Mesenchymal stem cell-derived exosomes protect against liver fibrosis via delivering miR-148a to target KLF6/STAT3 pathway in macrophages

Serum samples were collected from 24 patients with liver cirrhosis and 12 healthy controls at the Xijing Hospital of Digestive Disease. The diagnosis of cirrhosis followed the American Association for the Study of Liver Disease (AASLD) guideline [19]. The aspartate aminotransferase (AST)- to platelet ratio index (APRI) score [20] and Fibrosis-4 (FIB-4) score [21] were calculated according to relevant formulas. Informed consent was obtained from all participants, and the study protocol was approved by the Ethics Committee of Xijing Hospital. The detail information of patients is listed in Additional file 1: Table S1.

The male C57BL/6 J mice (6–8 weeks) were brought from and cared in animal center of The Fourth Military Medical University. The animal study protocol was approved by the Animal Welfare and Ethics Committee of the Fourth Military Medical University and performed according to the "Guidelines for the Care and Use of Laboratory Animals". The liver fibrosis model was established via intraperitoneal injection of 0.2 ml/20 g 20% (v/v) of CCL4 for 8 weeks twice a week. After then, the fibrotic mice were randomly into groups (n = 4–6) which received treatments as mentioned below via tail vein injection. The mice were continually injected with CCL4 during the observation period. After 2 weeks, mice were killed, and samples were collected for further analyses.

To investigate the role of MSC-EXOs in the liver fibrosis, MSC were pre-treated with 10 μM GW4869 (Sigma Aldrich) for 24 h to inhibit exosomes release. Mice were then randomly assigned to different groups which received PBS, 1 × 10⁶ MSCs or pre-treated MSCs (PBS group, MSC group and MSC + GW4869 group). For exosome treatment, 150 μg of EXOs diluted in 150μL PBS was injected intravenously to mice via tail vein as a single dose. The control group was injected with an equal volume of PBS. In addition, we further expanded the treatment groups, that is, the fibrotic mice were treated with EXOs derived from MSCs transfected with miR-148a-3p mimics (EXO-148 m group) or inhibitors (EXO-148i group). For miRNA agomir injection, mice were injected with negative control (NC) agomir (NC-agomir) or miR-148a agomir (148a-agomir group) via tail vein injection at a dose of 5 nmol for 2 weeks, once per week.

Human umbilical cord-derived MSCs (hUC-MSCs) were provided by the National Engineering Research Center (Tianjin AmCellGene Engineering Co., Ltd, China). The isolated MSCs were cultured with mesenchymal basal medium (Dakewe Biotech Co., Ltd. China) supplemented with a serum-free replacement at 37 °C in a 5% CO2 incubator. Cells in passage 4–7 were used for subsequent experiments. The positive (CD29, CD44, CD90 and CD105) and negative (CD34, CD45) cell surface markers of MSCs were identified by flow cytometer. Meanwhile, the multilineage differentiation potential of MSCs was tested in vitro for adipogenesis, osteogenesis as well as chondrogenesis (HUXUC-90021, HUXUC-90031, HUXUC-9004, Guangzhou, China).

RAW264.7 cells purchased from American Type Culture Collection (ATCC) were cultured in Dulbecco's modified Eagle's medium (DMEM, Gibco, NY) containing 10% fetal bovine serum (FBS, Gibco, NY) and 1% penicillin–streptomycin. Bone morrow-derived macrophages (BMDM) were generated from C57 BL/6 mice (6–8 week) as previously described [22]. In brief, bone marrow cells were harvested from the femur and tibia. Then, these cells were seeded at a density of 2 × 10⁶ cells/mL and cultured in DMEM supplemented with 10% fetal bovine serum and 1% penicillin–streptomycin plus 20 ng/mL macrophage colony-stimulating factor (M-CSF, PeproTech Inc. USA) for 7 days.

EXOs of hUC-MSCs were isolated and purified by differential centrifugation as previously described [23]. Briefly, when the cells have reached 70–80% confluence, the medium was replaced with the exosome-depleted medium (Umibio, Shanghai, China), and the cell supernatant was collected after additional 48 h. The cells and apoptotic bodies were removed after centrifugation at 300 g for 10 min and 2000 g for 20 min, respectively. The sample was then centrifuged at 10,000 g for 30 min to remove cell debris. Finally, the supernatant was centrifuged at 100,000 g at 4 °C for 70 min and additionally washed with PBS at 100,000 × g for 70 min to obtain exosome pellets. The EXOs were characterized by transmission electron microscopy (TEM), and their morphology was observed. Nanoparticle tracking analysis (NTA) was used to analyze the size distribution of EXOs and the concentration of nanoparticles. Flow cytometry was employed to detect exosome surface markers CD9 and CD81 (BioLegend, San Diego). Specific exosome markers (CD9, CD81 and TSG101) were used as a positive control, whereas GM130 was used as a negative control for western blot.

To detect the uptake of MSC-EXOs by macrophages, EXOs were incubated with 1 μM PKH26 (Sigma-Aldrich, MA, USA) for 10 min at 37 °C followed by centrifuged at 100,000 g for 70 min to remove unbounded dye. The labeled EXOs were then co-cultured with RAW 264.7 or BMDM for 6 h. After treatment, cells were washed twice with PBS and fixed with 4% paraformaldehyde. Following DAPI staining, the cells were observed under a fluorescence microscope.

For in vivo exosome tracking, 1 μM DiR (Invitrogen, MA, USA) solution was used to label EXOs at 37 °C for 30 min. Free dyes were removed by ultracentrifugation as mentioned above. The supernatant was collected as the control. Mice that received labeled EXOs or supernatant via tail-vein injection were killed at 24 h. The distribution of fluorescence in the whole body and relative organs (heart, liver, spleen, lungs and kidneys) was detected using small animal imaging system (IVIS Kinetics, Caliper Life Science). Signal intensity was quantified and processed using the Living Image software (V5.0, Caliper Life Science).

For phenotype induction, the M1 (classically activated) and M2 (alternatively activated) phenotypes were established by stimulating RAW 264.7 cells or BMDMs with LPS (100 ng/mL)/IFN-γ (20 ng/mL) and IL-4 (20 ng/mL) for 24 h, respectively. To investigate the effects of hUC-MSC-derived EXOs on macrophage polarization, macrophages were treated with PBS or EXOs for 30 min prior to the M1 induction. Then, the cells were collected for western blot, real-time PCR or flow cytometry analysis.

The mimics and inhibitors of miR-148a-3p, NC mimics or inhibitor were purchased from Guangzhou RiboBio Co., Ltd. KLF6 siRNA and its NC siRNA were purchased from Sangon Biotech Co., Ltd. Cells were transfected using the riboFECT CP Transfection Kit (RiboBio) according to the manufacturer's protocol. For the luciferase reporter assay, the 3′-UTR sequence of KLF6 (wild type) containing the possible sites binding with miR-148a-3p and the corresponding mutated 3′-UTR sequence were cloned and inserted into the pMIR-REPORT vector to generate luciferase reporter constructs. RAW264.7 cells were co-transfected with wild-type or mutated luciferase vector together with miR-148a-3p mimics and negative controls. The luciferase activity was measured using a luciferase reporter assay system (Promega, WI, USA).

The RNA-sequencing technology was provided by LC-Bio Co., Ltd. In brief, total RNA was extracted from MSC-EXOs using the TRIzol reagent. The purity and quantity of RNA were checked using a Bioanalyzer 2100 (Agilent, CA, USA). Small-RNA libraries were constructed using the TruSeq Small RNA Sample Prep Kits (Illumina, San Diego, CA USA) and sequenced on the Illumina HiSeq 2500 platform. Then, we downloaded the public data (GSE159814) from Gene Expression Omnibus (GEO) database and obtained the intersections with our sequencing data for further analysis. The potential target genes of miR-148a-3p were obtained through examination of the overlapped intersection from four databases (miRanda [24], miRDB [25], TargetScan [26] and CLIP [27]). In addition, to further investigate the signaling pathways that KLF6 was involved in, we performed pathway analysis using gene set enrichment analysis (GSEA) and gene set variation analysis (GSVA). Data were retrieved from GSE13476 in the GEO platform. A P value < 0.05 was considered statistically significant.

For quantification of macrophage infiltration, single-cell suspensions from liver tissues were treated with anti-mouse FcR blocking reagent and stained with mixed fluorescence-conjugated antibodies. The detail information of antibodies for flow cytometric is listed in Additional file: Table S2. Flow cytometric data were acquired on a FACSVerse flow cytometer (BD Bioscience, CA, USA) and analyzed with FlowJo software (TreeStar, Ashland, OR, US). Gating strategies are depicted in Additional file: Fig. S1. 1 1

The serum of mice was obtained at each time point. The levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and albumin (ALB) were analyzed on an automatic biochemistry analyzer in Xijing Hospital. Tissue samples were prepared as paraffin-embedded sections by YiKE Biotechnology, China. Then, sections were stained with hematoxylin and eosin (HE) for routine histological examination or with Sirius Red and Masson for fibrosis evaluation. Further quantification and analysis of collagen fiber were assessed using Image-Pro Plus software (v6.0, Media Cybernetics Inc.).

For immunohistochemical analyses, the sections were dewaxed and rehydrated by xylol and alcohol, respectively. Antigen retrieval was conducted by sodium citrate in microwave condition. After then, 1% H₂O₂ was used to block endogenous peroxidase activity. The slides were further blocked with goat serum for 30 min. The antibodies of anti-iNOS (AB15323, 1:1000, Abcam, MA) and anti-CD206 (AB64693, 1:1000; Abcam, MA) were incubated with the slides at 4 °C overnight. On the next day, after washing with PBS, the secondary antibody was added and diaminobenzidine (DAB) was applied to visualize the image. Further quantification and analysis of positive area were assessed using Image-Pro Plus software. (v6.0, Media Cybernetics Inc.)

Liver tissue samples were fixed with 4% paraformaldehyde for 30 min and washed with PBS. After antigen retrieval with sodium citrate in microwave condition, the sections were blocked with goat serum for 30 min. Then, primary anti-iNOS antibody (AB15323, 1:200, Abcam), anti-CD206 antibody (AB64693, 1:200, Abcam), anti-F4/80 antibody (AB6640,1:100, Abcam), anti-αSMA antibody (A2547, 1:500, Sigma) and anti-Col1a1 (AB270993,1:100, Abcam) were added and incubated with tissue sections at 4 °C overnight. On the next day, the sections were washed three times with PBS and incubated with secondary antibody labeled with Alexa Fluor 488 or Alexa Fluor 647 (Invitrogen Inc.) at room temperature for 1 h. After DAPI staining of cell nucleus, sections were mounted using anti-fade solution. Later, images were acquired by a laser scanning confocal microscope. Further analysis of positive percentage cells was conducted by Image-Pro Plus software (v6.0, Media Cybernetics Inc, Bethesda, MD).

Total RNA was extracted by RNAeasy Plus kit (TaKaRa Biotechnology Co., Ltd., Dalian, China), and reverse transcription was performed using PrimeScript™ RT Master Mix (RR036A, Takara, Tokyo). Then, amplification was conducted by TB Green Premix Ex Taq II (DRR820A, Takara, Tokyo) on a CFX96 Touch™ real-time PCR System (Bio-Rad, CA). PCR primers are shown in Additional file: Table S3. 1

Proteins of cells or tissue samples were obtained by RIPA lysis buffer (Beyotime biotechnology, China), which was added in proteinase inhibitors and phosphatase inhibitors (Roche, Basel, Switzerland). The Bradford method was applied to quantify the protein samples. After then, 30 μg of protein was loaded to SDS-PAGE before transferring to nitrocellulose membranes (Bio-Rad Biotechnology, America). The membranes were blocked by TBST buffer containing 2.5% skim milk for 30 min. Then, the membranes were incubated with primary antibody at 4 °C overnight. Next day, the peroxidase-conjugated secondary antibody was added to incubate the membrane, and blot images were acquired by an enhanced chemiluminescence kit.

The data were in the expression of mean values ± standard deviation. One-way analysis of variance and t-test were performed to identify the significant differences. A P value < 0.05 was considered significant. Statistical analysis was plotted by GraphPad Prism 7.0 (GraphPad Software, CA, USA).

Finally, to test whether miR-148a is essential for MSC-EXOs to regulate macrophage polarization, we treated M1 macrophages with MSC-EXOs or miR-148a inhibitor MSC-EXOs and then detected relevant markers of macrophage polarization. The results showed that MSC-EXOs treatment led to the downregulation of M1 markers (iNOS, TNF-α, IL-6) and upregulation of M2 markers (IL-10, CD206, Arg-1). However, these beneficial effects could be partially negated by miR-148a inhibitor MSC-EXOs (Additional file: Fig. S4). From these results, we could conclude that miR-148a, serving as an important effector in MSC-EXOs, plays a key role in the immunomodulation of macrophage phenotype. 1

To further define the molecular mechanism underlying the induction of inflammatory response by KLF6, we mined a GEO dataset containing two sets of sequencing data of BMDMs from Lyz2^cre and Klf6^fl/fl:Lyz2^cre mice. The heatmap showed the gene expression pattern of two groups (Additional file 1: Fig. S5A). GSEA and GSVA were conducted to reveal significant differential gene sets between two groups in response to inflammatory agent stimuli (Additional file 1: Fig. S5B and C). We then identified a total of 16 significantly enriched pathways commonly overlapping in both GSEA and GSVA (Additional file 1: Fig. S5D). Previous studies have reported that JAK/STAT3 and PI3K/AKT signaling pathways were associated with the macrophage polarization. So, the protein expression levels of p-STAT3 and p-AKT were detected by western blot in M1 macrophages. As shown in Fig. 9M, overexpression of miR-148a or knockdown of KLF6 reduced the expression of p-STAT3 but did not affect p-AKT. Together, these findings suggested that miR-148a modulated the polarization of macrophage by directly targeting KLF6 to regulate JAK/STAT3 signaling pathways.

Additional file 1. Supplementary figures and tables.