Peritoneal dialysis effluent-derived exosomal miR-432-5p: an assessment tool for peritoneal dialysis efficacy.

Background: Ultrafiltration (UF) volume and peritoneal solute transport rate (PSTR) are common parameters used to evaluate the efficacy of peritoneal dialysis (PD) on individual patients. It is unclear whether the level of exosomal microRNA (miRNA) in peritoneal dialysis effluent (PDE) can predict UF or PSTR. This study was designed to investigate if there is a correlation between PDE exosomal miRNA (miR-432-5p) levels and various UF volumes and PSTRs in PD patients. It also aimed to explore the underlying mechanism of water and dialytic sodium removal (DSR).
Methods: The PSTR was quantified using the 4-hour (4 h) 3.86% dialysate to plasma creatinine ratio. The PDE exosomes (PDE-exo) were isolated by ultracentrifugation. An miRNA assay was used to identify the different miRNA in the PDE-exo of patients in a high (H; PSTR >0.65, n=5) and low (L; PSTR <0.65, n=5) group. We focused on miR-432-5p as bioinformatic analysis had shown that it could be involved in sodium transport. We used mimic/inhibitor transfection and dual luciferase reporter assay to verify the target genes of miR-432-5p. We used PKH-67 stained PDE-exo to observe their interaction with human MeT-5A mesothelial cells.
Results: Our results showed that the PDE-exo-miR-432-5p level was higher in group H than in group L. The levels of PDE-exo-miR-432-5p were positively correlated with PSTR (r=0.391; P<0.05; n=40) and negatively correlated with the 4 h UF volume (r=-0.376; P<0.05; n=40) and 4 h DSR (r=-0.535; P<0.01; n=24). Epithelial sodium channel α subunit (α-ENaC) was revealed as a direct target gene of miR-432-5p and expressed on both human peritoneum and MeT-5A cells. Furthermore, we found the PKH67 labeled-PDE-exo could be internalized into MeT-5A cells. Conclusions: A high PDE-exo-miR-432-5p level was associated with poor UF volume and DSR. It may be that PDE-exo-miR-432-5p affects DSR through downregulating α-ENaC expression. 2022 Annals of Translational Medicine. All rights reserved.

Introduction

Peritoneal dialysis (PD) is an effective alternative renal replacement therapy for patients with end-stage renal disease. Patients undergoing PD vary significantly in terms of their peritoneal solute transport rate (PSTR), which is evaluated by the peritoneal equilibration test (PET) (1). Higher PSTRs are associated with declines in ultrafiltration (UF) volume and dialytic sodium removal (DSR), resulting in poor patient treatment. Adequate fluid clearance requires the removal of water coupled with sodium and is crucial for a high-quality PD (2). Overhydration and hypertension are commonly reported in PD patients (3). Both optimized UF and DSR are associated with a reduction in patient mortality (4). Sun et al. (5) demonstrated that excessive salt intake under uremic conditions induced functional and structural changes in the peritoneal membrane, leading to higher baseline PSTRs in subtotal nephrectomized mice. Our preliminary data showed that patients with different PSTRs have different dialytic sodium concentrations, but the relationship and molecular mechanisms remain unexplored.

Exosomes are small membrane-bound extracellular vesicles (EVs) with diameters of 30–150 nm. They are secreted by different types of cells, found in many biological fluids (6), and are composed of a lipid bilayer and abundant cellular cargos, such as nucleic acids [DNA, messenger RNA (mRNA), microRNA (miRNA), and other non-coding RNA (ncRNA)], proteins, and bioactive lipids that can be transferred between cells (7). Since the contents and functions of exosomes derived from different cell types are distinct, exosomes can act as biomarkers of various diseases, such as acute kidney injury (AKI) and renal cell carcinoma (8).

The miRNA are a class of small, endogenous, single-stranded ncRNAs which regulate gene expression at the post-transcriptional level (9). Actively secreted miRNAs are packaged into exosomes and involved in cell-cell communication (10).

Peritoneal dialysis effluent (PDE) is easily collected from PD patients for a variety of analyses. Although previous studies have illustrated that proteins in exosomes derived from PDE are related to PSTR and PD efficiency (11,12), there has been no determination that exosomal miRNA in PDE (PDE-exo-miRNA) has any effect on, or reflects, PSTR or PD efficiency. It is also unknown whether PDE-exo-miRNA is involved in UF or DSR.

The epithelial sodium channel (ENaC) is a non-voltage-gated sodium ion channel in the apical membrane of polarized epithelial cells (13). It plays an important part in maintaining the dynamic balance of sodium and water by regulating blood volume and blood pressure, renal sodium reabsorption, and the airway surface fluid clearance of lung epithelial cells (14-17). The expression and function of ENaC can be influenced by multiple factors, such as serum glucocorticoid-induced kinase 1, aldosterone, drugs, and miRNA (16-19). To date, the relationship between PDE-exo-miRNA and ENaC has not been explored.

In this study, we used miRNA array to investigate the miRNA profiles of PDE-exo in patients with different PSTRs, as well as the potential mechanism of DSR. We selected the significantly differential miR-432-5p, and its target gene ENaC α subunit (α-ENaC), according to bioinformatics analysis. We measured the expression of the significantly differential miR-432-5p in 40 patients and determined its correlation to PSTR, 4-hour (4 h) UF, and DSR. We also verified α-ENaC as a potential target gene of miR-432-5p. Consequently, we propose that miR-432-5p may target α-ENaC, thus affecting the processes of UF and DSR in PD.

We present the following article in accordance with the Materials Design Analysis Reporting (MDAR) checklist (available at https://atm.amegroups.com/article/view/10.21037/atm-21-3957/rc).

Methods

Sample collection

The PET is a key tool to assess a patient’s PSTR, which measures the dialysate/plasma ratio of creatinine after a standardized 4 h dwell (4 h D/PCr) of 3.86% glucose dialysate (20). In this study, patients underwent PET at 6-month intervals and were classified into 4 transport characteristics according to 4 h D/PCr: high (H; 0.82–1.03), high average (HA; 0.65–0.81), low average (LA; 0.50–0.65), and low (L; 0.34–0.49). The PSTR was considered stable if a patient returned the same H/HA or LA/L result 3 times consecutively. Our cross-sectional study aimed to investigate the relationship between PDE-exo-miRNA and UF volumes and PSTRs of PD patients. We enrolled 50 patients who received regular continuous ambulatory peritoneal dialysis (CAPD) at Shanghai Ninth People’s Hospital between January 2000 and July 2020. Patients (aged 18–80 years) with stable PSTRs and nocturnal dwell prescriptions were included in our study. None of the participants had peritonitis 3 months prior to, or during, the study. The exclusion criteria were malignancies, abdominal operation histories, autoimmune diseases, hepatitis, tuberculosis, pulmonary infections, or systematic inflammation. The patients selected for the study were divided into 2 groups, H (H/HA) and L (LA/L), according to the results of their PETs.

We explained the aim of our study to the patients included in this study and guaranteed their privacy. All individual participants signed and provided informed consent for the use of their PDE and clinical data for research purposes. The study was conducted according to the tenets of the Declaration of Helsinki (as revised in 2013) and approved by the Ethics Committee of Shanghai Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine (SH9H-2020-T23-1).

The nocturnal PDE was collected before PET was performed. We weighed and recorded the drained volumes after 4 h PET and the 24 h total UF volumes before PET. The sodium concentrations in PDE were also measured. The DSR was calculated using the following formula:

where Vd denotes the drained volume, Vi the infused volume, and Cd and Ci are the sodium concentrations in Vd and Vi, respectively.

Exosome isolation and characterization

We first collected the overnight PDE and shook it well to ensure the uniformity of its components. As EVs in the peritoneal cavity can be diluted in the PDE after 2 L of peritoneal dialysate infusion, we used a Stirred Ultrafiltration Cell with a 10 kDa NMWL, Ultracel regenerated cellulose membrane (MilliporeSigma, Billerica, MA, USA) to concentrate 500 mL of the PDE. Since the membrane was 10 kDa NMWL, the EVs were expected to remain in the liquid above the membrane, while most of the water and other small molecules would pass through the filter. The concentrate was centrifuged at 2,500 g for 15 min at 4 ℃ to remove cellular debris. After centrifugation at 18,000 g for 30 min at 4 ℃, the supernatants were filtered through a 0.22-µm filter to remove large substances. The samples were centrifuged at 120,000 g for 2 h at 4 ℃ to obtain exosomes using an Optima XE-90 ultracentrifuge (Beckman Coulter, Inc., Indianapolis, IN, USA). The pellets were then washed again with a phosphate buffer solution (PBS) at 120,000 g for 2 h. The exosomes were stored at −80 ℃ for further experiments.

To identify and observe the morphology of the exosomes, we used a Tecnai F20 transmission electron microscopy (TEM; FEI Company, Hillsboro, OR, USA). The particle sizes were evaluated using ZetaView Nanoparticle Tracking Analysis (NTA; Particle Metrix GmbH, Inning am Ammersee, Germany). Western blot was conducted to detect the specific markers that identify exosomes.

RNA extraction and microarray

We extracted RNA was from PDE-exos using an miRNeasy Kit (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions. To qualify all the extracted RNA, we performed quality control of the miRNA using an Agilent 4200 TapeStation (Agilent Technologies, Santa Clara, CA, USA) and measured the A260/A280 using NanoDrop Microvolume Spectrophotometers. We added 3.5 µL cel-miR-39 to the denatured exosomal mixture according to the manufacturer’s protocols (exoRNeasy Serum/Plasma Maxi Kits; QIAGEN). Exosomal miRNAs from 10 different patients (group H, n=5; group L, n=5) were purified and quality assessed. The differentially expressed miRNAs were screened and analyzed for more details using a GeneChip miRNA 4.0 Array (Affymetrix Inc., Sunnyvale, CA, USA).

Bioinformatics analysis

For bioinformatics analysis, we first used the miTarBase 7.0 (http://mirtarbase.mbc.nctu.edu.tw/php/index.php) database to obtain the target genes of miRNAs. We then used Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses on the Database for Annotation, Visualization, and Integrated Discovery (DAVID) 6.8 platform (https://david.ncifcrf.gov/) to determine the possible biological functions and signaling pathways. The GO analysis was classified into 3 categories, biological process (BP; red), molecular function (MF; blue), and cellular component (CC; green). The threshold of enrichment analysis was set as P<0.05. Lastly, Targetscan Human 7.2 (http://www.targetscan.org/vert_72/), microrna.org (http://www.microrna.org/microrna/home.do), miRDB (http://mirdb.org/), and miRWalk (http://mirwalk.umm.uni-heidelberg.de/) were used to determine the target genes of differentially expressed miRNAs.

Cell culture and transfection

Human MeT-5A mesothelial cells were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). They were cultured in an M199 medium (Hyclone Laboratories, Logan, UT, USA) with 10% fetal bovine serum (FBS) and 1% penicillin and streptomycin (Biological Industries, Kibbutz Beit HaEmek, Israel) at an atmosphere of 5% CO2 at 37 ℃. The cells were transfected with an miR-432-5p mimic, inhibitor, and respective negative control (NC) using a riboFECT CP Transfection Kit (RiboBio, Guangzhou, China). The dose and procedure were administered according to the manufacturer’s protocols. Each experiment was performed with 3 independent replicates.

Reverse transcription and quantitative real-time polymerase chain reaction

The miRNA extracted from PDE was polyadenylated and reverse transcribed into complementary DNA (cDNA) with the miDETECT A Track miRNA qRT-PCR Starter Kit (RiboBio). The miRNA extracted from cells was reverse transcribed using a stem-loop primer, and mRNA was reverse transcribed using a PrimeScript RT Reagent Kit (Takara Bio Inc., Shiga, Japan). The products of reverse transcription polymerase chain reaction (RT-PCR) were used to perform agarose gel electrophoresis. We performed SYBR Green PCR amplification on a real‐time PCR system (Thermo Fisher Scientific, Waltham, MA, USA) using a TB Green Premix Ex Taq kit (Takara Bio Inc.). The relative expression level of exosomal miRNA was normalized to cel-miR-39. After transfection, cell miRNA was normalized to U6 (common control gene), and the cell mRNA was normalized to housekeeping gene glyceraldehyde 3-phospahte dehydrogenase (GAPDH). We calculated the relative expression levels using the equation:

Western blot analysis

Samples of exosomes or MeT-5A cells protein lysate were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on 10% gels and transferred onto a polyvinylidene difluoride (PVDF) membrane. Blots were blocked with 5% nonfat milk and then incubated overnight at 4 ℃ with the primary antibodies CD9, CD63, CD81, Hsp70 (1:1,000; SBI Medical, Fort Lauderdale, FL, USA), α-ENaC (1:500; Alomone Labs, Jerusalem, Israel), or GAPDH (1:5,000; Proteintech Group Inc., Rosemont, IL, USA). They were then blotted at room temperature (RT) for 1 h with horse radish peroxidase (HRP)-conjugated Affinipure goat anti-rabbit immunoglobulin G (IgG) secondary antibody (1:5,000; Proteintech Group Inc.). The blots were visualized using an enhanced chemiluminescence detection system.

Immunohistochemistry (IHC)

Specimens of human peritoneum were obtained from patients who underwent elective abdominal surgery. After antigen retrieval, 3% hydrogen peroxide was added to block endogenous peroxidase activity. The slides were washed with PBS and incubated with the α-ENaC antibody at 4 ℃ overnight. Then, PBS was used in place of the primary antibody as NC. The slides were incubated with a secondary antibody at 37 ℃ for 30 min the next day. Immunohistochemical Strept Avidin-Biotin Complex (Solarbio, Beijing, China) was used according to the manufacturer’s protocols. After diaminobenzidine staining and hematoxylin counterstaining, the slides were observed under a microscope.

Dual luciferase reporter assay

To verify that SCNN1A is the target gene of miR-432-5p, HEK-293T cells were co-transfected with SCNN1A-wild-type (WT), mutant (MUT), or NC PmirGLO vectors, and miR-432-5p mimic or corresponding NC, using a Lipofectamine 3000 Reagent (Thermo Fisher Scientific) based on the manufacturer’s instructions. After 48 h transfection, a Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA) was utilized to measure the luciferase activities. The ratio of renilla luciferase activity to firefly luciferase activity was calculated. Each treatment was repeated 3 times.

Cell Counting Kit-8 (CCK-8) assay

The MeT-5A cells were seeded into 96-well plates and transfected with NC, mimic, or inhibitor for 48 h. Each group of cells were then incubated with 10 µL CCK-8 reagents (Biosharp, Shanghai, China) for 2 h at 37 ℃. The viability of cells was measured by absorbance values at 450 nm.

Fluorescent labeling of exosomes

To observe how PDE-exo-miRNA influences MeT-5A cells, we labeled the PDE-exo with a PKH67 Green Fluorescent Cell Linker Kit (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer’s instructions. The MeT-5A cells were incubated with the labeled exosomes or control solution overnight. The cell slides were washed with PBS and fixed with 4% formaldehyde solution at RT for 10 min. The cytoskeleton was then stained with TRITC-phalloidin (Solarbio), and the nuclei of the MeT-5A cells were stained by 4',6-diamidino-2-phenylindole (DAPI) (Beyotime Biotechnology, Shanghai, China). We observed the uptake and internalization of the labeled exosomes with the use of a fluorescence microscope.

Statistical analysis

All statistical analyses and graph generation were performed using SPSS Statistics 23 (IBM Corp., Armonk, NY, USA) and Graphpad Prism 8.0.1 (GraphPad Software, San Diego, CA, USA). Measurement data in accordance with Gaussian distribution were represented as mean ± SD, and an unpaired, 2-tailed Student’s t-test was used to compare the 2 groups. Non-Gaussian distribution data are represented as M (1/4, 3/4), and a Mann-Whitney U test was used to compare the 2 groups. Enumeration data were analyzed by Fisher’s exact test or chi-square test. One-way analysis of variance (ANOVA) was performed to compare multiple groups with different treatments. Spearman correlation was used for correlation analysis. A P value <0.05 was considered statistically significant. The enriched GO terms and KEGG pathways were visualized using RStudio software (RStudio, Boston, MA, USA). The Bingo plugin in the Cytoscape software (https://cytoscape.org/) was applied to describe the GO analysis network.

Results

PDE-exo identification

To identify whether the EVs isolated from concentrated overnight PDE by ultracentrifugation were exosomes, we first observed them under TEM and observed the characteristic cup-shaped vesicles with a bilayer structure (). The results of NTA showed that the peak diameter of purified particles was approximately 100 nm (). We found that the specific surface markers of exosomes, such as CD9, CD63, CD81, and Hsp70 were positive (). Taken together, our findings suggested that the PDE-derived EVs collected in our study were exosomes.

Figure 1

PDE derived exosome characteristics. (A) The typical shape and morphology of PDE exosomes isolated by ultracentrifugation under TEM. Scale bar: 100 nm. (B) The diameters of PDE exosomes measured by NTA. (C) Western blot of exosome markers CD9, CD63, CD81, and Hsp70. PDE, peritoneal dialysis effluent; PDS, peritoneal dialysis solution; TEM, transmission electron microscopy; NTA, nanoparticle tracking analysis.

Different PDE-exo miRNA profiles

The demographic and clinical characteristics of the 10 participants used for miRNA array are summarized in . Apart from 4 h D/PCr, there were no significant differences in age, gender, dialysis time, estimated glomerular filtration rate (eGFR), and Kt/V between the 2 groups.

Table 1

Demographic and clinical characteristics of participants in screening cohorts

Clinical characteristics	H group (n=5)	L group (n=5)	Test value	P value
Gender (male/female)	2/3	1/4	–a	>0.999
Age (years)	63.40±15.96	57.40±19.37	0.535b	0.608
Dialysis time (months)	38.20±12.58	39.80±15.27	0.181b	0.861
4 h D/PCr	0.77±0.04	0.58±0.07	5.364b	0.0007***
eGFR (mL/min/1.73 m2)	3.57±3.42	2.64±3.58	0.424b	0.683
Kt/V (mL/min)	2.00±0.47	2.17±0.89	0.386b	0.710

Values are means ± SD. a, Fisher’s exact test; b, t-test; ***, P<0.001. eGFR, estimated glomerular filtration rate; Kt/V, urea clearance index; SD, standard deviation.

The miRNA expression profiles of PDE-exo from patients with different PSTRs was characterized by miRNA array. We used the limma package in R software to screen different genes before evaluating the fold change (FC) and significance level of the differences between group H and L. The miRNA microarray found 590 miRNAs and 8 miRNAs were significantly different (P<0.05) in group H when compared to group L. Among them, the 5 miRNAs with FC >1.7 were upregulated ones, and 3 miRNAs with FC <0.63 were downregulated ones. Cluster analysis was shown based on the signal values of each differentially expressed miRNA in the samples (). The top 10 GO terms in each GO category and the significant pathway terms are listed in . The GO analysis of upregulated miRNAs showed that GO terms relevant to cellular complexes and protein binding were predominantly enriched in the CC and MF. Interestingly, the second and third items in the BP part were related to sodium transport. Therefore, we focused on the upregulated miRNAs to predict their target genes. The GO and KEGG analysis of downregulated miRNAs are provided in Figure S1.

Figure 2

Different PDE exosomal miRNA profiles. (A) Heatmap describing clustering analysis of 8 markedly different exosomal miRNAs in the H and L group. (B) Differentially regulated exosomal miRNAs among all miRNAs detected by miRNA array in volcano plot based on FC and P value. (C) GO analysis of upregulated miRNAs. (D) KEGG analysis of upregulated miRNAs. HA, high average; LA, low average; PDE, peritoneal dialysis effluent; miRNA, microRNA; FC, fold change; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Bioinformatic prediction of upregulated miRNA

Among the 5 upregulated miRNAs, miR-503-5p and miR-432-5p were the most enriched miRNAs according to the volcano plot (). The results showed that when compared with group L, a 1.72-fold higher level of miR-503-5p expression, and a 1.91-fold higher level of miR-432-5p expression, were observed in group H (). For this reason, we selected miR-432-5p for further analysis. A Venn diagram was used to identify the overlapping target genes of miR-432-5p predicted by Targetscan Human, mirnaorg, miRDB, and miRWalk (). The GO and KEGG pathway enrichment analyses were implemented with DAVID for genes present in 3 or 4 concurrent databases. The GO analysis revealed that the candidate targets of miR-432-5p were involved in metal ion binding, sodium channel regulator activity, and sodium ion transmembrane transport (). Furthermore, an enriched GO terms network was established using the Bingo plugin in Cytoscape software. Ion transport and ion channel activity were also illustrated by this network (). The whole image of the Bingo network is provided in the supplementary material (Figure S2). Based on the miRNA profiling data and bioinformatic predictions, we determined to investigate the role of miR-432-5p in sodium transport.

Figure 3

Prediction and bioinformatics analysis of target genes of miR-432-5p. (A) Relative expression level of PDE-exo-miR-503-5p and PDE-exo-miR-432-5p in the H and L group. (B) The intersection of putative target genes of miR-432-5p predicted by 4 databases are shown in a Venn diagram. (C) GO analysis of miR-432-5p target genes. (D) Bingo network of miR-432-5p target genes. PDE, peritoneal dialysis effluent; GO, Gene Ontology.

Validation of PDE-exo-miR-432-5p by quantitative RT-PCR (qRT-PCR)

To confirm miR-432-5p was differentially expressed in the H and L groups, qRT-PCR analysis was performed in the validation cohorts (20 patients in each group). There were no significant differences in gender, age, dialysis time, eGFR, Kt/V, hypertension, and diabetes between group H and L. However, a significant difference was found between the 2 groups in terms of the 4 h D/PCr (). The results of qRT-PCR analysis showed that there was a 1.86-fold higher level of relative PDE-exo-miR-432-5p expression (P=0.002) in group H compared with group L ().

Table 2

Demographic and clinical characteristics of participants in validation cohorts

Clinical characteristics	H group (n=20)	L group (n=20)	Test value	P value
Gender (male/female)	10/10	8/12	0.404a	0.525
Age (years)	59.10±15.33	58.20±13.47	0.197b	0.845
Dialysis time (months)	45.10±21.37	54.55±23.86	1.319b	0.195
4 h D/PCr	0.76±0.06	0.58±0.04	11.390b	<0.0001****
eGFR (mL/min/1.73 m2)	2.44 (0, 4.25)	0 (0, 3.84)	173.50c	0.446
Kt/V (mL/min)	2.00 (1.69, 2.41)	1.80 (1.60, 1.94)	0.386c	0.710
Hypertension (yes/no)	18/2	13/7	–d	0.127
Diabetes (yes/no)	8/12	6/14	0.300ª	0.584

Values are means ± SD or M (1/4, 3/4). a, chi-square test; b, t-test; c, Mann-Whitney U test; d, Fisher’s exact test; ****, P<0.0001. eGFR, estimated glomerular filtration rate; Kt/V, urea clearance index; SD, standard deviation.

Figure 4

The relative expression level of PDE-exo-miR-432-5p and transport status. (A) Different expression levels of PDE-exo-miR-432-5p in the H and L group. (B) PDE-exo-miR-432-5p positively correlated with 4 h D/PCr. (C,D) Relationship between PDE-exo-miR-432-5p and UF. (E,F) Relationship between PDE-exo-miR-432-5p and DSR. *, P<0.05; **, P<0.01. UF, ultrafiltration; DSR, dialytic sodium removal; PDE, peritoneal dialysis effluent.

PDE-exo-miR-432-5p is associated with PET, UF volume, and DSR

Since miR-432-5p was increased in group H, we hypothesized that miR-432-5p might correlate with 4 h D/PCr and UF volume. Spearman correlation analysis indicated that a relative miR-432-5p expression level was positively correlated with 4 h D/PCr (r=0.391; P<0.05) (). We observed a negative correlation between PDE-exo-miR-432-5p and 4 h UF (r=−0.376; P<0.05). However, PDE-exo-miR-432-5p was not correlated with 24 h UF (). We also observed a negative correlation between PDE-exo-miR-432-5p and 4 h DSR (r=−0.535; P<0.01), but there was no relation between PDE-exo-miR-432-5p and 24 h DSR ().

α-ENaC is expressed on both human peritoneum and MeT-5A cells

Bioinformatic prediction indicated that 12 target genes of miR-432-5p were associated with sodium ion transmembrane transport (). However, only SCNN1A has a broad expression in the epithelial tissues of various organs and encodes the α-subunit of ENaC, which is an indispensable component of nonselective channels contributing to alveolar fluid clearance (21).

Table 3

Target genes of miR-432-5p related to sodium ion transmembrane transport

Sodium ion transmembrane transport (GO:0035725)	Genes
Voltage-gated sodium channel (SCN)	SCN4A, SCN5A, SCN3B
Epithelial sodium channel (ENaC)	SCNN1A, SCNN1G
Solute carrier family (SLC)	SLC4A4, SLC12A3, SLC24A1, SLC24A2, SLC34A1
Others	ASIC2, HCN4

GO, Gene Ontology.

The expression of α-ENaC on human peritoneum was determined by IHC. It was revealed that α-ENaC was primarily localized to monolayer mesothelial cells of peritoneum (). The result of RT-PCR showed that SCNN1A (406 bp), which encodes α-ENaC, was present in Met-5A cells (). A band at 90 kDa corresponding to α-ENaC was revealed in Met-5A cells ().

Figure 5

Expression of α-ENaC in peritoneum and mesothelial cells. (A) Expression of α-ENaC in human peritoneum determined by IHC. Staining method: IHC. Scale bar: 20 µm. (B) α-ENaC gene expression in mesothelial cells detected by AGE. (C) Detection of α-ENaC protein expression in mesothelial cells by western blot. PBS, phosphate-buffered saline; IHC, immunohistochemistry; AGE, agarose gel electrophoresis.

SCNN1A is a direct target gene of miR-432-5p

Targetscan predicted a complementary sequence between miR-432-5p and SCNN1A (). Therefore, to verify whether SCNN1A is a target gene of miR-432-5p, MeT-5A cells were transfected with miR-432-5p mimic or inhibitor. When a mimic was transfected, the expression level of miR-432-5p relative to U6 significantly increased, but miR-432-5p expression did not change in either the blank or NC group. The levels of α-ENaC mRNA and protein exhibited a marked decrease in the mimic group compared to the control group (), while α-ENaC mRNA and protein were increased by inhibiting miR-432-5p (). We also found that Met-5A cell viability was not influenced by transfection ().

Figure 6

Verification of SCNN1A as a target gene of miR-432-5p. (A) Predicted alignment of miR-432-5p with the 3'UTR of SCNN1A mRNA. (B,D) The relative expression of miR-432-5p and α-ENaC mRNA measured by qRT-PCR. (C,E) Western blot analysis of α-ENaC protein expression level. (F) Cell viability of Met-5A after transfection detected by CCK-8. (G) Determination of the interaction between miR-432-5p and SCNN1A by the dual luciferase reporter assay. **, P<0.01; ***, P<0.001; ****, P<0.0001. 3'UTR, 3'untranslated region; WT, wide-type plasmid; MUT, mutant type plasmid; mRNA, messenger RNA; MB, mimic blank; MNC, mimic negative control; IB, inhibitor blank; INC, inhibitor negative control; qRT-PCR, quantitative real-time polymerase chain reaction; CCK-8, Cell Counting Kit-8.

In addition, a dual luciferase reporter assay was applied to elucidate the targeting relationship between miR-432-5p and α-ENaC. The 293T cells were transfected with plasmids and mimic/NC. We found that the relative luciferase activity in the WT + miR-432-5p mimic group was significantly lower than that in any other groups, which confirmed that miR-432-5p directly interacted with the binding sites at the 3'untranslated region (3'-UTR) of SCNN1A mRNA ().

PDE-exo uptake by MeT-5A cells

To investigate if PDE-exo-miR-432-5p affected DSR, we labeled the exosomes with PKH67, which is a green-fluorescent dye. We then added them to the cell medium to observe whether the PDE-exo could be taken into cells. We used PKH67-PBS and PBS controls as the reference. After 24 h incubation, the exosomes were incorporated into the MeT-5A cells and distributed in the cytoplasm. However, this did not occur for the control groups, which showed no fluorescence (). These observations clarified that PDE-exo can be internalized and is thus a potential regulator of genetic expression in MeT-5A cells.

Figure 7

Uptake of PDE-exo by MeT-5A cells. Fluorescent imaging analysis of PKH67 labeled exosomes derived from PDE internalized by MeT-5A cells. Exosomes (green) in the cytoplasm (red) and around the nuclei (blue). Staining method: fluorescent staining. Scale bar: 50 µm. DAPI, 4',6-diamidino-2-phenylindole; PBS, phosphate-buffered saline; PDE, peritoneal dialysis effluent.

Discussion

In this study, we isolated EVs from concentrated PDE and found they possessed a size and composition characteristic of exosomes (22). Their morphology and size distribution were also consistent with the findings of other studies (11,12). For this reason, we used the term exosome when describing these vesicles. Examining the exosomal miRNA profiles in patients with different PSTRs yielded evidence that the expression level of PDE-exo-miR-432-5p was higher in group H than that in group L, and this was negatively correlated to 4 h UF and DSR. Furthermore, we verified that SCNN1A is one of the target genes of miR-432-5p. Since PDE-exo can be internalized into mesothelial cells, they can therefore be regarded as carriers of miRNA, which inhibit the expression of α-ENaC ().

Figure 8

Potential mechanism for PDE-exo-miR-432-5p involved in DSR. The level of PDE-exo-miR-432-5p increases in the PDE of H/HA patients, which can bind to the 3'UTR of α-ENaC and decrease its expression in peritoneal mesothelial cells, affecting the sodium ion transport in PD. Red rectangle, toxin; green triangle, ion; blue ellipse, water; bilayer vesicle, exosome. PDE, peritoneal dialysis effluent; DSR, dialytic sodium removal; H/HA, high/high average group of patients; 3'UTR, 3'untranslated region; PD, peritoneal dialysis.

It is suggested that the criterion for assessing PD adequacy should include not only solute clearance, such as urea or creatine clearance, but also fluid and sodium removal, control of blood pressure, and adequate nutrition (4). Previous trials in PD have demonstrated that sodium and fluid overload result in a higher risk for cardiovascular events and technique failure (23). Patients with higher PSTRs are more prone to sodium and water retention than those with lower PSTRs, thus leading to UF failure and a poor prognosis (2). Although patients can achieve similar Kt/V by adjusting the prescriptions of PD to meet the standard dialysis adequacy, we found that the differences of DSR among patients with different PSTR may affect UF and result in volume overload. In recent years, many long non-coding (lnc)RNAs/miRNAs have been identified as diagnostic or prognostic biomarkers of human diseases. For example, urinary exosomal ncRNAs-miRNA-181a is a diagnostic biomarker for the early detection of chronic kidney disease (CKD). Both HCP5 and lncRNA NOP14-AS1 are potential biomarkers to predict CKD progression. Furthermore, there are various lncRNAs involved in experimental AKI that have been noted to improve renal injury by reducing inflammation, oxidative stress, and cell death, such as LRNA9884, LINC00520, and so on (24). A previous study found that lncRNA 6030408B16RIK silencing can prevent UF failure in PD via miRNA-326-3p-mediated WISP2 down-regulation (25). However, few miRNAs or lncRNAs can predict peritoneal transport characteristics or have been reported as being involved in UF and DSR. To extend the research on this topic, we used an miRNA array to analyze PDE-exo-miRNA profiles from patients with different PSTRs in this study.

Exosomes are well recognized as carriers of miRNA and protein delivery (26,27). Szeto et al. (28) found that a change in PDE-miR-21 and miR-589 levels over 1 year produced a corresponding change in 4 h D/PCr. However, there was no significant correlation between these miRNAs and UF volume. Unlike these previous studies, which extracted free miRNA directly from PDE (28,29), we extracted exo-miRNA, which has a higher stability and resistance to degradation due to the bilayer structure of exosomes. We first examined the differentially expressed PDE-exo-miRNAs in patients from H and L PSTR groups. We found the expression of miR-432-5p showed the most variation when compared to several other differential miRNAs measured. As GO and KEGG pathway analyses suggested that the target genes of miR-432-5p may mediate sodium transport, we enrolled 40 patients to investigate whether miR-432-5p is involved in the regulation of UF and DSR. Our study showed that PDE-exo-miR-432-5p was significantly higher in group H and had a weak positive correlation with 4 h D/PCr. However, PDE-exo-miR-432-5p had a moderate negative correlation with 4 h UF and 4 h DSR, but not with 24 h UF or 24 h DSR. Our results surrounding the correlation between PDE-exo-miR-432-5p and UF or DSR were thus inconsistent at 4 and 24 h dwell times. Because 24 h UF and 24 h DSR were also affected by PD conditions (glucose concentration, dialysis time, dialysis fluid dwell time), urine volume, and dietary sodium intake, we took the 4 h dwell time results as being more valid. It is yet to be determined if PDE-exo-miR-432-5p can be used as a reliable biomarker for measuring PSTR. Large sample clinical trials and more experiments are needed to test the sensitivity and specificity of miR-432-5p.

Previous studies have shown that miR-432-5p plays an important role in the regulation of the development and drug resistance of cancers (30,31); however, the function of miR-432-5p in sodium ion transport remains unclear. To eliminate the limitations of a single database, we used 4 different databases to predict the target genes and investigated whether the target genes of miR-432-5p could influence sodium ion transport in PD. Our analysis indicated that the primary target genes of miR-432-5p included the voltage-gated sodium channel (SCN), ENaC, solute carrier family (SLC), acid sensing ion channel (ASIC), and hyperpolarization activated cyclic nucleotide gated potassium channel 4 (HCN4) (). The SCN4A, SCN5A, and SCN3B genes are expressed in skeletal and cardiac muscle (32). The SLCs are among the target genes of miR-432-5p and encode different sodium channels or transporters which have restricted expressions in the kidney, rods, and cones of the retina (33-36). The ASIC gene mainly expresses in brain and endometrium (37), whereas HCN4 encoded proteins are necessary for cardiac pacing (38). Since the ENaC gene is the only one broadly expressed in the epithelium over the body, we focused our experiment on whether it was the target gene of miR-432-5p. Although both the α-ENaC gene SCNN1A and the γ-ENaC gene SCNN1G are putative target genes of miR-432-5p, we preferred SCNN1A for the potential it showed for binding better to miR-432-5p (Figure S3). In addition, α subunit is vitally important for ENaC composition. In a previous study, mice deficient in α-ENaC were found to die of respiratory distress more quickly than those lacking the γ subunit, indicating that α-ENaC plays a decisive role in lung fluid clearance (39). On account of these factors, we proposed that α-ENaC may be critical in determining the DSR in PD.

The channels and transporters on mesothelial cells are responsible for regulating the transport of fluids and electrolytes across serous membranes (40). Nie et al. (41) demonstrated that the ENaC channels were expressed in human pleural mesothelial cells. Peritoneal mesothelium is a part of the barriers responsible for solute transport in PD. In the present study, we detected the expression of α-ENaC on human peritoneum biopsies and mesothelial cells. Subsequently, we demonstrated that α-ENaC is the target of miR-432-5p. To investigate how PDE-exo-miR-432-5p participate in DSR, we observed PDE-exo internalization, suggesting that the PDE-exo can transfer information to the mesothelial cells. This may, in turn, inhibit the expression of α-ENaC and decrease DSR ().

The rate of technique failure in patients with high PSTR is much higher than that in those with low PSTR. Some studies have shown that increased levels of interleukin-6 (IL-6), vascular endothelial growth factor (VEGF), and matrix metalloproteinase-2 (MMP-2) are indicative of high transport characteristics. In our study, we found that patients have different PDE-exo-miRNA profiles depending on their PSTRs. However, whether PDE-exo-miR-432-5p can be used as a biomarker for measuring PSTR will require large sample clinical trials. In addition, PDE-exo-miR-432-5p may play an important role in DSR by inhibiting α-ENaC, but the sensitivity and specificity of miR-432-5p in predicting sodium and water clearance require further investigation. To this end, animal experiments could be conducted to characterize the role of ENaC in DSR and dissect the underlying molecular pathways. Future studies are also necessary to explore the mechanism of how miR-432-5p is regulated and what signal pathway it uses to effect sodium removal.

Conclusions

In conclusion, our study identified a unique profile for exosomal miRNAs in the PDE of patients with different peritoneal membrane transport characteristics. We found that the relative expression level of PDE-exo-miR-432-5p was higher in the H PSTR group, and that the PDE-exo-miR-432-5p expression level was negatively correlated with 4 h UF and 4 h DSR. We also demonstrated that α-ENaC is a gene for miR-432-5p targeted inhibition. Our findings provide evidence that PDE miR-432-5p could be used as a biomarker to determine the effectiveness of PD and provide a novel prospect for the mechanism of sodium removal during PD.

The article’s supplementary files as