t test was performed. D: Western blot imaging showed increased p-p38 expression and decreased synaptopodin expression at 1–4 h of plasma and EV stimulation. E: quantification of band intensity of p-p38 compared with α-tubulin by ImageJ analysis. Means and SDs of three independent experiments are shown. F: Quantification of band intensity of synaptopodin as fold induction (FI) to α-tubulin by ImageJ analysis. Columns represent means; error bars are defined as SDs of three independent experiments.

DISCUSSION

Idiopathic NS was postulated to be mediated by circulating factor(s) on the basis of observations that plasma collected from nephrotic subjects induced proteinuria in previously non-nephrotic subjects (10). To date, many plasma permeability factors have been suggested as causes of NS; however, none has been validated yet (3, 5). In this report, our hypothesis represents a new way of thinking about circulating permeability factors.

EVs are highly abundant in plasma and other body fluids, are highly stable, and participate in important biological functions by acting as a mode of communication between cells (16). Circulating EVs have been reported to be increased in many diseases and are widely accepted as “liquid biopsies” in cancer (17). There is an increasing body of evidence that suggests that the cargo of circulating plasma EVs reflects the ongoing pathophysiological processes in organisms (18). EV omics is increasingly considered a rich source of biomarkers for various disease states. However, there is no single method of plasma EV isolation, which causes confusion in the evaluation of study results (7). Here, we took the approach of differential ultracentrifugation, which is the most well-studied method of EV isolation and is an appropriate approach for functional studies. In addition, proteomics studies of plasma have shown that ultracentrifugation supplies fewer soluble proteins and lipoprotein complexes than newly developed isolation methods (19, 20). Strategies have been developed to decipher the purity of EVs. Nevertheless, none of these methods are accepted as a gold standard method, and there is no consensus with regard to obtaining pure EVs using a single method. One approach could be favored to the other in some studies, but always there is a trade off either from purity or from EV yield. Previous studies have shown that plasma and/or serum-derived EVs using various isolation methods showed a very similar proteome profile and concluded that certain plasma proteins such as lipoproteins, immunoglobulins, and coagulation and complement proteins are tightly bound to the EV surface, constituting a true protein corona rather than being mere coisolating contaminants (21). Our focus was to isolate enough EVs with acceptable purity so that we could pursue our functional assays.

Here, we showed that plasma EV numbers were also correlated with disease relapse and that their RAC content can be used as a potential biomarker. However, circulating EVs are very heterogeneous in nature and difficult to characterize (9). The vast majority of human tissue contributes to the plasma EV pool, which changes according to the disease. Our study confirmed previous studies showing that platelet and endothelial EVs constitute the major part of the plasma EV pool and that their percentages are significantly higher in relapse patients with NS compared with healthy controls. This finding was also showed in patients with increased intravascular coagulation (22). Moreover, the higher expression of podocalyxin, which is found on podocytes and as well as endothelial cells, may also reflect endothelial dysfunction and increased thrombosis risk in relapse patients with NS. Here, we used CD14 to reflect the EVs shed from monocytes. CD14 is a receptor of LPS and is found on activated monocytes and to a lesser extent on granulocytes. In a cohort of adult patients with NS, CD14 was found to be excreted in urine at higher levels from biopsy-proven patients with MCD compared with focal segmental glomerulosclerosis and membranous glomerulopathy (23). Increased CD19-positive EVs may also reflect the increased B cell pool in circulation of relapse patients with NS (24). The increased expression of podocalyxin in our patient cohort may also reflect podocyte injury as our proteomics study revealed specifically expressed proteins that suggested nephrosis.

Another argument is that under physiological conditions, EVs would not be expected to pass the highly efficient and selective glomerular filtration barrier. Only molecular complexes below 6.4 nm in diameter and under 70 kDa could transit into the lumen of the nephron. Intriguingly, fluorescently labeled EVs injected parentally into normal rats were shown in kidney sections and in urine (25). Moreover, transrenal EV passage may be facilitated under pathological conditions such as NS, which is characterized by an injured glomerular filtration barrier (26). Transrenal EV release could depend on a process similar to transcytosis, i.e., vesicular uptake followed by transcellular release, a well-known process for the selective transendothelial transport of plasma albumin and low-density lipoproteins as well as EV uptake and secretion from the blood-brain barrier (27).

Cell motility, which is coordinated by actin and microtubules, is involved in the pathological events of NS, and enhanced migration was shown to be correlated with induced proteinuria (3, 28). Increased podocyte motility is one of the features of cultured podocytes induced by proteinuric factors (3, 28). Cell motility and migration are complex processes, and recent studies in tumor models have shown that tumor EVs promote aggressive behavior and metastasis by modifying the complex interactions of the tumor microenvironment and promoting the motility and invasiveness of tumor cells (for reviews, see Refs. 29 and 30). In particular, EVs were shown to deliver extracellular matrix components, such as fibronectin, to promote adhesion formation, and exosome secretion was found to be crucial for directional and efficient cell migration (31). Moreover, EVs were shown to participate in the biogenesis and activity of invasive structures called invadopodia, which are cellular protrusions similar to lamellipodia, through MVB-dependent delivery of metalloproteinases and other cargo molecules, including RAC1 (32).

In our study, the most significantly enriched pathways in circulating EVs of relapsed patients with NS were the ILK, MAPK, Rho, and RhoGD pathways, which converge at RAC1 protein. RAC1 is a member of the Rac family of GTPases, which is a subfamily of Rho small GTPases that regulates the actin cytoskeleton. RAC1 and Cdc42 activation was shown to increase lamellipodia formation and promote cell motility in podocytes (33). Activation of RAC1 signaling in podocytes was shown to be a widespread cause of foot process effacement and proteinuria (34). In a transgenic mouse model expressing a doxycycline-inducible active form of RAC1, transient activation of RAC1 caused transient proteinuria resembling SSNS, leading to p38-MAPK activation and β₁-integrin redistribution (15). Moreover, Robins et al. showed that RAC1 was overexpressed in kidney biopsies from patients with MCD and idiopathic focal segmental glomerulosclerosis and that the sera of the same patients activated RAC1 downstream of p-p38 in cultured podocytes. These authors suggested RAC-p38 MAPK activation as a useful bioassay to assess the activity of potential permeability factors (15). In this report, we showed that relapse NS EVs had higher RAC-GTP than remission EVs and other proteinuric disease EVs and induced higher expression of RAC1 and its downstream molecule p38 in cultured podocytes than EVs obtained from healthy controls, EVs from the same patients during the remission period, and EVs from patients with other proteinuric glomerular diseases. In addition, relapse NS EVs induced higher RAC1 and p-p38 expression in podocytes than plasma and EV-depleted plasma. Based on the higher RAC-GTP levels in the EVs of relapsed patients with NS, EVs may be a mediator of podocyte dysfunction. The Rho-GTP pathway and RAC1 also have a role in the biogenesis of EV formation, and RAC1 was reported to be one of the most common top 100 proteins found in EVs (11, 35). Increased RAC1 may also reflect increased number of EVs in plasma, as shown in Fig. 1B. However, LPS and IL-13 stimulation of podocytes has also caused increased RAC-GTP content normalized to particle number, supporting that the increased motility of podocytes and the increased RAC levels after stimulation with relapse EVs may be attributed to active RAC1 contents in EVs (Fig. 6B). Recent studies have shown the effective transport of proteins by EVs, including another GTP-bound enzyme, RAS, between tumor cells, which may induce the malignant transformation of recipient cells (36). Gopal et al. (37) showed that ovarian cancer cells undergoing epithelial to mesenchymal transition (EMT) specifically produce exosomes containing RAC1 and PAK2 and that these exosomes activate the RAC1 pathway in endothelial cells, promoting angiogenesis.

EVs were shown to play a role in intranephronic communication, effective transport of proteins modulating physiological response of kidneys, and the progression of kidney diseases. Moreover, urinary EVs were shown to be increased in various kidney diseases and have been suggested to be potential biomarkers (38). The major limitation of our study is that modeling NS in vitro is in its infancy, and we have limited parameters to quantify changes in podocyte culture. However, human immortalized podocyte culture, which differentiates at 37°C, expresses a slit diaphragm-like structure and the vast majority of podocyte-specific proteins (39). Assessment of these proteins, such as synaptopodin, gives us important clues to podocyte injury, which is the main pathological event in idiopathic NS. However, an in vitro model representing the entire glomerular filtration barrier and the interactions between podocytes, the fenestrated endothelium, and the glomerular basement membrane would better reflect the role of EVs in the cross talk between cells and the in vivo situation (1, 39).

Perspectives and Significance

To the best of our knowledge, this study is the first to characterize the protein cargo of circulating EVs in patients with NS. Another limitation of our study is that EVs carry not only proteins but also many different types of cargo, such as mRNAs and miRNAs, and there is great potential for the discovery of potential pathological factors that play roles in the progression of kidney disease. This is a preliminary study with a small sample size, but we conclude that circulating relapse EVs are biologically active molecules that carry elevated levels of RAC1 as cargo and induce recapitulation of the NS phenotype in podocytes in vitro. Further studies are needed to completely characterize the cargo of EVs and their effects on the pathogenesis of NS.

SUPPLEMENTAL DATA

Supplemental Figs. S1–S6: https://doi.org/10.6084/m9.figshare.15153141.v2.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

F.K.E., T.G., E.C., D.K., M.B., and I.G., conceived and designed research; F.K.E., U.G., M.Y., T.Y., N.T., E.C.D., P.K., B.S., and I.G. performed experiments; F.K.E., V.Y., U.G., N.T., and I.G. analyzed data; F.K.E., T.Y., and I.G. interpreted results of experiments; F.K.E., V.Y., T.Y., N.T., E.C.D., P.K., and I.G. prepared figures; F.K.E., V.Y., U.G., and I.G. drafted manuscript; F.K.E. and I.G. edited and revised manuscript; F.K.E., V.Y., U.G., M.Y., T.Y., T.G., E.C., D.K., N.T., E.C.D., P.K., B.S., M.B., and I.G. approved final version of manuscript.