Exosomes derived from human umbilical cord blood mesenchymal stem cells improve hepatic ischemia reperfusion injury via delivering miR-1246

ABSTRACT

The purpose of this study was to explore the associated mechanism by which MSCs-derived exosomes exerted protective effect in hepatic ischemia/reperfusion injury (IRI). Human umbilical cord blood mesenchymal stem cells (hUCB-MSCs)-derived exosomes were administrated into LO2 cells exposed to hypoxia/reoxygenation (H/R) and mice subjected to IRI. Cell viability was assessed by CCK-8 assay. Apoptosis was analyzed by flow cytometry and TUNEL staining. The expression of miR-1246 and Wnt/β-catenin pathway-related proteins was detected by quantitative real-time PCR (qRT-PCR) and western blotting. The concentration of pro-inflammatory cytokines was determined by ELISA. Luciferase activity assay was performed to confirm the interaction between miR-1246 and glycogen synthase kinase 3β (GSK3β). Hepatic function was assessed by determining serum alanine amino transferase (ALT) and aspartate amino transferase (AST) levels. Histological changes were observed using hematoxylin-eosin (H&E) staining. MiR-1246 was significantly downregulated in H/R-treated LO2 cells. Treatment with exosomes derived from hUCB-MSCs led to miR-1246 upregulation. Furthermore, hUCB-MSCs-derived exosomes induced anti-apoptotic and pro-survival effects in LO2 cells and ameliorated IRI-induced hepatic dysfunction in mice, while treatment of exosomes from miR-1246 inhibitor-transfected hUCB-MSCs showed opposite effect, which was mediated by regulating GSK3β-Wnt/β-catenin pathway. Collectively, hUCB-MSCs-derived exosomes alleviated hepatic IRI by transporting miR-1246 via regulating GSK3β-mediated Wnt/β-catenin pathway.

Introduction

Ischemia/reperfusion injury (IRI) refers to the pathological process of tissue and organ dysfunction and structural damage caused by the restoration of hemoperfusion after short periods of ischemia. Hepatic IRI is a common pathological condition in liver surgery, which is also one of the leading causes of liver failure in patients who underwent liver transplantation and liver resection [1]. Currently, there are no effective therapeutic strategies for treatment with hepatic IRI.

Wnt signaling pathway is an evolutionarily conserved signaling pathway that plays an important role in the physiological and pathological processes of associated liver diseases, such as liver tumorigenesis, liver fibrosis, and liver cirrhosis [2]. β-catenin is a core component of the canonical Wnt signaling cascade. Under normal conditions, β-catenin binds to the polyprotein complex formed by glycogen synthase kinase-3β (GSK3β), adenomatous polyposis coli (APC), and Axin, leading to phosphorylation of β-catenin-catenin and degradation. The activated Wnt signaling reduces the activity of GSK3β, that contributes to β-catenin accumulation in the nucleus to form nuclear complexes with T-cell factor/lymphoid enhancer-binding factor (TCF/LEF) factors, leading to anti-inflammatory, anti-apoptotic, anti-oxidative stress properties [3]. Compelling evidence has delineated the hepatoprotective role of activated Wnt/β-catenin signaling in hepatic IRI [4,5]. Therefore, in-depth insight into the mechanism of Wnt/β-catenin signaling pathway on hepatic IRI could provide theoretical basis for the development of new drugs.

Mesenchymal stem cells (MSCs) are a subset of adult stem cells with high self-renewal capacity and multidirectional differentiation potentials, which are considered an ideal cell source for cell therapy of liver injury [6]. Early studies have shown that MSCs can in vitro differentiate into hepatocytes induced by growth factors to alleviate liver failure and promote hepatocyte regeneration [7]. Recent studies have demonstrated that MSCs preserve its function of tissue repair and regeneration via a paracrine mechanism. Exosomes, an important paracrine factor for stem cells has been shown to play a key role in the repair of damaged tissues [8]. Nong et al demonstrated that exosomes isolated from human-induced pluripotent stem cell-derived mesenchymal stromal cells attenuated hepatic IRI by inhibiting apoptosis, inflammatory response, and oxidative stress [9]. Moreover, Yan et al reported that MSCs-derived exosomes improved survival rate of carbon tetrachloride-induced mouse model of liver failure and promoted the recovery of liver oxidative injury [10]. To the best of our knowledge, MSCs-derived exosomes carry proteins, functional mRNA and miRNAs, which can be systemically delivered to recipient cells [11]. Recently, Ferguson et al gained a comprehensive view of exosomal miRNAs, and found that miR-1246 was highly expressed in exosomes produced by human bone marrow-derived MSCs [12]. In addition, Yang and colleagues uncovered that miR-1246 was involved in the regulation of Wnt/β-catenin pathway via targeting GSK3β [13]. Here, in the present study, we investigated whether MSCs could ameliorate hepatic IRI via exosome secretion, and whether this effect of MSCs-derived exosomes was mediated by miR-1246 delivery.

Materials and methods

Preparation of hUCB-MSCs

Fresh human umbilical cord blood was obtained following delivery of healthy newborn infants to healthy mothers, with the written consent of their parents. This experiment was approved by the Ethics Committee of The first Affiliated Hospital of Anhui Medical University. Mononuclear cells were obtained using Ficoll-Paque density gradient centrifugation (GE Healthcare, Madison, WI, USA) and cultured in low-glucose Dulbecco’s modified Eagle’s medium (DMEM; Life Technologies, Carlsbad, CA, USA) containing 10% exosome-depleted fetal bovine serum (FBS; Gibco, Carlsbad, CA, USA). The medium was replaced every 2 days.

Flow cytometry

The cultured hUCB-MSCs of the third generation were characterized by flow cytometry. Cells were digested by 0.25% tyrisin (Gibco) and re-suspended at a concentration of 1 × 10⁶ cells/ml. Cells were incubated with mouse anti-human monoclonal antibodies against receptors for hematopoietic cell markers (CD34 and CD45), extracellular matrix (CD29) and MSCs marker CD105 (all from BD Biosciences, San Diego, CA, USA) at room temperature for 30 min in the dark. Flow cytometry was performed using flow cytometer (Beckman Coulter, Brea, CA, USA).

Osteogenic and adipogenic differentiation

We further investigated the multipotent differentiation capacity of hUCB-MSCs. Briefly, hUCB-MSCs at passage 3 were seeded in high-glucose DMEM supplemented with 10% FBS, 1 mM dexamethasone and 50 mM ascorbic acid at 5 × 10³ cells/cm² and cultured with osteogenic and adipogenic differentiation kit (Invitrogen, Carlsbad, CA, USA). After 4 weeks, cells were subjected to Alizarin Red or oil red O staining to visualize calcium deposition in osteocytes and lipid droplets in adipocytes, respectively, as previously described [14].

Isolation and authentication of MSCs-derived exosomes

When hUCB-MSCs reached 70% confluence, the adherent cells were expanded in culture medium for 48 h and collected the culture supernatant. The obtained supernatant was then centrifuged at 20, 000 g for 30 min and 110, 000 g for 80 min. After ultracentrifugation, the supernatant was eventually discarded. The remaining precipitate was re-suspended in PBS and stored at −80°C. Ultrasonic homogenized exosomes (10 µl) were dripped on ultrathin carbon film, and standing for 1 min. After 60 sec, excess fluid was removed with filter paper and the additional of 5 µl polyformaldehyde was subsequently used for fixation. The presence of exosomes was morphologically confirmed by transmission electron microscopy after treatment with glutaraldehyde and phosphotungstic acid dye. Nanoparticle tracking analysis (NTA) was then performed to determine the size of isolated exosomes using a NanoSight LM10 microscope (NanoSight Ltd, Salisbury, UK) with the detection threshold set at 15. Exosomal markers CD9 and CD63 were also detected by western blotting.

Hypoxia-reoxygenation (H/R) induction

Human liver cell line LO2 purchased from American Type Culture Collection (ATCC, Rockville, MD, USA) was cultured in DMEM (Life Technologies) supplemented with 10% FBS (Gibco) and 100 μg/mL penicillin/streptomycin (Life Technologies) at 37°C with 5% CO₂. After that, cells were subjected to hypoxic treatment in an atmosphere containing 95% N₂ and 5% CO₂ at 37°C to mimic ischemia. Reoxygenation was subsequently achieved by placing these cells in a room air atmosphere containing 5% CO₂ and 95% O₂ for another 2 h.

Cell transfection and treatment

HUCB-MSCs were transfected with either inhibitor NC or miR-1246 inhibitor (100 nM) using Lipofectamine 2000 (Invitrogen) following the manufacturer’s instructions. After co-cultured with phosphate-buffered saline (PBS) or Dio-labeled exosomes (10 µg/ml) extracted from transfected hUCB-MSCs for 24 h, LO2 cells labeled with red fluorescent protein mCherry were exposed to H/R treatment. The uptake of exosomes was observed by laser scanning confocal microscope (Leica Microsystems, Wetzlar, Germany).

Luciferase activity assay

The binding sites of miR-1246 on the 3ʹUTR of GSK3β were identified using TargetScan online bioinformatics software (http://www.targetscan.org) and were verified by the dual luciferase reporter assay. LO2 cells were co-transfected with GSK3β recombinant plasmids (GSK3β-WT and GSK3β-Mut) and miR-1246 mimic or mimic NC. The luciferase activity was determined using the dual luciferase assay kit (Promega, Madison, WI, USA) after adding firefly or Renilla luciferase reagents, according to the manufacturer’s instructions.

In vivo experiments

C57BL/6 mice were averagely divided into several subgroups (n = 6 per group), namely, Sham, IRI, PBS, Exo, NCI-Exo and miR-1246I-Exo. To establish the hepatic IRI model, mice were anesthetized with intraperitoneally ketamine at a dose of 120 mg/kg and 1% pentobarbital sodium at a dose of 30 mg/kg. An atraumatic clip was used to interrupt the artery/portal vein blood supply to the left and middle liver lobes for 90 min. The clip was then removed to reperfusion. The sham-operated mice underwent the identical procedure except artery/portal vein occlusion. Mice in the PBS, Exo, NCI-Exo and miR-1246I-Exo group were respectively injected with PBS, hUCB-MSCs-exosomes or exosomes-derived from inhibitor NC/miR-1246 inhibitor-transfected hUCB-MSCs via portal vein immediately after the initiation of reperfusion. The exosomes (2.5 × 10¹² particles for each moue) were suspended in 500 μL PBS. The PBS group was only injected with 500 μL PBS. At last, mice were euthanasized and the blood and liver samples were obtained.

Quantitative real-time PCR (qRT-PCR) analysis

Total RNA was extracted from exosomes and LO2 cells using TRIzol reagent according to the manufacturer’s instructions. Real-time PCR was conducted using SYBR Premix ExTaq^TM (Takara, Shiga, Japan). For the qRT-PCR analysis, all samples were normalized to U6. The mean value in each triplicate was used to calculate relative miR-1246 expression. Expression fold changes were calculated using the 2^−ΔΔCt method. The primers were as follows: miR-1246-F, 5ʹ-TGTATCCTTGAATGGATTTTT; miR-1246-R, 5ʹ- TAACGATCGGATACCTAACT-3ʹ; U6-F, 5ʹ-GTGCTCGCTTCGGCAGCAC-3ʹ; U6-R, 5ʹ- TTTATACCTTGCGAAGTGC-3ʹ.

Cell proliferation and apoptosis detection

LO2 cells were seeded into 96-well plates at a density of 5 × 10³ cells/well. After culture for 48 h, CCK-8 reagent (Dojin Laboratories, Kumamoto, Japan) was added to each well and cells were incubated at 37°C for 4 h. The absorbance at 450 nm was measured to represent cell viability. Each experiment was repeated in triplicate.

For apoptosis detection, LO2 cells were seeded in 24-well plates (2 × 10⁵ cells/well) for 48 h, and stained using Annexin V-FITC apoptosis assay (Invitrogen). The stained cells we reanalyzed by flow cytometry (Beckman Coulter).

Western blot

With the lysis in RIPA buffer (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and centrifugation for 20 min, proteins were extracted from tissue or cells. Bradford method was used to detect the quality of the proteins. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was used to separate the proteins with equal amount, the proteins were then transferred onto PVDF membrane (Millipore, Bedford, MA, USA), with the primary antibodies against Wnt1 (ab15251; 1:500 dilution; Abcam, Cambridge, UK), Wnt3a (ab28472; 1:500 dilution; Abcam), GSK3β (#12456; 1:1000 dilution; Cell Signaling Technology, Boston, MA, USA), and β-catenin (#8480; 1:1000 dilution; Cell Signaling Technology) at 4ºC for 24 h, then the membrane was incubated with the secondary antibodies at room temperature for 2 h. β-actin was used as the internal control.

Cytokines measurement and liver function evaluation

Enzyme-linked immunosorbent assay (ELISA) method was performed to determine the concentration of tumor necrosis factor-α (TNF-α), interleukine (IL)-6 and IL-1β in LO2 cells and hepatic tissues using commercial ELISA kit (Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer’s instructions. Automatic biochemical analyzer (Beckman Coulter AU5800; Beckman Coulter, Fullerton, CA, USA) was applied to examine the serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST).

Hematoxylin and eosin (H&E) staining

Hepatic tissues were fixed in 4% paraformaldehyde, followed by dehydration in an ethanol gradient. After paraffin embedding, the samples were sectioned into 5-µm thickness slices and were processed for H&E staining using kit (Sigma-Aldrich).

Terminal dUTP nick end labeling (TUNEL) staining

The apoptotic liver tissue cells were assessed by TUNEL assay using In Situ Cell Death Detection Kit (Roche Applied Science, Mannheim, Germany). Paraformalde-fixed slices were treated with protease K (10 mM) for 15 min and stained with TUNEL reaction mixture at 37°C for 1h. After rinsing the slides three times, slides were incubated in DAB substrate for 10 min at room temperature and analyzed under light microscopy (Olympus Co., Tokyo, Japan). The apoptotic cells were quantified by a pathologist based on five randomly selected areas (×400 magnification). Apoptosis index was calculated as the percentage of positive nuclei (brown) in sections stained by TUNEL.

Statistical analysis

All data were represented as mean ± standard deviation (SD). Two-sided Student’s t test along with one-way analysis of variance (ANOVA) were used to analyze difference among groups. P values < 0.05 were considered significant.

Results

Morphology and characterization of hUCB-MSCs and exosomes

As shown in Figure 1(a), flow cytometry showed the phenotypes of purified cells were uniformly negative for reactivity to antigens CD34 or CD45 and positive for CD29 and CD105, which were consistent with the characteristics of MSCs, suggesting the existence of hUCB-MSCs. We then examined the property of hUCB-MSCs by inducing osteogenic and adipogenic differentiation. After induction, we observed the appearance of calcium deposits stained by Alizarin Red and intracytoplasmic lipid droplets stained by oil red O (Figure 1(b,c)), suggesting the potential of adipogenic and osteogenic differentiation of the obtained hUCB-MSCs. Transmission electron microscopy showed that that exosomes obtained from hUCB-MSCs were elliptical nanovesicles (Figure 1(d)). NTA showed a size distribution between 0 ~ 200 nm in diameter (Figure 1(e)). Additionally, these harvested particles showed positive for exosomal surface marker proteins CD9 and CD63 (Figure 1(f)). The above properties analysis identified these collected particles as hUCB-MSCs‐derived exosomes.

Figure 1.

Morphology and characterization of hUCB-MSCs and exosomes. (a) Identification of isolated hUCB-MSCs by flow cytometry using antibodies against CD34, CD105, CD29, and CD45; Osteogenic differentiation detected by calcium deposits stained by Alizarin Red (400×); (b) Adipogenic differentiation detected by oil red O staining (400×); (c) Electron microscopic images of hUCB-MSCs-derived exosomes (C); (d) Quantification of hUCB-MSCs-derived exosomes using NTA; (e) Detection of exosomal markers (CD9 and CD63) using western blotting.

Exosomal miR-1246 derived from hUCB-MSCs protected hepatocytes against H/R injury

To investigate the contribution of miR-1246 to the hUCB-MSCs-exosomes effect, we developed miR-1246-deficient exosomes by transfecting hUCB-MSCs with miR-1246 inhibitor followed by exosomes isolation. Successful knockdown of miR-1246 was confirmed by qRT-PCR in hUCB-MSCs (Figure S1(a)) and resultant exosomes (Figure 2(a)). To further evaluate the effect of hUCB-MSCs-exosomes on hepatic I/R injury, we constructed H/R-induced LO2 cell model in vitro to simulate hepatic I/R injury. As shown in Figure 2(b), miR-1246 expression was reduced in H/R-induced LO2 cells, and was upregulated in LO2 cells co-cultured with hUCB-MSCs-exosomes. When LO2 cells were incubated with miR-1246-deficient hUCB-MSCs-exosomes, the mRNA expression of miR-1246 was downregulated in LO2 cells (Figure 2(b)). CCK‐8 proliferation assay showed that treatment with hUCB-MSCs-exosomes effectively rescued cell viability under H/R condition, while the proliferation of LO2 cells was inhibited after co-culture with miR-1246-depletion hUCB-MSCs-exosomes (Figure 2(c)). hUCB-MSCs-exosomes implantation also significantly reduced H/R-induced increase in the percentage of apoptotic cells, whereas, apoptosis was promoted in miR-1246I-Exo (Figure 2(d)). Western blot analysis showed that hUCB-MSCs-exosomes significantly inhibited H/R-induced decrease in the expression of Wnt1, Wnt3a, and β-catenin in LO2 cells via downregulating GSK3β. In LO2 cells, miR-1246 silencing-exosomes led to downregulation of Wnt1, Wnt3a, and β-catenin and upregulation of GSK3β in response to H/R (Figure 2(e)). ELISA results showed that treatment with hUCB-MSCs-exosomes significantly attenuated the increase in pro-inflammatory mediators (TNF-α, IL-6 and IL-1β) in H/R group, while miR-1246 inhibitor showed adverse effect (Figure 2(f–h)). Fluorescence microscope further confirmed that the Dio-labeled exosomes had been taken up and transferred to LO2 cells (Figure 2(i)). Taken together, these results confirmed that the delivery of miR‐1246 in hUCB-MSCs attenuated hepatic H/R damage.

Figure 2.

HUCB-MSCs-derived exosomal miR-1246 derived from hUCB-MSCs protected hepatocytes against H/R injury (a) The exosomal miR-1246 in exosomes derived from hUCB-MSCs transfected with inhibitor NC (NCI-Exo) or miR-1246 inhibitor (miR-1246I-Exo). The relative expression of miR-1246 (b), CCK-8 cell viability assay (c), cell apoptosis detection using flow cytomertry (d), the protein levels of Wnt1, Wnt3a, GSK3β and β-catenin (Ee), the levels of TNF-α (f), IL-6 (g) and IL-1β (h) using ELISA in LO2 cells in the group of Control, H/R, PBS, Exo, NCI-Exo and miR-1246I-Exo. (i) The exosomes were labeled with the lipophilic fluorescent dye DiO (green) and co-incubated with LO2 cells transfected with mCherry plasmid (red). The intake of exosomes by LO2 cells was analyzed under a laser confocal microscope. Scale bar: 10 μm. *P < 0.05 vs. NCI-Exo; ^#P < 0.05 vs. Control; ^&P < 0.05 vs. PBS. n = 3.

miR-1246 activated Wnt/β-catenin signaling pathway via targeting GSK3β in hepatocytes

Since Targetscan software (http://www.targetscan.org) has predicted the interaction between miR-1246 and GSK3β, luciferase reporter assay was performed to explore whether GSK3β was a direct target of miR-1246. The results revealed that the miR-1246 mimic decreased the luciferase activity in GSK3β-WT co-transfected system, conversely, this GSK3β-MUT scarcely responded to miR-1246 mimic, suggesting that the direct binding existed between miR-1246 and GSK3β (Figure 3(a)). We also examine the effect of miR-1246 in the regulation of GSK3β-involved Wnt/β-catenin signaling in LO2 cells. We found that miR-1246 overexpression significantly decreased GSK3β protein levels and upregulated β-catenin expression, while miR-1246 knockdown led to an opposite effect (Figure 3(b)).

Figure 3.

MiR-1246 activated Wnt/β-catenin signaling pathway via targeting GSK3β in hepatocytes. (a) Schematic illustration of the miRNA binding sites of miR-1246 in GSK3β and the direct binding of miR-1246 to GSK3β by dual luciferase reporter assay. (b) The protein levels of Wnt1, Wnt3a, GSK3β and β-catenin in LO2 cells transfected with mimic NC, miR-1246 mimic, inhibitor NC and miR-1246 inhibitor. CCK-8 cell viability assay (c), cell apoptosis detection using flow cytomertry (d), the levels of TNF-α (e), IL-6 (f) and IL-1β (g) using ELISA in LO2 cells in the group of H/R+PBS, H/R+Exo, H/R+Exo+DMSO, H/R+Exo+IWR-1 endo, H/R+Exo+siRNA, H/R+Exo+si-GSK3β. LO2 cells were treated with an inhibitor of the Wnt/β-catenin pathway inhibitor (IWR-1-endo; 5 μM in DMSO)/DMSO or transfected with si-GSK3β/si-Ctrl prior to be exposed to hUCB-MSC-derived exosomes and H/R injury. *P < 0.05 vs. mimic NC; ^#P < 0.05 vs. inhibitor NC; ^&P < 0.05 vs. H/R+PBS; ^$P < 0.05 vs. H/R+Exo+DMSO; ^@P < 0.05 vs. H/R+Exo+siRNA. n = 3.

hUCB-MSCs delivered exogenous miR-1246 via exosomes to alleviate hepatic IRI via GSK3β-mediated Wnt/β-catenin pathway

To further explore the role of Wnt/β-Catenin signaling pathway and GSK3β in the hUCB-MSCs-exosomes-mediated protection of hepatocytes against H/R injury, we pretreated LO2 cells with an inhibitor of the Wnt/β-catenin pathway inhibitor (IWR-1-endo; 5 μM in DMSO)/DMSO or transfected LO2 cells with si-GSK3β/si-Ctrl prior to be exposed to hUCB-MSC-derived exosomes and H/R injury. Data revealed that IWR-1-endo treatment significantly attenuated the hUCB-MSCs-exosomes-mediated protection of hepatocytes against H/R injury (Figure 3(c – g)). In contrast, GSK3β silencing effectively enhanced the hUCB-MSCs-exosomes-mediated protection of hepatocytes against H/R injury (Figure 3(c – g)).

We continued to explore the protective mechanism of hUCB-MSCs-derived exosomes in hepatic IRI. Figure 4(a,b) showed that the increase of serum levels of ALT and AST induced by IRI was dramatically decreased following treatment with hUCB-MSCs-exosomes, which was partly recovered by administration of miR-1246 inhibitor-transfected hUCB-MSCs derived-exosomes. HE staining showed that hUCB-MSCs-exosomes administration alleviated histologic damage, which was aggravated by miR-1246I-Exo injection (Figure 4(c)). To further determine the status of hepatocyte apoptosis, livers were analyzed by TUNEL staining, which revealed that IRI mice showed increased the frequency of TUNEL-positive cells in hepatic tissues, which was abolished by hUCB-MSCs-exosomes implantation. However, miR-1246I-Exo transplantation enhanced apoptosis index (Figure 4(d)). As expected, hUCB-MSCs-exosomes significantly suppressed I/R-induced decrease in the expression of Wnt1, Wnt3a and β-catenin, but led to GSK3β downregulation. Conversely, the administration of miR-1246I-Exo overturned the effect of exosomes on GSK3β-mediated Wnt/β-catenin pathway (Figure 4(e)). Moreover, the higher production of TNF-α, IL-6 and IL-1β induced by IRI were observably reduced in Exo group, while miR-1246I-Exo infusion increased the release of these proinflammatory cytokines in IRI model (Figure 4(f,h)). Collectively, our results demonstrated that exosomal miR-1246 exerted hepatoprotective effect against IRI via direct targeting GSK3β through the activation of Wnt/β-catenin signaling pathway.

Figure 4.

HUCB-MSCs delivered exogenous miR-1246 via exosomes to alleviate hepatic IRI via GSK3β-mediated Wnt/β-catenin pathway. (a and b) The serum levels of ALT, AST at 6, 12 and 24 h post-surgery in C57BL/6 mice in the groups of Sham, IRI, PBS, Exo, NCI-Exo and miR-1246I-Exo. (c and d) hepatic histopathology and hepatocyte apoptosis of different groups using H&E staining and TUNEL assay (original magnifcation 200×); The protein levels of Wnt1, Wnt3a, GSK3β and β-catenin (e), the levels of TNF-α (f), IL-6 (g) and IL-1β (h) in livers obtained from different groups. *P < 0.05 vs. NCI-Exo; ^#P < 0.05 vs. Sham; ^&P < 0.05 vs. PBS. n = 6.

Discussion

In this study, we investigated whether the paracrine mechanism by which MSCs alleviated hepatic IRI was associated with exosomes-mediated miRNAs transfer. Our data indicated that hUCB-MSCs released exosomes that transferred miR-1246 to hepatocytes and subsequently produced protective effect against hepatic IRI via regulating GSK3β-mediated Wnt/β-catenin signaling pathway. These findings suggested that MSCs-derived exosomes are promising candidates for liver-directed cell therapy.

Exosomes are emerged as membranous nanovesicles secreted by almost cell types, especially MSCs. The protective effects of MSCs-derived exosomes have been highlighted by multiple studies in various experimental I/R models. For instance, Liu et al suggested that bone marrow MSCs-derived exosomes could attenuate myocardial IRI by inducing autophagy via AMPK/mTOR and Akt/mTOR pathways [15]. Li et al confirmed that bone marrow MSCs protected against lung IRI by secreting exosomes loaded with miR-21-5p [16]. Wang et al manifested that MSCs ameliorated renal IRI in an exosomal miR-199a-5p-dependent manner [17]. Although a growing body of evidence demonstrated the protection of MSCs against hepatic IRI by secreting exosomes [9,18], the underlying mechanisms have rarely been reported. We showed, in the present study that hUCB-MSCs-derived exosomes H/R exposure-induced proliferation inhibition and apoptosis promotion of hepatocytes in vitro but also attenuate hepatic IRI in vivo by improving liver function parameters, alleviating the degree of damage and necrosis of hepatocytes, decreasing hepatocyte apoptosis and the levels of pro-inflammatory mediators. Previous studies have indicated that MSCs-derived exosomes activated the Wnt/β-catenin pathway to prevent myocardial IRI [19] and carbon tetrachloride-induced liver fibrosis [20] in vivo. Another report showed that Wnt/β-catenin pathway resisted hypoxia-induced oxidative stress and apoptosis in hepatocytes [4]. These studies hinted that Wnt/β-catenin activation might be associated with the MSCs-derived exosomes-mediated protective mechanism. In fact, our study showed that compared with exposure to H/R alone, treatment with hUCB-MSCs-derived exosomes significantly enhanced the protein levels of Wnt1, Wnt3a, and β-catenin in LO2 cells. Moreover, the activation of Wnt/β-catenin in liver tissues was observed after injection with exosomes in an in vivo mice IRI model. The aforementioned findings provided novel insight into the role of exosomes in hepatic IRI.

Abundant studies exist with regard to the importance of exosomal miRNAs in cell-to-cell communication, which highlight their potential for applications in miRNA-based therapeutics. In this study, we focused on exosomal miR-1246 in the mechanism of hUCB-MSCs-derived exosomes in treatment of hepatic IRI, since miR-1246 was significantly enriched in exosomes and was responsible for tumor progression. For example, Li et al breast cancer cells-derived exosomal miR-1246 enhanced the viability, migration and chemotherapy resistance of nonmalignant HMLE cells [21]. Kanlikilicer et al manifested that exosomal miR-1246 promoted ovarian cancer progression via M2-type oncogenic macrophages [22]. Our data confirmed that miR-1246 was significantly downregulated in H/R-treated LO2 cells and was abundant in LO2 cells treated with hUCB-MSCs-derived exosomes. Additionally, miR-1246 has been shown to directly target GSK3β in non-small cell lung cancer cells, which in turn activated Wnt/β-catenin signaling [13]. Consistently, our data indicated direct targeting of GSK3β by exosomal miR-1246 resulted in enhanced β-catenin protein expression and inhibition of apoptosis and inflammation. When incubated with hUCB-MSCs-derived exosomes, hepatocytes could take up the exosomes, and the exosomal miR-1246 could further exert its potential effect. Consistent with the results from in vitro studies, the effects of miR-1246 in MSC-exosomes on hepatic IRI were proved by in vivo experiments. Infusion of exosomes from hUCB-MSCs transfected with miR-1246 inhibitor reversed the functional and histological protection of exosomes.

To conclude, the present study for the first time demonstrated that delivery of hUCB-MSCs-derived exosomes moderated hepatic IRI via transfer of miR-1246 targeting GSK3β/Wnt/β-catenin pathway, which might represent a therapeutic tool in acute liver injury.

Acknowledgments

Thanks to Qiang Fang, Xue Zhang and Guodong Liu for their help in the experiment.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemented data of this article can be accessed here.