Subjects and sample collection
Consecutive LT patients were included in this prospective study which was conducted over the period from July 2020 to February 2021 at the Liver Transplantation Center of the Beijing Friendship Hospital, Capital Medical University. Patients with malignant tumors, concurrent autoimmune disease, any type of infection, or combined organ transplantation were excluded. The normal control group consisted of donors from living donor LT. To harvest plasma, blood samples were centrifuged at 2000 × g for 10 min at 4°C and 1-ml aliquots from the supernatant was transferred to a fresh 1.5-ml tube and stored at −80°C prior to use. Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood by density-gradient centrifugation using the Lymphocyte Separation Medium (Cat: 7111011, DAKEWE) and were analyzed with use of flow cytometry. Livers from donors as obtained during surgery and liver biopsy tissues of patients were used for histopathological studies. This protocol was approved by the Ethics Committee of the Beijing Friendship Hospital. Written informed consent was obtained from all participants or their guardians for the collection of plasma/liver samples and their clinical data to be used in this study.
Hematoxylin–eosin (HE) staining
Liver tissues were fixed in 10% neutral buffered formalin and embedded in paraffin. Sections (4-μm) were stained with hematoxylin–eosin (HE) for histological examination. The histological classification of hepatic HE staining was graded according to the Banff scheme. A rejection activity index (RAI) was calculated from 3 individual scores (venous endothelial inflammation, bile duct damage and portal inflammation) as performed by a single-blinded pathologist. Acute rejection was defined as an RAI score of ≥3.
Immunohistochemistry (IHC)
The 4-μm sections were deparaffinized and rehydrated, followed by microwave antigen retrieval in sodium citrate buffer (pH 6.0) for 15 min and cooling at room temperature. The slides were incubated in 3% H2O2 for 10 min and blocked with 5% bovine serum albumin for 10 min. Primary antibodies against CD4 (1:200, ab133616; Abcam), CD8α (1:200, ab237709; Abcam), forkhead box (Fox)p3 (1:200, ab22510; Abcam), IL-17A (1:200, ab79056; Abcam), and NLRP3 (1:200; ab214185; Abcam) were added and incubated at 4°C overnight. The sections were stained using a DAB kit (ZLI-9017, ZSGB-BIO, China).
Multiplex immunofluorescent (mIF) and image analysis
After testing all of the markers using chromogenic immunohistochemistry and uniplex IF staining, these markers were validated in liver samples as achieved using mIF staining. Tissue multiplex immunofluorescent (mIF) staining was performed with use of the Opal Polaris 7 Color IHC Detection Kit (Akoya Biosciences). Five-micrometer-thick formalin-fixed, paraffin-embedded tissue sections were baked for 2 h at 60°C before staining. Slides were rehydrated with graded ethanol in deionized water. Antigen retrieval was performed at pH 6 for 20 min at 95°C. The slides were serially stained with primary antibodies, consisting of CD4 (1:400, ab133616; Abcam), CD8 (1:200, ab178089; Abcam), forkhead box (Fox)p3 (1:200, ab215206; Abcam) and IL-17A (1:200, ab79056; Abcam) with incubation times being 1 h per primary antibody. Subsequently, anti-rabbit polymeric horseradish peroxidase (Opal IHC Detection Kit; Akoya Biosciences) was applied as a secondary label with an incubation time of 10 min. Following incubation of the slides for 10 min, antibody signaling was visualized with use of the corresponding Opal Fluorophore (Akoya Biosciences). Slides were mounted with anti-fade mounting medium (P36965; Life Technologies) and stored at 4°C before imaging. Image acquisition was performed using the Vectra Polaris multispectral imaging platform (Akoya Biosciences). Whole slide images were scanned and 5–7 representative regions of interest at ×200 resolution were chosen by the pathologist as multispectral images. Image analysis was performed using the InForm 2.4.8 Image Analysis Software (Akoya Biosciences).
Exosome isolation and characterization
The procedure for isolation of exosomes is shown in Figure 1A. Plasma samples or cell culture supernatants were centrifuged at 3,000 × g for 10 min at 4℃. Supernatants were transferred to a new tube and centrifuged at 90,000 × g for 30 min at 4℃. The supernatant was then filtered (0.22 μm) and the filtered culture solution was centrifuged at 100,000 × g for 150 min at 4℃. The pellets (exosomes) were resuspended in 200 μl phosphate-buffered saline. The concentration of exosomes was determined by analyzing protein concentrations as measured using the BCA Protein Assay Kit (PC0020, Solarbio). Nanoparticle tracking analysis (NTA) with ZetaView (Particle Metrix, GmbH, Meerbusch, Germany) was used to track the number and size of exosomes. The morphology of exosomes was observed under transmission electron microscopy (TEM). Exosomes were deposited onto copper grids and negatively stained with 2% phosphotungstic acid for 2 min. After air-drying for 15 min at room temperature, the samples were observed under a transmission electron microscope (Hitachi High-Technologies, Tokyo, Japan) at 80 kV.
miRNA library construction and sequencing
Total RNAs were extracted from exosome pellets. Adaptors were added to the 3' and 5' ends, and then reverse transcribed into cDNA, followed by reverse transcription and polymerase chain reaction (PCR) amplification with a low number of cycles. The PCR products derived from the 18–30-base RNA molecules were purified by PAGE using the NEBNext® Multiplex Small RNA Library Prep Set for Illumina® (Illumina, San Diego, USA). Purified library products were evaluated using the Agilent 2200 TapeStation, which were sequenced using the Illumina HiSeqTM 2500 platform. Adaptor sequences of the raw reads were trimmed and low-quality reads were excluded to obtain purified miRNAs for subsequent analysis. The clean reads were compared with the databases (miRBase version 22, Rfam12.1, piRNABank) to obtain non-coding(NC) RNAs. miRDeep2 was used to identify known mature miRNAs based on miRBase21 (www.miRBase.org) and predict novel miRNAs.
Differential expression miRNAs target prediction and pathway enrichment analysis
Differential expression (DE) miRNAs between two sets of samples were calculated using edgeR algorithm according to the criteria of |log2(Fold Change)|≥1 and Q (corrected P) < 0.05. TargetScan 7.1, miRDB v21.0, miRTarBase and miRWalk were used to predict targets for the DE miRNAs. Prediction results from the four programs were further screened and sorted, and the prediction results of miRTarBase or miRDB, miRWalk and TargetScan were taken as candidate target genes of miRNA. When candidate target genes were not shared by two programs, the prediction results of a single database were selected. KOBAS3.0 (http://kobas.cbi.pku.edu.cn/) was used to performed KEGG and GO enrichment analysis of DE miRNA targets with a P < 0.05 required as the significance threshold.
Exosomal miR-193b-3p overexpression in dendritic cells
A lentiviral system was employed to overexpress miR-193b-3p in mouse DCs as a means to produce exosomes enriched in miR-193b-3p. DC2.4 was purchased from Procell Life Science & Technology (CL-0545). The plasmid pHBLV-CMV-ZsGreen-T2A-Puro coupled with GFP (green) was used to construct the lentivirus vector containing miR-193b-3p according to the manufacturers’ instructions (Hanbio Biotechnology, China). In brief, DC2.4s were plated at 30–50% confluence and transfected with a previously prepared lentivirus consisting of a negative control vector or reconstructed vector containing a miR-193b-3p fragment. Following a 12 h incubation the culture medium containing the lentivirus was replaced with complete RPMI 1640 medium. Puromycin (15μg/ml) was used as a means to screen the DCs that were successfully transfected. At 48–60 h later, the DC2.4s transfected with lentivirus and exosomes secreted into the supernatant were collected for further analysis.
LT mouse model
SPF-grade male C3H and C57BL/6j mice (7-8 weeks old) served as LT recipients and donors, respectively. The mice were purchased from SiPeiFu (Beijing) Biotechnology Co., LTD [license SYXK (Beijing) 2017-0010]. The orthotopic LT model employed was generated using the established “two-cuff” technique as described previously[28]. Samples from these mice were obtained at seven and ten days after treatment. Mice were housed in SPF animal rooms in the Experimental Animal Center at the Beijing Friendship Hospital. All experiments in this study were performed according to the Beijing Friendship Hospital Guide for Laboratory Animals.
Preparation of cells
Mice were euthanized using carbon dioxide and the blood, liver and spleen were removed. Isolation of PBMC was the same as that described above. Lymphocytes from the liver and spleen were obtained following procedures contained in a previous report[28]. Erythrocytes were lysed using erythrocyte lysis buffer (R1010, Solarbio, China). The cells were filtered through a cell strainer (70 μm). Lymphocytes from the spleens of normal mice were plated at a density of 2×105 cells/ml in a 96-well plate, which was previously coated with anti-CD3 (1μg/ml, BioLegend, USA) and anti-CD28 (1μg/ml, BioLegend) antibodies. IL-2 (100U/ml, #0717108, PEPROTECH) and TGF-β (20ng/ml, abs04222, absin, China) were added into the culture to induce Treg. These cells were then divided into three groups: 1) Exo-miR-193b-3p - cells co-cultured with 20 μg/ml Exo-miR-193b-3p or exo-NC, 2) miR-193b-3p mimics - cells transfected with 50 nM miR-193b-3p mimics or mimic-NC using RiboFECTTM CP Reagent according to the manufacturer’s protocol and 3) NLRP3-siRNA - cells treated with NLRP3 siRNA or NC.
Flow cytometry
The flow cytometry data were determined using Attune (Thermo, USA). Fluorescein isothiocyanate (FITC)-conjugated anti CD3 (300306, Biolegend), V500-conjugated anti CD45 (560777, Bioscience), Alexa fluor 700A (AF700)-conjugated anti CD4 (344622, Biolegend), V450-conjugated anti CD25 (560355, Bioscience), BV421-conjugated anti CD127 (351316, Biolegend), Allophycocyanin (APC)-conjugated anti CD196 (CCR3) and Phycoerythrin/Cyanine7 (PE-Cy7)-conjugated anti CD183 (CXCR6) were used to distinguish human Treg and Th17 cells. Phycoerythrin (PE)-conjugated anti CD11b (24965S, CST), allophycocyanin (APC) conjugated anti CD11c (33293S, CST), fluorescein isothiocyanate (FITC)-conjugated anti CD45 (62307S, CST), BV421-conjugated anti CD80 (104725, BioLegend) and BV605-A conjugated anti CD86 (105037, BioLegend) were used to characterize DCs. PerCP-Cy5.5-conjugated anti CD3 (100218, BioLegend), PE-conjugated anti CD4 (26589S, CST), AF700A-conjugated anti CD8α (100730, BioLegend), APC-conjugated anti CD25 (36055S, CST), PE-Cy7-conjugated anti Foxp3 (65210S, CST) were used to distinguish mouse Treg cells. FITC conjugated anti Ki-67 (11-5698-80, eBioscience) was used to evaluate cell proliferation, while the Live/DeadTM Fixable Aqua Dead Cell Stain Kit (L34965, Invitrogen) distinguished dead cells. For Foxp3 staining, the eBioscience™ Foxp3 fixation/permeabilization buffer kit was used following instructions of the manufacturer (00-5523-00; ebioscience). FlowJo version 10.8 (Tree Star Inc., Ashland, OR, USA) was used to analyze the results.
Western blot
Cell proteins were extracted on ice for 30 min using RIPA buffer with a protease inhibitor (P0013B; Beyotime, Shanghai, China). Protein samples were separated with 8–12% SDS-PAGE (Bio-Rad, Redmond, WA, USA), electroblotted onto polyvinylidene difluoride membranes (Billerica, MA, USA) and then incubated with primary anti-CD81 rabbit antibody (1:2000, ab109201; Abcam), anti-Alix rabbit antibody (1:2000, ab186429; Abcam), anti-TSG101 rabbit antibody (1:2000, ab125011; Abcam), anti-NLRP3 rabbit antibody (1:2000, ab263899; Abcam), anti-STAT3 rabbit antibody (1:2000, #30835; CST), anti-phosph-STAT3 rabbit antibody (1:2000, #34911; CST), anti-Caspase1 rabbit antibody (1:2000, ab138483; Abcam) or anti-β-actin rabbit antibody (1:2000, #4970; CST) at 4℃ overnight. Horseradish-peroxidase-conjugated anti-rabbit IgG was used as the secondary antibody (1:5000; Cell Signaling Technology). Antibody binding was detected using a chemiluminescence system (Tanon-5200 Multi; Shanghai, China).
Quantitative reverse transcription (qRT)-PCR
For qRT-PCR, the ABI7500 system (Applied Biosystems) was used. Extracted miRNA (100 ng) was reverse transcribed into cDNA using the miRcute Plus miRNA First-Strand cDNA Kit (KR211-02; Tiangen, Beijing, China) and qRT-PCR was performed with use of the miRcute Plus qPCR Kit (FP411; Tiangen). Total RNA (1500 ng) was reversed using the TRUEscript 1st Strand cDNA Synthesis Kit (PC1802; Aidlab, China) and qRT-PCR was performed with use of 2x Sybr Green qPCR Mix (low ROX) (PC60; Aidlab). mRNA expression was calculated using the 2−ΔΔCt method with ΔCt calculated as Ct (mRNA of target)–Ct (reference gene). Primers are listed in Supplementary Data Table S9.
Enzyme-Linked Immunosorbent Assay
Cytokine levels of IL-17A, IL-2, IL-10, IL-6 and IL-1β and transforming growth factor (TGF)-β were assayed using ELISA according to instructions provided by the manufacturer (all ELISA kits from Anoric Bio-technology, Tianjin, China). All samples were measured in duplicate.
Statistical analysis
Data are shown as means ± SEMs. Comparisons among the three groups were performed using the Kruskal–Wallis test with the Mann-Whitney U-test or ANOVA with the Tukey test used for post-hoc pairwise comparisons of subgroups. Student’s t tests were used for comparisons of data involving two groups. Data were tested for normality and log normality before use of parametric assessments. All analyses were conducted using GraphPad Prism version 8.0 (GraphPad Software). A P < 0.05 was required for results to be considered as being statistically significant.