Microscopic polyangiitis plasma-derived exosomal miR-1287-5p induces endothelial inflammatory injury and neutrophil adhesion by targeting CBL

Subjects and sampling

We obtained blood samples from nine MPA patients in the Department of Nephrology of the Second Affiliated Hospital at Guangxi Medical University. All of the MPA patients fulfilled the criteria in the 2012 revised International Chapel Hill Consensus Conference Nomenclature of Vasculitis (Jennette et al., 2013) and were positive for IgG anti-MPO antibodies. We excluded patients with serious infections and patients who were positive for anti–PR3 and anti–glomerular basement membrane IgG antibodies. The blood samples were collected before MPA patients received glucocorticoid or immunosuppressants therapy. Nine healthy control individuals (HC) matching the MPA group with respect to age and sex were selected as HC group. The demographic characteristics of the MPA patients and HC in this study are shown in Table 1. The Ethical Committee of the Second Affiliated Hospital at Guangxi Medical University approved this study (Number: KY-0018) in accordance with the Declaration of Helsinki, and we obtained written informed consent from all participants.

Isolation and purification of exosomes

We purified exosomes from the plasma of HC (HC-exo) and MPA patients (MPA-exo) as previously described with minor changes (Zhang et al., 2020a). Briefly, we collected peripheral blood samples in anticoagulant tubes containing EDTA and centrifuged them at 3,000×g for 15 min at 4 °C to obtain plasma. We then centrifuged the plasma at 12,000×g for 45 min before filtering it through a 0.22-µm filter. Next, we subjected the collected supernatant to ultracentrifugation at 120,000×g for 70 min at 4 °C. We washed the exosome pallet with ice-cold phosphate-buffered saline (PBS), centrifuged it at 120,000×g for 70 min at 4 °C, and finally resuspended it in 100 µL of PBS.

Identification of exosomes

After washing the exosome suspension in PBS, we adsorbed the exosomes onto Formvar/carbon support film copper-mesh grids and fixed them with 3% uranyl acetate for negative staining. We observed the ultrastructure of the exosomes using transmission electron microscopy (TEM), detected the expression of exosomal markers via Western blot, and analyzed the size distribution of the exosomes via nanoparticle tracking analysis (NTA) using a Zetasizer Nano-ZS. We diluted the exosome samples 10,000×with sterilized PBS, injected it into the chamber and quantified the exosomes using NTA software.

Western blot

We harvested the collected exosomes and the cells with RIPA lysis buffer (Beyotime Biotechnology, Shanghai, China) containing 1% PMSF (a protease inhibitor) for protein extraction and examined the protein concentrations using a BCA protein assay kit (Beyotime Biotechnology, China). We then mixed the proteins in the different groups with loading buffer, denatured the proteins at 100 °C for 5 min and then subjected the proteins to SDS/PAGE. After electrophoresis, we incubated the polyvinylidene difluoride membranes with the following antibodies: anti-CD9 (1:2000, Abmart, Shanghai, China), anti-CD63 (1:2000, Abmart), anti-TSG101 (1:2000, Abmart), anti-CBL (1:2000, Abmart), and anti β-tubulin (1:2000, Abmart). After incubating the membranes overnight at 4 °C, we incubated the membranes with conjugated secondary antibodies (1:10000, Abmart) for 1 h at room temperature. We visualized the images with enhanced chemiluminescence reagent (Epizyme, Shanghai, China) and quantitated them using ImageJ. We compared the grayscale value of each protein signal to that of β-tubulin.

Culture of HUVEC and treatment with plasma-derived exosomes

We obtained human umbilical vein endothelial cells (HUVEC) from the cell bank of American Type Culture Collection (ATCC) and cultured them at 37 °C with 5% CO₂ in endothelial cell medium (ECM) containing 5% foetal bovine serum (FBS), 1% endothelial cell growth supplement, and 100 U/ml of penicillin and streptomycin (HyClone, Logan, UT, USA). We planted the cells with exosomes-depleted FBS and allowed them to reach approximately 90–95% confluence before use. We prepared exosomes-depleted serum by ultracentrifugation at 120,000×g at 4 °C for 18 h followed by passage through a 0.22-µm filter. We treated the HUVEC with HC–exo or MPA–exo for 24 or 48 h. Then, we subjected the treated HUVEC to the following experiments.

Exosome uptake

To trace the cellular uptake of exosomes, we stained exosomes with the green dye PKH67 (Sigma Aldrich, St. Louis, MO, USA) for 5 min. After terminating the labelling reaction with 5% bovine serum albumin (BSA), we harvested the exosomes and washed them with PBS by centrifugation (110,000×g for 70 min). We resuspended the PKH67–labeled exosomes in FBS–free DMEM and cultured the HUVEC with 50 µg of exosomes for 24 h on confocal dishes at 37 °C with 5% CO₂. After incubation, we washed the cells twice with PBS, fixed them with 4% paraformaldehyde for 10 min, and counterstained them with 4′,6-diamidino-2-phenylindole (DAPI) for 5 min to label the nuclei at room temperature. We then observed the cellular uptake of exosomes using a laser scanning confocal microscope (Leica Microsystems, Inc., Wezlar, Germany) and analyzed the absorption efficiency between HC–exo and MPA–exo with ImageJ.

Transfection

The mimic negative control (MNC), miR-1287-5p mimic, inhibitor negative control (INC) and miR-1287-5p inhibitor were synthesized by GenePharma (Shanghai, China). When the density reached approximately 50–70%, we transfected HUVEC with miR-1287-5p mimic at a concentration of 50 nM and miR-1287-5p inhibitor at a concentration of 100 nM. We used Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) as a transfection reagent following the manufacturer’s recommendations. After transfection for 6 h, we replaced the antibiotic–free medium with complete medium and cultured the HUVEC for 24 or 48 h. Then, we transfected si–001, si–002, and si–003 (si–CBL) (RiboBio, Guangzhou, China) into HUVEC to determine which siRNA most strongly decreased the mRNA transcription of CBL. We performed the transfection process as described above. Subsequently, we used qRT–PCR or Western blot to evaluate transfection efficiency by examining the mRNA or protein expression of CBL. After determining the most efficient siRNA for CBL silencing, we transfected HUVEC with siRNA negative control (si-NC), si-CBL, or si-CBL combined with miR-1287-5p inhibitor for 24 or 48 h. Table 2 displays the related sequences.

Exosome small-RNA sequencing and bioinformatics analysis

To conduct small-RNA sequencing, we lysed plasma exosomes isolated from HC (n = 3) and MPA patients (n = 3) and extracted total RNA using a miRNeasy Micro Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. We generated sequencing libraries using an NEBNext^® Multiplex Small RNA Library Prep Set for Illumina^® (NEB, Ipswich, MA, USA). We clustered the index-coded samples on a cBot Cluster Generation System using a TruSeq SR Cluster Kit v3-cBot-HS (Illumina) according to the manufacturer’s instructions. After cluster generation, we sequenced the library preparations sequenced on an Illumina HiSeq 2500/2000 platform. The Software Novomagic was used to obtain the potential miRNA and volcano map. We considered each miRNA with a p value<0.05 and a log₂—(fold change)—>2 to be physiologically relevant. A volcano map showed the differential miRNA expression profiles among exosome samples. The statistical power of experimental design, calculated in RNASeqPower is (depth = 15, cv = 0.4, effect = c (0.8,1.1), alpha = 0.05, power = c(0.8,0.9)).

RNA extraction and qRT-PCR

We extracted miRNA from the exosomes with a miRNeasy Serum/Plasma Kit (Qiagen, Germany) according to the manufacturer’s instructions. We reverse transcribed the extracted miRNA with a miRcute Plus miRNA First-Stand cDNA kit and detected it with a miRcute Plus miRNA SYBR Green qPCR Kit (Tiangen, Beijing, China). We extracted total RNA from HUVEC using TRIzol Reagent (Invitrogen) and performed qRT–PCR with HiScriptII Q RT SuperMix and ChamQ Universal SYBR qPCR Master Mix kits (Vazyme, Shanghai, China). We used an ABI StepOne Real-time PCR instrument (Applied Biosystems, Waltham, MA, U.S.A.) for quantitative analyses. We measured the relative expression levels of miRNA and mRNA using the 2^−ΔΔCt method and standardized them to the expression levels of cel-miR-39 and GAPDH, respectively. Sangon (Shanghai, China) synthesized the primers for miRNA detection, while Vazyme synthesized the primers for mRNA examination. The primer sequences are listed in Table 2.

Neutrophil adhesion assay

We performed a neutrophil adhesion experiment with minor modifications as previously described (Choi et al., 2017; Nistri et al., 2003). We isolated neutrophils from the peripheral venous blood of healthy volunteers by using Ficoll-Paque density gradient centrifugation (Tiangen, China) and determined the purity of neutrophils by Giemsa staining. We primed freshly prepared neutrophils (2 ×10⁵) with 10 µg/mL lipopolysaccharide (LPS, Sigma Aldrich, USA) for 30 min at 37 °C and added them to HUVEC. We maintained the cocultures for 120 min, rinsed them with PBS three times to remove unbound neutrophils, fixed them in 4% formaldehyde and counterstained them with haematoxylin for light microscopic screening. We performed each group experiment three times and counted the number of adherent neutrophils and HUVEC in five randomly chosen microscopic fields at a 200 × final magnification. Finally, we analyzed the ratio of the neutrophils number to HUVEC number.

Dual-luciferase reporter assay

We predicted miR-1287-5p target genes using the TargetScan database available on the internet (http://www.targetscan.org). We performed a dual-luciferase reporter assay to verify whether CBL is the target mRNA of miR-1287-5p. Briefly, we constructed a wild-type CBL 3′-untranslated region (3′-UTR) reporter plasmid (CBL Wt) and a mutated-type CBL 3′-UTR reporter plasmid (CBL Mut) with the pMIR-promoter vector by OBiO Technology (Shanghai, China). We plated HEK293T cells in 96-well plates and transfected them with CBL Wt or CBL Mut in the presence of miR-1287-5p mimic or negative control (NC) in strict accordance to the manufacturer’s specifications. At 48 h after transfection, we quantified luciferase activity in the cell lysates with a luciferase assay kit (Promega, Madison, WI, USA).

Statistical analysis

We performed all statistical analyses using SPSS software and GraphPad Prism 7. We express the data as the mean ± SD. We performed comparisons between two groups using independent-sample t-tests. We considered p < 0.05 to indicate statistical significance.

Characterization and internalization of exosomes derived from plasma

We collected blood samples from both MPA patients and HC. We identified plasma-derived exosomes by TEM, NTA and Western blot. We observed typical cup-shaped vesicles, ∼100 nm in diameter, using TEM (Fig. 1A). We confirmed the expressions of exosomal marker proteins in the two groups, including CD9, CD63 and TSG101, by Western blot analysis (Fig. 1B). NTA showed that the particle size peaks from MPA patients and HC were 80 nm and 100 nm, respectively (Fig. 1C). These results revealed that we successfully isolated exosomes from plasma.

To further determine whether HUVEC absorbed the plasma-derived exosomes, we cocultured exosomes labelled with PKH67 (a green fluorescent cell linker) with HUVEC. After 24 h of incubation, the HUVEC had efficiently internalized PKH67–labeled exosomes, as indicated by confocal laser scanning microscopy (Fig. 1D), which indicated that HUVEC could take up plasma-derived exosomes. Moreover, there was no significant difference between HC–exo and MPA–exo in the average grey value (Fig. 1E).

Treatment of HUVEC with MPA-exo induced endothelial inflammation and neutrophil adhesion

To investigate whether MPA-exo induced endothelial inflammation, we cocultured HUVEC with HC-exo and MPA-exo for 24 h and measured the expression of inflammatory factors by qRT–PCR. As shown in Fig. 2A, the relative mRNA levels of IL-6, IL-8, and MCP-1 (fold changes: 1.44, 1.82 and 2.07, respectively) and of the adhesion molecules ICAM-1 and E-selectin (fold changes: 1.81 and 1.86, respectively) were significantly higher in HUVEC treated with MPA-exo than in HUVEC treated with HC-exo (p < 0.05).

To assess whether MPA-exo attracted neutrophils to the endothelium, we first isolated neutrophils by using Ficoll-Paque density gradient centrifugation. We observed that the cytoplasm was light red, the nuclei was rod-like or 2∼5 lobulated, and the purity of cells ranged from 92 to 96% (Fig. 2B). The results indicated that we have successfully isolated neutrophils and the isolated neutrophils could be used for experiments. Then, we cultured HUVEC in the presence of HC-exo or MPA-exo and subsequently incubated the HUVEC monolayer with primed LPS-stimulated neutrophils. As shown in Figs. 2C and 2D, treatment with MPA-exo enhanced the adhesion of neutrophils to HUVEC, and the number of neutrophils adhering to HUVEC was increased three times that in the HC-exo group according to the ratio of neutrophil/HUVCE (HC-exo: 0.12 ± 0.03; MPA-exo: 0.24 ± 0.01; p = 0.002). Our results indicated that MPA-exo treatment significantly increased proinflammatory cytokine release in HUVEC and neutrophil adhesion to HUVEC.

The miRNA expression profiles of plasma exosomes in MPA patients

To characterize whether the miRNA in MPA-exo play an important role in the development of MPA, we used RNA sequencing to explore the miRNA spectrum between HC and MPA patients. We screened a total of 1,077 miRNA in exosomes purified from plasma samples. According to the criteria of a p value < 0.05 and a log₂—(fold change)—>2, we identified nine differentially expressed miRNA. Of these, six miRNA were upregulated and three miRNA were downregulated in MPA (Table 3). The volcano map displays the overall distribution of differentially expressed miRNA (Fig. 3A). Moreover, we randomly selected three differentially expressed exosomal miRNA (miR-1303, miR-129-1-3p, and miR-1287-5p) for qRT–PCR to validate the sequencing data. We normalized the expression levels of these miRNA to that of cel-miR-39 as a control. The mRNA expressions of miR-1287-5p and miR-1303 were significantly upregulated in MPA samples compared with HC samples, whereas that of miR-129-1-3p was significantly downregulated (p < 0.05, Fig. 3B).

MiR-1287-5p enhanced HUVEC inflammatory response and neutrophil adhesion

For further study, we first examined the basal expression of miR-1303, miR-129-1-3p and miR-1287-5p in HUVEC (Fig. 4A). We found that the basic expression of miR-1303 was relatively high in HUVEC, which indicated that exogenous exosomal miR-1303 had difficulty affecting the endogenous expression level of miR-1303 in HUVEC. The basal expression of miR-129-1-3p in HUVEC and the expression of miR-129-1-3p in plasma-derived exosomes (HC-exo and MPA-exo) were not high, which also suggested that exogenous exosomal miR-129-1-3p and the miR-129-1-3p inhibitor had little influence on the expression of miR-129-1-3p in HUVEC. Therefore, we tested only the changes in miR-1287-5p expression after the uptake of exosomes by HUVEC after 24 h, and found that the expression levels of miR-1287-5p increased in the HUVEC (p < 0.05, Fig. 4B). Subsequently, we determined the effects of miR-1287-5p on HUVEC’s inflammatory response and neutrophil adhesion. The qRT–PCR results shown in Fig. 4C suggested that the miR-1287-5p mimic predominantly increased the mRNA expression of IL-6, IL-8 and MCP-1 (fold changes: 2.29, 1.99, and 2.50, respectively), while the miR-1287-5p inhibitor significantly inhibited the mRNA expression of these factors compared to that in the control group (fold changes: 0.50, 0.36 and 0.31, respectively, (p < 0.05); we obtained marginal evidence for MCP-1 (p = 0.081). We also investigated the mRNA expression of adhesion molecules by qRT–PCR. In comparison with the control group, ICAM-1 and E-selectin mRNA levels were markedly elevated after treatment with miR-1287-5p mimic (fold changes: 2.36 and 1.17), but were reduced after treatment with the miR-1287-5p inhibitor (fold changes: 0.53 and 0.26, (p < 0.05, Fig. 4C).

The neutrophil adhesion assay results shown in Fig. 5A revealed that miR-1287-5p mimic promoted the adhesion of primed neutrophils to the HUVEC, which was suppressed by the miR-1287-5p inhibitor (Fig. 5B, (p < 0.05). Similarly, compared with MNC group, the number of neutrophils adhering to HUVEC increased 2.36 times in miR-1287-5p mimic group according to the ratio of neutrophil/HUVEC (MNC: 0.5 ± 0.14; mimic: 1.18 ± 0.14; p = 0.004); while the ratio of neutrophil/HUVEC in miR-1287-5p inhibitor group was decreased in contrast to INC group (INC: 0.53 ± 0.12; inhibitor: 0.31 ± 0.04, p = 0.036).

CBL is a target gene of miR-1287-5p

We next investigated the downstream mechanism by which plasma-derived exosomal miR-1287-5p induces HUVEC inflammatory response and neutrophil adhesion. We predicted a binding site in CBL-mRNA for miR-1287-5p using TargetScan (Fig. 6A). We constructed CBL Wt and CBL Mut in OBiO Technology (Shanghai, China, Fig. 6B) and used them to transfect HEK393T cells together with miR-1287-5p mimic or MNC. The results of the dual luciferase reporter assay suggested that cotransfection with miR-1287-5p mimic decreased the luciferase activity of CBL Wt (p < 0.001), but not that of CBL Mut (p = 0.0031, Fig. 6C). Subsequently, we determined the CBL-mRNA levels using qRT–PCR after transfecting HUVEC with a miR-1287-5p mimic or a miR-1287-5p inhibitor for 24 h. As shown in Fig. 6D, CBL-mRNA was downregulated in HUVEC treated with the miR-1287-5p mimic (p < 0.05). The CBL-mRNA level in HUVEC increased by 1.35 times after transfection with the miR-1287-5p inhibitor, but the difference was not significant (p = 0.058). Western blot analysis also demonstrated that HUVEC treated with miR-1287-5p mimic for 48 h exhibited marked reductions in CBL protein expression; however, HUVEC treated with miR-1287-5p inhibitor exhibited upregulation of CBL protein expression (Fig. 6E). In brief, these results indicated that CBL-3′ UTR-mRNA contains a binding site targeted by miR-1287-5p.

Downregulation of CBL promoted inflammatory response and neutrophil adhesion in HUVEC

To determine whether MPA-exo affect CBL protein expression in HUVEC, we cultured HUVEC in the presence of NC-exo or MPA-exo for 48 h. As shown in Fig. 7A, treatment with MPA-exo decreased CBL protein expression in HUVEC (p < 0.05). Subsequently, to explore the effect of CBL gene on inflammation and neutrophil adhesion in HUVEC, we used siRNA (si-001, si-002, si-003) to silence CBL-mRNA expression. Given the results of both qRT–PCR and Western blot, the most efficient siRNA for CBL silencing was si-003 (Figs. 7B&7C). HUVEC transfected with si-CBL (si-003) exhibited significantly increased the mRNA expression of IL-6, IL-8, MCP-1, ICAM-1 and E-selectin after 48 h compared with that in the si-NC group (fold changes: 1.44, 3.44, 2.14, 3.87, and 5.03, respectively, p < 0.05, Fig. 7D). Blocking CBL protein expression also contributed to neutrophil adhesion on HUVEC, the ratio of neutrophil/HUVEC was increased in si-003 group compared with si-NC group (si-NC: 0.55 ± 0.06; si-003: 1.66 ± 0.14; p < 0.05, Figs. 8A&8B). In addition, the dysfunction induced by si-CBL was reversed by the miR-1287-5p inhibitor. Compared with si-003 group, the mRNA expression of IL-6, IL-8, MCP-1, ICAM-1 and E-selectin was decreased in si-003+miR-1287-5p inhibitor group (fold changes: 0.74, 0.43, 0.71, 0.39, and 0.28, respectively, p < 0.05). The ratio of neutrophil/HUVEC in si-003 group and si-003+miR-1287-5p inhibitor group was 1.66 and 0.48, respectively. Taken together, the results revealed that CBL protein inhibition was associated with an increased inflammatory response and increased neutrophil adhesion.