Patients and tissue
This study was approved by the Ethics Committee of Xinhua Hospital (Shanghai, China) and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients before the study began.
Tumor and adjacent normal tissues were obtained from 27 patients who underwent radical resection between 2015 and 2017 at the Xinhua Hospital. Each tissue sample was snap-frozen in liquid nitrogen for further analyses. All the patients in this study belonged to the same ethnic group. The patients were selected according to the following criteria: (1) All clinicopathological diagnoses were confirmed by two pathologists. (2) None of the patients had received any treatment before surgery. (4) Availability of complete follow-up data. (5) No death occurred during the perioperative period. (6) No history of other synchronous malignancies.
RNA sequencing
Total RNA from tissues and cells was isolated using the Hipure Total RNA Mini Kit (Magen), according to the manufacturer’s protocol. RNAs were eluted with 50μl of RNase-free water, and the concentration and integrity of the extracted total RNA were estimated a Qubit 3.0 Fluorometer (Invitrogen, Carlsbad, California) and Agilent 2100 Bioanalyzer (Applied Biosystems, Carlsbad, CA), respectively. RNA samples with a RIN value of at least 7.0 or higher were used for further processing. The RNA-seq library was prepared with approximately 1μg of total RNA using the KAPA Stranded RNA-Seq Kit with RiboErase (HMR) for Illumina Platforms (Kapa Biosystems, Inc., Woburn, MA, USA). Briefly, ribosomal RNA was removed from the total RNA. Next, first-strand and directional second-strand syntheses were performed. Then, tailing and adapter ligation were performed using the purified cDNA. Finally, the purified adapter-ligated DNA was amplified. Library quality and concentration were assessed using a DNA 1000 chip on an Agilent 2100 bioanalyzer. Accurate quantification for sequencing applications was performed using a qPCR-based KAPA Biosystems Library Quantification Kit (Kapa Biosystems, Inc., Woburn, MA, USA). Each library was diluted to a final concentration of 10 nM and pooled to equimolar concentrations prior to clustering. Paired-end (PE) sequencing was performed on all samples. For circRNA expression analysis, the read was mapped to the genome using STAR, and DCC was used to identify the circRNAs and estimate circRNA expression. Trimmed mean of M-values (TMM) was used to normalize gene expression. Differentially expressed genes (DEGs) were identified using EdgeR.
Cell cultures
Human cholangiocarcinoma cell lines (RBE, HCCC-9810, and QBC-939) and a human normal bile duct epithelial cell line (H69) were obtained from the Cell Bank of the Shanghai Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Cells were maintained at 37℃ in a 5% CO2 humidified incubator and cultured in RPMI-1640 (Gibco) (RBE and QBC-939) or DMEM (Gibco) (HCCC-9810) supplemented with 10% fetal bovine serum (Gibco) and 1% antibiotics(Gibco). The cells were not cultured for longer than 2 months.
RNAi and plasmid construction
siRNA duplexes were synthesized by GenePharma (Shanghai, China) and transfected into cells using Lipofectamine 3000 (Invitrogen), according to the manufacturer’s protocol. Lentiviruses for knockdown of circMBOAT2 and a plasmid for circMBOAT2 overexpression were obtained from GeneChem (Shanghai, China). The target sequences for constructing the lentiviral shRNAs and siRNAs are listed in Additional File 2.
RNA extraction and qRT-PCR analysis
Total RNA derived from ICC tissues and cells was extracted using TRIzol reagent (TaKaRa, Dalian, China), according to the manufacturer’s instructions. RNA was reverse-transcribed into cDNA using a Primer-Script one-step RT-PCR kit (TaKaRa, Dalian, China). The Hieff UNICON® qPCR SYBR® Green Master Mix (Yeasen, Shanghai, China) was used for qRT-PCR. The circRNA and mRNA levels were normalized by GAPDH, β-actin or U3. The fold-change in the relative expression of RNAs was calculated using the 2−ΔΔCt method. The oligonucleotide sequences are listed in Additional File 2.
RNase R treatment
Two micrograms of total RNA were incubated 30 min at 37 °C in the absence or presence of 5 U/μg RNase R (Geneseed, Guangzhou, China), and the resulting RNA was subsequently purified the RNeasy Mini Kit (Qiagen, Germany) and then analyzed by qRT-PCR.
RNA fluorescence in situ hybridization (FISH)
Oligonucleotide-modified probe sequences for circMBOAT2 and FASN were synthesized by Sangon Biotech (Shanghai, China). The fixed cells were then washed with PBS. The cell suspension was pipetted onto autoclaved glass slides, followed by dehydration with 70, 80%, and 100% ethanol. Then Hybridization was performed at 37 °C overnight in a dark moist chamber. After washing twice in 50% formamide/2 × SSC for 5 min, the slices were incubated with the reagents in Alexa FluorTM 488 Tyramide SuperBoost™ Kits (Thermo Fisher Scientific, Waltham, USA) for 30 min and sealed with parafilm containing DAPI. Images were acquired using a fluorescence microscope (OLYMPUS, Tokyo, Japan). The probe sequences were as follows:
circMBOAT2: 5’- Cy3- CACTACAAAGTTGACTTGTGCATGTTCTCCACT- 3’
FASN: 5’- digoxin- GCGTAGGATGGAATCTCGGAAGCGGTC- 3’
In vitro cell phenotypic assays
For the CCK-8 proliferation assay, 2×103 cells were seeded in 100 μl complete culture media in 96-well plates for various time periods. The Cell Counting Kit-8 assay was performed to measure cell viability, according to the manufacturer’s instructions. For the 5-Ethynyl-2’-deoxyuridine (EdU) proliferation assay, BeyoClickTM EdU proliferation assay (Beyotime, Shanghai, China) was used according to the manufacturer's protocol. The cells were incubated with 10μM EdU for two hours, stained with DAB, and visualized under a light microscope.
For cell cycle analysis, after 48 h of incubation, transfected ICC cells were washed with cold phosphate-buffered saline (PBS) and incubated in ice-cold 70% ethanol at 4 °C overnight. Cells were then incubated with propidium iodide and Rnase A for 30 min, and the cell cycle distribution was analyzed using a flow cytometer (FACS Calibur, BD Biosciences, USA). Data were analyzed using FlowJo 10.6.2 software and presented as the percentage of cell phase distribution, including the G0/G1, S, and G2/M phases. For cell apoptosis analysis, cultured cells were digested by trypsin and washed by phosphate-buffered saline, then stained with annexin V-fluorescein isothiocyanate and propidium iodide for flow cytometric analysis (BD Biosciences, USA).
Tumor xenograft models
The animal experiments were approved by the Ethics Committee of Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine and carried out in accordance with its guidelines. Male BALB/c nude mice (4–6 weeks old) were housed in a specific pathogen-free animal room under 12-hour light-dark cycles, with water and food available. To establish a subcutaneous ICC model, 1×106 RBE cells in 0.1 ml of phosphate-buffered saline were subcutaneously injected into the flanks. After 4-5 weeks of feeding, the animals were euthanized, and the subcutaneous tumors were removed for further research.
Immunofluorescence assays
The cultured cells were fixed using 4% paraformaldehyde for 15 minutes at least. The samples were then incubated with primary antibody against PTBP1 at 4 °C overnight and then incubated with goat anti-rabbit IgG with a red or green fluorescent label (Invitrogen, Carlsbad, CA). After washing twice in phosphate-buffered saline, the slices were sealed with parafilm containing DAPI. Images were acquired using a fluorescence microscope (OLYMPUS, Tokyo, Japan).
Western blot
Proteins were extracted from transfected cells using RIPA lysis buffer. Equal amounts of protein samples were loaded and separated by SDS-PAGE and transferred onto PVDF membranes (Merck Millipore, Germany). Membranes were blocked with 5% skim milk in TBST for 1 h at room temperature. Then, Membranes were incubated with diluted primary antibodies anti-PTBP1 for western blot (1:1000 dilution, 12582-1-AP, Proteintech Group, USA), anti-FASN for western blot (1:1000 dilution, 10624-2-AP, Proteintech Group, USA), anti-c-myc for western blot (1:1000 dilution, ab32072, Abcam, UK), anti-Bcl-2 for western blot (1:1000 dilution, 4223, Cell Signaling Technology, USA), anti-cyclin D1 for western blot (1:1000 dilution, 55506, Cell Signaling Technology, USA), anti-Bax for western blot (1:1000 dilution, 5023, Cell Signaling Technology, USA), anti-NRF2 for western blot (1:1000 dilution, 16396-1-AP, Proteintech Group, USA) at 4℃ overnight. Then, membranes were washed with TBST three times for 15 min/wash, followed by incubation with secondary antibody (Beyotime, Shanghai, China) for an hour, and washed again with TBST. Finally, the protein bands were visualized using Gel Doc 2000 (Bio-Rad, USA), and the gray values were measured using Image J software.
Transcription in vitro and RNA pull-down assay
For in vitro transcription, plasmids containing two T7 promoters were digested using a single restriction endonuclease. The T7 High Yield RNA Transcription Kit (Vazyme, Nanjing, China) was used to transcribe forward and reverse linear DNA templates to RNA (sense and antisense probe), which was subsequently purified using the RNeasy Mini Kit (Qiagen, Germany). The Pierce™ RNA 3' End Desthiobiotinylation Kit (Thermo Fisher Scientific, Rockford, USA) was used to label biotin to the 3' end of RNA. For the RNA pull-down assay, 1×107 cells were washed in ice-cold PBS, lysed in 500 μl co-IP buffer (Thermo Scientific) supplemented with a cocktail of proteinase inhibitors, phosphatase inhibitors, and RNase inhibitors (Invitrogen), and then incubated with 3 μg biotinylated DNA oligo probes against sense or antisense for 2 h at room temperature. A total of 50 μl washed streptavidin magnetic beads (Thermo Fisher Scientific, Rockford, USA) were added to each binding reaction and further incubated for another hour at room temperature. The beads were briefly washed five times with elution buffer. Finally, the retrieved proteins were subjected for mass spectrometry or western blot analysis.
Silver staining and mass spectrometry analysis
Silver staining was performed using the Fast Silver Stain Kit (Beyotime, Shanghai, China), as described previously, while MS was performed by BGI Genomics (Shenzhen, China). Protein identification uses experimental MS/MS data and aligns them with theoretical MS/MS data from the database to obtain results. The entire process starts by converting raw MS data into a peak list and then searching for matches in the database. The search results were subjected to strict filtering and quality control, and possible protein identification was performed. Finally, from the final protein identification list, functional annotation analyses, such as GO, COG/KOG, and pathway analysis, were performed.
RNA immunoprecipitation
RIP experiments were performed using an RNA Immunoprecipitation kit (Geneseed, Guangzhou, China), according to the manufacturer’s instructions. Co-precipitated RNA was detected using qRT-PCR.
Lipidomics
Lipid extraction and mass spectrometry-based lipid detection of cell pellets were performed as described by experimental operation of Huang et al52. A quality control sample was prepared by mixing equal parts of the samples. During the experiments, we processed the analytical sample using the same parameters to evaluate the stability of the analytical performance and reliability of the data. Ultra-high-performance liquid chromatography-mass spectrometry analysis was performed using a Q Exactive Plus high-resolution mass spectrometer (Thermo Scientific, USA) equipped with an Ultimate 3000 UHPLC system (Thermo Scientific, USA). Lipid identification (structural identification) and a peak table containing the retention time, m/z, and peak area (peak matching) were assessed using MS-DIAL software. The data combination of the positive and negative ion modes was defined as the relative lipid abundance for subsequent statistical analysis.
Actinomycin D assays
RBE and HCCC-9810 cells were seeded in six-well plates (1 × 106 cells per well). Twenty-four hours later, cells were exposed to 2 μg/ml Actinomycin D (Sigma) and collected at indicated time points. The RNA stability was analyzed using qRT-PCR and normalized to the values measured in the mock treatment group (the 0 h group).
Co-Immunoprecipitation
To detect protein–protein interactions, cells were lysed in 500 μl co-IP buffer supplemented with a cocktail of proteinase inhibitors, phosphatase inhibitors, and RNase inhibitors. The lysates were added to the same IgG species used for immunoprecipitation as the normal IgG and agarose beads (Beyotime, Shanghai, China) and shaken slowly at 4ºC for 30 min to 2 h. The supernatant is centrifuged at 1000g for 5 min and used for immunoprecipitation with agarose beads, which were pre-incubated with the corresponding antibodies. After incubation at 4 °C overnight, beads were washed 5 times with PBS. SDS sample buffer was added to the agarose beads, and immunoprecipitates were used for western blot analysis.
Lipid peroxidation labeling
4,4-difluoro-5-(4-phenyl-1,3-butadienyl)-4-bora-3a,4a-diaza-s-indacene-3-undecanoic acid (BODIPY® 581/591 C11) (Invitrogen, Carlsbad, CA, USA), a fluorescent fatty acid analogue, was used. This dye shifts from red to green upon oxidation. Freshly cells were incubated with BODIPY 581/591 as a free radical sensor for 30 min at 37ºC. Then, cells were rinsed twice with PBS. The fluorescence inside the cells was analyzed using flow cytometry. Cells were selected by FSC/SSC gate. Subsequently, BODIPY C11 non-oxidized positive hepatocytes (red cells, PE filter) were gated, then positive green cells were selected (BODIPY oxidized positive cells) (FITC filter). The limits of the markers were established using unstained cells (negative control). Then, they were rinsed twice with PBS and measured on a FACSCanto II flow cytometer (FACS Calibur, BD Biosciences, USA). Data analysis was performed using the FlowJo 10.6.2 software.
Statistical analysis
Results are shown as the mean ± SD. SPSS 22.0 (IBM Corp., Armonk, NY, USA), R 4.2.0, and GraphPad Prism 8.0 were used for comparison analysis. ImageJ was used to analyze the colocalization between circMBOAT2 and PTBP1 by the function of coloc-2. Student’s t-test was used for comparisons between the two groups. The Fisher's exact test was used to determine the association between the expression of circMBOAT2 with patients’ clinic pathological parameters. Partial least squares discriminant analysis (PLS-DA) including loading plot and variable importance in projection (VIP) value were performed by MetaboAnalyst 5.0 using the online server. Statistical significance was set at P < 0.05.