Clinical data and tissue specimens
The medical records of all consecutive patients diagnosed and treated for EOCs from January, 2010 to December, 2018 in Beijing Chao-Yang hospital. Patients who had the primary malignant tumor in another part of body were not included in this study. Patients without complete surgery and pathology reports or who were lost to follow-up within one month after the initial surgery were also excluded from this study. Patient information, including demographic and pathological characteristics, and disease status at the last contact, were collected and evaluated. Additionally, paraffin-embedded tissues from the enrolled patients were obtained from the pathology center. In our study, tumor staging was based on the 2015 International Federation of Gynecology and Obstetrics (FIGO) system23. The progression-free survival (PFS) was defined as the data of surgery to the data of recurrence; patients who hadn’t recurred at their last visit were censored. The overall survival (OS) time was defined as the data of surgery to the data of patient death because of the disease, and patients died due to other conditions, disease and the patients who survived at their last visit were censored.
Tissue microarray (TMA)
TMAs were executed by matched primary tumor lesions. Two gynecological pathologists independently reexamined the pathology slides and ensure EOC diagnosis, marking the accurate the localization of the malignant lesions on the paraffin-embedded tissue samples. Round tissue samples with a diameter of 1 millimetre were obtained from the tumor located in the donor block using a manual tissue array instrument (TMArrayer), transferred into the TMA block. Sequential sections were cut from the paraffin-embedded TMA blocks at a thickness 4 um and placed on blank slides.
Immunohistochemical (IHC) staining
IHC staining was performed as previously described10. Briefly, baking the sections in 70℃ for 60 min, then deparaffinized and rehydrated sections in the dimethylbenzene and graded ethanol. Using the 3% hydrogen peroxide to block endogenous peroxidase activity. Ten percent goat serum was treated at room temperature for 1h. The sections were incubated with the primary antibody (Rabbit anti-HDAC3, 1:20 and mouse anti-FOXA1, 1:100) at 4℃ over night. After recovering to room temperature, adding the horseradish peroxidase-conjugated goat anti-rabbit/mouse antibody for 1 h, tissue sections were subjected to diaminobenzidine staining for color development. Subsequently, sections were subjected to hematoxylin counterstaining and dehydration, and then sealed with neutral resin.
Evaluation of the IHC staining of the TMA
Utilizing a digital pathological section scanner (Pannoramic MIDI/P250) to capture images of the TMA slides, then displaying the imaging at 400x magnification by panoramic Viewer 1.15.4 software. Two gynecologists independently evaluated the images and they didn’t know the clinical data. A histochemistry score (H-score) based on a combination of the percentage of stained cells and staining intensity was calculated for the semiquantitative analysis. H-score = Σ(percentage [0–100%] × intensity [1–3]) = (percentage of cells with weak intensity × 1) + (percentage of cells with moderate intensity × 2) + (percentage of cells with strong intensity × 3)24. The relationship between HDAC3 and FOXA1 was analyzed, in addition, the relationship between them and the patients’ prognosis was also analyzed.
Western blotting analysis
Using RIPA buffer to collect the total protein. Ten percent SDS-PAGE gels was performed and loading 30ug protein onto it. Then transferring the protein to a polyvinylidene difluoride （PVDF）membranes（Millipore）through 100mV for 90min. All membranes were treated with 0.5% BSA for 2h and then incubated with 1:1000 dilution of rabbit anti-HDAC3 monoclonal antibody (Abcam), a 1:1000 dilution of mouse anti-FOXA1 monoclonal antibody (Abcam), 1:1000 dilution of rabbit anti-β-catenin monoclonal antibody (Cell Signaling Technology), 1:1000 dilution of rabbit anti-MMP2 monoclonal antibody (Cell Signaling Technology), 1:1000 dilution of rabbit anti-Cyclin D1 monoclonal antibody (Cell Signaling Technology) , and 1:2000 dilution of rabbit anti-GAPDH monoclonal antibody (abcam) in PBS containing 0.1% Tween-20 and 5% bovine serum (BSA) overnight at 4℃. After incubation with the secondary antibody at 1:4000 dilution (Golden Bridge International, Inc., Beijing, China) for 1 h at room temperature, the blots were detected by the enhanced chemiluminescence (ECL) substrate kit (Thermo USA). The experiments were repeated three times.
Quantitative real-time polymerase chain reaction (qRT-PCR)
Total RNA was isolated from ovarian cell lines or tissues by the TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. cDNA of each sample was reverse transcribed by approximately 2μg of RNA. Synthesized cDNA was amplified on the ABI Biosystems 7500 Fast Real-Time PCR System with SYBR Green I Master. The cycling conditions were as follows: 95 °C for 30 s followed by 40 cycles of 95 °C for 5 s and 60 °C for 34 s for telomere PCR. The experiments were repeated three times with triplicates of each sample. The expression levels of HDAC3 and HE4 were calculated by the 2-ΔCt method. This experiment was repeated three times independently. Primer sequences were as follows: HDAC3: F: 5′-GCAAGGCTTCACCAAGAGTCT-3′, R: 5′-AGATGCGCCTGTGTAACGC-3’; FOXA1: F: 5′-GCAATACTCGCCTTACGGCT-3′, R: 5′-TACACACCTTGGTAGTACGCC-3’; and GAPDH: F: 5′-GGAGCGAGATCCCTCCAAAAT-3′, R: 5′-GGCTGTTGTCATACTTCTCATGG-3’.
The human ovarian cancer cell lines (OVCAR3, A2780, SKOV3 and ES-2) were purchased from the Peking Union Medical College Cell Resource Center (Beijing, China). The OC cell line A2780 was kindly provided by Dr. Deng (Tsinghua University). A2780, OVCAR3 and SKOV3 cells were cultures in complete medium RPMI 1640 medium (Corning) containing 10% heat-inactivated FBS (Gibco), penicillin(100U/ml)/streptomycin(100ug/mL). ES-2 was cultured in McCoy's 5A Medium (Gibco) with 10% heat-inactivated FBS (Gibco), penicillin(100U/ml)/streptomycin(100ug/mL). Incubating in a humidified incubator at 37 °C under 5% CO2 in air.
OC cells was treated with ice-cold RIPA buffer (1ml) and incubation on ice for 30 min. The cells suspension was centrifuged at 12,000g for 30min at 4°C, then discarded the supernatant and collected supernatant fractions. Anti-FOXA1 antibody (10μl) (Abcam, rabbit monoclonal) or anti-HDAC3 antibody (CST, mouse monoclonal) was added in the supernatant fractions overnight at 4 °C. The mixture was treated with Protein A/G PLUS-Agarose (20μl; Santa Cruz) and incubation them on the rocker platform overnight at 4°C. The negative control was 10ul anti-Rabbit IgG (abcam). Western blot was used to analyze the immunoprecipitates with FOXA1 monoclonal (Abcam, Rabbit) and HDAC3 monoclonal (Abcam, Mouse) antibodies. ECL reagent (Amersham ECL Prime detection) was used to visualize the proteins.
The lentiviral named Lv-HDAC3-shRNA and Lv-FOXA1-shRNA was purchased for knockdown the expression of HDAC3 and FOXA1, respectively. The lentiviral named Lv-HDAC3 and Lv-FOXA1 was purchased for increasing the expression of HDAC3 and FOXA1, respectively. The cells were seeded in 24-well plates with the density of 5 x 104 cells/well. After 72 h of infection, lentiviral vectors encoded enhanced green fluorescent protein (eGFP) and infected cells can be observed through fluorescence microscopy. The fluorescence intensity represented transduction efficiency and eGFP expression. Cells with 80%-90% infection efficiency and better cellular growth behavior were chosen and expanded via puromycin treatment. The efficiency of the infection was confirmed by western blot and reverse transcription PCR (RT-PCR).
Transwell invasion assay
Transwell inserts (8um pore size) were used for invasion assays and were placed in 24-well plates. Matrigel was melt before the experiment and diluted with serum-free medium at 1:20. Forty microliter Matrigel was added in the upper chambers and put the plates in a humidified incubator at 37 °C. Cells were resuspended in serum-free medium in the upper chamber at 4*104 cells/well. Medium containing 10% FBS was added in the lower chamber. Culturing 16h in humidified incubator, the medium was discarded and the migration cells was fixed by 4% paraformaldehyde for 15min. Then the membrane was washed 3 times with PBS and stained with 0.1% crystal violet for 20 min after drying. The stained cells were counted and observed in the bright field of a microscope. Positive cells were numbered at least 5 random microscopic fields and the statistical analysis was performed.
Cell colony forming experiment
The cells in the log phase were spread into a 6-well plate with 1,000 cells per well, culture in a CO2 cell incubator for 14 days, discard the medium, stain with 0.01% crystal violet, and count the colonies under a microscope to calculate the formation of cell colonies rate. Number of colonies = average number of cell colonies in N wells, colony formation rate% = number of colonies/total number of cultured cells × 100%.
5-Ethynyl-2'-deoxyuridin (EDU) assay
EdU assay (RIBOBio Co, Guangzhou, China) was used to measure cells’ abilities to proliferate. After incubation with EdU for 1 h, the cells were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100. Then, the Apollo reaction cocktail (reaction buffer and Apollo® 567 fluorescence) was added to medium for another 30 mins in the dark. After washed with PBS for three times, and the nuclei were stained with Hoechst 33342 and immediately viewed under fluorescence microscopy. Cell proliferation ratios were calculated using the formulation of (Edu-positive cells/Hoechst-stained cells) × 100%. The number of EdU-positive cells were calculated by counting at least three random separate fields.
The assay was performed as instruction of the PE Annexin V apoptosis detection kit I (BD Pharmingen). In brief, cells in the logarithmic growth phase were washed once with PBS, then EDTA-free trypsin was added, and all the target cells attached and suspended were collected by centrifugation at 1000 rpm for 5 min. Wash cells twice with cold PBS and then resuspend cells in 1X Binding Buffer at a concentration of 1 x 10^6 cells/ml. Transfer 100 μl of the solution (1 x 10^5 cells) to a 5 ml culture tube. Add 5 μl of PE Annexin V and 5 μl 7-AAD. Gently vortex the cells and incubate for 15 min at RT (25°C) in the dark. Add 400 μl of 1X Binding Buffer to each tube. Analyze by flow cytometry within 1 hr.
All procedures performed in animal studies were approved by the Animal Research Ethics Committee of Capital Medical University. Female BALB/c nu/nu mice (4-6 weeks old; Shanghai Institute of Material Medicine, Chinese Academy of Science) were raised in specific pathogen-free conditions at 22°C and 55% humidity. Cells at the exponential phase of growth were digested by 0.25% trypsin and resuspended in PBS. The mice were divided into four groups randomly which group had 10 mices, namely NC, HDAC3 overexpression, NC and HDAC3 knockdown groups in mouse tumorigenicity assay. PBS containing 5*106 cells/200ul was subcutaneously injected into the right flank of each nude mouse. Digital calipers were used to measure the tumor width and length every 5 days. The formula of formula (width)2 x length/2 was calculated the tumor volumes. The nude mice were sacrificed via broking the neck without any anesthetic inhalation after 20 days of observation. The tumors were isolated, fixed with 10% formalin, and embedded in paraffin for further pathological analyses. In The tumors were isolated, fixed with 10% formalin, and embedded in paraffin for further pathological analyses.
Mouse tail-vein injection experiments and peritoneal metastasis assays were performed. The mouses were divided into Mock and HDAC3 overexpression groups with 5 mices independently. A volume of 200ul (1x106) was injected into the tail-vein of nu/nu mice and the mice were observed for 21 days. Cells (5x106 cells/200ul PBS) were injected intraperitoneally for peritoneal metastatic formation and the mice were observed for 21 days. The mice were then killed humanely by broking the neck without any anesthetic and an autopsy was performed and the lungs and the peritoneal were examined for tumors separately. The tissues were dehydrated, processed, and embedded in paraffin wax and stained with haematoxylin and eosin (HE).
Statistical analyses were performed using the SPSS 22.0 statistical package. The quantitative data are presented as mean ± SD and were analyzed using ANOVA or two tailed Student’s t-tests. The Wilcoxon signed rank test was used to compare the HDAC3 and FOXA1 expression in the same patients. The two-sample rank sum test was used to analyze the relationship between HDAC3 and FOXA1expression and the prognosis. A two-sided P-value < 0.05 was considered statistically significant25.