Differentially Expressed microRNAs in Peritoneal Dialysis Effluent-Derived Exosomes from the Patients with Ultrafiltration Failure

Background Ultrafiltration failure remains one of the most severe complications of long-term peritoneal dialysis (PD), which results in death. This study aimed to characterize the circulating exosomal microRNA (miRNA) profiles associated with ultrafiltration failure and explore its underlying mechanisms. Methods Exosomes were isolated from the peritoneal dialysis effluent (PDE) of patients with ultrafiltration failure or success using the ultracentrifugation method, and then transmission electron microscopy (TEM), nanoparticle tracking analysis (NTA), and western blot were used for exosome characterization. After that, the isolated exosomes were sent for small RNA sequencing, and eight differentially expressed miRNAs (DE-miRNAs) were chosen for RT-qPCR validation. Results TEM, NTA, and western blot revealed that exosomes were successfully isolated. After sequencing, 70 DE-miRNAs involved in ultrafiltration were identified, including 41 upregulated ones and 29 downregulated ones. Functional analyses revealed that these DE-miRNAs were significantly enriched in pathways of cancer, ubiquitin-mediated proteolysis, axon orientation, and the Rap1 and Ras signaling pathways. In addition, the consistency rate of RT-qPCR and sequencing results was 75%, which indicated the relatively high reliability of the sequencing data. Conclusions Our findings implied that these DE-miRNAs may be potential biomarkers of ultrafiltration failure, which would help us to discover novel therapeutic targets/pathways for ultrafiltration failure in patients with end-stage renal disease.


Introduction
Chronic kidney disease (CKD), which can result in the gradual loss of kidney function, is the leading cause of endstage renal disease (ESRD). Peritoneal dialysis (PD) is one of the major forms of renal replacement therapy for patients with ESRD. Currently, CKD affects approximately 700 million people worldwide, and individuals with late stages of CKD are at a high risk of developing kidney function failure, requiring dialysis or transplantation [1]. In many countries, increasing evidence have demonstrated that the outcomes of patients undergoing PD are comparable to or better than those undergoing hemodialysis [2]. PD is generally more convenient, more cost effective, and preserves the residual renal function better than hemodialysis [3]. However, continuous exposure to glucose in the peritoneal dialysate inevitably induces injury that negatively affects peritoneum function, resulting in ultrafiltration failure and PD discontinuation [4].
Peritoneal fibrosis is the main pathological condition of ultrafiltration failure and is regulated by processes of inflammation, angiogenesis, and epithelial-mesenchymal transition (EMT) [5]. TGFβ1-meditated EMT has been confirmed as the central regulator of organ fibrosis. Most PD solutions contain glucose. Glucose degradation products activate TGFβ1 signaling and downstream profibrotic molecules such as the smad family and snail [6,7]. TGFβ1 increases VEGF-A production in mesothelial cells and fibroblasts, which promote the EMT process and fibroblast proliferation, leading to the expansion of peritoneal fibrosis [8]. Currently, there are no specific biomarkers or wellestablished therapies for peritoneal fibrosis. Peritoneal functional assessment mainly relies on the peritoneal equilibration test (PET). Considering that ultrafiltration failure occurs only in the advanced stages of peritoneal fibrosis, PET data in clinical practice might be available too late to monitor peritoneal function. Although invasive peritoneal biopsy could be used for the pathological diagnosis of PD-related peritoneal injury, the results do not reflect global peritoneal structural changes [9,10]. us, it is highly desirable to establish suitable biomarkers and therapeutic targets for disease progress prediction and to allow patients to improve their survival during long-term PD.
Exosomes serve as key mediators in intercellular communication by transferring diverse components, including DNA, microRNA (miRNA), proteins, and lipids [11]. Cytokine stimulation induces the release of abundant exosomes from original cells and affects peritoneal membrane function, thereby contributing to the development of fibrosis [12]. Recent studies have demonstrated the importance of exosomes in peritoneal dialysis effluent (PDE) samples, and they might be used as the ideal source of biomarkers for PD patients owing to their noninvasiveness and convenience in collection [13][14][15]. Moreover, miRNAs have attracted attention as reliable biomarkers and potential targets for precise therapies. miRNAs are small noncoding RNAs (21-25 nt) that function as major post-transcriptional regulators through interactions with mRNAs and induce their destabilization and translation [16]. Many miRNAs display aberrant expressions in dialysate effluents or serum and are implicated in the activation of profibrotic signaling in PD patients [17]. For instance, miR-302c is observed to be downregulated in peritoneal mesenchymal cells (PMCs) isolated from PDE, and it modulates PD-related fibrosis via the modulation of connective tissue growth factors [18]. miR-21 and miR-589 levels are also dysregulated in PDE, which are correlated with changes in peritoneal transport characteristics [19]. However, few studies report the specific miRNA expression status in PDE exosomes of patients with ultrafiltration failure. A recent study revealed that the miRNAs enclosed in exosomes were more stable than circulating miRNAs as the exosomal membrane structure acts as a barrier against enzyme degradation [20]. is observation provides a basis that exosomal miRNAs might be ideal biomarkers for the diagnosis of certain diseases. erefore, in this study, we aimed to investigate the underlying miRNA-related mechanisms in peritoneal ultrafiltration failure. Exosomes were isolated from the PDE samples of patients with ultrafiltration failure and success and then were submitted for small RNA sequencing. Differentially expressed miRNAs (miRNAs) were screened with the thresholds of |log 2 fold change (FC)| > 1 and P < 0.05, and functional analyses were performed. After that, eight DE-miRNAs, namely, hsa-miR-1273c-p3, hsa-miR-125a-5p, hsa-miR-1277-5p, hsa-miR-132-3p, hsa-miR-296-3p, hsa-miR-25-5p, hsa-miR-155-5p, and hsa-miR-708-5p, were selected for real-time quantitative PCR (RT-qPCR) verification. Our work will improve our understanding of ultrafiltration failure and provide potential targets for the treatment of ultrafiltration failure in patients with end-stage of renal disease. Approximately 500 mL of overnight PDE (abdominal retention >8 h) was collected from all subjects and was immediately processed for exosome separation. General clinical information of patients was recorded, including PDE routine testing, assessment of the residual renal function, peritoneal equilibrium test parameters, residual urine volume, ratio of dialysate creatinine to serum creatinine at 4 h (4 h D/Pcr), a sodium sieving ratio at 1 h (1 h D/Pna), material transport area coefficient of creatinine (MTAC), and ultrafiltration volume (UF) ( Table 1). All the patients diagnosed with primary nephropathy were in accordance with the diagnosis of ESRD and received regular PD replacement therapy for more than 3 months. e exclusion criteria were as follows: incomplete data; historical peritonitis at the time of enrollment or 3 months prior to enrollment; abdominal bleeding or other basic diseases such as neoplastic diseases, liver cirrhosis, serious heart, and lung disease; secondary nephropathy such as diabetic nephropathy, lupus nephritis, ANCA-associated vasculitis, and systemic amyloidosis; and the patients who were taking hormones and immunosuppressants were also excluded in this study. e diagnosis of type I peritoneal ultrafiltration failure was based on the definition of the International Society for Peritoneal Dialysis in the year 2000 [21,22]. e enrolled patients underwent a modified method of PET with 4.25% dextrose fluid. All subjects underwent 4 h of dialysis exchange using 2 L of dialysis solutions at 3.86% glucose. Patients with ultrafiltration capacities of <400 mL and dialysate/serum creatinine concentration ratios (D/Pcr) of >0.81 could be diagnosed with ultrafiltration failure. e possible causes of ultrafiltration failure should be excluded to affect the judgment of the results, such as PD-associated peritonitis, improper adjustment of diet and dialysis management, hyperglycemia, posterior peritoneal leakage, and mechanical failure of the dialysis catheter (drift tube, leakage, etc.). Prior to PET experiments, we pre-exchanged overnight stays of 8-12 h.

Exosome Isolation and Characterization.
e isolation of exosomes by ultracentrifugation was accomplished as previously described [23]. Briefly, the PDE samples were centrifuged at 300g for 10 min. e supernatant was transferred to a new tube and centrifuged at 3, 000g for 15 min, followed by centrifugation at 12, 000g for 30 min to remove cell fragments. After filtration with a 0.22 μm filter, the supernatant was collected and centrifuged at 12, 000g for 60 min to obtain the sediments of the membrane pellets. e pellets were resuspended with PBS and centrifuged at 12, 000g for 70 min. e supernatant was then carefully removed, and the reserved precipitates were resuspended with 200 μL of PBS, that was exosomes. e characterization of the isolated exosomes was performed according to the ISEV suggestions in 2014 [24]. e concentrations of the isolated exosomes were detected using a BCA protein assay kit (Boster, Wuhan, China). ereafter, a transmission electron microscope (TEM) was used to visualize the morphology of the exosomes, and nanoparticle tracking analysis (NTA) was performed to evaluate the size and distribution of the exosomes [25,26]. Finally, the expressions of HSP70, CD63, and CD81, which are the specific markers of exosomes, were detected by western blot with their corresponding antibodies [27].

Small RNA Sequencing and Bioinformatic Analysis.
e isolated exosomes were then sent to Yanzai Biotechnology (Shanghai) Co. Ltd (Shanghai, China) for small RNA sequencing. Firstly, total RNA was extracted from the exosomes using RNAiso Plus reagent (TAKARA, Shiga, Japan), following the manufacturer's instructions. For each 250 μL of exosome sample, 750 μl of Trizol reagent was added for lysis, followed by 200 μL of chloroform separation liquid. en, isopropyl alcohol was added to the upper liquid to precipitate RNA from the exosomal suspension. e precipitates were resuspended with 75% ethanol and centrifuged at 12, 000g at 4°C for 5 min. Finally, the precipitates were dissolved in 20 μl of DEPC, and the purity and concentration of the extracted RNA were assessed by calculating the 260/280 optical density ratio using a microplate reader. After that, the extracted RNA was used for small RNA sequencing.
TruSeq small RNA sample prep kits (Illumina, San Diego, USA) were used for the construction of RNA libraries. Data preprocessing was performed using the ACGT101-miR software (LC Sciences, Texas, USA) [28]. Subsequently, miRNAs in the exosomes from PDE were annotated according to the miRbase database (https://www. mirbase.org/). Based on the criteria of |log 2 FC| > 1 and P < 0.05, DE-miRNAs were screened between the exosomes from the failed ultrafiltration PDE and successful PDE. en, global analysis and hierarchical clustering analyses were performed for these DE-miRNAs. Afterward, the target genes of these screened DE-miRNAs were predicted using the target scan and miRDB software programs. Functional analyses of these DE-miRNAs, including Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, were carried out based on the Database for Annotation, Visualization, and Integrated Discovery (DAVID) [29]. e threshold of the significantly enriched GO terms and KEGG pathways was P < 0.05. e raw data of small RNA sequencing were uploaded to the Gene Expression Omnibus database, and the ID was GSE142819.

Statistical Analysis.
All the data in this study were expressed as mean ± standard deviation. Graphpad Prism 6 (San Diego, CA, USA) and SPSS 22.0 (IBM, Armonk, NY, USA) were utilized for statistical analysis. Student's t-test was used for comparative analysis between the two groups. P < 0.05 was considered as the threshold for statistical significance.

Characterization of the Isolated Exosomes.
Exosomes were isolated from the PDE samples and then characterized by TEM, NTA, and western blot. TEM results showed that the exosomes isolated from the PDE samples exhibited a cup-shaped or round morphology with a diameter of approximately 100 nm (Figure 1(a)). en, NTA analysis revealed that the major peak of the isolated substance was approximately 141 nm (Figure 1(b)). Besides, western blot revealed that the exosome-specific markers HSP70, CD63, and CD81 were expressed in the isolated exosomes (Figure 1(c)). ese results indicated that exosomes were successfully isolated from the PDE samples using the ultracentrifugation method.

Identification of DE-miRNAs between the Exosomes from the Failed Ultrafiltration PDE and Successful PDE.
After small RNA sequencing, a total of 686 miRNAs were annotated in the exosomes from all PDE samples. According to the criteria of |log 2 FC| > 1 and P < 0.05, 70 DE-miRNAs were identified between the exosomes from the failed ultrafiltration PDE and successful PDE, including 41 upregulated miRNAs and 29 downregulated miRNAs (Figures 2(a) and 2(b)). Based on the threshold of P < 0.01, 19 DE-miRNAs were found between the two groups, including 10 downregulated miRNAs and 9 upregulated miRNAs (Figure 2(a)). e hierarchical clustering of these 19 DE-miRNAs is shown in Figure 2(c), which implied that these DE-miRNAs could properly distinguish the failed ultrafiltration PDE-derived exosomes from the successful ultrafiltration PDE-derived exosomes.

GO Terms and KEGG Pathway Analyses.
After that, these DE-miRNAs were submitted to predict the target genes using the target scan and miRDB databases, and 13176 target genes of DE-miRNAs were predicted. en, these genes were used for functional analyses. Functional analyses revealed that these genes were mainly enriched in various GO terms and KEGG pathways. Figures 3(a) and 3(b) show the most enriched GO terms in biological process, molecular function (MF), and cellular component (CC), such as "protein binding," "nucleotide binding," "cytoplasm," "cytosol," "transferase activity," "oxidation-reduction process," and "ATP binding." Additionally, the KEGG pathway enrichment analysis indicated that these genes were also significantly enriched in "ubiquitin-mediated proteolysis," "Rap1 signaling pathway," "pathways in cancer," "PI3K-Akt signaling pathway," "p53 signaling pathway," "axon guidance," and the "Ras signaling pathway." Table 2: e sequences of all primers.

Discussion
Peritoneal ultrafiltration failure is associated with long-term PD and repeated peritoneal inflammation, which is the primary cause of PD withdrawal and is closely related to the poor prognosis of patients. MiRNAs, which serve as important post-transcriptional regulators, play important roles in tissue fibrosis and fibrosis-related diseases [31]. In our study, exosomes were isolated from the PDE of failed ultrafiltration patients and successful ultrafiltration patients and sent for small RNA sequencing. After sequencing, 70 DE-miRNAs were identified, including 41 upregulated miRNAs and 29 downregulated miRNAs. Functional analyses revealed that these DE-miRNAs were significantly enriched in pathways of cancer, ubiquitin-mediated proteolysis, axon orientation, the Rap1 signaling pathway, and the Ras signaling pathway. Finally, RT-qPCR showed that miR-1273c-p-3, miR-1277-5p, miR-132-3p, and miR-125a-5p were upregulated, while miR-296-3p and miR-708-5p were downregulated in the exosomes from the failed ultrafiltration PDE. ese results would provide a basis for the potential mechanisms of exosomes associated with miRNAmediated regulation of ultrafiltration failure in PD. Accumulating studies have shown that patients undergoing PD begin to develop peritoneal structure changes, such as fibrosis and angiogenesis. Previous studies have reported that miRNAs displayed significant effects on peritoneal fibrosis both in vitro PMC assays and rodent models [32].
Similarly, current studies on miR-708-5p and miR-296-3p were also mostly focused on different cancer types. Zhao et al. found that miR-708-5p could promote the EMT process by targeting the ZNF549 gene and activating the PI3K/AKT signaling pathway in the adenocarcinoma cells of the colon [43]. Another study on gliomas demonstrated that the silencing of miR-708 promoted cell growth and EMT transition by activating the SPHK2/AKT/β-catenin pathway [44]. Tian et al. [45] reported that miR-296-3p could inhibit the Wnt/β-catenin pathway by targeting SOX4 and exert antitumor effects in triple-negative breast cancer. Wang et al. demonstrated that miR-296-3p, acted as a tumor suppressor, could inhibit the migration and invasion of nonsmall-cell lung cancer cells by targeting APEX1 and regulating the PI3K/AKT/mTOR signaling pathway {[34] # 1145}. ese indicated that miR-708-5p and miR-296-3p may be closely related to the EMT process and the development of diseases. Furthermore, miR-708 can impact the immunoreaction in  Figure 4: Verification of eight DE-miRNAs (four downregulated and four upregulated, one DE-miRNA failed to be detected due to its low abundance) in the exosomes from the failed ultrafiltration PDE and successful ultrafiltration PDE by RT-qPCR analysis. e levels of miR-125a-5p (a), miR-132-3p (b), miR-1273c-p-3 (c), miR-1277-5p (d), miR-296-3p (e), miR-708-5p (f ), and miR-155-5p (g). * : P < 0.05, compared with the normal PDE-Exos group. 8 Genetics Research the progression of diseases. It was found that the miR-708-5p level was increased in bacteria-infected human macrophages, and miR-708-5p mimics could reduce inflammatory cytokines secretion by targeting TRL4 [46]. Besides, several miRNAs, including miR-708-5p, could exert anti-inflammatory effects on endothelial cells in injured arteries by inactivating the NF-κB pathway [47]. MiR-708 was identified as a negative regulator of TNFα and IL-1β molecules, and exacerbated TNFα and IL-1β signaling was involved in diverse inflammatory diseases [48]. Taken together, it can be inferred that miR-708-5p and miR-296-3p may participate in the regulation of inflammation and the EMT process in peritoneal fibrosis. According to a literature search, our study is the first to report the associated between exosomal miR-1273c-p-3/miR-1277-5p/miR-132-3p and ultrafiltration failure. e specific roles of miR-125a-5p, miR-1273cpc-3, miR-1277-5p, miR-132-3p, miR-296-3p, and miR-708-5p in peritoneal fibrosis require further investigations.
In addition, these DE-miRNAs were significantly enriched in several major pathways, including ubiquitinmediated proteolysis, axon orientation, the Rap1 signaling pathway, and the Ras signaling pathway. e ubiquitin-proteasome system (UPS) was a major regulatory mechanism of intracellular protein degradation, and controlled diverse cellular functions. UPS dysfunction can contribute to the development of cancer, autoimmune diseases, and organ fibrosis [49,50]. In PD mice models, the upregulation of nestin proteins could stimulate peritoneal fibrosis by protecting HIF1-α from proteasomal degradation [51]. UPS participates in the pathogenesis of organ fibrosis mainly by regulating the TGFβ/Smad and Wnt/β-cantenin pathways [52,53], and the increased TGFβ level could induce EMTand activate the STAT3 signaling in human PMCs [54]. Interestingly, axon orientation was also identified as a key pathway in our study. Classical neural guidance molecules have been implicated as regulators of vascular remodeling and vessel navigation, such as netrins, semaphorins, and ephrins [55,56]. In particular, secreted class 3 semaphorins (SEMA3) were confirmed as effective normalizing agents of cancer vasculature [57]. e activation of Ras triggered by GTP binding ultimately leads to mitosis in fibroblasts and epithelial cells via the RAF/MEK/ERK cascade and the PI3K/AKT downstream pathway. Oncogenic Ras could directly induce the upregulation of EMT-activated transcription factors, and promote cell invasion and migration [58]. On the other hand, the phosphorylation of the TGFβ receptor provided docking sites for SH2 domain-containing proteins such as PI3K, GRB2, and SOS, thus linking it to the PI3K and Ras pathways [59,60]. e cooperation of the Ras and TGFβ-non-Smad signaling pathways can induce the EMT process and tissue fibrosis in tumors. Furthermore, Ras-mediated ERK-MAPK activation promoted angiogenic homeostasis by increasing the levels of pro-angiogenic factors in endothelial cells [61].
Rap1, a member of the Ras superfamily, has been reported to participate in the regulation of cell adhesion, polarity, and cellular interactions during fetal development [62]. Previous studies have shown that Rap1 affects cell adhesion and related EMT process by targeting two major factors, cadherin and integrin [63]. e downregulation of miR-708 in ovarian cancer cells resulted in the suppression of Rap1b, thereby impairing integrin-mediated cellular junction formation, migration, and invasion [64]. Additionally, Rap1 could promote VEGFR2 activation in endothelial cells, and induce angiogenesis by regulating integrin αvβ3 [65]. erefore, the abnormal level of miR-708 and aberrant signaling of Rap1 might be vital factors affecting peritoneal structural changes or fibrosis in patients with ultrafiltration failure. Of note, Rap1 has a high sequence identity with Ras, and can competitively bind with Raf1 effectors to attenuate Ras-mediated ERK activation [66]. With these reports, together with our results, it can be speculated that ubiquitin-mediated proteolysis, axon orientation, the Rap1 signaling pathway, and the Ras signaling pathway may be associated with the peritoneal structural changes and ultrafiltration failure in patients with ESRD.
However, this study has certain limitations. First, our conclusions need to be verified in another study with a larger sample size, and more experiments need to be conducted to validate the target genes of these DE-miRNAs through the dual-luciferase reporter gene assay of functional assays. e specific roles of miR-708-5p, miR-1273c-p-3, miR-1277-5p, miR-132-3p, miR-296-3p, and miR-125a-5p in ultrafiltration failure should also be explored, and more evidence is needed to support the identification of these DE-miRNAs as biomarkers used for the diagnosis of ultrafiltration failure. Additionally, the biological functions and key pathways of candidate miRNAs in peritoneal fibrosis will be investigated in subsequent studies and further experimental verifications also need to be performed.

Conclusions
In this study, exosomes were isolated from the PDE of failed and successful ultrafiltration and sent for small RNA sequencing. According to the sequencing results, 70 DE-miRNAs involved in the ultrafiltration failure were identified, including 41 upregulated and 29 downregulated ones. Besides, miR-125a-5p, miR-1273c-pc-3, miR-1277-5p, miR-132-3p, miR-296-3p, and miR-708-5p were found to be associated with peritoneal fibrosis by regulating angiogenesis and EMT. Functional analyses showed that ubiquitin-mediated proteolysis, axon orientation, the Rap1 signaling pathway, and the Ras signaling pathway may be associated with the peritoneal structural changes and ultrafiltration process. Our findings implied that these DE-miRNAs could be used as potential biomarkers of ultrafiltration failure, which would help us to discover novel therapeutic targets/ pathways for ultrafiltration failure in patients with ESRD.