Cell culture
Rat BMSCs (Procell, Cat no: CP-R131) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 50 U/mL penicillin and 0.1 mg/mL streptomycin (P/S), and 10% foetal bovine serum (FBS) before differentiation. The human ovarian carcinoma cell line SKOV3 (Procell, Cat no: CL-0215) was cultured in DMEM supplemented with P/S and 10% FBS. All the cells were cultured at 37 °C with 5% CO2.
Adipogenesis of BMSCs
Inducers including 1 μM dexamethasone, 5 mg/L insulin, and 0.5 mM IBMX (all purchased from Sigma) were added into the medium in which BMSCs were cultured when the cells reached about 80% confluency. After two days of culture, the medium was replaced with DMEM supplemented with 50 U/mL P/S, 10% FBS, and 5 mg/L insulin. The medium was changed every two days until differentiated adipocytes were developed. All the cells were cultured at 37 °C with 5% CO2.
ORO staining of BMSC lipids
BMSC-derived adipocytes were rinsed gently with phosphate-buffered saline, and subsequently stained with fresh diluted ORO with final concentration of 60% (Sigma) to evaluate their lipid content. Briefly, the cells were stained with ORO solution for 10 min at 60 °C and then washed in 75% ethanol solution. Then, the samples were thoroughly washed under the running tap water for 2 min followed by counterstaining with Gill’s haematoxylin solution (Servicebio, Cat no: G1004) for 30 s. The samples were mounted using an aqueous solution. The prepared slides were visualised using an Axiovert microscope (Perkin Elmer) at ×200 magnification. The relative steatosis area is expressed as the percentage of ORO-stained areas.
CCDC3 overexpression and knockdown
Rat Ccdc3 cDNA was synthesised and cloned into a cloning expression vector pcDNA3.1 (+) at the BamHI-XhoI site to generate a CCDC3-expression vector. CCDC3-packaged plasmids were transiently transfected into BMSC-derived adipocytes for the generation of the CCDC3-containing CM (hereinafter referred to as CCDC3 CM). Different siRNAs (siRNA1, siRNA2, and siRNA3) targeting three distinct sequences of rat Ccdc3 gene (Additional Table 1) were purchased from Jima Pharmaceutical Company. These siRNAs were used for knocking down CCDC3 and an NC siRNA was used. BMSC-derived adipocytes were transfected with siRNA1 for the generation of knocked down CCDC3-containing CM (hereinafter referred to as si-CCDC3 CM).
SKOV3 co-cultured with CM
SKOV3 cells were co-cultured with eight groups of CM: CCDC3 CM; NC of CCDC3 CM (NC CCDC3); CCDC3 CM treated with 20 μM SKL2001 (Selleck, S1180) for 48 h (hereinafter referred to as CCDC3-agonist CM); CCDC3 CM treated with 5 μM XAV939 (Selleck, S8320) for 48 h (CCDC3-inhibitor CM); si-CCDC3 CM; NC of si-CCDC3 CM (NC si-CCDC3); si-CCDC3 CM treated with 20 μM SKL2001 for 48 h (si-CCDC3-agonist CM); and si-CCDC3 CM treated with 5 μM XAV939 for 48 h (si-CCDC3-inhibitor CM). Subsequently, cells were harvested for RT-qPCR, western blotting, and cell proliferation, migration, and invasion assays.
RNA extraction and RT-qPCR
Total RNA was extracted using MiniBEST Universal RNA Extraction Kit (Takara, Cat no: 9767,), following the manufacturer's instructions. cDNA was generated using the Reverse Transcriptase Kit (Primescript RT Reagent Kit with gDNA Eraser perfect Real time), following the manufacturer's instructions. Real-time qPCR was performed using an ABI12K Real Time PCR System. The sequences of the primers used are listed in Table 2. Relative expression levels of the candidate genes were calculated using the expression level of GAPDH as the reference using the 2−ΔΔCT method. All primers were synthesised by Tsingke Biotechnology (Beijing, China). All the experiments were conducted in triplicate.
Western blotting
Cultured cells were lysed with lysis buffer containing radioimmunoprecipitation assay buffer (RIPA buffer). The proteins concentration in the cell lysates was quantified using the DC protein assay (Bio-Rad). Equal amounts of proteins were subjected to sodium dodecyl sulphate-polyacrylamide gel electrophoresis, and then transferred to polyvinylidene difluoride membranes. The membranes were incubated with the following antibodies: anti-CCDC3, anti-E-cadherin, anti-N-cadherin, anti-vimentin, anti-fibronectin, anti-β-catenin, anti-c-myc, and anti-cyclin D1 (Additional Table 2). Bands were imaged using a ChemiDoc™XRS+ scanner (Bio-Rad).
Proliferation assays
SKOV3 cells were seeded in 96-well plates at a density of 5,000 cells/well, and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays were performed after 24, 48, and 72 h to evaluate cell viability and proliferation. Specifically, MTT was first prepared as a stock solution of 5 mg/mL in phosphate-buffered saline (PBS, pH 7.2) and filtered. At the end of the treatment period (24, 48, and 72 h), 10 µL of MTT solution was added to each well. After incubation for 4 h at 37 °C, discard the supernatant, add 150 μl DMSO to each well, shake for 10min to dissolve the crystals, and then the 96-well plate was read by an enzyme-linked immunosorbent assay (ELISA) reader at 490 nm for absorbance density values to determine the cell viability. All the experiments were conducted in triplicate.
Cell invasion and migration assays
To measure the cell invasion ability, transwell chambers (8-μm pore size) were coated with Matrigel (50 μL/well with a solution containing four volumes of serum-free medium to one volume of Matrigel solution) in 24-well plates. Cell migration assay was performed with transwell chambers (8-μm pore size) coated with serum-free medium in 24-well plates. Cells at a density of approximately 1.5×105 cells/well were seeded into the upper chamber of the insert. After incubation for 48 h at 37 °C with 5% CO2, any cells that had invaded the lower cavity were fixed by 4% paraformaldehyde solution (Beyotime, P0099) for 10 min and then stained with 5% crystal violet solution (Beyotime, C0121) for 10 min. The cells were counted in three randomly selected visual fields using an inverted microscope (IX71, Olympus). All the experiments were performed in triplicate.
Bioinformatic analysis
Data acquisition
The raw data of CCDC3 mRNA expression in 359 ovarian cancer samples were extracted from TCGA (https://portal.gdc.cancer.gov/).
Kaplan–Meier survival curve analysis
The relationship between CCDC3 mRNA expression (high vs. low) and the survival of patients with ovarian cancer was assessed using the online tool Kaplan–Meier plotter (kmplot.com/analysis) (44). The log-rank p-value and HR with 95% CI were calculated.
Co-expression analysis
The association between CCDC3 and the expression level of other genes was measured using Pearson’s correlation analysis. Pairwise genes with a correlation coefficient larger than 0.3 (p < 0.05) were considered to be co-expressed. Data are presented using a circle diagram and heatmap using R.
Correlation between CCDC3 expression level and tumour-infiltrating immune cells
The CIBERSORT (45) package in R was used to estimate the infiltration score of 22 tumour-infiltrating immune cells. The chi-square test was performed to analyse the correlation between the CCDC3 expression level (high vs. low) and immune cells infiltration. A Spearman correlation coefficient with p < 0.05 was considered statistically significant.
Drug sensitivity analysis
The pRRophetic package in R was used to predict the drug sensitivity of each tumour sample. The ridge regression method was applied to estimate the concentration of IC50 of each specific therapeutic drug. GDSC (https://www.cancerrxgene.org/), the largest pharmacogenomics database available nowadays, was used as the training set for 10-fold cross-validation to evaluate the performance of the constructed regression mode. Default values were selected for all parameters, and batch effect was removed.
GSVA and GSEA
The GSVA (46) package in R was utilised to estimate the signalling pathways associated with CCDC3 expression. The clusterprofiler (47) package in R was used to perform GSEA of CCDC3 with RNA-sequencing data. The statistical significance of the category enrichment was assessed using the p-value computed from 1,000 permutations. The population was divided into two groups according to the CCDC3 mRNA expression level (high vs. low). The acceptable level of significance was p ≤ 0.01. The gene set “c2.cp.kegg.v6.2.symbols.gmt” was selected as the reference gene set (MSigDB, http://software.broadinstitute.org/gsea/msigdb/index.jsp).
GeneMANIA analysis
To better understand the function of CCDC, a protein–protein interaction (PPI) network was constructed using GeneMANIA (http://www.genemania.org), an online tool commonly used to identify functions and interactions and visualise the potential PPIs.
Statistical analysis
Analysis of variance (ANOVA, one-way), Wilcoxon test, and the Student's t-test were employed to compare the variables. Survival curves were evaluated by the log-rank test using the Kaplan–Meier method. A p-value < 0.05 was considered statistically significant. Statistical analyses were performed using R (R Core Team, version 3.6).