Human umbilical cord mesenchymal stem cells facilitate the up‐regulation of miR‐153‐3p, whereby attenuating MGO‐induced peritoneal fibrosis in rats

Abstract MiRNAs contribute greatly to epithelial to mesenchymal transition (EMT) of peritoneal mesothelial cells (PMCs), which is a crucial step in peritoneal fibrosis (PF). In this study, we tried to profile whether miRNA expression differences exist after human umbilical cord mesenchymal stem cells (hUCMSCs) treatment in PF rats and investigate the possible role of miR‐153‐3p involved in anti‐EMT process. We randomly assigned 34 rats into three groups: control group (Group Control), MGO‐induced PF rats (Group MGO) and hUCMSCs‐treated rats (Group MGO + hUCMSCs). MiRNA microarrays and real‐time PCR analyses were conducted in three groups. α‐SMA, Snail1 and E‐cadherin expression were detected by Western blot. Luciferase reporter assays were used to detect the effects of miR‐153‐3p overexpression on Snai1 in rat peritoneal mesothelial cells (RPMCs). We identified differentially expressed miRNAs related to EMT, in which miR‐153‐3p demonstrated the greatest increase in Group MGO + hUCMSCs. Transient cotransfection of miR‐153‐3p mimics with luciferase expression plasmids resulted in a significant repression of Snai1 3′‐untranslated region luciferase activity in RPMCs. These studies suggest that miR‐153‐3p is a critical molecule in anti‐EMT effects of hUCMSCs in MGO‐induced PF rats. MiR‐153‐3p might exert its beneficial effect through directly targeting Snai1.

angiogenesis. 2,3 Peritoneal fibrosis (PF), which is lack of effective prevention and control countermeasures, is one of the most important factors that force patients with ESRD to interrupt from longterm PD, hence restricts the application and development of PD. 4 How to delay or even block PF process has become an urgent issue in research field.
Mesenchymal stem cells (MSCs) from the early development of mesoderm and ectoderm belong to the pluripotent stem cells, which have been suggested as promising candidates for various cell-based therapeutic approaches including treatments for PF. 5 Human umbilical cord mesenchymal stem cells (hUCMSCs) are one of the typical MSCs, which have advantages such as low immunogenicity, painless collection procedure, faster expansion in vitro and ethical access compared to MSCs from other sources. 6 Therefore, the administration of hUCMSCs might be considered as a potential strategy for the intervention point of PF. Despite this potential, the mechanisms and signalling pathways involved in hUCMSCs-mediated PF repair are poorly understood.
MicroRNAs (miRNAs) are a group of endogenous, non-coding small RNAs that have recently been implicated in the regulation of gene expression on the post-transcriptional level in multiple biological processes. 7 Recently, several reports have demonstrated the association between miRNAs and PF, which indicated the emerging roles of miRNAs in the pathogenesis of PF. 8,9 However, further studies are needed to determine whether and which miRNA is involved in hUCMSCs-mediated PF healing in rats.
In this study, we tested the protective effects of hUCMSCs treatment on methylglyoxal (MGO)-induced PF in rats and tried to profile whether differences existed in peritoneal miRNA expression of control rats, MGO-induced PF rats and hUCMSCs-treated rats.
We hypothesized that the greatest increased miR-153 is overlapped among the three groups contributed to the attenuation of MGOinduced PF in rats.

| Animals
A total of 34 male Wistar rats, 180-220 g body weight, were purchased from Center of Radiation Medicine of Peking Union Medical College. The peritoneal dialysate fluid (PDF) with 20 mmol/L MGO was prepared by adding MGO to 2.5% PDF (Baxter, Guangzhou, China) and free of endotoxin for in vivo delivery. Twenty-four rats were injected intraperitoneally daily with 100 mL/kg PDF with 20 mmol/L MGO. The rest rats, which also underwent intraperitoneal injection, were prepared as control treated with normal saline (Group Control, n = 10). Two weeks after injection, intraperitonealinjected rats were randomly divided into two groups. One group received 2 9 10 6 hUCMSCs infusion via a tail vein only one time to intervent PF (Group MGO + hUCMSCs, n = 12). Another group was given only normal saline instead of hUCMSCs (Group MGO, n = 12).
Two weeks after hUCMSCs administration, rats were anaesthetized by intraperitoneal injection with sodium pentobarbital 50 mg/kg body weight. After that, 20 mL of 2.5% PDF was intraperitoneally injected and kept in peritoneal cavity for 2 hour. All dialysate in the peritoneal cavity was carefully collected through a midline incision. Blood sample was obtained by a direct cardiac puncture. Plasma creatinine concentration (P), dialysate creatinine concentration (D), dialysate fluid glucose concentration (D2) and PDF glucose concentration (D0) were determined by automatic biochemical analyser (HITACHI 7170A, Japan). Ultrafiltration volume (UF) was counted as followed: UF = the resulting dialysate (mL) -20(mL). The peritoneum of individual rats was also sampled for histological analysis, gene expression analysis and miRNA microarray analysis. All procedures were carried out in accordance with the approval of the ethics committee of Tianjin Medical University.
2.2 | Isolation, cultivation, expansion, differentiation and identification of hUCMSCs hUCMSCs were isolated and cultured with the tissue block attachment method as described previously. 10 Fifteen fresh human umbilical cords were obtained from healthy mothers at Tianjin Central Hospital of Gynecology Obstetrics following their informed consent, and the experimental procedures were approved by the ethics committee of Tianjin Medical University. hUCMSCs were cultured for expansion in DMEM/F12 supplemented with 10% FBS and 100 U/mL penicillin/streptomycin, at 37°C in a 5% CO 2 incubator. hUCMSCs between the fourth and sixth passages were collected for cell infusion. hUCMSCs (the 6th passages, 1 9 10 6 ) were stained with anti-CD34, CD45, CD90 and CD105 antibodies conducted at room temperature for 30 minutes in dark (1:500; BD Pharmingen, San Diego, CA, USA). Flow cytometry (Gilson, Middleton, WI, USA) was used to analyse antibody binding. Multipotency of hUCMSCs was confirmed by osteogenic and adipogenic differentiation.

| Histological and immunohistochemical analysis
Peritoneum was removed from rats and fixed in 10% phosphate-buffered formalin. The fixed tissue was then embedded in paraffin at 60°C and cut transversely into 2-to 3-lm-thick sections. Some sections were stained with haematoxylin and eosin (HE) to analyse cell type and identify collagen fibres with Sirius Red histochemistry.
Peritoneal thickness was measured using image analysis software Image-Pro Plus 5.0. Thickness was measured at 30 points per site (at 0.5-mm intervals within a range of 1.5 cm), and the average was     (Group Normal hUCMSCs and Group hUCMSCs + TGF-b1).

| Preparation of CM
To study the influence of soluble factors produced by hUCMSCs, CM from the culture of hUCMSCs at passage 2-4 was prepared.
hUCMSCs in 80%-90% confluent were washed three times with PBS and then fed with serum-free medium for 24 hour. The medium was collected, and the cellular debris was removed through centrifugation at 671g for 10 minutes. This supernatant was subsequently used as hUCMSCs-CM.

| Liposome-mediated plasmid transfection
MiR-153-3p mimics, control mimics, miR-153-3p negative control and anti-miR-153-3p molecules were synthesized by GenePharma 2.11 | Statistical analysis SPSS 16.0 software (SPSS Inc., Chicago, IL, USA) was used for the data processing. P < .05 was required for results to be considered statistically significant. Data were presented as Mean AE SEM and were compared by Student's t test or ANOVA followed by a post-SNK q test as appropriate.

| Characterization of cell cultures
After 14 days in culture, scattered spindle-shaped cells were found around the tissue blocks. Cell density gradually reduced from tissue blocks, Figure 1A. Four to 6 days later, cell confluency reached 80%-90% and a swirl colony growth form was represented, Figure 1B. Cells after the first passage were used to detect their surface markers by flow cytometry. The results suggested that CD105 and CD90 were strongly expressed, Figure 1D-E, whereas CD45 and CD34, the specific markers of hematopoietic cell surface, were hardly expressed, Figure 1F-G. Figure 1C was served as isotype control.  Comparing with Group Control, we found that the collagen type I/III ratio was significantly increased in Group MGO (P < .05) by Sirius Red staining. Administration of hUCMSCs significantly decreased the collagen type I/III ratio (P < .05), Figure 2B  Therefore, miR-153-3p was chosen to explore the association between miRNAs and EMT during the repairing process of hUCMSCs on MGO-induced PF rats.

TGF-b1-induced EMT in RPMCs
Stimulation of RPMCs with 2.5 ng/mL TGF-b1 up-regulated a-SMA ( Figure 6A and D) and Snail1 ( Figure 6B and E) expression and down-regulated E-cadherin ( Figure 6C and F) expression as comparing to that of quiescent cells (both P < .05), which was inhibited by pre-treatment with hUCMSCs-CM and coculture with hUCMSCs (P < .05).

| 3 0 -UTR of Snai1 is the direct target of miR-153-3p
Computational microRNA target predictive database (TargetScan 7.0) and luciferase reporter assay were used to examine the relationship between miR-153-3p and Snai1. As shown in Figure 7, RPMCs Patients undergoing long-term PD develop increased peritoneal thickness thus PF characterized by uncontrolled deposition of ECM. 14 It has been proposed that collagen type I/III ratio is an important marker of ECM overexpression and accumulation. 15 The current data indicate that MSCs represent a promising candidate in direct antifibrotic treatment of various fibrotic diseases. 16,17 In our present study, after treatment of hUCMSCs, peritoneal thickness and type I/III collagen ratio were remarkedly decreased in MGO-induced PF rats. Epithelial to mesenchymal transition of mesothelial cells plays an important role in the induction of fibrosis and early event during PD, which leads to PF. 18   and SOX4 during PD. 22 MiR-30a negatively regulated Snai1mediated EMT during PD in vitro and in vivo. 9 In this study, we discovered that, after hUCMSCs treatment in PF rats, miR-153-3p was significantly increased accompanied by decreased EMT markers a-SMA and Snail1 and increased E-cadherin. In vitro study showed that miR-153-3p directly targeted and inhibited Snai1 by binding to its 3 0 -UTR region in RPMCs, which suggested that miR-153-3p acted as a negative modulator during the process of EMT in PD.
TGF-b1 is involved in EMT processes and favours the formation of PF. 23 33 Recent reports have demonstrated that the protective potential of MSCs was mediated by paracrine pathways that involved antioxidants and anti-inflammatory agents. 34,35 Hence, in our present study, the mechanism of which hUCMSCs exert their beneficial effects might be that hUCMSCs injected intravenously secrete soluble cytokines into the blood to enhance the repair of injured peritoneum by suppressing inflammatory reactions which is independent from hUCMSCs differentiation.
Collectively, these studies suggest that miR-153-3p is a critical molecule in the protective effects of hUCMSCs during PD therapy.
Our findings also suggest that miR-153-3p exerts its beneficial effect perhaps through direct targeting of Snai1. It is anticipated that overexpression of miR-153-3p may represent an effective therapeutic potential for PF and guide the development of therapeutic innovations in PD.

CONF LICT OF I NTEREST
None of the authors have any potential financial conflict of interest related to this manuscript.