The HuCCT1, RBE, KKU-M213, and KKU-M156 human iCCA cell lines were purchased from ATCC or RIKEN. Regular monitoring with the MycoFluorTM Mycoplasma Detection Kit (ThermoFisher Scientific, Waltham, MA) was performed to obtain a mycoplasma-free cell culture environment. The first two cell lines were cultured in DMEM (Dulbecco's Modified Eagle Medium), while the last two were cultured in RPMI, both supplemented with Sodium Pyruvate, Antibiotic–Antimycotic, Hepes, and 10% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, MA). Crenigacestat (LY3039478, Selleckchem Chemicals, Houston, TX) was used in in vitro and in vivo studies as GSI. Stock solutions were prepared in dimethylsulfoxide (DMSO) (Thermo Fisher Scientific, Waltham, MA), and aliquots were stored at −20 °C.
Mycoplasma-free KKU-M213, RBE, and KKU-M156 iCCA cell lines, after validation (Genetica DNA Laboratories, Burlington, NC, USA), were used for the in vitro studies. Cell lines were maintained as monolayer cultures in Dulbecco's modified Eagle medium supplemented with 10% fetal bovine serum (FBS, Gibco, Grand Island, NY, USA), 100 U/mL penicillin, and 100 g/mL streptomycin (Gibco, Grand Island, NY, USA). Cells were grown for 12 h, then serum-deprived for 24 h and treated with siRNA against NOTCH1 (# s453558, ThermoFisher Scientific), following the manufacturer's recommendations. Effects at 48h after siRNA transfection are shown.
- Stable THY1 silencing and overexpression
For loss of function studies, the KKU-M213 cell line was transduced with Human shRNA lentiviral particles carrying 4 THY1 specific sequences (A to D) and scramble control-shRNA sequence (OriGene Technologies, Inc., Rockville, MD 20850, USA). Scramble control-shRNA sequence and THY1-shRNA sequence B were used in all the experiments involving THY1 downregulation at a multiplicity of infection (MOI) 100.
For gain of function studies, the HuCCT1 cell line was transduced with human THY1-CMV-GFP and Lenti-CMV-GFP-2A-Puro-Blank Lentivirus (Aurogene, Italy) at MOI 40.
Both genetically modified cell lines were selected with puromycin dihydrochloride (Thermo Fisher Scientific, Waltham, MA) at 1 µg/ml and 0.625 µg/ml for KKU-M213 and HuCCT1, respectively, to obtain stable THY1 silencing and THY1 overexpression, according to the manufacturer's instructions.
-
Immunohistochemistry and immunofluorescence
For immunohistochemistry (IHC), formalin-fixed, paraffin-embedded (FFPE) tissues from 174 patients with iCCA, routinely collected for diagnosis purposes, were studied. Tumor sections of 4 μm were freshly cut and dried at 60 °C for 30 min. IHC analysis was carried out in sections after deparaffinization for 30 min and then rehydration in grades of alcohol. Antigen retrieval was performed at 90 °C for 20 min with Tris-borate-EDTA Buffer. An automated stainer (cat. K5007, Dako, Glostrup, Denmark) was used to stain the iCCA sections with the primary anti-CD90 antibody diluted 1:150 (Abcam, Cambridge, MA). The Real Envision DAB Substrate Kit (DAKO) was used according to the manufacturer's instructions. CD90 expression was scored for all staining patterns, according to the staining intensity and the percentage of positively stained cells, by two independent, blinded pathologists.
Additionally, two groups (n=10) of KKU-M213 scramble shRNA tumors treated with vehicle or Crenigacestat and two groups (n=10) of HuCCT1 THY1++ tumors treated with vehicle or Crenigacestat were stained by immunohistochemistry with AKT and pAKT (1:100 Cell Signaling Technologies, Massachusetts, USA).
For immunofluorescence (IF) staining, frozen tissues sections were fixed in a 1:1 acetone:chloroform solution, blocked with 2% bovine serum albumin solution, and incubated with anti-CD90, anti-αSMA, and anti-EPCAM antibodies (1:500, Cell Signaling Technologies, Massachusetts, USA). After washing, slides were incubated with secondary goat anti-rabbit immunoglobulin G H&L (Alexa Fluor 488, Thermo Fisher Scientific, Waltham, MA).
Two million stable CD90-silenced KKU-M213 cell lines or CD90 overexpressed HuCCT1 cell lines, or corresponding cells that received the empty vector, were subcutaneously injected into the flanks of 4–5-week-old females CD1 nude mice. Each mouse was monitored daily for clinical signs and mortality, and body weight was recorded twice a week. Tumor growth was assessed with a caliper once a week, evaluating tumor volume with the formula (mm3) = [length (mm) × width (mm)2]/2, where width and length are the shortest and the longest diameters. When the tumor masses volume reached approximately 70–100 mm3, the mice were randomly divided into 8 experimental groups of ten animals and administered Crenigacestat (8 mg/kg) or vehicle by oral gavage every 2 days for 35 days. At the end of the study, mice were sacrificed by cervical dislocation, and tumor masses were collected and frozen to -80°C.
Total RNA was extracted using TRIzol® (Thermo Fisher Scientific, Waltham, MA) in combination with the TissueLyser homogenizer (Qiagen, Hilden, Germany), according to the manufacturer's instructions. The RNA concentration was determined with the NanoDrop Spectrophotometer (Thermo Fisher Scientific, Waltham, MA).
-
Quantitative reverse-transcription real-time PCR (qRT-PCR)
cDNA was obtained starting from 1 microgram of total RNA, using the iScript Reverse Transcription Supermix (Bio-Rad Laboratories, Hercules, CA, USA) according to the manufacturer's instructions. Quantitative PCR reactions were performed using SsoAdvanced SYBR green (Biorad Laboratories, Hercules, CA, USA) and the primers THY1 Human PrimePCR™ SYBR® Green Assay ID: qHsaCED0036661 (Biorad) and Hs_GAPDH_1_SG QuantiTect Primer Assay ID: QT00079247 (Qiagen, Hilden, Germania). Gene Expression Assays for human NOTCH1 (Hs01062014_m1), HES1 (Hs00172878_m1), THY1 (Hs06633377_s1), and β-actin (4333762T) were purchased from Applied Biosystems (Foster City, CA, USA). Quantitative values for each gene were calculated by using the PE Biosystems Analysis software and expressed as number target (NT). NT = 2−ΔCt, wherein the ΔCt value of each sample was calculated by subtracting the average Ct value of the target gene from the average Ct value of the β-Actin gene. Experiments were repeated three times in triplicate. For the screening of AKT Signaling, PrimePCR arrays (Cat #10025059, Bio-Rad Laboratories) was used. Real-time PCR was performed on the CFX96 System (Biorad Laboratories, Hercules, CA, USA). Comparative real-time PCR was performed in triplicate, including no-template controls. Relative expression was calculated using the 2-ΔΔCt method. The ΔCt value of each sample was calculated by subtracting the average Ct value of the target gene from the average Ct value of the GAPDH gene.
-
Chromatin immunoprecipitation
Chromatin isolated from KKU-M213, KKU-M156, and RBE cells was subjected to immunoprecipitation using the MAGnify Chromatin Immunoprecipitation System (492024, Thermo Fisher Scientific, Waltham, MA) according to the manufacturer's instructions. Chromatin was sonicated to a fragment length of about 500 bp and immunoprecipitated with 1 μg of RBPJ (Cell Signaling Technologies, Massachusetts, USA, Active Motif, Belgium, Germany), H3K4me3 (Abcam, Cambridge, MA), H3K27me3 (Abcam, Cambridge, MA), and IgG (Cell Signaling Technologies, Massachusetts, USA) antibodies. The set of primers used for ChIP allows the amplification of target regions, including RBPJ consensus motifs (1 and 2) sites at - 1000bp (genomic position 119,424,535 – 119,424,508) from TSS at the 5' UTR region of the THY gene locus. Primer sequences can be provided upon request.
-
Statistical and bioinformatic analysis
All experimental results are expressed as mean ± standard error of the mean (SEM). GraphPad Prism version 5.03 (GraphPad Software, San Diego, CA, USA) was used to calculate Unpaired Student's t-test and one-way analysis of variance (ANOVA). Differences were considered statistically significant at p values * P < 0.05, ** P < 0.01, and *** P < 0.001.
Each biomarkers distribution was compared using a graphical plot that shows the scatter distribution of the quantities across medians and interquartile ranges. Due to the non-normal distribution of the standard deviations of biomarker distribution, the Wilcoxon test on the median was used to compare the expression of the biomarkers in the tumor tissues vs. surrounding areas. Pearson correlation analysis was applied to reveal a linear association between each biomarker in the following combinations: CD90 vs. NOTCH1, CD90 vs. HES1, HES1 vs. NOTCH1. To assess the statistical significance of monotonic associations between the variables, a P<0.05 was used as a cut-off to reject the null hypothesis.
The whole patients' population was subdivided into two groups according to the median values of each biomarker, assayed both separately and summed. Then the non-parametric Kaplan–Meier method was used to explore the survival probability in months, comparing the two groups of patients (below median vs. above median) for each biomarker and all the biomarkers together. Log-rank test was applied to evaluate the equality of survival among categories, setting a statistical significance threshold of < 0.05.
RPBJ putative transcription factor binding motifs on the THY1 gene promoter were predicted using the Eukaryotic Promoter Database New (EPDnew [20], https://epd.epfl.ch).
The Ingenuity pathway analysis (IPA) software (Qiagen, Hilden, Germania) was used to uncover the canonical pathways and molecular networks in which modulated genes are involved.