Tanshinone IIA inhibits epithelial-mesenchymal transition in ovarian cancer by promoting NR3C2 transcriptional activation of ING1 Authors: Xiaoqin Wang, Duoyi Zhang, Tingting Wang, Xin Li, Jing Zhang Published: 2026 ABSTRACT SummaryOvarian cancer (OV) is among the most lethal forms of gynecological cancers. This work aims to investigate the roles and underlying mechanisms of Tanshinone IIA (Tan IIA) in the epithelial-to-mesenchymal transition (EMT) process in OV. Tan IIA inhibited OV cell proliferation, migration, invasion, and EMT in a dose-dependent manner. Knockdown of nuclear receptor subfamily 3 group C member 2 (NR3C2) weakened the efficacy of Tan IIA, while combined overexpression of inhibitor of growth protein 1 (ING1) rescued the OV progression promoted by NR3C2 knockdown. NR3C2 occupied the ING1 promoter to exert transcriptional activation. Patients with low NR3C2/ING1 expression were associated with more advanced FIGO staging, larger ascites volume, and peritoneal metastasis, and had significantly poorer survival outcomes. Collectively, Tan IIA upregulates NR3C2 expression and activates ING1 transcription to exert its anti-metastatic effects by inhibiting EMT in OV. The study also underscored the prognostic potential of NR3C2/ING1 for OV. Introduction According to global epidemiological reports, there has been an observed increase in the incidence of ovarian cancer (OV).1 A significant proportion of patients are diagnosed at a late stage (FIGO stage III or IV) with peritoneal metastasis, with 70% of cases falling into this category.2,3 While a majority of patients diagnosed with the condition experienced a positive response to initial treatment, the majority of these patients experienced relapse within three years, and treatment outcomes after relapse were poor due to significant drug resistance.4 Consequently, the development of effective therapeutic strategies must be a priority. Given that most OV cells are of epithelial origin, the epithelial-to-mesenchymal transition (EMT) is the most probable transition to occur, during which cancer cells gain a more spindle-like shape and an enhanced migratory capacity.5 Tanshinone, a lipophilic compound derived from the roots of Salvia miltiorrhiza (S. miltiorrhiza), has been proven to be highly effective in combating tumors.6 Tanshinone IIA (Tan IIA) represents a significant advancement in the field of traditional medicine and plays a pivotal role in the digestive, respiratory, and circulatory systems, as well as in other tumor diseases.7 For instance, Tan IIA compound has demonstrated noteworthy anticancer properties through multiple mechanisms, including the inhibition of tumor cell growth and proliferation, metastasis, invasion, and angiogenesis.8 Moreover, Tan IIA reversed EGF- and TGF-β1-mediated EMT in HepG2 cells.9 However, its specific role in EMT in OV has not been revealed. Nuclear receptor subfamily 3 group C member 2 (NR3C2, mineralocorticoid receptor) is a member of the family of steroid receptors and functions primarily as a nuclear receptor to regulate gene expression.10 Overexpression of NR3C2 has been recently reported to repress EMT in vitro and lung metastasis in vivo in colorectal cancer.11 Our preliminary prediction results revealed NR3C2 as a target protein of Tan IIA that might relate to EMT in OV, and inhibitor of growth protein 1 (ING1) as a target of the transcription factor NR3C2. The ING proteins (ING1, ING2, ING3, ING4, and ING5) are a family of epigenetic regulators, and their downregulation in different cancers has led to the identification of the ING proteins as gatekeeper tumor suppressors.12 Interestingly, ING1 knockdown increased expression of Twist1, Zeb1, Zeb2, and Snai1, thereby promoting EMT and cell motility in breast cancer.13 This leads us to hypothesize that Tan IIA might promote the protein expression of NR3C2, thereby increasing ING1 expression to repress EMT in OV. Results Tan IIA inhibits OV cell proliferation, migration, invasion, and EMT in a dose-dependent manner We incubated SK-OV-3 and OVCAR-3 cells with different concentrations of Tan IIA and found through CCK-8 assay that as the concentration of Tan IIA increased, cell viability gradually decreased and reached the viability inhibition threshold at 150 μM. Therefore, higher doses of Tan IIA (200 and 250 μM) were not used in subsequent experiments (Figure 1A). Clonogenic assays also showed that Tan IIA dose-dependently inhibited the colony-forming ability of OV cells (Figure 1B). Subsequently, we also employed the EMT inhibitor SB431542 as a positive control to investigate the effects of Tan IIA on the migration and invasion of OV cells and on EMT markers. Wound healing experiments (Figure 1C), Transwell migration (Figure 1D), and Matrigel invasion experiments (Figure 1E) demonstrated that the repressive effects of Tan IIA on OV cell migration and invasion capacity were dose-dependent. Immunofluorescence staining showed that as the Tan IIA dose increased, E-cadherin expression in OV cells gradually increased, while N-cadherin expression gradually decreased (Figure 1F). Notably, the EMT inhibitor SB431542, used as a positive control, produced EMT inhibitory effects comparable to and more pronounced than those of 150 μM Tan IIA. This confirms the reliability of the experimental system for detecting EMT-related phenotypes and suggests that Tan IIA may possess anti-metastatic potential comparable to that of targeting this pathway.Figure 1Tan IIA inhibits OV cell proliferation, migration, invasion, and EMT in a dose-dependent manner(A) The effects of different concentrations (50, 100, 150, 200, and 250 μM) of Tan IIA on the viability of SK-OV-3 and OVCAR-3 cells were examined using CCK-8.(B) The effects of 50, 100, and 150 μM Tan IIA on the colony-forming ability of OV cells were examined using a clonogenic assay.(C) The effects of 50, 100, and 150 μM Tan IIA on the migration of OV cells were examined using the wound healing assay (scale bars, 100 μm).(D and E) The effects of 50, 100, and 150 μM Tan IIA on the migration and invasion of OV cells were examined using the Transwell assay (scale bars, 50 μm).(F) The MFI of E-cadherin and N-cadherin in OV cells was examined using immunofluorescence (scale bars, 20 μm). n = 3 biologically independent experiments. One-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. Tan IIA inhibits OV cell proliferation, migration, invasion, and EMT in a dose-dependent manner (A) The effects of different concentrations (50, 100, 150, 200, and 250 μM) of Tan IIA on the viability of SK-OV-3 and OVCAR-3 cells were examined using CCK-8. (B) The effects of 50, 100, and 150 μM Tan IIA on the colony-forming ability of OV cells were examined using a clonogenic assay. (C) The effects of 50, 100, and 150 μM Tan IIA on the migration of OV cells were examined using the wound healing assay (scale bars, 100 μm). (D and E) The effects of 50, 100, and 150 μM Tan IIA on the migration and invasion of OV cells were examined using the Transwell assay (scale bars, 50 μm). (F) The MFI of E-cadherin and N-cadherin in OV cells was examined using immunofluorescence (scale bars, 20 μm). n = 3 biologically independent experiments. One-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. Given that the 150 μM dose of Tan IIA exhibited the strongest inhibitory effects on OV cell proliferation, migration, invasion, and EMT phenotypes, this dose was used for subsequent in vitro experiments. Tan IIA causes moderate cytotoxicity to OV cells Flow cytometry analysis revealed that the positive control STS induced an extremely potent apoptotic response. Compared to the PBS group, 150 μM Tan IIA significantly induced apoptosis in OV cells, yet its intensity was markedly lower than that observed in the positive control STS group (Figure S1A). Meanwhile, LDH release data indicated that 150 μM Tan IIA did not induce the extensive membrane disruption observed with the Triton X-100 treatment (positive control) (Figure S1B). As shown in Figure S1C, the ROS increase induced by 150 μM Tan IIA was relatively limited compared to the dramatic ROS surge caused by H2O2 (positive control). The above findings suggested that the changes induced by 150 μM Tan IIA did not reach the level of acute cytotoxicity, widespread necrotic cell death, membrane rupture, or oxidative stress burst. Tan IIA inhibits tumor growth and metastasis in nude mice We selected SK-OV-3 cells to establish a xenograft tumor model and treated the mice with low-dose (10 mg/kg) and high-dose (30 mg/kg) Tan IIA. Tan IIA inhibited subcutaneous tumor growth (Figure 2A) and weight (Figure 2B) in a dose-dependent manner. IHC staining showed that Tan IIA treatment reduced the IHC scores of EMT-related proteins (Twist1, Snail, and Slug) in xenograft tumors (Figure 2C). TUNEL analysis revealed that Tan IIA induced apoptosis in tumor tissues (Figure 2D).Figure 2Tan IIA inhibits metastasis in OV by blocking EMT(A) Tumor volume was measured every 5 days.(B) Mice were euthanized after 25 days, and their tumors were weighed.(C) IHC scores of Twist1, Snail, and Slug in xenograft tumor tissues (scale bars, 20 μm).(D) The apoptosis rate of tumors was examined using TUNEL (scale bars, 20 μm).(E) Lung tissue was collected from nude mice with intravenous injection of SK-OV-3 cells and analyzed by H&E staining to determine the number of metastatic nodules (Scale bars: 50 μm).(F) Peritoneal tumor burden in C57BL/6J mice with intraperitoneal injection of ID8 cells, with quantitative assessment of metastatic tumor nodule count and tumor weight. n = 5 mice/group. One-way or two-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. Tan IIA inhibits metastasis in OV by blocking EMT (A) Tumor volume was measured every 5 days. (B) Mice were euthanized after 25 days, and their tumors were weighed. (C) IHC scores of Twist1, Snail, and Slug in xenograft tumor tissues (scale bars, 20 μm). (D) The apoptosis rate of tumors was examined using TUNEL (scale bars, 20 μm). (E) Lung tissue was collected from nude mice with intravenous injection of SK-OV-3 cells and analyzed by H&E staining to determine the number of metastatic nodules (Scale bars: 50 μm). (F) Peritoneal tumor burden in C57BL/6J mice with intraperitoneal injection of ID8 cells, with quantitative assessment of metastatic tumor nodule count and tumor weight. n = 5 mice/group. One-way or two-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. We also constructed a lung metastasis model by injecting SK-OV-3 cells into the tail vein of BALB/c nude mice and found through H&E staining that Tan IIA limited the number of lung metastasis nodules (Figure 2E). We further established an ID8 cell peritoneal dissemination model in C57BL/6J mice. Macroscopic examination of mouse peritoneal cavities revealed extensive tumor nodules throughout the mesenteric and parietal peritoneal layers in the saline group. The Tan IIA treatment groups demonstrated a significant reduction in both the number of metastatic nodules within the peritoneal cavity and the total weight of peritoneal metastases (Figure 2F). Additionally, to assess the safety of Tan IIA, we systematically monitored changes in body weight in mice and performed histopathological analysis of major organs. Compared with the saline group, neither the 10 mg/kg nor the 30 mg/kg of Tan IIA caused significant changes in mouse body weight (Figure S2A). H&E staining demonstrated that no significant structural abnormalities or pathological damage were observed in the hearts, livers, lungs, or kidneys of mice across all groups (Figure S2B). Tan IIA treatment significantly upregulates NR3C2 To identify the molecular targets of Tan IIA, a search was conducted in PubChem (https://pubchem.ncbi.nlm.nih.gov/) using the keyword “Tanshinone IIA” to download the SMILES structure formula CC1 = COC2 = C1C(=O)C(=O)C3 = C2C = CC4 = C3CCCC4(C)C) (Figure 3A). Based on the SMILES structure, we set AUC ≥0.7 and Fingerprint type: ECFP4 in TargetNet (http://targetnet.scbdd.com/home/index/) to download the potential targets of Tan IIA. At the same time, we predicted the potential targets of Tan IIA in Super-PRED (https://prediction.charite.de/). Genes related to EMT and OV were downloaded from GeneCards (https://www.genecards.org/) using the keywords “Epithelial-mesenchymal transition” and “Ovarian cancer”, respectively. Gene expression profiles were analyzed in the GSE36668 for serous OV tissue and superficial scraping from normal ovary (four cases each), as well as in GSE66957 for 57 cases of primary OV and 12 ovarian normal samples. The criteria adj. p value < 0.01 and |log2FoldChange| > 1 were set, differentially expressed genes between cancer and normal tissues were identified, and the volcano plot tool in Hiplot Pro (https://hiplot.com.cn/) was used to visualize the differentially expressed genes (Figures 3B and 3C). In Jvenn (https://jvenn.toulouse.inrae.fr/), the intersection of these six sets of data was taken to obtain the only intersecting gene: NR3C2 (Figure 3D). NR3C2 expression was downregulated in the OV tissues of GSE36668 (log2FoldChange = −3.04075981) and GSE66957 (log2FoldChange = −1.3508498) datasets. We speculate that Tan IIA may inhibit EMT in OV by promoting NR3C2 expression.Figure 3Tan IIA treatment significantly upregulates NR3C2(A) Chemical structure of Tan IIA.(B) The volcano plot of differentially expressed genes in the GSE36668 dataset.(C) The volcano plot of the differentially expressed genes in the GSE66957 dataset.(D) Common intersection of potential targets of Tan IIA predicted by TargetNet, potential targets of Tan IIA predicted by Super-PRED, GSE36668-DEG, GSE66957-DEG, EMT-related genes, and OV-related genes.(E) NR3C2 expression in SK-OV-3 and OVCAR-3 cells after treatment with 50, 100, and 150 μM Tan IIA was examined using western blot analysis (n = 3 biologically independent experiments).(F) NR3C2 expression in subcutaneous xenograft tumors after Tan IIA injection at low (10 mg/kg) and high (30 mg/kg) doses was examined using western blot analysis (n = 5 mice/group).(G) The predicted binding mode of Tan IIA within the NR3C2 protein binding pocket.(H) NR3C2 sensitivity to protease degradation following treatment with different concentrations (50, 100, and 150 μM) of Tan IIA or PBS was examined using the DARTS assay (n = 3 biologically independent experiments).(I) NR3C2 protein thermal stability following Tan IIA treatment was examined using CETSA (n = 3 biologically independent experiments).(J) NR3C2 protein enriched by Biotin-Tan IIA-conjugated magnetic beads was examined using a pull-down assay (n = 3 biologically independent experiments). One-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. Tan IIA treatment significantly upregulates NR3C2 (A) Chemical structure of Tan IIA. (B) The volcano plot of differentially expressed genes in the GSE36668 dataset. (C) The volcano plot of the differentially expressed genes in the GSE66957 dataset. (D) Common intersection of potential targets of Tan IIA predicted by TargetNet, potential targets of Tan IIA predicted by Super-PRED, GSE36668-DEG, GSE66957-DEG, EMT-related genes, and OV-related genes. (E) NR3C2 expression in SK-OV-3 and OVCAR-3 cells after treatment with 50, 100, and 150 μM Tan IIA was examined using western blot analysis (n = 3 biologically independent experiments). (F) NR3C2 expression in subcutaneous xenograft tumors after Tan IIA injection at low (10 mg/kg) and high (30 mg/kg) doses was examined using western blot analysis (n = 5 mice/group). (G) The predicted binding mode of Tan IIA within the NR3C2 protein binding pocket. (H) NR3C2 sensitivity to protease degradation following treatment with different concentrations (50, 100, and 150 μM) of Tan IIA or PBS was examined using the DARTS assay (n = 3 biologically independent experiments). (I) NR3C2 protein thermal stability following Tan IIA treatment was examined using CETSA (n = 3 biologically independent experiments). (J) NR3C2 protein enriched by Biotin-Tan IIA-conjugated magnetic beads was examined using a pull-down assay (n = 3 biologically independent experiments). One-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. To demonstrate that NR3C2 is a protein target of Tan IIA, we conducted Western blot analysis. As the concentration of Tan IIA increased, the protein expression of NR3C2 in OV cells also gradually increased (Figure 3E). In addition, the upregulation of NR3C2 was also found in the tumor tissues from mice injected with Tan IIA (Figure 3F). We performed molecular docking using the DeepMice platform and calculated the binding free energy between Tan IIA and NR3C2 based on AutoDock and Vina scoring functions, predicting a value of −9.823 kcal/mol (Figure 3G). This indicates a strong binding affinity between the two and a stable complex. We also employed the DARTS assay to examine the interaction between Tan IIA and NR3C2. The addition of protease significantly reduced NR3C2 protein levels, while the introduction of Tan IIA led to a dose-dependent increase in NR3C2 protein concentration, suggesting that Tan IIA enhanced NR3C2’s resistance to protease degradation (Figure 3H). CETSA results showed that NR3C2 expression in the PBS group gradually decreased with increasing temperature, while Tan IIA inhibited temperature-induced NR3C2 protein degradation (Figure 3I). The final pull-down experiment demonstrated that incubation with magnetic beads conjugated to biotin-labeled Tan IIA (Biotin-Tan IIA) resulted in detectable enrichment of NR3C2 protein, whereas no enrichment was observed on control beads (Figure 3J). These findings indicate that Tan IIA can directly interact with the NR3C2 protein. Combined NR3C2 knockdown promotes malignant progression of OV in the presence of Tan IIA To validate the above hypothesis, we constructed three groups of NR3C2 knockdown (KD) cells by infecting them with hNR3C2 [shRNA] and verified the successful downregulation of NR3C2 expression in OV cells (Figures 4A and 4B). After treatment with 150 μM Tan IIA, clonogenic formation assays (Figure 4C), wound healing assays (Figure 4D), Transwell migration assays (Figure 4E), and Matrigel invasion assays (Figure 4F) revealed that OV cells with combined knockdown of NR3C2 showed significantly increased colony formation, wound healing rates, migration, and invasion. Furthermore, combined knockdown of NR3C2 reduced E-cadherin expression but enhanced N-cadherin expression in OV cells (Figure 4G).Figure 4NR3C2 knockdown promotes the malignant phenotypes of OV cells in the presence of Tan IIA(A and B) NR3C2 expression in SK-OV-3 and OVCAR-3 cells treated with NR3C2 knockdown (KD) #1-#3 was examined using RT-qPCR and western blot.(C) The effects of 150 μM Tan IIA and NR3C2 knockdown on the colony-forming ability of OV cells were examined using a clonogenic assay.(D) The effects of 150 μM Tan IIA and NR3C2 knockdown on the migration of OV cells were examined using the wound healing assay (scale bars, 100 μm).(E and F) The effects of 150 μM Tan IIA and NR3C2 knockdown on the migration and invasion of OV cells were examined using the Transwell assay (Scale bars: 50 μm).(G) The MFI of E-cadherin and N-cadherin in OV cells was examined using immunofluorescence (scale bars, 20 μm). n = 3 biologically independent experiments. One-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. NR3C2 knockdown promotes the malignant phenotypes of OV cells in the presence of Tan IIA (A and B) NR3C2 expression in SK-OV-3 and OVCAR-3 cells treated with NR3C2 knockdown (KD) #1-#3 was examined using RT-qPCR and western blot. (C) The effects of 150 μM Tan IIA and NR3C2 knockdown on the colony-forming ability of OV cells were examined using a clonogenic assay. (D) The effects of 150 μM Tan IIA and NR3C2 knockdown on the migration of OV cells were examined using the wound healing assay (scale bars, 100 μm). (E and F) The effects of 150 μM Tan IIA and NR3C2 knockdown on the migration and invasion of OV cells were examined using the Transwell assay (Scale bars: 50 μm). (G) The MFI of E-cadherin and N-cadherin in OV cells was examined using immunofluorescence (scale bars, 20 μm). n = 3 biologically independent experiments. One-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. NR3C2 silencing compromises Tan IIA-inhibited tumor growth and metastasis Subsequently, we subcutaneously injected SK-OV-3 cells from the KD-NR3C2 #1 group, which has the optimal NR3C2 efficiency, into the right flank of nude mice. After high-dose (30 mg/kg) Tan IIA treatment, exogenous knockdown of NR3C2 significantly accelerated tumor development (Figures 5A and 5B). IHC scores for Twist1, Snail, and Slug were increased (Figure 5C), while cell apoptosis rates were significantly reduced in xenograft tumors formed by SK-OV-3 cells with NR3C2 knockdown (Figure 5D). In the lung metastasis model, combined knockdown of NR3C2 also accelerated tumor lung metastasis, with an increase in the number of lung metastasis nodules (Figure 5E).Figure 5Exogenous knockdown of NR3C2 rescues Tan IIA-inhibited tumor growth and metastasis(A) Starting from the beginning of Tan IIA treatment, the volume of subcutaneous xenografts in nude mice was measured every 5 days (n = 5 mice/group).(B) Mice were euthanized after 25 days, and their tumors were weighed (n = 5 mice/group).(C) IHC scores of Twist1, Snail, and Slug in xenograft tumor tissues (n = 5 mice/group) (Scale bars: 20 μm).(D) The apoptosis rate of tumors was examined using TUNEL (n = 5 mice/group) (Scale bars: 20 μm).(E) Lung tissue was collected from nude mice with intravenous injection of SK-OV-3 cells and analyzed by H&E staining to determine the number of metastatic nodules (n = 5 mice/group) (scale bars, 50 μm).7(F) Nr3c2 mRNA levels in ID8 cells following infection with Nr3c2-knockdown lentivirus were examined using RT-qPCR (n = 3 biologically independent experiments).(G) Nr3c2 protein levels in ID8 cells following infection with Nr3c2-knockdown lentivirus were examined using western blot (n = 3 biologically independent experiments).(H) Peritoneal tumor burden in C57BL/6J mice with intraperitoneal injection of ID8 cells, with quantitative assessment of metastatic tumor nodule count and tumor weight (n = 5 mice/group). Unpaired t test for two comparisons and two-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. Exogenous knockdown of NR3C2 rescues Tan IIA-inhibited tumor growth and metastasis (A) Starting from the beginning of Tan IIA treatment, the volume of subcutaneous xenografts in nude mice was measured every 5 days (n = 5 mice/group). (B) Mice were euthanized after 25 days, and their tumors were weighed (n = 5 mice/group). (C) IHC scores of Twist1, Snail, and Slug in xenograft tumor tissues (n = 5 mice/group) (Scale bars: 20 μm). (D) The apoptosis rate of tumors was examined using TUNEL (n = 5 mice/group) (Scale bars: 20 μm). (E) Lung tissue was collected from nude mice with intravenous injection of SK-OV-3 cells and analyzed by H&E staining to determine the number of metastatic nodules (n = 5 mice/group) (scale bars, 50 μm).7 (F) Nr3c2 mRNA levels in ID8 cells following infection with Nr3c2-knockdown lentivirus were examined using RT-qPCR (n = 3 biologically independent experiments). (G) Nr3c2 protein levels in ID8 cells following infection with Nr3c2-knockdown lentivirus were examined using western blot (n = 3 biologically independent experiments). (H) Peritoneal tumor burden in C57BL/6J mice with intraperitoneal injection of ID8 cells, with quantitative assessment of metastatic tumor nodule count and tumor weight (n = 5 mice/group). Unpaired t test for two comparisons and two-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. ID8 cells stably knocked down for Nr3c2 (Figures 5F and 5G) were intraperitoneally injected into mice. Knockdown of Nr3c2 also reversed the inhibitory effect of high-dose Tan IIA on peritoneal metastasis. Compared with the control group, mice with combined Nr3c2 knockdown exhibited substantial tumor nodules in the peritoneal cavity, mesentery, and parietal peritoneum. The weight of metastatic tumors was significantly increased (Figure 5H). NR3C2 binds to the ING1 promoter and activates its transcription We downloaded the potential target genes regulated by NR3C2 from hTFtarget (https://guolab.wchscu.cn/hTFtarget//#!/) and performed a cross-screening analysis with the differentially expressed genes obtained from GSE36668 and GSE66957 datasets, as well as the EMT- and OV-related genes from GeneCards, identifying 9 intersecting genes: SCRIB/CFL1/S100A11/ZMYND8/ENO1/ING1/GSTP1/GAPDH/S100A6 (Figure 6A). We displayed the expression of the corresponding probes for these intersecting genes in the GSE36668 and GSE66957 datasets as heat maps using SRplot (https://www.bioinformatics.com.cn/SRplot) (Figures 6B and 6C). The only downregulated expressed gene in the OV tissues of GSE36668 is: 244177_at (ING1: log2FoldChange = −1.69857405), and its corresponding probe in GSE66957, merck2-BF510437_at (ING1: log2FoldChange = −1.3247472), also shows downregulation in the OV tissue. Considering that most nuclear receptors contain a variable amino-terminal domain that contributes to transcriptional activation functions,14 we identified ING1 as a downstream target of NR3C2. Correlation analysis using GEPIA2 (http://gepia2.cancer-pku.cn/#index) showed that NR3C2 expression was positively correlated with ING1 expression in OV (Figure 6D). ChIP-seq from UCSC (https://genome.ucsc.edu/cgi-bin/hgGateway) showed that NR3C2 bound to the promoter region of ING1 (chr13: 110713411–110713760) (Figure 6E). We thus hypothesized that Tan IIA may inhibit EMT in OV by promoting NR3C2 transcription activation of ING1.Figure 6NR3C2 binds to the ING1 promoter and activates its transcription(A) Common intersection of potential target genes regulated by NR3C2, GSE36668-DEG, GSE66957-DEG, EMT-related genes, and OV-related genes.(B) The heatmap of nine intersecting genes in the GSE36668 dataset.(C) The heatmap of nine intersecting genes in the GSE66957 dataset.(D) The expression correlation between NR3C2 and ING1 in OV tumors was predicted using the GEPIA2 Website.(E) The binding peaks of NR3C2 and the ING1 promoter (chr13: 110713411–110713760) were predicted using the UCSC ChIP-seq database.(F) ING1 expression levels in SK-OV-3 and OVCAR-3 cells with NR3C2 knockdown were measured using RT-qPCR.(G) The specific binding of NR3C2 to the ING1 promoter (chr13: 110713411–110713760) in OV cells was measured using ChIP-qPCR.(H) Luciferase activity in the OV cells stably infected with NR3C2 knockdown and co-transfected with the constructed plasmid containing the ING1 promoter (chr13: 110713411–110713760) and the internal reference plasmid. n = 3 biologically independent experiments. Unpaired t test for two comparisons. Data are shown as mean ± SEM. NR3C2 binds to the ING1 promoter and activates its transcription (A) Common intersection of potential target genes regulated by NR3C2, GSE36668-DEG, GSE66957-DEG, EMT-related genes, and OV-related genes. (B) The heatmap of nine intersecting genes in the GSE36668 dataset. (C) The heatmap of nine intersecting genes in the GSE66957 dataset. (D) The expression correlation between NR3C2 and ING1 in OV tumors was predicted using the GEPIA2 Website. (E) The binding peaks of NR3C2 and the ING1 promoter (chr13: 110713411–110713760) were predicted using the UCSC ChIP-seq database. (F) ING1 expression levels in SK-OV-3 and OVCAR-3 cells with NR3C2 knockdown were measured using RT-qPCR. (G) The specific binding of NR3C2 to the ING1 promoter (chr13: 110713411–110713760) in OV cells was measured using ChIP-qPCR. (H) Luciferase activity in the OV cells stably infected with NR3C2 knockdown and co-transfected with the constructed plasmid containing the ING1 promoter (chr13: 110713411–110713760) and the internal reference plasmid. n = 3 biologically independent experiments. Unpaired t test for two comparisons. Data are shown as mean ± SEM. ING1 expression was similarly downregulated in OV cells with NR3C2 knockdown (Figure 6F). ChIP-qPCR results also showed that anti-NR3C2 bound to the ING1 promoter (chr13: 110713411–110713760) in OV cells (Figure 6G). In addition, knockdown of NR3C2 downregulated luciferase activity reporters with the ING1 promoter in OV cells (Figure 6H). Combined overexpression of ING1 represses the EMT in vitro We infected OV cells with a lentiviral vector expressing hING1 and validated the successful upregulation of ING1 expression (Figure 7A). Based on Tan IIA and NR3C2 knockdown, reactivation of ING1 decreased the number of clones formed by OV cells (Figure 7B), wound healing rate (Figure 7C), migration (Figure 7D), and invasion (Figure 7D) ability. The upregulation of E-cadherin and the downregulation of N-cadherin were observed in OV cells overexpressing ING1 (Figure 7F).Figure 7Overexpression of ING1 suppresses the malignant phenotypes of OV cells(A) After stable infection with OE-ING1 or OE-NC, ING1 protein levels in SK-OV-3 and OVCAR-3 cells were analyzed by western blot analysis.(B) The effects of KD-NR3C2 #1 and OE-ING1 on the colony-forming ability of OV cells were examined using a clonogenic assay.(C) The effects of KD-NR3C2 #1 and OE-ING1 on the migration of OV cells were examined using the wound healing assay (scale bars, 100 μm).(D and E) The effects of KD-NR3C2 #1 and OE-ING1 on the migration and invasion of OV cells were examined using the Transwell assay (scale bars, 50 μm).(F) The MFI of E-cadherin and N-cadherin in OV cells was examined using immunofluorescence. n = 3 biologically independent experiments (scale bars, 20 μm). Unpaired t test for two comparisons. Data are shown as mean ± SEM. Overexpression of ING1 suppresses the malignant phenotypes of OV cells (A) After stable infection with OE-ING1 or OE-NC, ING1 protein levels in SK-OV-3 and OVCAR-3 cells were analyzed by western blot analysis. (B) The effects of KD-NR3C2 #1 and OE-ING1 on the colony-forming ability of OV cells were examined using a clonogenic assay. (C) The effects of KD-NR3C2 #1 and OE-ING1 on the migration of OV cells were examined using the wound healing assay (scale bars, 100 μm). (D and E) The effects of KD-NR3C2 #1 and OE-ING1 on the migration and invasion of OV cells were examined using the Transwell assay (scale bars, 50 μm). (F) The MFI of E-cadherin and N-cadherin in OV cells was examined using immunofluorescence. n = 3 biologically independent experiments (scale bars, 20 μm). Unpaired t test for two comparisons. Data are shown as mean ± SEM. To further validate the functional role of ING1 in inhibiting EMT in OV, we established stable OV cell lines with ING1 knockdown. The efficiency of genetic intervention was verified by RT-qPCR (Figure S3A) and western blot (Figure S3B). Cells from the KD-ING1 #1 group, which exhibited the highest knockdown efficiency, were selected for combined treatment with 150 μM Tan IIA. In the absence of Tan IIA treatment, ING1 knockdown (PBS + KD-ING1 #1) significantly enhanced the EMT compared to the control group (PBS + KD-SCR). This was manifested by markedly increased migration (Figure S3C) and invasion (Figure S3D) of OV cells, accompanied by a significant downregulation of E-cadherin expression and a marked upregulation of N-cadherin expression (Figure S3E). Furthermore, ING1 knockdown attenuated the inhibitory effect of Tan IIA on the migration, invasion, and EMT of OV cells. In summary, downregulating ING1 independently enhanced EMT in OV cells, and ING1 played a crucial regulatory role in the anti-EMT effect mediated by Tan IIA. Overexpression of ING1 represses tumor growth and metastasis in mice Consistently, SK-OV-3 cells overexpressing ING1 or not in the presence of NR3C2 knockdown were injected into mice, followed by Tan IIA treatment. Forced activation of ING1 significantly inhibited tumor volume and weight (Figures 8A and 8B). IHC scores for EMT-related proteins (Twist1, Snail, and Slug) were also downregulated in xenografts formed by SK-OV-3 cells overexpressing ING1 (Figure 8C), along with enhanced apoptosis (Figure 8D). In the lung metastasis model, ING1 overexpression also significantly inhibited lung metastasis (Figure 8E). After confirming the stable overexpression of Ing1 in ID8 cells (Figure 8F), we established a peritoneal metastasis model using these cells. Forced activation of Ing1 reduced the number of metastatic nodules in the mesenteric and parietal peritoneal regions of mice, while also further decreasing the total weight of peritoneal metastatic tumors (Figure 8G).Figure 8Overexpression of ING1 suppresses tumor growth and metastasis in the presence of knockdown of NR3C2(A) Starting from the beginning of Tan IIA treatment, the volume of subcutaneous xenografts in nude mice was measured every 5 days (n = 5 mice/group).(B) Mice were euthanized after 25 days, and their tumors were weighed (n = 5 mice/group).(C) IHC scores of Twist1, Snail, and Slug in xenograft tumor tissues (n = 5 mice/group) (scale bars, 20 μm).(D) The apoptosis rate of tumors was examined using TUNEL (n = 5 mice/group) (scale bars, 20 μm).(E) Lung tissue was collected from nude mice with intravenous injection of SK-OV-3 cells and analyzed by H&E staining to determine the number of metastatic nodules (n = 5 mice/group) (scale bars, 50 μm).(F) Ing1 protein levels in ID8 cells following infection with Ing1-overexpressing lentivirus were examined using a western blot (n = 3 biologically independent experiments).(G) Peritoneal tumor burden in C57BL/6J mice with intraperitoneal injection of ID8 cells, with quantitative assessment of metastatic tumor nodule count and tumor weight (n = 5 mice/group). Unpaired t test for two comparisons and two-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. Overexpression of ING1 suppresses tumor growth and metastasis in the presence of knockdown of NR3C2 (A) Starting from the beginning of Tan IIA treatment, the volume of subcutaneous xenografts in nude mice was measured every 5 days (n = 5 mice/group). (B) Mice were euthanized after 25 days, and their tumors were weighed (n = 5 mice/group). (C) IHC scores of Twist1, Snail, and Slug in xenograft tumor tissues (n = 5 mice/group) (scale bars, 20 μm). (D) The apoptosis rate of tumors was examined using TUNEL (n = 5 mice/group) (scale bars, 20 μm). (E) Lung tissue was collected from nude mice with intravenous injection of SK-OV-3 cells and analyzed by H&E staining to determine the number of metastatic nodules (n = 5 mice/group) (scale bars, 50 μm). (F) Ing1 protein levels in ID8 cells following infection with Ing1-overexpressing lentivirus were examined using a western blot (n = 3 biologically independent experiments). (G) Peritoneal tumor burden in C57BL/6J mice with intraperitoneal injection of ID8 cells, with quantitative assessment of metastatic tumor nodule count and tumor weight (n = 5 mice/group). Unpaired t test for two comparisons and two-way ANOVA for comparisons of three or more groups. Data are shown as mean ± SEM. Low expression of NR3C2 or ING1 is related to peritoneal metastasis and poor outcomes in OV patients We analyzed the expression of NR3C2 and ING1 proteins in cells isolated from primary tumor tissues and malignant ascites in 34 OV patients. Compared with tumor cells isolated from the patients’ primary tumor tissues, the expression of NR3C2 and ING1 in the patients’ ascites tumor cells was downregulated (Figures 9A and 9B). Correlation analysis using protein expression data from tumor cells in patients’ ascites revealed a positive correlation between NR3C2 and ING1 expression (Figure 9C). Subsequently, we divided the patient cohorts into high- and low-expression groups based on the mean expression of NR3C2 (mean = 0.367) and ING1 (mean = 0.35) in ascites tumor cells. Further analysis of survival rates also revealed that patients with low expression of NR3C2 or ING1 in ascites tumor cells had significantly poorer prognosis (Figure 9D). Moreover, patients with low expression of NR3C2 or ING1 in ascites tumor cells were associated with more advanced FIGO staging, larger ascites volume, and peritoneal metastasis (Table 1).Figure 9Low expression of NR3C2 or ING1 is associated with peritoneal metastasis and poor prognosis in OV(A) NR3C2 expression in primary tumor cells and matched ascites from 34 OV patients was examined using western blot analysis (only five representative images are shown). Data are shown as mean ± SEM.(B) ING1 expression in primary tumor cells and matched ascites from 34 OV patients was examined using western blot analysis (only five representative images are shown). Data are shown as mean ± SEM.(C) The Pearson’s correlation analysis between NR3C2 and ING1 expression in ascites tumor cells (n = 34).(D) The survival of OV patients with high/low expression of NR3C2 or ING1 was analyzed using the log rank (Mantel-Cox) test. Paired t test for two comparisons.Table 1Correlation between high/low expression of NR3C2 or ING1 and clinical characteristics of OV patientsCharacteristicPatient numbersNR3C2 expressionING1 expressionLow (n = 19)High (n = 15)p valueLow (n = 21)High (n = 13)p valueAge ≥55141040.170695>0.9999 <5520911128FIGO stage I/II10280.0098∗370.0223∗ III/IV24177186Histological subtype Serous3118130.57141912>0.9999 Other31221Ascites ≥500 mL221750.0011∗1930.0001∗ <500 mL12210210LNM Positive151050.31411140.2955 Negative19910109Peritoneal metastasis Positive231670.0301∗1850.0078∗ Negative113838OV, ovarian cancer; NR3C2, nuclear receptor subfamily 3 group C member 2; ING1, inhibitor of growth protein 1; FIGO, International Federation of Gynecology and Obstetrics; LNM, lymph node metastasis. Fisher’s exact test was used to analyze the clinical-pathological correlations; ∗p < 0.05 was considered statistically significant. Low expression of NR3C2 or ING1 is associated with peritoneal metastasis and poor prognosis in OV (A) NR3C2 expression in primary tumor cells and matched ascites from 34 OV patients was examined using western blot analysis (only five representative images are shown). Data are shown as mean ± SEM. (B) ING1 expression in primary tumor cells and matched ascites from 34 OV patients was examined using western blot analysis (only five representative images are shown). Data are shown as mean ± SEM. (C) The Pearson’s correlation analysis between NR3C2 and ING1 expression in ascites tumor cells (n = 34). (D) The survival of OV patients with high/low expression of NR3C2 or ING1 was analyzed using the log rank (Mantel-Cox) test. Paired t test for two comparisons. Table 1Correlation between high/low expression of NR3C2 or ING1 and clinical characteristics of OV patientsCharacteristicPatient numbersNR3C2 expressionING1 expressionLow (n = 19)High (n = 15)p valueLow (n = 21)High (n = 13)p valueAge ≥55141040.170695>0.9999 <5520911128FIGO stage I/II10280.0098∗370.0223∗ III/IV24177186Histological subtype Serous3118130.57141912>0.9999 Other31221Ascites ≥500 mL221750.0011∗1930.0001∗ <500 mL12210210LNM Positive151050.31411140.2955 Negative19910109Peritoneal metastasis Positive231670.0301∗1850.0078∗ Negative113838OV, ovarian cancer; NR3C2, nuclear receptor subfamily 3 group C member 2; ING1, inhibitor of growth protein 1; FIGO, International Federation of Gynecology and Obstetrics; LNM, lymph node metastasis. Fisher’s exact test was used to analyze the clinical-pathological correlations; ∗p < 0.05 was considered statistically significant. Correlation between high/low expression of NR3C2 or ING1 and clinical characteristics of OV patients OV, ovarian cancer; NR3C2, nuclear receptor subfamily 3 group C member 2; ING1, inhibitor of growth protein 1; FIGO, International Federation of Gynecology and Obstetrics; LNM, lymph node metastasis. Fisher’s exact test was used to analyze the clinical-pathological correlations; ∗p < 0.05 was considered statistically significant. Discussion EMT is characterized by a series of phenotypical changes within the cell, resulting in enhanced drug resistance, increased cell plasticity, and heightened metastatic potential.15 Currently, we have found that Tan IIA attenuates EMT and metastasis in OV. Mechanistically, the NR3C2/ING1 axis plays an essential role in Tan IIA-mediated suppressive effects on metastasis in OV. Tanshinones represent the primary active constituents of S. miltiorrhiza, exhibiting notable pharmacological activities, including antioxidant, anti-inflammatory, cardiovascular, and antitumor properties.16 In breast cancer, Tan IIA has been revealed to elicit anti-tumor effects by repressing stemness or promoting ferroptosis.17,18 A seminal study by Einbond et al. revealed a pivotal finding: Tan IIA exhibited the capacity to both augment the expression of p53 and impede the expression of HPV16 E1, E2, E4, E6, and E7 viral transcripts in W12 integrant cells, indicating its potential role in cervical cancer.19 In the context of bladder cancer cells, Tan IIA has been observed to induce the expression of the epithelial marker E-cadherin and to concomitantly reduce the levels of the mesenchymal markers N-cadherin and Vimentin in a time- and dose-dependent manner.20 Consistent observations were also obtained in our in vitro studies using OV cells. In addition, Tan IIA derivative 4 demonstrated the capacity to impede the proliferation, metastasis, and angiogenesis of triple-negative breast cancer cells, which may be attributed to the stabilization of multiple G4 DNAs.21 Snail and Slug have been shown to repress the E-cadherin gene CDH1, a process that has been linked to the loss of cell-cell adhesion and the increased motility of cancer cells.22 Twist, another pivotal regulator of EMT, has been observed to promote cytoskeletal reorganization and increase the expression of mesenchymal markers, such as N-cadherin.23 In the present study, we found that the tumor burden in mice treated with Tan IIA was lighter than that in saline-treated mice, with downregulation of these three proteins. Our integrated prediction results showed that NR3C2 is a potent target of Tan IIA in regulating EMT during OV. Furthermore, our rescue experiments showed that additional knockdown of NR3C2 led to a compromise in the anti-tumor effects of Tan IIA. Likewise, NR3C2 silencing abated the effects of circSTK39 on EMT of bladder cancer.24 Overexpression of NR3C2 has been demonstrated to exert a dual effect on colorectal cancer cells: it suppresses proliferation, and it induces G2/M phase arrest.25 Among the targets of NR3C2 related to OV, we focused on ING1 since it was the only downregulated gene in the included GEO datasets. The downregulation of ING1 has also been identified by Calanca et al. in juvenile nasopharyngeal angiofibroma.26 Its transcription has been revealed to be regulated by a transcription factor, RUNX3, in gallbladder cancer.27 About its functional role, ING1, in a variant-dependent manner, exhibited the capacity to inhibit hepatoma cell proliferation and colony formation and postpone tumor formation in nude mice.28 Rusu et al. found that ING1 knockdown disrupted E-cadherin distribution at cell-cell junctions, and loss of ING1 affected the expression of several junctional and cytoskeletal modulators, confirming ING1 as an invasion suppressor and a major modulator of cell-cell junction integrity.29 The transfer of exosomal microRNA-1825 to vascular endothelial cells has been shown to result in the suppression of ING1, which, in turn, promotes angiogenesis and liver metastasis in colorectal cancer.30 Here, the overexpression of ING1 was found to repress OV cell EMT and lung metastasis in mice in the presence of knockdown of NR3C2 and Tan IIA. Moreover, knockdown of ING1 compromised the anti-EMT properties of Tan IIA. Lastly, we reported the possible prognostic role of NR3C2 and ING1 in OV. NR3C2 was downregulated in clear-cell renal cell carcinoma and breast cancer tissues and was correlated with several clinicopathological parameters, including the histological or clinical stages.31,32 Concurrently, a correlation was identified between reduced ING1 expression and unfavorable prognoses in patients diagnosed with glioblastoma, as indicated by outcomes related to both overall survival and disease-free survival.33 Analysis of our cohort showed that OV patients with lower expression of NR3C2 or ING1 tended to have poorer outcomes and higher FIGO stage, greater ascites volume, and peritoneal metastasis. In this study, we demonstrate that Tan IIA activates the NR3C2/ING1 axis to suppress EMT and metastasis in OV, thereby establishing an unrecognized mechanistic pathway with potential therapeutic insights. Limitations of the study In addition to the NR3C2/ING1 axis examined in this study, Tan IIA has been reported to modulate other signaling pathways relevant to OV, including PI3K/AKT, JNK, and microRNAs.34,35 These findings suggest that Tan IIA may exert its anticancer effects through multiple mechanisms, and future studies will be required to delineate the relative contributions of these pathways. Furthermore, translational barriers persist due to the incomplete characterization of Tan IIA pharmacokinetics, bioavailability, and safety profile in humans, and the potential impact of variability across OV subtypes on therapeutic efficacy. Subsequent endeavors involving patient-derived models and clinical validation will be crucial to address these limitations and to enhance the translational potential of our findings. Resource availability Lead contact Further information and requests concerning resources and reagents should be directed to and will be answered by the lead contact, Jing Zhang (zj_gynecology@163.com). Materials availability This study did not generate any unique new reagents. All reagents used in this study are commercially available. Data and code availability •All data reported in this paper will be shared by the lead contact upon request.•This paper does not report original code.•Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. •All data reported in this paper will be shared by the lead contact upon request. All data reported in this paper will be shared by the lead contact upon request. •This paper does not report original code. This paper does not report original code. •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request. Acknowledgments We thank the Ningbo Municipal Health Commission Project (2023Y24), the Zhejiang Provincial Traditional Chinese Medicine Project (2024ZL956), the Ningbo Natural Science Foundation Project (2024J381), and the Ningbo Top Medical and Health Research Program (2024021020) for funding support. Author contributions X.W., conceptualization, methodology, investigation, formal analysis, and writing – original draft; D.Z., investigation, data curation, methodology, and software; T.W. and X.L., resources, software, and validation; J.Z., supervision, project administration, and writing – review and editing. Declaration of interests The authors declare no competing interests. STAR★Methods Key resources table REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesN-cadherinProteinTech GroupCat# 22018-1-AP; RRID: AB_2813891E-cadherinProteinTech GroupCat# 20874-1-AP; RRID: AB_10697811goat anti-rabbit IgG secondary antibody, Alexa Fluor 647InvitrogenCat# A-21244; RRID: AB_2535812Twist1InvitrogenCat# PA5-49688; RRID: AB_2635141SnailInvitrogenCat# PA5-23482; RRID: AB_2540982SlugInvitrogenCat# PA5-85522; RRID: AB_2792662goat anti-rabbit IgG secondary antibodyInvitrogenCat# G-21234; RRID: AB_1500696NR3C2ProteinTech GroupCat# 21854-1-AP; RRID: AB_10858785ING1BiossCat# BS-1291R; RRID: AB_10857108rabbit IgGMedChemExpressCat# HY-P80879; RRID: AB_3102988Bacterial and virus strainsLentivirus-expressing NR3C2 shRNAThis paperN/ALentivirus-expressing ING1 shRNAThis paperN/ALentivirus-expressing Nr3c2 shRNAThis paperN/ALentivirus-expressing Ing1This paperN/ABiological samplesPrimary tumor tissues from OV patientsThis paperN/AMalignant ascites samples from OV patientsThis paperN/AChemicals, peptides, and recombinant proteinsRPMI-1640 mediumZhong Qiao Xin ZhouCat# ZQ-200collagenase IIMeilunbioCat# MB26651× GlutaMAX supplementGibcoCat# 350500611% HEPESGibcoCat# 15630080HeparinSelleckCat# S1346McCoy’s 5A mediumCellverseCat# iCell-0011DMEMGibcoCat# 11965092Tan IIAMedChemExpressCat# HY-N0135SB431542MedChemExpressCat# HY-10431CCK-8 solutionSelleckCat# B34304mitomycin CSelleckCat# S8146crystal violet staining solutionSangonCat# E607309MatrigelCorningCat# 356237StaurosporineMedChemExpressCat# HY-15141DAB chromogen concentrateBioLegendCat# 926506antifade mounting mediumMedChemExpressCat# HY-K1042RIPA lysis bufferProteinTech GroupCat# PR20035PVDF membranesThermo Fisher ScientificCat# 88585SuperSignal West Pico PLUS chemiluminescent substrateThermo Fisher ScientificCat# 34580TRIzol reagentCYTOCHCat# NP0001Critical commercial assaysTumor Cell Isolation Kit, humanMiltenyiCat# 130-108-339Annexin V-FITC/PI Apoptosis Detection KitMedChemExpressCat# HY-K1073LDH Cytotoxicity Assay KitYeasenCat# 40209ES76One-step TUNEL In Situ Apoptosis KitElabscienceCat# E-CK-A320HE staining kitMeilunbioCat# MB9898Pierce BCA protein assay kitThermo Fisher ScientificCat# A65453RevertAid RT Reverse Transcriptase KitThermo Fisher ScientificCat# K1691Magna ChIP A/G ChIP KitMerckCat# 17-10085Renilla-Firefly luciferase dual assay kitMedChemExpressCat# HY-K1013Experimental models: Cell linesSK-OV-3MingzhoubioCat# MZ-0169OVCAR-3Zhong Qiao Xin ZhouCat# ZQ0365ID8Merck MilliporeCat# SCC145Experimental models: Organisms/strainsBALB/c nude miceVital RiverN/AC57BL/6J miceVital RiverN/AOligonucleotidesNR3C2-F (Human)This paper5′-AAATCACACGGCGACCTGTCGT-3′NR3C2-R (Human)This paper5′- ATGGCATCCTGAAGCCTCATCC-3′ING1-F (Human)This paper5′- GGCTCGGAGACAGTTTCAGGC-3′ING1-R (Human)This paper5′- GACTGAAGCGCTCGTAGCACTC-3′GAPDH-F (Human)This paper5′- GTCTCCTCTGACTTCAACAGCG-3′GAPDH-R (Human)This paper5′- ACCACCCTGTTGCTGTAGCCAA-3′Nr3c2-F (Mouse)This paper5′-AACCACACGGTGACCTGTCATC-3′Nr3c2-R (Mouse)This paper5′- AGTGACACCCAGAAGCCTCATC-3′Gapdh-F (Mouse)This paper5′- CATCACTGCCACCCAGAAGACTG-3′Gapdh-R (Mouse)This paper5′- ATGCCAGTGAGCTTCCCGTTCAG-3′Software and algorithmsImageJNIHhttps://imagej.nih.gov/ij/Image Pro PlusMedia CyberneticsN/APrism softwareGraphPadhttps://www.graphpad.com/OtherPolyplus jetPRIME transfection reagentShanghai Nonin Biological TechnologyCat# 101000046 REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesN-cadherinProteinTech GroupCat# 22018-1-AP; RRID: AB_2813891E-cadherinProteinTech GroupCat# 20874-1-AP; RRID: AB_10697811goat anti-rabbit IgG secondary antibody, Alexa Fluor 647InvitrogenCat# A-21244; RRID: AB_2535812Twist1InvitrogenCat# PA5-49688; RRID: AB_2635141SnailInvitrogenCat# PA5-23482; RRID: AB_2540982SlugInvitrogenCat# PA5-85522; RRID: AB_2792662goat anti-rabbit IgG secondary antibodyInvitrogenCat# G-21234; RRID: AB_1500696NR3C2ProteinTech GroupCat# 21854-1-AP; RRID: AB_10858785ING1BiossCat# BS-1291R; RRID: AB_10857108rabbit IgGMedChemExpressCat# HY-P80879; RRID: AB_3102988Bacterial and virus strainsLentivirus-expressing NR3C2 shRNAThis paperN/ALentivirus-expressing ING1 shRNAThis paperN/ALentivirus-expressing Nr3c2 shRNAThis paperN/ALentivirus-expressing Ing1This paperN/ABiological samplesPrimary tumor tissues from OV patientsThis paperN/AMalignant ascites samples from OV patientsThis paperN/AChemicals, peptides, and recombinant proteinsRPMI-1640 mediumZhong Qiao Xin ZhouCat# ZQ-200collagenase IIMeilunbioCat# MB26651× GlutaMAX supplementGibcoCat# 350500611% HEPESGibcoCat# 15630080HeparinSelleckCat# S1346McCoy’s 5A mediumCellverseCat# iCell-0011DMEMGibcoCat# 11965092Tan IIAMedChemExpressCat# HY-N0135SB431542MedChemExpressCat# HY-10431CCK-8 solutionSelleckCat# B34304mitomycin CSelleckCat# S8146crystal violet staining solutionSangonCat# E607309MatrigelCorningCat# 356237StaurosporineMedChemExpressCat# HY-15141DAB chromogen concentrateBioLegendCat# 926506antifade mounting mediumMedChemExpressCat# HY-K1042RIPA lysis bufferProteinTech GroupCat# PR20035PVDF membranesThermo Fisher ScientificCat# 88585SuperSignal West Pico PLUS chemiluminescent substrateThermo Fisher ScientificCat# 34580TRIzol reagentCYTOCHCat# NP0001Critical commercial assaysTumor Cell Isolation Kit, humanMiltenyiCat# 130-108-339Annexin V-FITC/PI Apoptosis Detection KitMedChemExpressCat# HY-K1073LDH Cytotoxicity Assay KitYeasenCat# 40209ES76One-step TUNEL In Situ Apoptosis KitElabscienceCat# E-CK-A320HE staining kitMeilunbioCat# MB9898Pierce BCA protein assay kitThermo Fisher ScientificCat# A65453RevertAid RT Reverse Transcriptase KitThermo Fisher ScientificCat# K1691Magna ChIP A/G ChIP KitMerckCat# 17-10085Renilla-Firefly luciferase dual assay kitMedChemExpressCat# HY-K1013Experimental models: Cell linesSK-OV-3MingzhoubioCat# MZ-0169OVCAR-3Zhong Qiao Xin ZhouCat# ZQ0365ID8Merck MilliporeCat# SCC145Experimental models: Organisms/strainsBALB/c nude miceVital RiverN/AC57BL/6J miceVital RiverN/AOligonucleotidesNR3C2-F (Human)This paper5′-AAATCACACGGCGACCTGTCGT-3′NR3C2-R (Human)This paper5′- ATGGCATCCTGAAGCCTCATCC-3′ING1-F (Human)This paper5′- GGCTCGGAGACAGTTTCAGGC-3′ING1-R (Human)This paper5′- GACTGAAGCGCTCGTAGCACTC-3′GAPDH-F (Human)This paper5′- GTCTCCTCTGACTTCAACAGCG-3′GAPDH-R (Human)This paper5′- ACCACCCTGTTGCTGTAGCCAA-3′Nr3c2-F (Mouse)This paper5′-AACCACACGGTGACCTGTCATC-3′Nr3c2-R (Mouse)This paper5′- AGTGACACCCAGAAGCCTCATC-3′Gapdh-F (Mouse)This paper5′- CATCACTGCCACCCAGAAGACTG-3′Gapdh-R (Mouse)This paper5′- ATGCCAGTGAGCTTCCCGTTCAG-3′Software and algorithmsImageJNIHhttps://imagej.nih.gov/ij/Image Pro PlusMedia CyberneticsN/APrism softwareGraphPadhttps://www.graphpad.com/OtherPolyplus jetPRIME transfection reagentShanghai Nonin Biological TechnologyCat# 101000046 Experimental model and study participant details Cell lines Ovarian serous cystadenocarcinoma SK-OV-3 (MZ-0169, Mingzhoubio, Ningbo, Zhejiang, China) cells were cultured using McCoy’s 5A medium (iCell-0011, Cellverse Bioscience Technology Co., Ltd., Shanghai, China) containing 10% FBS and 1% penicillin/streptomycin. High-grade ovarian serous adenocarcinoma cells OVCAR-3 (ZQ0365, Zhong Qiao Xin Zhou, Shanghai, China) were transferred to RPMI-1640 medium (ZQ-200, Zhong Qiao Xin Zhou) containing 20% FBS and 1% penicillin/streptomycin for cultivation. The mouse ovarian surface epithelial cell line ID8 (SCC145, Merck Millipore, Darmstadt, Germany) was cultured in DMEM (11965092, Gibco) supplemented with 10% FBS and 1% penicillin/streptomycin. Cells were cultured with 5% CO2 at 37°C. The SK-OV-3 and OVCAR-3 cell lines were identified using short tandem repeat genotyping upon receipt. All cells were tested for mycoplasma contamination using PCR before the experiment, and the results were negative. Clinical specimen All clinical samples were approved by the Institutional Review Board of Affiliated Women and Children’s Hospital of Ningbo University (approval number: EC2022-008). All subjects provided consent for research participation. Primary tumor tissues and matched malignant ascites samples were collected from 34 patients with OV during surgery. All patients included in the study were women, ranging in age from 31 to 81 years. Patients included in this study must meet all the following criteria: (1) Pathologically confirmed primary OV; (2) No prior anticancer treatment (including radiotherapy, chemotherapy, or targeted therapy) before sample collection and initial surgery; (3) Availability of matched primary tumor tissue and malignant ascites samples; (4) Complete clinical and follow-up data. Patients were excluded if any of the following conditions applied: (1) History of other malignant tumors; (2) Prior receipt of any form of antitumor therapy before surgery; (3) Presence of severe comorbidities that could interfere with treatment or follow-up assessments; (4) Incomplete clinical information or loss to follow-up. All patients receive regular follow-up visits for three years (every six months) from the date of diagnosis. All 34 paired samples were used for Western blot analysis of NR3C2 and ING1 protein expression. Based on the quantitative Western blot results for NR3C2 and ING1 in tumor cells from malignant ascites, patients were divided into high-expression and low-expression groups using their respective mean expression values as cutoffs. These groups were then used for subsequent survival analysis and to assess correlations with clinical and pathological parameters such as FIGO stage, histological subtype, ascites volume, lymph node metastasis (LNM), and peritoneal metastasis. Since all participants in this study were female, no between-group comparisons were made regarding biological sex differences; therefore, it was not possible to assess the impact of sex on the study results, which constitutes a limitation of this study. This study did not systematically collect or analyze information on the participants’ race, ethnicity, ancestry, or gender. The primary tumor tissues of the patients were minced and then homogenized using a syringe barrel. The homogenate was detached in RPMI-1640 medium containing 2.5 mg/mL collagenase II (MB2665, Meilunbio, Dalian, Liaoning, China), 1× GlutaMAX supplement (35050061, Gibco, Carlsbad, CA, USA), 1% HEPES (15630080, Gibco), and 1% penicillin/streptomycin at 37°C for 25 min, filtered through a 70 μm filter, and centrifuged at 300 ×g to precipitate the cells. Malignant ascites from patients was added with heparin (S1346, Selleck, Shanghai, China). The samples were then centrifuged at 500 ×g for 10 min at 4°C to collect the cell pellet. Cancer cells were separated from cell pellets using the Tumor Cell Isolation Kit, human (130-108-339, Miltenyi Biotec, Bergisch Gladbach, Germany). In vivo assays Five-week-old female BALB/c nude mice (401) were procured from Vital River (Beijing, China). All mice were housed in an SPF-grade laboratory animal facility, where they had free access to food and water under constant temperature and humidity conditions and a 12-h light/dark cycle. The Ethics Committee of Affiliated Women and Children’s Hospital of Ningbo University approved animal experiments (approval number: No.14930). The use of female mice in this study was because OV is a disease specific to the female population. Therefore, sex-based comparisons were not applicable. For the xenograft model, mice were randomly assigned to groups (n = 5 per group) using a random number table before the start of the experiment. SK-OV-3 cells (1 × 106) treated with different conditions were suspended in 200 μL of physiological saline and injected into the right flank of the mouse. The tumor volume was assessed once every 5 days using a vernier caliper. The tumor volume was calculated as length × width2 × 0.5. When the subcutaneous tumor volume reached about 100 mm3, Tan IIA (low dose 10 mg/kg, or high dose 30 mg/kg) was administered via intraperitoneal injection every 2 days, while the control group received injections of saline solvent. Twenty-five days after the start of Tan IIA treatment, mice were euthanized with 150 mg/kg sodium pentobarbital, and tumor tissues were weighed. Part of the subcutaneous tumor was cryopreserved, while the other part was fixed, dehydrated, and preserved in paraffin. For the lung metastasis model, mice were randomly divided into groups, and 1 × 106 SK-OV-3 cells dissolved in 200 μL of saline were injected into nude mice via the tail vein. After cell inoculation, treatment was initiated via intraperitoneal injection of 10 or 30 mg/kg Tan IIA or saline (vehicle solvent control) every 2 days. Five weeks after injection, the mice were euthanized, and the lung tissue was removed and paraffin-embedded. To establish an experimental mouse model of OV metastasis, 1 × 107 ID8 cells with different genetic interventions were suspended in 100 μL PBS and intraperitoneally injected into 6-week-old female C57BL/6J mice (219, Vital River). Tan IIA was administered via intraperitoneal injection every 2 days at either a low dose of 10 mg/kg or a high dose of 30 mg/kg. The control group received injections of physiological saline as the vehicle solvent. Mice were euthanized on day 60 (experimental endpoint). Tumor burden was determined by analyzing the number of peritoneal tumor nodules and weighing the total mass of metastatic tumors.36 To assess the toxicity of Tan IIA, we established a separate group of mice that underwent weekly weight monitoring. Subsequently, mice were euthanized at the experimental endpoint, and heart, liver, lung, and kidney tissues were collected. Following paraffin-embedding and sectioning, these tissues underwent hematoxylin-eosin (HE) staining for systematic histopathological evaluation. Method details Cell culture, treatment, and infection SK-OV-3 and OVCAR-3 cells were cultured at 5 × 104 cells/well in a six-well plate. They were infected with the human NR3C2 shRNA, human ING1 shRNA lentiviral vectors, and the corresponding scrambled control lentiviral vector (the target shRNA sequences are listed in the table below), as well as the human ING1 gene expression lentiviral vector (vector components: hING1[NM_005537.5]-CDS sequence, see File S1) or control vector. The multiplicity of infection (MOI) was set to 10. For the mouse ID8 cells, infection was performed using mouse Nr3c2 shRNA lentiviral vectors, mouse Ing1 gene expression lentiviral vectors (vector construct: mIng1[NM_011919.5]-CDS sequence, see File S2), along with scrambled control lentiviral vectors and overexpression lentiviral control vectors. The MOI was set at 20. Puromycin or neomycin was used for screening to achieve stable transfection. The efficiency of knockdown was verified using RT-qPCR and Western blot analysis, and the efficiency of overexpression was verified using Western blot analysis.shRNA informationshRNA nameTarget sequencesRegionSiteKD-NR3C2 #1CGGATTCCCAGTGGGTATTAACDS1824–1844KD-NR3C2 #2ACACGCCCTTGAGATCATTTACDS836–856KD-NR3C2 #3GCAAAGGTACTTCCAGGATTTCDS2713–2733KD-ING1 #1TTGTTCCTCTAGTAGTATATT3′UTR2540–2560KD-ING1 #2GCCATGATGTATATCCATATA3′UTR2133–2153KD-ING1 #3GAGGGTGGACGAGTTGATTTGCDS574–594KD-Nr3c2CCAAGGTACTTCCAGGATTTACDS2374–2394shRNA, short hairpin RNA; KD, knockdown; CDS, coding sequences. shRNA informationshRNA nameTarget sequencesRegionSiteKD-NR3C2 #1CGGATTCCCAGTGGGTATTAACDS1824–1844KD-NR3C2 #2ACACGCCCTTGAGATCATTTACDS836–856KD-NR3C2 #3GCAAAGGTACTTCCAGGATTTCDS2713–2733KD-ING1 #1TTGTTCCTCTAGTAGTATATT3′UTR2540–2560KD-ING1 #2GCCATGATGTATATCCATATA3′UTR2133–2153KD-ING1 #3GAGGGTGGACGAGTTGATTTGCDS574–594KD-Nr3c2CCAAGGTACTTCCAGGATTTACDS2374–2394shRNA, short hairpin RNA; KD, knockdown; CDS, coding sequences. shRNA information shRNA, short hairpin RNA; KD, knockdown; CDS, coding sequences. Except for the CCK-8 proliferation assay, SK-OV-3 and OVCAR-3 cells were treated with 50, 100, or 150 μM Tan IIA (HY-N0135, MedChemExpress, Monmouth Junction, NJ, USA) or PBS for 24 h. EMT inhibitor SB431542 (HY-10431, MedChemExpress) was used as the positive control at 5 μM. In the joint experiment, only 150 μM Tan IIA or PBS was used for treatment. Cell proliferation and clonogenic assays SK-OV-3 and OVCAR-3 cells in the exponential phase were seeded at 1,000 cells/well in a 96-well plate, pre-cultured for 24 h, and incubated with 50, 100, 150, 200, and 250 μM Tan IIA or control PBS for 24 h.35 After the addition of 10 μL of CCK-8 solution (B34304, Selleck), incubation was conducted at 37°C for 2 h. Finally, the optical density at 450 nm was read. SK-OV-3 and OVCAR-3 cells were seeded at 1,000 cells per well in a 6-well plate and cultured for 14 days. The cells were fixed with 4% PFA in PBS for 30 min and stained with 0.1% w/v crystal violet at room temperature (RT). Photos of the colonies were taken using a microscope, and colonies with more than 50 cells were analyzed using ImageJ software. Wound healing SKOV3 and OVCAR-3 cells (2 × 106) were seeded into a six-well plate containing complete medium, cultured until a confluent monolayer was formed, and treated with 0.4 μg/mL mitomycin C (99.37%, S8146, Selleck) for 30 min. The cell monolayer was scratched with a sharp tip to create wounds, rinsed with PBS, and cultured in medium containing 0.7% FBS with the addition of 0.04 μg/mL mitomycin C. Finally, images were captured using a microscope at 0 h and 48 h after wound creation. All images were randomly renamed by researchers unaware of the experimental groups (blinded). Each experimental group comprised three independent biological replicates, with three fields of view analyzed per replicate. The wound closure distance was measured using Image Pro Plus software. Transwell migration and invasion assay SKOV3 and OVCAR-3 cells (5 × 104) were resuspended in FBS-free medium, treated with 0.4 μg/mL mitomycin C for 30 min, transferred to the upper chamber of a Transwell chamber (3428, Corning, Corning, NY, USA), and medium containing 10% FBS was added to the lower chamber. After 48 h, cells that had migrated into the lower chamber were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet staining solution (E607309, Sangon, Shanghai, China) for 10 min. For the invasion assay, Matrigel (356237, Corning) was pre-coated in Transwell chambers, and the rest of the method was the same as the migration assay. All images were randomly numbered blindly before analysis. Cell counting was performed using ImageJ software, with the migration/invasion cell count in each well averaged from five fields of view. Immunofluorescence SKOV3 and OVCAR-3 cells were fixed with PBS containing 4% paraformaldehyde for 15 min, permeated with 0.5% Triton X-100 for 15 min, and blocked with 5% goat serum for 1 h. Primary antibodies (N-cadherin: 22018-1-AP; E-cadherin: 20874-1-AP, ProteinTech Group, Chicago, IL, USA) were applied and incubated at 4°C overnight. Goat anti-Rabbit IgG secondary antibody, Alexa Fluor 647 (A-21244, Invitrogen Inc., Carlsbad, CA, USA) was applied and incubated at RT for 1 h. DAPI was used for nuclear staining. Images were captured using a fluorescence microscope and analyzed using ImageJ software to determine the mean fluorescence intensity (MFI). Flow cytometry Apoptosis in SK-OV-3 and OVCAR-3 cells was assessed using the Annexin V-FITC/PI Apoptosis Detection Kit (HY-K1073, MedChemExpress). OV cells (5 × 105 cells) were treated with Tan IIA at concentrations of 50, 100, and 150 μM, with PBS as the control; 500 ng/mL (approximately 1.07 μM) Staurosporine (STS) (HY-15141, MedChemExpress) served as the positive control. After 24 h, the cells were resuspended in 195 μL of binding buffer and incubated with 5 μL of Annexin V-FITC and 10 μL of PI stain at RT in the dark for 15 min. Finally, the stained cell populations were analyzed using flow cytometry. Lactate dehydrogenase (LDH) release assay Membrane integrity of OV cells was analyzed using the LDH cytotoxicity assay kit (40209ES76, Yeasen, Shanghai, China). In short, SK-OV-3 and OVCAR-3 cells were seeded at 2 × 104 cells per well in a 96-well plate. After treatment with Tan IIA at concentrations of 50, 100, and 150 μM, PBS (negative control) or 1% Triton X-100 (positive control) for 24 h, 120 μL of supernatant from each well was transferred to a fresh 96-well plate and incubated with 60 μL of LDH assay working solution at RT in the dark for 30 min. The absorbance at 490 nm was read. Cellular reactive oxygen species (ROS) assay Fluorescent probe DCFH-DA (HY-D0940, MedChemExpress) was used to assess ROS levels in SK-OV-3 and OVCAR-3 cells. OV cells were treated with 50, 100, or 150 μM Tan IIA, with PBS as the negative control, or with 10 μM H2O2 as the positive control for 24 h. The cells were washed with PBS and collected after trypsin digestion. Ten micromoles per liter of DCFH-DA were added, and the mixture was incubated at RT in the dark for 20 min. Images were acquired using a fluorescence microscope, and ImageJ was used to calculate the relative fluorescence intensity of DCF to measure ROS levels. Immunohistochemistry (IHC) Paraffin-embedded xenograft tumors were sectioned at 5 μm, dewaxed in xylene, and hydrated in a series of ethanol solutions of decreasing concentration. The sections were incubated in 3% (v/v) hydrogen peroxide for 10 min to inactivate endogenous peroxidase. Antigen retrieval was performed on the tumor sections under high pressure, and the sections were sealed with 2.5% goat serum for 20 min and incubated at RT with antibodies to Twist1 (1:100, PA5-49688, Invitrogen), Snail (1:200, PA5-23482, Invitrogen), and Slug (1:500, PA5-85522, Invitrogen) for 2 h. After washing with PBS containing 0.1% Tween 20, incubation continued with goat anti-rabbit IgG secondary antibody (1:1000, G-21234, Invitrogen) for 30 min and with DAB chromogen concentrate (926506, BioLegend, San Diego, CA, USA) for 4 min. After staining with hematoxylin, the sections were dehydrated, cleared, and fixed with neutral resin. The stained images were viewed using a microscope. The IHC staining results were evaluated by two professional pathologists. The two experts scored the percentage of positive cells and the staining intensity separately. Positive cell percentage scoring rules: 0 points (<5% positive cells), 1 point (6–25% positive cells), 2 points (26–50% positive cells), 3 points (51–75% positive cells), or 4 points (>75% positive cells). Staining intensity scoring rules: 3 points (strong positive staining); 2 points (moderate positive staining); 1 point (weak positive staining); 0 points (negative staining). The final score was equal to the percentage score multiplied by the intensity score. TUNEL assay The One-step TUNEL In Situ Apoptosis Kit (Green, FITC) (E-CK-A320, Elabscience, Wuhan, Hubei, China). Sections of xenograft tumors from nude mice were dewaxed and hydrated. The sections were reacted with 100 μL of 1× proteinase K working solution for 20 min, equilibrated with 100 μL of TdT Equilibration Buffer for 15 min, and incubated with 50 μL of labeled working solution in the dark for 60 min (all at 37°C). After the treatment with DAPI working solution at RT in the dark for 5 min, the slides were covered with antifade mounting medium (HY-K1042, MedChemExpress). The staining results were obtained using a fluorescence microscope, and ImageJ was used to calculate the TUNEL-positive staining rate in a random field of view of the section. HE staining According to the manufacturer’s instructions, lung tissue was stained using an HE staining kit (MB9898, Meilunbio). Paraffin-embedded mouse lung tissues were sectioned at 5 μm, dewaxed, and hydrated. The sections were incubated in hematoxylin for 2 min, with differentiation solution for 30 s, washed, and transferred to eosin for 2 min. The stained tissue sections were dehydrated, permeabilized, and fixed with neutral resin. Finally, photographs were taken using a microscope, and the number of metastatic nodules was analyzed using ImageJ software. Western blot assay SKOV3, OVCAR-3, and ID8 cells, as well as cells isolated from clinical samples, were lysed with RIPA lysis buffer (PR20035, ProteinTech Group, Chicago, IL, USA). After centrifugation to collect the supernatant, the protein content was determined using the Pierce BCA protein assay kit (A65453, Thermo Fisher Scientific Inc., Waltham, MA, USA). Equal amounts of protein (25 μg) were separated by 10% SDS-PAGE, transferred to PVDF membranes (88585, Thermo Fisher Scientific), and sealed for 1 h at RT using 5% non-fat milk. Blots were incubated with primary antibodies overnight at 4°C: anti-NR3C2 (1:4000, 21854-1-AP, ProteinTech Group) and anti-ING1 (1:1000, BS-1291R, Bioss Biotech, Beijing, China). Membranes were then incubated in goat anti-rabbit horseradish peroxidase-linked secondary antibody at a 1:10000 dilution (G-21234, Invitrogen) for 1 h at RT. Antibody binding was detected by adding SuperSignal West Pico PLUS chemiluminescent substrate (34580, Thermo Fisher Scientific). Molecular docking The DeepMice platform (https://www.deepmice.com/) and its AutoDock Vina docking engine were used for molecular docking. The three-dimensional structure of the target protein NR3C2 was obtained in PDB format from Uniprot (https://www.uniprot.org/), and the SMILES structure of the ligand (Tan IIA) was imported into the docking model. Both protein and ligand structures were processed according to DeepMice’s default preprocessing workflow. Subsequently, docking simulations were performed to predict the binding affinity and binding mode between Tan IIA and NR3C2. The Vina scoring function was used to evaluate the binding energy of each docked conformation. The conformation with the lowest predicted binding free energy was selected as the best binding conformation (best_vina_score) for subsequent analysis. Drug affinity responsive target stability (DARTS) assay The OV cell (SK-OV-3 and OVCAR-3) lysate was added with 1/10 volume of Tris-NaCl-CaCl2 buffer and incubated with PBS or Tan IIA at low (50 μM), medium (100 μM), and high (150 μM) concentrations at RT with gentle shaking for 1 h. The lysate was added with or without the protease and incubated at RT for 30 min. The reaction was terminated by adding loading buffer. Finally, Western blot analysis was performed to detect NR3C2 protein expression. Cellular thermal shift assay (CETSA) Lysates from SK-OV-3 and OVCAR-3 cells were incubated with 150 μM Tan IIA (PBS as control) at RT for 2 h. Subsequently, CETSA was conducted at 41°C, 44°C, 47°C, 50°C, 53°C, 56°C, 59°C, 62°C, and 65°C for 3 min each. The resulting NR3C2 protein levels in the supernatants were analyzed via Western blot. Pull-down assay OV cells were homogenized in binding buffer (10 mmol/L Tris-HCl, pH = 7.4, 50 mmol/L KCl, 5 mmol/L MgCl2, 1 mmol/L EDTA, and 0.1 mmol/L Na3VO4). After centrifuging the cell lysate, the supernatant was incubated with Pierce High-Capacity NeutrAvidin Agarose (29202, Thermo Fisher Scientific) at 4°C for 1 h for pre-clearing. The pre-cleared cell lysate was incubated with biotin-labeled Tan IIA magnetic beads (Biotin-Tan IIA-Beads) or with biotin-saturated magnetic beads (Biotin-Beads) overnight at 4°C. For the Input group, an equal volume of pre-cleared cell lysate was mixed directly with the loading buffer. The following day, proteins bound to Biotin-Tan IIA were precipitated using NeutrAvidin magnetic beads. After washing with the washing buffer, the beads were eluted. The sample was boiled for 10 min, and Western blot analysis was performed to detect NR3C2 protein levels. RT-qPCR Total RNA from SKOV3, OVCAR-3, and ID8 cells was extracted using TRIzol reagent (NP0001, CYTOCH, Shanghai, China) and reverse transcribed using the RevertAid RT Reverse Transcriptase Kit (K1691, Thermo Fisher Scientific). qPCR was carried out using PowerTrack SYBR Green master mix (A46109, Thermo Fisher Scientific) and analyzed using the 2−ΔΔCt method, with GAPDH/Gapdh serving as the internal control. The forward primer for NR3C2 was 5′-AAATCACACGGCGACCTGTCGT-3′, and the reverse primer was 5′- ATGGCATCCTGAAGCCTCATCC-3’. The forward primer for ING1 was 5′- GGCTCGGAGACAGTTTCAGGC-3′, and the reverse primer was 5′- GACTGAAGCGCTCGTAGCACTC-3’. The forward primer for GAPDH was 5′- GTCTCCTCTGACTTCAACAGCG-3′, and the reverse primer was 5′- ACCACCCTGTTGCTGTAGCCAA-3’. The forward primer for Nr3c2 was 5′-AACCACACGGTGACCTGTCATC-3′, and the reverse primer was 5′- AGTGACACCCAGAAGCCTCATC-3’. The forward primer for Gapdh was 5′- CATCACTGCCACCCAGAAGACTG-3′, and the reverse primer was 5′- ATGCCAGTGAGCTTCCCGTTCAG-3’. ChIP Magna ChIP A/G ChIP Kit (17–10085, Merck) was used to verify the transcriptional factor NR3C2 binding at the ING1 promoter. Approximately 1 × 107 SKOV3 and OVCAR-3 cells were fixed with 1% formaldehyde at RT for 10 min and treated with glycine for 5 min. The cells were resuspended in cell lysis buffer and incubated on ice for 10 min, followed by centrifugation. The samples were resuspended in nuclear lysis buffer. Ultrasonication yielded DNA fragments ranging from 100 to 400 bp. A 1% sample of the ultrasonicated DNA was used as input. The samples were immunoprecipitated using anti-NR3C2 (1:200, 21854-1-AP, ProteinTech Group) or rabbit IgG (1:20, HY-P80879, MedChemExpress). The immune complexes were harvested using magnetic protein A/G beads. After eluting and purifying the DNA, subsequent qPCR analysis was performed using primers for the ING1 promoter (chr13: 110713411–110713760) (Left Primer: 5′-GACCCTCGCCTCTGGAAAA-3’; Right Primer: 5′-ACACTTGGGAAACATTCAGCC-3′). Luciferase reporter assay The ING1 promoter sequence (chr13: 110713411–110713760) was subcloned into the pGL3-basic vector (VT1554, Youbio, Changsha, Hunan, China) to construct the plasmid. SKOV3 and OVCAR-3 cells were co-transfected with 250 ng of the constructed plasmid and 25 ng of the pRL-SV40 vector (VT1571, Youbio) using Polyplus jetPRIME transfection reagent (101000046, Shanghai Nonin Biological Technology Co., Ltd., Shanghai, China). After 48 h, the Renilla-Firefly luciferase dual assay kit (HY-K1013, MedChemExpress) was used to determine the ratio of firefly luciferase activity to Renilla luciferase activity. Quantification and statistical analysis Survival curves were analyzed according to the Log rank (Mantel-Cox) test. The correlations were analyzed by Pearson’s correlation coefficient. Data analysis was performed using GraphPad Prism software (10.4.2, GraphPad, San Diego, CA, USA). The data were expressed as mean ± standard error of the mean (SEM) from at least three independent experiments unless otherwise specified. Fisher’s exact test was used to analyze the clinical-pathological correlations. The two groups were compared using a paired or unpaired t test, and the comparisons of more than two groups were evaluated using one-way or two-way ANOVA. p < 0.05 was taken as statistically significant.