Cell lines and cell culture
All cell lines used in this study were listed using the official cell line name and its Research Resource Identifier (RRID) as available in the ExPASy Cellosaurus database (https://web.expasy.org/cellosaurus/). Human ovarian epithelial cancer cell lines HEY (RRID:CVCL_0297), HEY A8 (RRID:CVCL_8878, the isogenic cell lines of HEY)15, SK-OV-3 (RRID:CVCL_0532), A2780 (RRID:CVCL_0134), OVCA433 (RRID:CVCL_0475), OVCA429 (RRID:CVCL_3936) and lentivirus packaging cell line HEK293T (RRID:CVCL_0063) were obtained from the American Tissue Culture Collection (ATCC), immortalized normal human ovarian surface epithelial cell line T29 (RRID:CVCL_2G67)4, and normal human ovarian surface epithelial cell line NOE (also called HOSE) were established in our lab 16. All human cell lines have been authenticated using STR profiling by Genetic Testing Biotechnology (Suzhou, China) within the last three years. All experiments were performed with mycoplasma-free cells. All cell lines were cultured using Roswell Park Memorial Institute (RPMI) 1640 in a humidified incubator with an atmosphere of 5% CO2 at 37°C, with the exception of 293T which was maintained in Dulbecco's modified Eagle’s medium (DMEM). 10% fetal bovine serum (FBS), 2 mM L-glutamine, penicillin (100 units/ml) and streptomycin (100 μg/ml) were added to the culture media.
Stable cell lines construction
To knockdown GMFG and ECM1, synthesized DNA oligos were designed to generate small hairpin RNA (shRNAs) targeting GMFG-specific mRNAs and ECM1-specific mRNAs, the specific shRNAs and a scrambled shRNA (negative control) were inserted into plko.1/puromycin vectors separately. To overexpress GMFG and ECM1a, the cDNA of GMFG and ECM1a were cloned and inserted into PCDH-neomycin vectors along with a Flag-tag and a HA‐tag respectively, and an empty PCDH-neomycin vector was used to generate control cell lines. All related primers were listed in Supplementary Table 1, plasmids and stable cell lines were constructed following previous research pipeline4.
Quantitative real‐time polymerase chain reaction (qPCR)
To perform qPCR, total RNA was extracted from cells and isolated with TRIzol reagent (Invitrogen) according to the manufacturer’s instructions, and RNA reverse transcription was conducted to generate cDNA using a PrimeScript™ RT Master Mix (TaKaRa) following the manufacturer’s instructions. To test the mRNA expression levels of specific genes, qPCR was carried out on the Applied Biosystems 7500 Real-Time PCR Systems (Applied Biosystems) using FastStart Universal SYBR Green Master (ROX) kit (Roche) with the generated cDNA and specific qPCR primers. GAPDH was applied as the internal standard, and all qPCR primers used in this study were showed in Supplementary Table 1. The relevant mRNA expression values were measured using the ΔΔCt method.
Western blot (WB) analysis
Protein samples were collected from cell lysates using RIPA buffer added with a protease inhibitor cocktail, and then the expression levels of different proteins were detected by standard WB procedures4. The detailed information of the primary antibodies applied in the present study was summarized in Supplementary Table 2. The secondary antibodies against mouse (#7076) or rabbit IgG (#7074) were purchased from Cell Signaling Technology (Massachusetts, USA). The immunoblots in our study were visualized using Fluorchem E (ProteinSimple, USA) and enhanced chemiluminescence reagent (Thermo Scientific, USA).
Co-Immunoprecipitation (Co-IP)
Co-IP assay was conducted according to the standard protocol using protein G‐agarose (Roche). After several washes in ice-cold PBS, the crude extract of cells was suspended using ice-cold lysis buffer at a recommended concentration. 25 μL protein G‐agarose was used to decrease the background for no less than 30 minutes at 4°C, then a specific antibody at a recommended amount was added to the cell supernatant for incubation overnight at 4°C, and the same amount of a normal IgG of the same species was used as the negative control. After antibody incubation, 50 μL protein G‐agarose was used to form agarose‐antibody‐antigen complex, and the mixture was washed using the recommended washing buffer for several times. Finally, bound proteins were detected by routine WB analysis. The antibodies used in Co-IP assay were listed in Supplementary Table 2, and normal IgG applied in Co-IP assay was either rabbit IgG (SC#2027, Santa Cruz Biotech, CA, USA) or mouse IgG (SC#2025, Santa Cruz Biotech, CA, USA).
Immunofluorescence (IF) staining
IF staining was performed using published method17. Briefly, an appropriate density of cells was seeded and incubated overnight in 24-well plates with cover slides. The cover slides were washed using PBS and fixed by precooled methanol for 15 minutes. 0.5% Triton X-100 was used to permeabilize the cells for 5 min, then cells were blocked by 5% BSA. All cell samples were treated with double staining in this study, cells were incubated with a mixture of two different primary antibodies at an appropriate concentration and then secondary antibodies under dark condition. Finally, cells were washed again with PBS and stained with DAPI. The information of applied primary antibodies was showed in Supplementary Table 2. The secondary antibodies used in IF staining were Alexa Fluor 594 AffiniPure donkey anti-rabbit IgG (green) and Fluor 594 AffiniPure donkey anti-mouse IgG (red). All stained cells were observed using a fluorescence microscope and photographed by a fluorescent microscope camera.
Cell migration and invasion assay
Wound healing assay was carried out to detect cell migration speed. Cells were placed and incubated in six-well plate overnight. A sterile P200 pipette tip was used to form a scratch, and warm PBS was used to wash the floating cells away. The migration speed was recorded by taking photographs at 0 hour and 24 hours.
Besides, cell migration and invasion abilities were also measured using the transwell chamber assay. For migration assay, an appropriate number of cancer cells in 100 μL FBS‐free RPMI 1640 was seeded directly into the upper chambers, and about 600 μL RPMI 1640 supplemented with 10% FBS was added into the lower chambers. For invasion assay, the upper chambers were firstly added with Matrigel (BD Biosciences) following the manufacturer’s protocol and the cells were then placed on top of the Matrigel in the upper chamber. After culturing for about 24 hours in incubator, the invaded cells were fixed by methanol, stained by 0.1% crystal violet for about 30 minutes, and the images were captured by a light microscope. Experiments were performed in triplicate.
Plate colony formation assay
500 cells were seeded in six-well plates at a single-cell density. Fresh RPMI 1640 containing 10% FBS was added every 3-5 days to allow cell growth until visible cell colonies could be observed by unaided eye. Then colonies were fixed using methanol, stained by 0.1% crystal violet, manually counted and analyzed.
Soft agar anchorage-independent growth assays
2.5 ml RPMI 1640 containing 20% FBS and 2.5 ml 1.5% agarose (Sigma-Aldrich, United States) were added to 6 cm2 plate as the lower agar. Then 5,000 cancer cells in 0.5 ml RPMI 1640 medium containing 20% FBS with 0.5 ml 0.7% agar were mixed and placed onto the lower agar. After incubation for about 3 weeks, the colonies were photographed and counted under microscope. Colony formation assays were performed in triplicate.
Cell proliferation assay
A total of 1 × 103 cells in 100 μL RPMI 1640 containing 10% FBS were placed in 96‐well culture plates. Cell proliferation capacity was detected at the 1, 2, 3, and 4 days by the cell counting kit-8 (CCK-8) (Dojindo, Tokyo, Japan) following the manufacturer's instruction at 450 nm (Tecan Infinity 200PRO multi-well plate reader). The assays were independently conducted three times.
Cell treatment and cell viability detection
1×104 cells in 100 μL RPMI 1640 per well were seeded in 96-well culture plates and incubated overnight, then fresh RPMI 1640 containing different concentrations of paclitaxel (Haosen pharmaceutical company, Jiangsu, China) was used to culture cells. After culturing with paclitaxel for 48 hours, cell viability was detected by CCK-8 method described above, and IC50 value was calculated.
To inhibit the phosphorylation of FAK at the Tyr397 site, three different specific inhibitors (Defactinib hydrochloride, MedChemExpress, HY-12289A; GSK2256098, Selleckchem; and PF573228, Selleckchem) were applied to HEY-G cells for 8, 12 and 24 hours respectively. Then inhibitor-treated HEY-G cells, DMSO-treated HEY-G cells (diluent, the same volume of DMSO was used as negative control) and blank HEY-G cells (no treatment) were collected for WB analysis to confirm that FAK signaling pathway was inhibited. After that, Defactinib-treated, DMSO-treated, and blank HEY-G cells were used to conduct plate colony formation experiment, cell proliferation assay and wound healing assay according to the methods described previously. The concentrations of Defactinib and DMSO were maintained during the experimental period.
Xenograft tumor model establishment
The animal assays approved by the Institutional Animal Care and Use Committee of East China Normal University (reference number: m20200801) were carried out in accordance with the ARRIVE guidelines. 6-week-old female BALB/c nude mice from Slac Laboratory Animal (Shanghai, China) were subjected to intraperitoneal injection of 1 × 107 cells to establish xenograft tumor models. mice were observed regularly and killed before natural death occurred, then the mice were weighed and tumor nodules were taken out, counted, weighed and calculated.
Immunohistochemistry (IHC)
Murine tumor tissues were fixed in 10% formalin and embedded in paraffin, later were cut into 5-μm thick. Tissue sections were stained according to our previously published method4. The relevant information of the primary antibodies used to stain tissues were summarized in Supplementary Table 2, and the secondary antibodies were from EnVision™ Detection Kit (GENE, USA). The images were captured at 200× magnification.
Kaplan–Meier analysis
Kaplan–Meier plot was generated by Kaplan-Meier Plotter website (http://www.kmplot.com) 18 to assess the prognostic values of GMFG in ovarian cancer, hazard ratio (HR) with 95% confidence interval (95%CI) and log-rank P value were calculated. Patients were collected from TCGA database and were divided into high and low GMFG expression groups according to the best cutoff value.
Statistical analysis
All data were analyzed by an unpaired two-tailed Student’s t test, and statistical significance was set at *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.