Differential expression profile of urinary exosomal microRNAs in patients with mesangial proliferative glomerulonephritis

Clinical characteristics

The MsPGN patient and control groups participating in the sequencing and validation study and their clinical characteristics are summarized in Table 1. No statistically significant differences were observed between the MsPGN patient and control groups in age, sex, blood pressure, Alanine transaminase (ALT), Aspartate transaminase (AST), Blood urea nitrogen (BUN), Serum creatinine (Scr), urinary β2-microglobulin (β2-MG), urine red blood cell (URBC), Immunoglobulin G (IgG), and complement C3 (C3). However, there were significant differences in Total Protein (TP), albumin (ALB), Urine Albumin to. Creatinine Ratio (ACR), Urine Total Protein to. Creatinine Ratio (TCR), 24 hours urine protein (24hUpro), microalbuminuria (mAlb), uric acid (UA), Immunoglobulin A (IgA), Immunoglobulin M (IgM), and cystatin C (CysC) between the MsPGN patient and control groups.

Identification of isolated exosomes in urine

Exosomes from human urine are characterized by their morphology, diameter, concentration, and the presence of exosome-rich protein markers such as CD9, CD63, and TSG101. Transmission electron microscopy revealed exosomal bilayers of membranous structure and oval shape, with a typical cup-shaped structures, ranging in size from 30–100 nm. The results of particle size analysis of urinary exosomes in normal humans showed that the average particle size of urinary exosomes was 75.97 nm after 100-fold dilution, and a total of 5,985 particles were detected, with 99.78% of the particles being 30–150 nm and were at an average concentration of 2.64 × 10¹⁰ particles/mL. In patients with MsPGN, the mean particle size of urinary exosomes was 76.89 nm, and a total of 4,462 particles were detected, with 99.89% being 30–150-nm particles at a mean concentration of 3.93 × 10¹¹ particles/mL. In addition, western blotting revealed that these exosomes were positive for CD9, CD63, and TSG101 proteins (Figure 1). CD9 and CD63 are exosome-enriched proteins of the tetraspanin family, whereas TSG101 is a protein essential for multivesicular body biogenesis.

Figure 1. (A) Electron microscopic images of extracted exosomes revealed cup-shaped structures with a diameter of about 30–160 nm. Scale bar: 100 nm. (B) Nanoparticle tracking analysis (NTA) revealed the diameter of isolated extracellular vesicles (EVs) is consistent with that of exosomes. (C) Western blotting revealed CD9, CD63 and TSG101 proteins in exosome samples.

By comparing the TargetScan, miRDB, and miRWalk databases, 334, 445, and 279 miRNAs were found to be expressed in the normal control group, MsPGN patients, and co-expressed in both groups, respectively. The 50 significantly differentially expressed miRNAs (Table 2) were derived based on the log2 (fold change) absolute value ≥ 1, q-value < 0.05 screening criteria, among which 24 and 26 miRNAs were upregulated and downregulated, respectively (Figure 2A, 2B).

Figure 2. (A) volcano plot. (B) Heat map of 50 differentially expressed miRNAs (P < 0.05).

Validation of differentially expressed miRNAs

The validation cohort comprised a sample size of 32 (16 MsPGN patients and 16 healthy controls). To ensure the accuracy of the sequencing results, miRNAs for the subsequent PCR validation were screened according to the following criteria: (1) the mean expression (TPM) within at least one of the two groups of samples was > 2; (2) the absolute value of log2 (fold change) was ≥ 2; (3) at least one sample in the MsPGN group had a TPM > 0. The final selection of upregulated miR-125b -2-3p, miR-1262, miR-508-3p, let-7b-3p, miR-548o-3p, and miR-193b-3p and down-regulated miR-205-5p and miR-891a-5p were used for PCR validation. After statistical analysis, there were significant differences in the expression of miR-125b-2-3p, miR-1262, let-7b-3p, miR-548o-3p, miR-205-5p, and miR-193b-3p in the MsPGN group compared with that in the normal control group (P < 0.01); let-7b-3p expression was generally significant (P < 0.05), whereas miR-508-3p and miR-891a-5p expression differences were not statistically significant (P > 0.05). Among them, the miR-193b-3p expression differences were inconsistent with the direction of sequencing results, although they were significant. Therefore, miR-125b-2-3p, miR-1262, let-7b-3p, miR-548o-3p, and miR-205-5p were finally selected for the next step of target gene prediction and KEGG pathway and GO enrichment analyses. The relative expression levels of the five differentially expressed miRNAs are shown in Figure 3C.

Figure 3. (A) Top 20 statistically significantly enriched pathways in Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis. (B) Gene Ontology (GO) analysis for the target genes of five differentially expressed miRNAs. (C) The relative expression level of five differentially expressed miRNAs.

KEGG and GO analyses of differentially expressed miRNAs

Based on RNA sequencing results and subsequent validation, we revealed that five miRNAs (miR-125b-2-3p, miR-1262, let-7b-3p, miR-548o-3p, and miR-205-5p) were differentially expressed between the two groups. The target genes of the five miRNAs were predicted using TargetScan, miRDB, and miRWalk databases. The main biological processes, molecular functions, and cellular components of the target genes of the five miRNAs were identified by GO analysis. A total of 246 meaningful GO functions were obtained, including 47 molecular functions (mainly ubiquitin-protein ligase, protein serine/threonine kinase, DNA-binding transcription factor, and protein kinase activities), 44 cellular components (mainly nucleoplasm, cytoplasm, and post-emphasis density), and 155 biological processes (mainly positive and negative regulation of RNA polymerase II promoter transcription, positive and negative regulation of gene expression, cell migration). The top 30 GO functions identified by GO annotation are shown in Figure 3B. To better understand the biological functions of the predicted target genes, KEGG analysis was performed to reveal the main pathways in which the candidate target genes might be involved. The top 20 significantly enriched pathways were identified, including the PI3K/Akt, MAPK, Rap1, and mTOR signaling pathways (Figure 3A), and classified into environmental information processing (six), molecular processes (three), organismal systems (four), and human diseases (seven).

Correlation between clinical characteristics and urinary exosome miRNA in MsPGN patients

Spearman order correlation analysis was used to analyze the correlations between the five exosomal differential miRNAs (miR-125b-2-3p, miR-1262, let-7b-3p, miR-548o-3p, and miR-205-5p) and several clinical indicators in MsPGN patients. A P-value < 0.05 indicated correlation, and the larger the r-value, the greater the correlation. Both ACR and 24hUpro were inversely correlated with miR-125b-2-3p expression. Additionally, there was a positive correlation between uric acid levels and miR-548o-3p. miR-205-5p was positively correlated with urinary microalbumin. TP, ALB, TCR, IgA, and CysC were not correlated with any of the differentially expressed miRNAs (Table 3 and Figure 4). Meanwhile, fluorescence in situ hybridization was performed to detect miRNAs expression in renal tissues of MsPGN patients. The expression of miR-205-5p, let-7b-3p, miR-1262, and miR-548o-3p was elevated in the MsPGN patient group, and miR-125b-2-3p was decreased in the MsPGN patient group (Figure 5).

Figure 4. (A) simple linear regression between differentially expressed miRNAs and several clinical indicators in MsPGN patients. (B) Scatter diagram and linear correlation analysis. (C) Receiver operating characteristic (ROC) analysis of the combination of urinary exosomal miR-125b-2-3p, miR-1262, let-7b-3p, miR-548o-3p, and miR-205-5p for discriminating MsPGN. Plasma miRNAs yield an area under the ROC curve of 0.916.

Figure 5. According to fluorescence in situ hybridization, all five miRNAs were expressed in renal tissues of MsPGN patients. Nuclei are stained blue (DAPI) and miRNAs are stained red. Magnification, x200.

Study participants

All MsPGN patients were from the wards and outpatient clinics of the Department of Nephrology, the First Affiliated Hospital of Anhui University of Chinese Medicine (October 2019–October 2021), and baseline demographic and clinical data were recorded at the time of renal biopsy. The control group was derived from sex- and age-matched healthy subjects (from healthy volunteers at the medical screening center). Patients were all aged 18–70 years; five were MsPGN patients, and five were controls enrolled as the discovery cohort for exosomal miRNA sequencing. An additional 16 participants were enrolled in each group, increasing the validation cohort to 21 MsPGN patients vs. 21 controls. Data for all participants were obtained with their informed consent, and the study was approved by the ethical review committee of the Anhui University of Chinese Medicine.

Exosome isolation and RNA extraction

The morning urine of patients with MsPGN and that of normal healthy subjects was collected in 50-mL centrifuge tubes and sent to the laboratory within 1 h for preparation for centrifugation. Subsequently, the urine samples were centrifuged at 500 × g at 4° C for 10 min, after which the sediment was removed and the remaining supernatant transferred to new 50-mL sterile centrifuge tubes. The supernatant from the previous step was then centrifuged at 2,500 × g at 4° C for 30 min. After centrifugation, cells and debris were removed, the sediment discarded, and the supernatant transferred to new 50-mL Beckman special centrifuge tubes (Beckman Coulter, Brea, CA, USA). These were then centrifuged at 17,000 × g at 4° C for 30 min to remove large vesicles, after which the sediment was discarded. The resulting supernatant was transferred to new Beckman centrifuge tubes and centrifuged at 110,000 × g for 80 min at 4° C; the granular precipitate obtained at the bottom of the tube was the exosome. After resuspending the precipitate from the previous step with PBS, it was centrifuged at 110,000 × g for 80 min at 4° C, and the pellet-like precipitate at the bottom of the tube was the purified exosome. This was carefully blown with 100 μL of PBS for resuspension, after which the exosome suspension was aspirated into an EP tube and stored at -80° C in a refrigerator.

Total RNA was extracted using a mirVana miRNA isolation kit (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s protocol and quantified using a Nanodrop2000 Spectrophotometer (Thermo Fisher Scientific). RNA integrity was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA).

MicroRNA library construction and sequencing

One microgram of total RNA per sample was used for small RNA library construction using a TruSeq small RNA sample preparation kit (Cat. No. RS-200-0012; Illumina, San Diego, CA, USA). Briefly, total RNA was ligated to the adapters at both ends, after which the splice ligated RNA was reverse transcribed to cDNA and PCR amplified. Thereafter, the small RNA libraries were isolated and purified from the 140–160-bp PCR products. Library quality was assessed using an Agilent Bioanalyzer 2100 system (Agilent Technologies) with a DNA high sensitivity chip. The libraries were finally sequenced using an Illumina HiSeq X Ten platform (Illumina), and 150-bp paired-end reads were generated. The sequencing of small RNA library construction and data analysis were partially conducted by OE Biotech Co., Ltd. (Shanghai, China).

Western blot

Gel configuration was conducted according to the Takara catalog (Takara Bio Inc., Shiga, Japan). The sample stock solution was diluted to the desired loading volume, and then 5× SDS-PAGE protein loading buffer was added to a sample dilution of 1:4. Next, the proteins were fully denatured in a boiling water bath for 15 min. After cooling to room temperature, an appropriate amount (approximately 100 μg) was subjected to SDS-PAGE electrophoresis and transferred to a polyvinylidene difluoride (PVDF) membrane. Subsequently, the membrane was blocked with 5% skimmed milk. Thereafter, primary antibodies were added and diluted with primary antibody diluent at the appropriate ratios [CD9 (#Ab92726, 1:500; Abcam), CD63 (#Ab216130, 1:500; Abcam), and TSG101 (#Ab125011, 1:3,000; Abcam)] and incubated overnight at 4° C with slow shaking. Secondary antibodies included HRP-labeled Goat Anti-Mouse IgG(H+L) antibody (A0216, 1:1,000; Beyotime) and HRP-labeled Goat Anti-Rabbit IgG(H+L) antibody (A0208, 1:1,000; Beyotime). Finally, the PVDF membrane was immersed in the developing solution to develop the exposure, and the images were acquired using a UVP gel imaging analyzer (US, GelDoc-It Ts3 Imaging System).

Quantitative real-time PCR

Total RNA was extracted from urinary exosomes using TRIzol reagent (Life Technologies, USA) and reverse transcribed to cDNA using a PrimeScript^™ RT Reagent Kit with a gDNA Eraser (RR047A, AL21115A; Takara Bio Inc.) for miRNA expression verification. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the internal reference. All the primer sequences were designed and synthesized by Sangon Biotech Company (Shanghai, China) and data were analyzed according to the 2^-ΔΔCt method. Primer design as shown in Table 4.

Transmission electron microscopy

The exosome samples were removed from the -80-° C refrigerator, placed in an ice box, dissolved, and centrifuged, after which 15 μL were pipetted onto a copper grid for 1 min. Next, the exosome samples were blotted dry on the copper grid using filter paper, and then 15 μL of 2% UO2 acetate staining solution were pipetted onto the copper grid for 1 min at room temperature. Finally, the finished samples were baked under a lamp for 10 min, and the final images were observed using a transmission electron microscope at 80 kV (Tecnai, G2 spirit; FEI Company, Hillsboro, OR, USA), photographed, and saved.

Nanoparticle tracking analysis

The sample cells were washed with ultrapure water, and the instrument (NanoFCM, N30E) was calibrated using polystyrene microspheres (100 nm). Next, the exosomes were removed and diluted to the appropriate magnification. The instrument should be tested with the standard before the sample is loaded, and the sample should be diluted in a gradient to avoid blocking the injection needle. The particle size and concentration information of exosomes can then be obtained after the sample is tested.

Fluorescence in situ hybridization

The expression levels and localization of 5 miRNAs in renal tissues were detected by fluorescence in situ hybridization (FISH). Five probes for Cy3-labeled miRNAs were designed (hsa-miR-125b-2-3p, 5'-GTCCCAAGAGCC+TGACT+TGTGA-3'; hsa-miR-1262, 5'-ATCCT+ TCTACAAAT+TCACCCAT-3'; hsa-let-7b-3p, 5'-GGGAAGGCAG+TAGGT+TGTATAG-3'; hsa-miR-548o-3p, 5'-GCAAAAGTAAC+TGCAGT+TTTGG-3'; hsa-miR-205-5p, 5'-CAGAC+TCCGG+TGGAATGAAGGA-3', purchased from GenePharma (Shanghai, China). Each probe was hybridized overnight according to the manufacturer's instructions.

Bioinformatics analysis

To predict the target genes of differential miRNAs and take their intersection as candidate target genes, TargetScan Human version 7.2 (http://www.targetscan.org/vert_72/), MicroRNA Target Prediction Database (miRDB; http://mirdb.org/), and miRWalk (http://mirwalk.umm.uni-heidelberg.de/) were used. Gene ontology (GO; http://www.geneontology.org) analysis was then performed to analyze the main functions of the putative target genes, and Kyoto Encyclopedia of Genes and Genomes (KEGG; https://www.genome.jp/kegg) analysis was used to identify potentially altered molecular pathways.

Statistical analyses

Statistical analyses were performed using SPSS (version 25.0; SPSS Inc., Chicago, IL, USA). Continuous variables were tested for abnormality using the Kolmogorov–Smirnov test. To determine statistical significance, normally distributed variables were analyzed using Student’s t-test, and nonparametric data were assessed using the Wilcoxon–Mann–Whitney test. Receiver operating characteristic (ROC) curve analysis was used to assess the diagnostic potential of plasma exosomal miRNAs to differentiate MsPGN samples from controls. Logistic regression analysis was used to assess the diagnostic efficacy of miRNA combinations. Spearman correlation analysis was performed to assess the correlation between miRNA expression and MsPGN phenotype. P < 0.05 was considered statistically significant.