2.1 Specimen source and clinical data

Eighty-six malignant ovarian epithelial tumor samples and 40 normal ovarian tissue samples were analyzed for immunohistochemistry. These samples were obtained from the archived paraffin blocks of surgical specimens from inpatients at Shengjing Hospital of China Medical University between 2014 and 2016. Informed consent was received, and all procedures adhered to the guidelines and regulations set forth by the Ethics Committee of Shengjing Hospital of China Medical University (2021PS823K).Normal ovarian tissue samples were sourced from patients undergoing hysterectomy and preventive oophorectomy due to benign cervical or uterine tumors. They were pathologically confirmed to be normal ovarian tissues without lesions. The 86 OC patients had not received chemotherapy, radiotherapy, or hormone therapy before surgery, and their records were complete, including age, histological type, FIGO stage, histological differentiation grade, lymph node metastasis, survival time, and survival status. Regarding clinical application, surgical pathological staging adheres to the guidelines established by the FIGO in 2009. Patients were followed up until September 1, 2021. OS was the time from the initial diagnosis to death or the last follow-up (censored data for live patients).

2.2 Cell culture and transfection

In this study, normal human ovarian surface epithelial cells (HOSEpiCs) were grown at 37 °C in an ovarian surface epithelial cell medium. Ovarian cancer cells, including A2780, SKOV3, OVCAR3, and COC1, were procured from iCell Bioscience (China). COC1, A2780, and OVCAR3 were cultured in RPMI-1640 medium (Solarbio, China), while SKOV3 cells were cultured in McCoy's 5A medium (Procell, China) with 10% FBS at 37 °C in 5% CO2. SKOV3 and OVCAR3 cells were selected for overexpression transfection. To overexpress OSR1, the pcDNA3.1( +) vector (GenScript Biotech, China) was made by inserting the coding sequences of OSR1. The pcDNA3.1( +) empty vector was used as a negative control. G418 (400 µg/mL) was added to screen cell lines with stable transfection.

For OSR1 knockdown, the OSR1 interfering sequences (siRNA1: 5’-GUGUCAAGAGUGUGGGAAATT-3'; siRNA2: 5’-AGAAGGAAUUCGUCUGCAATT-3'; siRNA3: 5’-CCAGAAAAGAAGCCCACAATT-3') and negative control s-iNC:(5’-UUCUCCGAACGUGUCACGUTT-3') were designed and obtained from General bio (China). Cells were transfected with OSR1 overexpression plasmid, the empty vector, OSR1-siRNA, and si-NC using Lipofectamine 3000 (Invitrogen, USA). The Bay 11-7082 was purchased from Aladdin Industrial Corporation (China) and used in the investigation at a concentration of 20 µM [21].

2.3 Immunohistochemistry

After routine deparaffinizing, rehydrating, hydrogen peroxide blocking, and retrieving tissue antigen with a microwave, the sections were incubated with rabbit polyclonal antibody to OSR1 (A18272, 1:50; Abclonal, China) at 4 °C overnight. They were stained with goat anti-rabbit IgG (H + L) HRP (31460, 1:500; Thermo Fisher Scientific, US). Diaminobenzidine solution (DA1010, Solarbio, China) was employed to counterstain, and the sections were subsequently treated with hematoxylin (H8070, Solarbio) for 1 min and dehydrated. Two pathologists, unaware of the clinical context, evaluated the immunostained tissue sections. One hundred cells were counted in each of the five view fields on each slide at 400 × magnification. The intensity of IHC staining was visualized and scored as 3 (strong stain), 2 (medium), 1 (weak), and 0 (no stain). The extent of staining ranged from 0 to 4, corresponding to the immune-reactive tumor cells percentage (76–100%, 51–75%, 26–50%, 1–25%, 0%), with 4 being the highest. Based on the staining intensity and staining scores, each sample was assigned a score between 0 and 12, categorizing it into two categories: OSR1 low expression (0–6) and OSR1 high expression (8–12) [22].

2.4 Quantitative real-time PCR

Total cell RNA was obtained using TRIpure lysate (BioTeke, China). Quantification and reverse transcription of RNA was carried out using the PCR system (Bioneer, Korea). The 2 –ΔΔCT method was employed for quantifying gene expression, with β-actin as the internal control. Primer sequences utilized in this study comprised OSR1-F: CTCCTTCCTTCAGGCAGTG; OSR1-R: ATCTCGGGCTTGGGTTG; β-actin-F: CAGCAAGCAGGAGTATGACG; and β-actin-R: TTAGGATGGCAAGGGACTTC.

2.5 western blot (wb)

OC cells lysis with PMSF containing lysis buffer (Beyotime, China) and then centrifugation at 10,000 g for 5 min at 4 °C was done to collect the supernatant. A BCA Protein Assay Kit (Beyotime) was employed for protein quantification. The protein separation was performed via SDS-PAGE (Beyotime, China), followed by PVDF membranes (Millipore, MA, USA) transfer and blocking for 1 h with non-fat milk. Primary antibodies were incubated with membranes at 4 °C overnight. After TBST washing, secondary antibodies (1:5000) were incubated with membranes for 45 min at 37 °C, then washed with TBST. ECL reagents (Beyotime, China) were used for blot visualization. Primary antibodies included OSR1, PCNA, and cyclin D1 from ABclonal, China; caspase-3, cleaved-caspase-3, Bcl-2, and Bax from CST, USA; p-p65, p65, p-IκBα, and IκBα from Wanleibio, China. Secondary antibodies used were goat anti-rabbit IgG and goat anti-mouse IgG from Beyotime.

2.6 CCK-8 cell viability assay

OSR1-overexpressing SKOV3 and OVCAR3 cells (3 × 103) in 96-well plates were grown for 12, 24, 36, and 48 h. Each well was then supplemented with 10 µL of CCK-8 reagent (Beyotime). Cells were then incubated for 2 h. Cell viability was calculated by the optical density at 450 nm in a microplate reader.

2.7 Cell cycle assay

Cell Cycle Detection Kits (Beyotime) were used to assess cell cycle progression. 5 × 105 cells were grown in 6-well plates and fixed for 12 h at 4 °C in 70% cold ethanol, followed by staining with PI staining solution containing RNase A in the dark for 30 min. Cells were counted using a flow cytometer.

2.8 Migration assay

A wound-healing test was conducted to assess OC cells' migration capacity. The media was replaced with 20 µg/mL of mitomycin C (Sigma, MO, USA) containing serum-free medium (SFM). A 200-µL pipette tip was used to create the wounds. The cells were then cultured for 48 h after being washed with SFM. The migration rate was calculated using images taken at 0 and 48 h.

2.9 Transwell invasion assay

The invasive potential of transfected OC cells was assessed using a Matrigel-coated Transwell chamber (Corning, NY, USA). The cells were added to the upper compartment, while the lower section was filled with the 10% FBS-containing cell culture medium. Invaded cells were counted following the fixation and staining with 0.5% crystal violet (Amresco, USA) after 48 h. Under an inverted microscope (200x), cells were counted as they invaded the lower layer. For each sample, 5 fields were chosen to quantify the number of cells, and the average value was calculated.

2.10 ELISA assay

MMP-9 and MMP-2 expression levels were measured using the human MMP-2 and MMP-9 ELISA kit (MultiSciences, China) following the manufacturer's instructions.

2.11 Flow cytometry for apoptosis detection

The Annexin V-FITC Apoptosis Detection Kit (KeyGEN, Nanjing, China) was employed to examine cell apoptosis. A binding buffer containing AnnexinV-FITC was used to resuspend cells (5 × 105) cultured in 6-well plates. The PI staining solution was used to stain the cells for 15 min in the dark. The flow cytometer was used to analyze the levels of apoptosis.

2.12 Hoechst staining assay

After fixing for 10 min and PBS washing, the transfected cells were stained with Hoechst staining solution (Beyotime) for 5 min. Cells were mounted after being washed in PBS, then images were captured using a Fluorescent microscope (IX53, OLYMPUS, Japan).

2.13 Statistical analysis

Statistical analysis was conducted using SPSS 26.0 and GraphPad Prism version 9.0. A two-tailed student's t-test was employed to compare two groups, and a one-way single factor analysis of variance (ANOVA) for more than two groups. The chi-square test was used to analyze the differences between counting data groups. Survival curves were constructed using the Kaplan–Meier (KM) method in SPSS, with the log-rank test comparing curve differences. Through univariate and multivariate analysis, the Cox regression analysis was applied to examine the effect of clinicopathological factors on the prognosis for OC. A p-value below 0.05 indicated statistical significance (P < 0.05, *; P < 0.01, **; P < 0.001, ***; P < 0.0001, ****).