As mentioned previously (Yan et al., 2018; Zhou et al., 2019) several bioinformatics network resources were used to reanalyze the molecular profiles of ADARB1 in OC patients (Supplementary Table S1). Gene Expression Profiling Interactive Analysis (GEPIA) (Tang et al., 2017) and University of Alabama Cancer Database (UALCAN) (Chandrashekar et al., 2017) were used to identify the expression profiles of ADARB1 in OC tissues. For the prognostic analysis, the Kaplan–Meier Plotter (Lánczky et al., 2016) and DRUGSURV database (Amelio et al., 2014) were utilized to describe the relationship between ADARB1 expression and patients’ prognosis.

From the cBioPortal web tool (Gao et al., 2013), the genes coexpressed with ADARB1 in OC were downloaded (Supplementary Table S2). Then, the STRING (Szklarczyk et al., 2017) and Cytoscape software (Reimand et al., 2019) were used to complete the protein–protein interaction (PPI) network of these coexpression genes. Then, we utilized the WebGestalt (Wang et al., 2017) to conduct the GO and KEGG pathway analysis of ADARB1 coexpression genes in OC samples.

The ovarian cancer tissues and paired adjacent tissues were obtained from the Xiangya hospital, central south university. The ethical approval number is 20210205. Specimens were deparaffinized in xylene and rehydrated in a series of graded alcohol. Antigen retrieval was completed after heating in citrate buffer. Endogenous peroxidase was blocked using 3% H₂O₂ for 20 min. The sections were incubated with the primary antibody against ADARB1, p-AKT, CXCL12 and KDR at 37°C for 1 h. Horseradish peroxidase-conjugated secondary antibody was added and incubated at room temperature for 30 min, and 3,3′-diaminobenzitine (DAB) solution was used for color reaction (ZSGB-BIO, China).

Human OC cells OVCAR3, HO8910pm, A2780 and normal ovarian epithelial cells IOSE were obtained from department of Pathology, School of Basic Medical Sciences, Central South University. OVCAR3, HO8910pm and IOSE were maintained in Roswell Park Memorial Institute (RPMI)-1640 medium (BI, Israel Beit Haemek Ltd.) with 10% fetal bovine serum (FBS), while A2780 was grown in DMEM (BI, Israel Beit Haemek Ltd.) supplemented with 10% fetal bovine serum FBS at 37°C and 5% CO2. MK2206 was purchased from Selleck Chemicals and dissolved in dimethylsulfoxide (DMSO) (Sigma, United States), The exposed concentrations of MK2206 were 5 mM. Antibodies against ADARB1 and p-AKT were purchased from Proteintech (United States, 22248-1-AP and 66444-1-Ig). Antibodies against GAPDH and CXCL12 were purchased from Abclonal (China, AC002 and A1325), Antibody against KDR was purchased from Affinity (China, AF6281). Antibodies against AKT and p-AKT were from Cell Signaling Technology (United States). The second antibody were purchased from ZSGB-BIO.

Cells were lysed in RIPA buffer and protein concentrations were determined by a BCA protein assay kit (PP1002, BioTeck, Beijing, China). A total of 30–40 μg of protein was separated by 10% SDS-PAGE and electroblotted onto polvinylidene fluoride membranes (Millipore, Merck, United States) for detection using indicated antibodies. Immunoreactions were detected by an imaging system (Alpha FluorChem FC3).

Total RNA was extracted employing TRIzol reagent (Invitrogen) according to the manufacturter’s instruction and reverse transcribed to cDNA using the PrimeScript™ RT reagent kit (Abclonal, China). The RT-PCR assay was conducted through iTaqTM Universal SYBR green Supermix (Abclonal, China), with GAPDH as the internal control. The forward and reverse primer sequences were used as follows: ADARB1: 5′-CGC​GCC​TTT​GTT​TGT​CAT​GTC-3′ and 5′-GGA​AAC​TGA​ACG​AAA​GAC​CTC​AA-3′; GAPDH: 5′-CAG​CAA​GAG​CAC​AAG​AGG​AA-3′ and 5′-ATG​GTA​CAT​GAC​AAG​GTG​CGG-3′. Relative expression levels were decided using the 2-ΔΔCT method.

The logarithmic growth phase cells were cultured in a 96-well culture plate for 6 days with five replicates per group. MTT assay was performed 48 h after transfection. Add 10 μl MTT solution to each well at the end of each day. After 4 h of continuous culture, the culture medium was discarded, and DMSO (150 μl/well) was added 6 days later, and the medium was shaken for 10 min. The cell proliferation rate was calculated by measuring the optical density (OD) of each well at 490 nm on a Microplate Reader.

A total of 1,000 cells were seeded in 6-well plates and incubated at 37°C with 5% CO2 for 10 days. Culture plates were performed in duplicates. After a wash with phosphate buffer saline (PBS), cultures were fixed with methanol for 20 min and stained with crystal violet for 15 min. Colonies were examined and calculated.

The cells were inoculated into 6-well plates and cultured in complete medium (37°C, 5% CO2) to at least 95% confluence before wounds were created. To measure the cell migration, a plastic 100 ul pipette tip was used to scrape cells in a monolayer to creating wounds. Then, washed them three times in PBS and incubated with FBS-free RPMI-1640 medium. Subsequently, cells were cultured in either medium for 0, 12, 24 h. At the end of the incubation period, phase-contrast microscopy was employed to photograph the wounded area and migration cells at the wounded area. Finally, the extent to which the wound had closed over 24 h was calculated and expressed as a percentage of the difference between times 0 and 24 h using ImageJ software.

For invasion assays, Transwells (corning, United States) with 8-μm pore size filters covered with matrigel were inserted into 24-well plates. The cells were serum-starved overnight and then added in the upper chamber (5 × 10⁴ cells per insert) and the complete culture medium supplemented with 10% FBS was used as a chemoattractant in the lower chamber. After 24 h of incubation, non-invading cells that remained on the upper surface of the filter were removed, and the cells that had passed through the filter and attached to the bottom of the membrane were fixed in methanol and stained with 0.2% Crystal violet. Numbers of the invasive cells in seven randomly selected fields from triplicate chambers were counted in each experiment under a phase-contrast microscope.

Statistical analyses were performed using SPSS 22.0 for Windows. Data were presented as mean ± standard deviation (SD) of at least three independent experiments. Statistical analysis was performed using Student’s t test and one-way ANOVA. The Statistical significance is shown in the figures with *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001.

To evaluate changes in ADARB1 expression in OC and normal ovarian tissues, we analyzed the expression levels of ADARB1 through several bioinformatics databases. The GEPIA analysis revealed that ADARB1 mRNA expression was significantly decreased in OC tissues (Figure 1A). From TCGA-OV, we discovered the ADARB1 expression was clearly reduced in OC compared with the normal tissues (Figure 1B). Similarly, using the TNM plot (Figure 1C) and UALCAN (Figure 1D), we further confirmed the downregulation of ADARB1 in OC tissues. In addition, we found that OC cells OVCAR3, HO8910pm and A2780 displayed a lower level of ADARB1 compared to the normal ovarian cell IOSE (Figures 1E,F). Moreover, the expression of ADARB1 in tumor tissues was lower than that in adjacent tissues (Supplementary Figure S1). Next, we analyzed the effects of ADARB1 expression on the clinical characteristics of patients with OC. From DRUGSURV database, we found that the patients with low level of ADARB2 displayed poor prognosis in GSE14764 (Figure 2A). In addition, Kaplan-Meier Plotter database was employed to further evaluate the effects of ADARB1 expression on survival, revealing that low levels of ADARB1 expression were correlated with shorter progression-free survival (PFS) (Figure 2B). Therefore, these results suggested that ADARB1 expression levels might serve as an indicator for the clinical prognosis of OC patients. Taken together, the decreased expression of ADARB1 in OC tissues and cell lines suggested its anti-oncogenic roles.

To further identify the function of ADARB1 in OC, we overexpressed of ADARB1 in HO8910pm and OVCAR3 ovarian cancer cells (Figures 3A,B). After then, we tested the roles of ADARB1 overexpression in cell proliferation, migration and invasion. The MTT assay and colony formation assay showed that ADARB1 overexpressed significantly decreased the proliferation ability in HO8910pm and OVCAR3 cells (Figures 3C–F). Similarly, we observed that ADARB1-OE inhibited the metastasis and invasion ability in HO8910pm and OVCAR3 cells, as indicated by the wound healing and transwell assays (Figures 4A–D). These results suggest that ADARB1 could suppress the malignant biological behaviors of OC cells.

The PI3K-AKT pathway is a well-known tumor signaling pathway that is involved in the development of various tumors, including glioma (Dai et al., 2017), ovarian cancer (Wu et al., 2021), etc. Moreover, phosphorylation activation of AKT signaling can promote the malignant progression of ovarian cancer (Samartzis et al., 2020; Zhai et al., 2020). We found that the phosphorylation levels of AKT (p-AKT) in tumor tissues was higher than that in adjacent tissues (Supplementary Figure S1). And ADARB1-OE can effectively inhibit p-AKT without affecting the total AKT level. Then, we investigated whether the effect of ADARB1 on OC cell proliferation and metastasis was associated with AKT activity. Western blot analysis showed that p-AKT was obviously decreased when OC cells were overexpressed ADARB1 or treated with MK2206, a specific inhibitor of AKT. The inhibitory effect on p-AKT levels was further increased by combinational treatment (Figure 5A). We performed the MTT assay and colony formation assay and observed that OC cells treated with both MK2206 and ADARB1-OE had a significantly suppression effect of cell proliferation compared with the control group and either of the individual treatment groups (Figures 5B–D). Moreover, we examined cell migration and invasion capacity of HO8910pm and OVCAR3 cells under ADARB1-OE and MK2206 treatment. Compared to the control group, OC cells showed a significantly weaker migration and invasion ability when treated with MK2206 or ADARB1-OE (Figures 6A–D). The above results indicated that anti-cancer activity by ADARB1 overexpression might be due to the inhibition of AKT signaling.

To further investigate the biological functions of ADARB1, functional enrichment analysis was performed using ADARB1 related coexpressed genes. By using the cBioPortal database, we screened 1399 ADARB1-associated coexpressed genes with the criteria of p value ≤ 0.05 and |Spearman’s correlation|≥0.3 (Supplementary Table S2). Next, a PPI network was performed using STRING and Cytoscape software (Figure 7A). Meanwhile, the GO annotation (Figure 7B) and KEGG pathway (Figure 7C) were further analyzed using WebGestalt database. Biological processes indicated that these coexpressed genes are mainly involved in metabolic processes and biological regulation. For cell component, these coexpressed genes were mainly located on the nucleus and membrane. In the aspect of molecular function, the coexpressed genes were primarily enriched in protein binding and ion binding. The KEGG pathway demonstrated that these coexpressed genes were mainly related to the ribonucleotide metabolic process and mitochondrial transport.

Increasing evidence demonstrated that ADAR family members play important roles in immune regulation (Heraud-Farlow and Walkley, 2020; Yanai et al., 2020). Therefore, from TISIDB database, we investigated the relationship between ADARB1 expression and immune infiltration. As shown in Figure 8A, the expression levels of ADARB1 positively correlated with several immune cells, such as natural killer cell (NK cells), central Memory T cell (Tcm), T effector memory (Tem) and eosinophils. Figure 8B further confirmed the positive correlation between ADARB1 expression and NK cells (Spearman r = 0.546), Tcm (Spearman r = 0.45), Tem (Spearman r = 0.35) and eosinophils (Spearman r = 0.24). Similarly, ADARB1 was positively correlated with eosinophil and NK cells by using the TISIDB database (Supplementary Figure S2). Moreover, we analyzed the associations between ADARB1 expression and the immunomodulators and chemokines. Supplementary Figure S3 showed the positive correlations between ADARB1 expression and several immunomodulators, including CD160 (Spearman r = 0.211), CSF1R (Spearman r = 0.231), KDR (Spearman r = 0.3) and TGFBR1 (Spearman r = 0.208). Supplementary Figures S4A,B showed the positive correlations between ADARB1 expression and several chemokines, including CCL14 (Spearman r = 0.153), CCL21 (Spearman r = 0.136), CXCL12 (Spearman r = 0.224) and CXCL14 (Spearman r = 0.18), while the ADARB1 expression levels were negative correlation with CCL13 (Spearman r = −0.13), CXCL8 (Spearman r = −0.143), CXCL13 (Spearman r = −0.129) and XCL1 (Spearman r = −0.194). In addition, we found the low level of ADARB1, KDR and CXCL2 in tumor tissues compared with the adjacent tissues by immunohistochemistry, which indicated that the expression of ADARB1 was positively correlated with the expression of KDR and CXCL12 (Supplementary Figure S1). We hypothesized that ADARB1 might have a significant effect on immune regulation in OC since it is obviously associated with various types of tumor-infiltrating lymphocytes, immunomodulators, and chemokines in OC.